case study 1 government intervention

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• Int J Environ Res Public Health

Re-Thinking the Role of Government Information Intervention in the COVID-19 Pandemic: An Agent-Based Modeling Analysis

1 Dong Fureng Economic and Social Development School, Wuhan University, Beijing 100010, China; nc.ude.uhw@7009oayul

2 National School of Development, Southeast University, Nanjing 210000, China; ude.olaffub@hziqoaix (X.Z.); ude.dccws@3210801fk (H.L.)

Xiaoqi Zhang

Yanqiao zheng.

3 School of Finance, Zhejiang University of Finance and Economics, Hangzhou 310018, China; moc.liamtoh@oaiqnaygnehz

Associated Data

The data presented in this study are available on request of the authors.

The COVID-19 pandemic imposes new challenges on the capability of governments in intervening with the information dissemination and reducing the risk of infection outbreak. To reveal the complexity behind government intervention decision, we build a bi-layer network diffusion model for the information-disease dynamics that were intervened in and conduct a full space simulation to illustrate the trade-off faced by governments between information disclosing and blocking. The simulation results show that governments prioritize the accuracy of disclosed information over the disclosing speed when there is a high-level medical recognition of the virus and a high public health awareness, while, for the opposite situation, more strict information blocking is preferred. Furthermore, an unaccountable government tends to delay disclosing, a risk-averse government prefers a total blocking, and a low government credibility will discount the effect of information disclosing and aggravate the situation. These findings suggest that information intervention is indispensable for containing the outbreak of infectious disease, but its effectiveness depends on a complicated way on both external social/epidemic factors and the governments’ internal preferences and governance capability, for which more thorough investigations are needed in the future.

1. Introduction

The COVID-19 pandemic has attacked the whole world over the past few months. The key features of the Novel Coronavirus, such as long incubation period, high infectiousness, and asymptomatic transmission, were not perceived at the beginning until they were gradually unveiled [ 1 , 2 , 3 , 4 ]. The WHO and governments keep disclosing epidemic information, but the disclosure is based on their own endowments, preferences, and perceptions, resulting in misleading information at least in the early stage of COVID-19 outbreak, such as “Masks work? NO” (quoted from Scott Atlas, the White House coronavirus task force member), “This is a flu. This is like a flu” (quoted from Donald Trump, the president of the US), and “There is some immune system variation with Asian people”(quoted form Taro Aso, the Deputy Prime Minister of Japan), etc. This information failed to alert the public but let their guards down instead. Then, the high mortality rate and emergency announcements subsequently incited widespread fear and exacerbated the epidemic situation. Theoretically, a systematic provision of timely and effective information from the government can mitigate the downsides [ 5 ]. However, in the real world, speed entails inaccuracy and cognitive uncertainty that keep government away from accomplishing such a tough mission [ 6 ]. Thus, in the early stage of an epidemic with strong externalities like COVID-19, the government’s choice between timeliness and effectiveness of intervention strategies raises a theoretical challenge for the management of urgent public health crisis.

The key to successfully contain the spread of an unexpected disease like COVID-19 is to understand the complicated two-way interaction between the dynamics of disease and those of information (and the human behavior response to information) [ 7 ]. Information might either amplify or diminish the public’s response to a risk event, depending on the transmission of risk information and public’s reactions at the time it occurs [ 8 ]. At the micro-level, one’s behavior depends on the epidemiological status of the disease, the individual’s knowledge about it (information accessed), misinformation, and the individual’s education and income level [ 9 ]. Along with the spread of disease in social life (physical level), information spreads in a virtual network, which brings the awareness of crisis for people [ 10 , 11 , 12 ], leading them to take preventive measures to stay healthy [ 13 , 14 ]. Therefore, the spread of disease facilitates the spread of information, which in turn inhibits the spread of disease [ 15 , 16 ]. However, on the other hand, people usually get illogical, fail to discern falsity, and disregard the truth during information dissemination [ 17 , 18 ]. Misleading information seems to have a natural disposition to resonate with public opinions, which causes spontaneous misrepresentation in transmission [ 19 ]. In addition, discussions on epidemic bring panic [ 20 ] and aggravate the harm of the epidemic [ 21 ], which will be further exaggerated by social media [ 22 ]. Meanwhile, increasing uncertainty about the disease makes people feel loss of control and boost people’s anxiety [ 23 ], usually accompanied by psychological distress [ 24 ]. Therefore, information is critical to fighting against the COVID-19-crisis [ 25 , 26 ], and improper information management strategy may lead to systematic failure [ 27 ].

The government, as the main governing body, is the most critical (information) node in the entire network, since it can intervene in information by “blocking” [ 28 , 29 ] and “disclosing” [ 5 , 30 ]. The minimal blocking implies a free and open information environment, which stimulates information to be widely diffused and induces more people to take self-protective measures [ 31 ], but “rumors” might also proliferate at the mean time [ 24 ]. Even if (under certain premises) some “rumors” are accurate [ 7 , 32 , 33 ], it might still interrupt the prevention efforts to the epidemic. It is always believed that government should perform as a central node to disclose accurate and up-to-date information to the entire society, so as to keep the public away from untruthful information and prompt the public to make informed decisions about health protection [ 30 , 34 ]. However, in the real world, governments do face time constraints and the trade-off between being accurate and being up-to-date in terms of information disclosing, which is not considered in classical information theory. Because a highly infectious disease caused by unknown viruses with great externality, such as COVID-19, spreads together with information of varying qualities (truthfulness, accuracy, etc.), it is highly probable that the disease has already contaminated the society before low-quality information is purged. In this case, the government has no way to disclose accurate information in time, resulting in the loss of public trust and raising the doubt of the public on the governing capacity of the government, which will accelerate epidemic outbreak [ 35 , 36 , 37 ]. Therefore, governments need to not only decide when to inject information into the network, but also whether to follow the tenet that governments do not and should not block information spreading at any circumstance [ 38 ].

For information blocking, studies have been conducted in theoretical [ 39 , 40 , 41 ], in empirical [ 42 , 43 ], in case studies [ 43 , 44 ], and other perspectives. These studies argued that, even if governments have the power to control information [ 45 ], they should not do that because free spread of information is essential to welfare-maximizing [ 38 ]. This argument is based on two underlying assumptions: (1) publishers are completely competitive to reach an equilibrium of disclosing accurate information; (2) there is no time constraint. These two assumptions do not apply for COVID-19 because, in the age of Internet media, people are not incentive compatible to spread accurate information. Moreover, such highly externalized infectious diseases caused by unknown viruses might have already infected a considerable amount of people before low-quality information is purified, so the government should not simply adhere to the tenet of not blocking information when facing an unknown health crisis [ 38 ]. As a result, we will discuss the complexity and diversity in information blocking and broaden current information control theory.

By the discussion so far, we notice that the successful containment of epidemic outbreak relies on the successful management on the information dissemination process, which, however, is hard to achieve in the real world. To better understand the failure in containing the COVID-19 pandemic, we here construct an information-behavior bi-layer model by adding a parallel layer of information transmission to the classical SI (Susceptible-Infected) model of infectious diseases. We intend to describe the effect of heterogeneous virtual information (at information layer) on heterogeneous nodes’ behaviors (at physical layer) [ 46 ]. The government, as the key node in information network, can influence the entire network through information disclosing and blocking. Based on this, we assign values to key variables such as the medical awareness level of the virus and the public’s health awareness level and then conduct computer simulation experiments under different scenarios. We reveal the pattern of government information intervention based on the simulation results. In addition, we only focus on the emergence stage of the infectious disease, during which the recovery from the infected status, such as self-healing and cure of the disease, is omitted [ 47 ]; therefore, the base model is the SI model rather than the SIR (Susceptible-Infected-Recovery) model.

Based on the bi-layered network model, we explore two main themes: how disease spreads affect information spreads and how information affects the efficiency of controlling the epidemic. We introduce the non-dualism of information and the heterogeneity of nodes’ behaviors into the epidemic model and conduct a simulation to reveal the information intervention dilemma faced by the government between information disclosing and blocking. We find that governments face a trade-off between speed and accuracy in information disclosing; and the optimal strategy is contingent on varying conditions in information blocking. The optimal combination of disclosing and blocking is highly sensitive to the government preference and its governance capacity. Governments that are only responsible for the outcome of intervention will focus unilaterally on the accuracy at the expense of speed; a risk-averse government that intends to minimize the maximum infection rate under uncertain scenarios will impose a more restrictive blocking; and the most restrictive blocking strategy might be the best for governments with lower capability and credibility.

In summary, this paper makes several important contributions to the literature. First, existing studies did not pay sufficient attention to the spread and evolution of rumors during a public crisis [ 48 , 49 , 50 ], which is considered in our study. We expounded the impacts of information dissemination on epidemic evolution in scenarios with different levels of medical awareness of the virus, public health awareness, and government preferences and credibilities, which complements the research in this field. Second, most current studies regard information and disease transmissions as simultaneously happened and jointly induced by the physical movement of an agent [ 46 ], while this is not the case during COVID-19 pandemic as the internet obviates the needs for physical contact in information transmissions [ 51 ]. Thus, in our paper, we separate the information and disease transmissions and investigate the impact of heterogeneous information on the individual behaviors and disease dynamics. Third, unlike previous research on government information interventions with known risk [ 5 , 28 , 29 , 30 ], ours are on government information interventions with unknown risk. The lack of prior knowledge on the Corona-SARS-2 is the most striking feature of the COVID-19 pandemic, which weakens the usefulness of government action and calls for a reassessment of government information intervention under a crisis environment with high uncertainty. To this end, this paper demonstrates a couple of intervention dilemmas faced by government, which complements the existing theories.

2.1. The Model of Information Disclosing

The information dissemination system resp. behavioral response system is embedded in the information network resp. physical network. Both networks are given as follows.

Information network: the network has ( N + 1 ) nodes, first N are individual nodes representing N individuals denoted as i , i = 1 , 2 , ⋯ N , and one government information node denoted as j . The degree of an individual node i is denoted as y i , which obeys a power-law distribution, that is, F y i ∝ y i − v , where F ( · ) is the CDF and y i satisfies ϵ ≤ 1 ∕ y i ≤ 1 , where ϵ is a small constant to avoid the degree to blow up. Degree and degree distribution are concepts used in graph theory and network theory. A graph (or network) consists of a number of vertices (nodes) and the edges (links) that connect them. The number of edges (links) connected to each vertex (node) is the degree of the vertex (node). The degree distribution is a general description of the number of degrees of vertices (nodes) in a graph (or network), and, for random graphs, the degree distribution is the probability distribution of the number of degrees of vertices in the graph, which usually assumes a power-law distribution. Throughout the following analysis, we take v = − 1 and ϵ = 0.01 . The government node j (representing real-world government) discloses information to every individual node and can only obtain information from n 1 ( n 1 ≪ N ) (The notation “≪” means that the number n 1 must be far less than the number N.) random nodes. The neighborhood of an individual node i is the set of all other nodes (including j ) it connects with, denoted as O i .

Physical network: the physical network has M nodes, including n 2 “special” nodes defined as the “gathering spots”, which predisposes these nodes to this epidemic. M t denotes the distribution of locations of all N individuals during period t , and M 0 is the initial distribution that can be viewed as the “home” for every individual (node), thus at the beginning of each period t the individuals move from M 0 to M t and return back to M 0 at the end of period t . Home coordinates M 0 and gathering spots are randomly assigned and different from each other, so we have N + n 2 < M . Suppose there are n 3 random nodes, each with identical initial information ξ , who disseminate information at the outbreak of disease; n 4 random nodes are initially affected by the public crisis, representing the “patient zero”.

Without loss of generality, we unitize the information between 0 and 1. The rules for information dissemination in each period are as follows.

Stage i. Individual nodes send information to neighbors . Each node that has information at the beginning of each period sends its information to all its neighbors, so all ( N + 1 ) nodes might receive information from others. As information is spontaneously [ 19 ], rapidly, and extensively [ 22 ] misrepresented during transmission, and most people do not send more accurate information than they receive [ 17 , 18 ], we assume that information gets distorted and misrepresented during each transmission. Thus, the actual amount of information received is δ x i due to information decay, where δ ∼ U ( 0 , 1 ) , and we assume x i ∈ 0 , 1 without loss of generality.

Stage ii. Individual nodes receive information from neighbors . Each node might have multiple information sources, and it merges the information from all its neighbors weighted by their degrees (and including itself). Each individual updates its information based on Equation ( 1 ) at each period before the government intervenes:

Stage iii. The government node censors and screens the information . The government has a threshold X D once it receives information from individuals (otherwise, the government would not act in this stage), the government will screen out all individuals with above-threshold information at the beginning of current period, among whom the government pinpoints the nearest ones and takes the maximum amount of information they carry denoted as x d .

Stage iv. Government node discloses information . The government is not able to intervene until it censors and screens the information; thus, there is a lag between receiving information and disclosing, which as we can see in Figure 4e, increases with X D . After the lag (otherwise, the government would not act in this stage), the government shall disclose x d to all nodes in each period with a weight of λ , where λ ∈ 0 , 1 . The higher the λ , the more credible the information.

Stage v. Individual nodes update information again . The government intervention switches the updating rule to

which is also the final amount of the information after government intervention. In addition, we assume that the amount of information of initial information holders (those who have information in period 0) is constant, i.e., they do not apply for Equations ( 1 ) and ( 2 ).

In short, in the first period, only a few people disseminate information, which will be randomly decayed in each subsequent period, this process simulates the misrepresentation of information.

Twitter data show that there was a significant heterogeneity in the behavioral response to the COVID-19 epidemic [ 52 ]. Some people, once informed about the epidemic, wear a mask and practice social distance to not expose themselves to the virus—while others panicked, herded, and behaved irrationally because of bad news, exemplified by flocking to churches for psychological comfort [ 53 ], to supermarkets for daily supplies [ 54 ], and taking radical actions like repeated hospital visits [ 55 ]. Thus, in this paper, we group the population by susceptibleness to irrational behavior caused by information described by an exogenous parameter—individual threshold X I that distinguishes whether an individual is panic-prone or non panic-prone by comparing it with the amount of information the individual has. An above-threshold (under-threshold) information denotes a (non) panic-prone individual. For a panic-prone node, we assume its probability of going to gathering spots instead of maintaining the original trajectory is 1 − x · , a , where x · , a is the amount of information it has. For a non panic-prone node, we assume that its probability of not moving is r · , N = x · , a . Thus, the behavioral routine is as follows (see Figure 1 for a simplified example): a node moves along with its path with a maximum radius d 1 , and the actual distance it moves obeys a uniform distribution in ( 0 , d 1 ) ; this node will randomly choose one of the gathering spots if intending to go to one in this period; every individual node follows this routine, then we have an evolving geographical distribution M t of the population moving in period t . The uninfected will contact everyone within the maximum infection radius d 2 and there is a probability μ of being infected for each contact.

A diagram for individuals’ behavioral routine. Consider a world with 36 nodes ( d 1 = 2 , d 2 = 1 ), among which there are only one infected node (S), one health node (H) and one gathering spot (G).

Throughout the simulation analysis, we focus on the impact of three key parameters, the initial information ( ϵ ), individual threshold ( X I ), and disclosing threshold ( X D ), which are the most important quantities to measure the impact of government intervention on the coupled information-disease dynamics. The initial information is the source of all information, which denotes the medical awareness of the virus; the individual threshold is a parameter to distinguish the population by groups set above, the smaller it is, the higher the level of public health awareness. Disclosing the threshold, chosen by the government, measures its relative priority to speed and accuracy in information dissemination. One of the objectives of our experiment is to ascertain the optimal disclosing threshold. Government prioritizes speed more as its threshold is lower, “0“ means that government discloses the information immediately upon receipt; “1” means that government only discloses completely accurate information.

2.2. The Experiments of Information Disclosing

The simulation steps will be: ( Figure 2 brief overviews these steps):

• Generate a random information network and a random physical network, the former illustrates the information relationship between people, and the latter records the coordinates of people M 0 and gathering spots on the map.
• Assign values to initial information and individual threshold. The initial information is the source of all information, which denotes the medical awareness of the virus; the individual threshold is a parameter to distinguish the population by groups set above; the smaller it is, the higher the level of public health awareness.
• Assign values to the disclosing threshold. The disclosing threshold, chosen by the government, measures its relative priority to speed and accuracy in information dissemination. One of the objectives of our experiment is to ascertain the optimal disclosing threshold. Government prioritizes speed more as its threshold is lower.
• Generate random individual nodes with initial information and random initial infected nodes.
• (a) Each individual node with information sends out information to neighbors.
• (b) Each individual node will update its information (weighted) based on Equation ( 1 ).
• (c) The government initiates a censoring and screening and enters stage d after a lag period, only for the first time does it receive the above-threshold information. If the government never receives above-threshold information, skip c , d , and go to e .
• (d) Government discloses information to the public, which induces another round of information update for individual nodes based on Equation ( 2 ).
• (e) The population is grouped into infected and healthy people by health status, and into panic-prone and non panic-prone by how much information one has compared with the individual threshold.
• (f) Each individual node moves in a physical layer following the routine of the subgroup it is in with probability based on its final information.
• (g) Reset the infection status of healthy individual within the transmission radius of an infected one according to the infection probability.
• Return to step 5, initiate a new round for 50 times, that is, run the experiment for 50 periods. The data show a stability after 40 periods, so we stopped at 50.
• Output the final overall infection rate at the end of period 50.
• Repeat steps 4–7 for 50 times to reduce the randomness, record the mean, and standard deviation of the final infection rate.
• Reassign for the disclosing threshold discrete values that equally divide the interval 0 , 1 into 11 parts, and repeat steps 3–8 for each value, that is, 11 times, to find the final infection rates for different disclosing threshold scenarios.
• Reassign for initial information a discrete array 0.4 , 0.6 , 0.8 , 1.0 , and reassign for an individual threshold the same values reassigned for the disclosing threshold in the previous step. Then, repeat steps 2–9, that is, 44 times.

Flowchart of our simulation. In step ii, we assigned 4 × 11 × 11 = 484 different values. Then, we repeated steps iii and iv each for 50 times, respectively.

Now, we have conducted an experiment with a full parameter space for each initial condition. A total of 484 different conditions were simulated for 24,200 repetitions of the experiment, each lasts for 50 periods, which adds up to a total of 1,210,000 periods of experiments. They essentially cover all possible scenarios under different external constraints. Table 1 lists the definitions, values, and distributions of all parameters in the model.

Definitions, values, and distributions of variables in the model.

2.3. The Model and Experiments of Information Blocking

We assume the government will suppress any transmission of information under X B ; thus, X B = 0 denotes the special case in the previous discussion. Combining the blocking threshold with other initial conditions above, we have a new parameter space simulation with a total of 5324 different scenarios simulated and a total of 266,200 repeated experiments. Each group lasts for 50 periods, for a total of 13,310,000 experiments.

3. Results and Discussion

3.1. modeling framework.

Our model consists of two main systems: information dissemination system and behavioral response system. In the information dissemination system, each individual sends (receives) information to (from) its neighbors through an information network. Given that information will always be rapidly, extensively [ 22 ], and spontaneously [ 19 ] misrepresented during transmission, and that most people do not send more accurate information than they receive [ 17 , 18 ], we assume information gets distorted and misrepresented during each transmission. In the behavioral response system, each individual makes a move according to its information (with probability). Once informed about the epidemic, some people behave rationally such as practicing social distancing, while others behave irrationally such as flocking to churches [ 53 ], to supermarkets [ 54 ], and taking radical actions like repeated hospital visits [ 55 ].

The information dissemination system affects the behavioral response system. The government might intervene in the information dissemination to reduce infections by either disclosing or blocking information.

For information disclosing, the government discloses information to all individuals to make them behave rationally (or at least not behave irrationally). Obviously, the more accurate the information is and the earlier it is disclosed, the public can be better guided which lowers the final infection rate. However, it takes time for government to censor and screen information before disclosing, which brings an accuracy-speed trade-off. We use a disclosing threshold to measure the government’s preference on speed or accuracy; a higher threshold indicates a higher preference on accuracy: a threshold “1” means the government would not disclose any information unless it is completely accurate; while “0” indicates an immediate disclosure without any censoring and screening.

For information blocking, the government blocks less-accurate information transmissions between individuals. Obviously, a stringent blocking leads to a transmission of information with higher accuracy. However, blocking will slow down the overall information dissemination in the network, and then slow down the government’s censoring and screening of information. Thus, there is a trade-off between disclosing and blocking. We use a blocking threshold to measure the blocking stringency, the government would block any transmission of any under-threshold information: a threshold “0” means no blocking at all; while “1” means that government blocks all information transmissions.

We analyze both of the optimal thresholds for government. More details on the settings of our model can be found in the Methods section. In reality, the government has a great influence on the information dissemination. Thus, in our model, we assume that the government node is the most critical one and the government-disclosed information highly outweighs individuals’ information (except for the further discussion of a government with low credibility in a later section).

3.2. Intervention Dilemma in Disclosing Information

In this part, we will discuss the speed-accuracy trade-off results and analyze the mechanism in information disclosing.

First, Figure 3 summarizes the results of the simulations with 44 different external constraints, and we find that there is seldom a single dominant disclosing threshold (the government’s preference on speed or accuracy), i.e., seeking either speed or accuracy alone will not result in the lowest infection rate, and the optimal strategy (corresponding to the lowest infection rate at the end of the last period) is somewhere in between, which implies a speed-accuracy trade-off.

Distribution diagram of disclosing threshold. These four subfigures ( a – d ) correspond to initial information of 1, 0.8, 0.6, and 0.4, respectively, each one has 11 curves representing individual thresholds of all 11 values, which shows the optimal disclosing thresholds under all 44 different external constraints. The horizontal axis is all possible values of disclosing thresholds, the vertical axis is the infection rate of the whole society.

Specifically, the optimal disclosing threshold lies between both ends in about 84.09 % of the cases, and their distributions vary in different external constraints. Figure 3 a,b show that (1) if the initial information is 1 or 0.8 and the individual threshold is in [ 0 , 0.7 ] , the optimal disclosing threshold has a 91.75 % probability of being in the middle, and mostly ( 66.89 % ) falls within [ 0.6 , 0.8 ] ; (2) if the initial information is 0.6 and the individual threshold is in [ 0 , 0.5 ] , the government has a 93.00 % probability of dealing with a trade-off ( Figure 3 c), and the optimal disclosing threshold mostly ( 77.78 % ) falls in [ 0.4 , 0.6 ] ; (3) if the initial information is 0.4 , the optimal disclosing threshold will almost certainly be greater than 0.4 ( 99.98 % ), but the distribution is too scattered to give a specific interval ( Figure 3 d). The mode of optimal disclosing threshold is 0.8 but only with 20.30 % frequency.

In addition, we can see that, if the virus is medically well-known ( ξ ≥ 0.8 ) and public health awareness is low ( X I ≥ 0.7 ), the government shall prioritize accuracy over speed; if the virus is medically medium-known ( ξ = 0.6 ) and public awareness is high ( X I ≤ 0.5 ), the government shall balance speed and accuracy, which almost equally signifies; and if the virus in medically less-known ( ξ = 0.4 ), the government shall probably prioritize accuracy over speed.

Second, we will dissect the underlying logic and mechanism of the government’s trade-off. We take one of the curves in Figure 3 a that is denoted by ξ = 1 and X I = 0.5 , as an example, to find the relationship between disclosing the threshold infection rate, then we have Figure 4 . In all 550 ( 11 × 50 ) experiments, the disclosing threshold for the lowest final infection rate usually lies between 0.7 and 0.9 , and the final infection rate first falls then rises as the disclosing threshold increases, with the inflection point being at 0.8 ( Figure 4 a); the amount of final information per capita and the duration of government intervention both increase monotonically with the disclosing threshold ( Figure 4 b,e); the number of people infected after government intervention, the number of people infected by a panic after government intervention and the number of uninfected people remaining at the time of government intervention all negatively correlate with the disclosing threshold ( Figure 4 c,d,f).

Simulation results for government’s trade-off in disclosing under the special case of ξ = 1 , X I = 0.5 . ( a ) depicts final infection rate; ( b ) conveys the impact of disclosing threshold on information; ( c , d ) represent the positive effective of a larger disclosing threshold by showing the new infections overall or from panic after government intervention; and ( e , f ) describe the negative effect by showing when the government intervenes and how many health people remained at intervention.

One of the fundamental reasons for the government to balance speed and accuracy is the precipitous fall in the marginal contribution of accuracy as the disclosing threshold exceeds a certain “point”, while speed hardly affects the final infection rate. The following is a detailed analysis on the effects of both accuracy and speed.

With respect to accuracy, the effect comes from two perspectives: (1) accurate information lowers the infection from panic in a healthy panic-prone population ( Figure 4 d); (2) accurate information changes the behavior routine in the panic-prone population, which reduces the spread of the disease. Both (1) and (2) are in play until the disclosing threshold exceeds 0.5 ( X I in our example); after that, there is no longer a panic-prone group, neither is infection from panic, which explains the precipitous fall in the marginal contribution of accuracy.

When it comes to speed, the effect comes from two perspectives as well: (1) the time the government spends on censoring and screening information, which we call a lag; (2) the number of remaining uninfected people at the time of government disclosing information. The more accurate information the government seeks, the longer the lag ( Figure 4 e) and the fewer uninfected people remain at the time of disclosing ( Figure 4 f). Notice that both have roughly the same slope with respect to disclosing threshold, which explains a roughly constant marginal cost of pursuing accuracy.

The combination of constant marginal costs and abrupt fall in marginal benefits leads to an inflection point in disclosing threshold, which explains the heterogeneity in the distribution of optimal disclosing threshold: as disclosing threshold exceeds individual threshold, a sudden fall of benefits occurs, which theoretically makes the optimal disclosing threshold slightly greater than the individual threshold, which explains what we discussed above that the intervals in which the optimal disclosing threshold mostly lays differ.

In most cases, the government’s premature disclosing of inaccurate information will contaminate the overall network, while obsession with accuracy may have the government miss the disclosure window before too many people are infected, which is intolerable to government who requires a low infection rate. Thus, there is a trade-off.

3.3. Intervention Dilemma in Blocking Information

We assume the government will block any transmission of under-the-blocking-threshold ( X B ) information between individuals; thus, X B = 0 denotes the special case in previous discussion. Other settings are the same as above. In this part, we will discuss the optimal blocking strategies and analyze the mechanism.

As shown in Figure 5 , the optimal blocking threshold varies from case to case. Overall, a small blocking threshold ( [ 0.1 , 0.3 ] ) is necessarily ( 100 % ) not optimal; a strict blocking threshold ( X B ≥ 0.8 ) is usually ( 50.41 % ) optimal, experimental data show a value between 45 % and 55 % in most external conditions; but 0 is the optimal threshold still in 20.25 % of cases, and usually ( 89.80 % ) occurs when ξ ≥ 0.8 and X I ≤ 0.7 . When the initial information is low ( ξ ≤ 0.6 ), not blocking is seldom ( 0.83 % ) optimal.

Simulation results for government’s trade-off in blocking. ( a ) conveys the distributions of optimal blocking threshold under 484 different external scenarios (4 pieces of initial information, 11 individual thresholds, and 11 disclosing thresholds). We show how the figure works by taking the first column as an example: there are 98 conditions in which 0 is the lowest infection rate; ( b ) conveys the relationship between initial information and blocking threshold; and ( c ) conveys the relationship between individual threshold and blocking threshold.

We have our key findings from the above analysis. First, minor blocking is not an option for government because it is dominated by stricter blocking in a deteriorated or being deteriorated information environment and undermines the efficiency of information dissemination in a benevolent environment. Second, in the age of the Internet, information is extremely interconnected and low-quality information is more easily disseminated, thus stricter information blocking might be an option worth considering in the early stages of an outbreak of an unknown infectious disease. Finally, if the virus is well-known at the medical level, plus the public has a certain level of health awareness, free spread of information might improve the situation; while, otherwise, as in the case of COVID-19, governments should intervene in the spread of information in social networks.

From the simulation results, we can see that, in most cases, the optimal strategy will be either highly stringent blocking or free spread. Blocking low-quality information not only increases the overall information of the whole population, but causes side effects under certain external conditions. Thus, not blocking can be an optimal strategy in some cases. In this section, we provide an in-depth analysis of the data and a mechanistic analysis.

Figure 5 b reveals in general the optimal blocking threshold negatively correlates with initial information: as initial information drops from 1 to 0.4 , the probability of optimal blocking threshold taking 0 will be 52.07 % , 27.27 % , 1.65 % , and 0 % , respectively. Figure 5 c shows a positive correlation between optimal blocking threshold and individual threshold.

Governments block information mainly by suppressing less-accurate information, but which, once implemented, will slow down the overall information dissemination in the network anyway. Therefore, blocking can neither be too stringent nor too liberal, an optimal one usually lies in between. However, mild blocking is necessarily not optimal as it fails to purify the information environment.

Furthermore, when the virus is well-known ( ξ = 1 ), especially when public health awareness is high ( X I ≤ 0.3 ), not blocking dominates most of the time ( 69.70 % ). While stringent blocking ( X B ≥ 0.8 ) is necessarily not ( 0 % ) an optimal strategy because higher-quality information, which helps to slow the spread of the disease with high public health awareness, is also blocked. Thus, when both the initial information and the level of public health awareness are at a high level, not blocking is optimal; otherwise, information that would not cause panic might do now. In addition, when medical awareness of the virus declines, so does the proportion of valuable information, which necessitates blocking as well.

3.4. Optimal Intervention under Different Government Types

In previous sections, our study was based on the neutral government assumption that governments only seek the lowest infection rate. However, in reality, a government is not a personalized organization pursuing social optimum because it is often checked by various inside and outside nodes. In addition, government credibility makes a difference as well. In this section, we will discuss the optimal strategy for non-neutral governments and low-credible governments.

An unaccountable government that evades responsibilities would only care for lower new infections after intervention rather than global infections, which digresses from the objective described above. As shown in Figure 6 a, the later the government discloses information, the less that will be newly infected after disclosure. There are two underlying reasons: (1) late disclosed information will indeed be more accurate, which reduces the infection rate; and (2) there are less uninfected people at the time of disclosing. Therefore, a blame-evading government would delay the disclosing to avoid being held accountable.

Simulation results for non-neutral governments and low-credible governments. ( a ) portrays the unaccountable governments based on the settings in Figure 4 a, the vertical axis is the number of newly infected people after government intervention; ( b ) the distribution of worst blocking threshold (highest infections) based on the settings of Figure 5 portrays the risk-averse government; ( c , d ) describe the low-credible governments by reassigning for government’s credibility 10 % while keeping other variables unchanged.

A conservative government that prefers the least error-prone strategy (minimizing maximum loss) rather than the optimal one (the loss minimization strategy) would block all the information ( Figure 6 b). Since the optimal strategy would not be accessed until all external conditions are fully judged and scrutinized, which is not feasible for COVID-19, complete blocking would be optimal for such a government to avoid the worst case scenario. Our experiment of 484 different scenarios manifests a complete blocking will never lead to the highest infections.

4. Conclusions

In this paper, we introduce the non-dualism (by non-dualism, we mean the information is neither absolutely accurate nor absolutely not but partially accurate) of information and the heterogeneity of nodes’ behaviors into the epidemic model and conduct a simulation to reveal the information intervention dilemma faced by the government and to explore the trade-offs among corresponding strategies. Our experiments highlight that:

• For information disclosing, governments face a trade-off between speed and accuracy. A better medical understanding of the virus and an inadequate public health awareness make accuracy outweigh speed; otherwise, a quick one is better.
• For information blocking, the optimal strategy is contingent on varying conditions: no blocking is usually optimal for a well-known virus and a higher public health awareness; otherwise, blocking is preferred.
• The optimal combination of disclosing and blocking is highly sensitive to the government preference and its governance capability. A government that is only responsible for the outcome of intervention will focus unilaterally on accuracy at the expense of speed; a risk-averse government that intends to minimize the maximum infection rate in uncertain scenarios will impose a more restrictive blocking; and the most restrictive blocking strategy might be best for governments with lower capability and credibility.

These findings reveal the complexity in government decision-making about dissemination of disease information: neither allowing free flow of information nor disclosing it as early as possible is optimal. Under extreme conditions, they are even harmful to the goal of controlling disease outbreak. The interaction between information and infectious disease deepens our knowledge about public health crisis governance, enriches the existing theories in public economics and public management, and provides useful social and policy implications.

In reality, some governments are not as capable and credible as assumed. A lower credibility will discount the effects of disclosing information or even annul it, which makes a total blocking optimal as shown in Figure 6 c,d. The bankruptcy of government credibility originates in two ways: (1) the government’s past mediocre performance; (2) the public’s inherent belief in “small government”. Meanwhile, a similar experience in the past also affected government responses and effects, as we can see with the horrible painful memories of SARS inducing vigilance for COVID-19 in East Asia countries, while the U.S. and Europe were indifferent in the early stage of this pandemic.

In the preceding discussion, we relaxed one assumption at a time, whereas the government’s preferences are more complex in reality. In a broader context, the government’s preferences (objective function) are affected by two things: the government’s perception and judgment of the epidemic (decision-making base), and the government’s priorities in different objectives (decision-making objectives); both change over time.

This paper also has some limitations. For instance, our discussion focuses mainly on the theoretical mechanisms behind the joint spreading process of information and epidemic, and the proposed intervention strategies have not yet been analyzed with the real-world data. One reason for the lack of empirical analysis is the complex set-up of the bi-layered network model. The information dissemination network and the physical-layer contact network are not precisely observable in the real world, which makes it challenging for acquiring sufficient data for model fitting. On the other hand, the observed infection and information dissemination process are often already intervened in by the government; therefore, it is hard to separate the net effect of government intervention from the ex-post spreading data. Then, it is technically difficult to quantify the key parameters of intervention. To this end, we believe more sophisticated empirical techniques have to be introduced for the implement data-oriented analysis of our model, such as the network reconstruction and the causal detection techniques, which forms a promising direction for future investigation.

Acknowledgments

The authors are grateful to Sheng Hua (Wuhan University) for helpful discussions and assistance.

Author Contributions

Conceptualization and methodology, Y.L., Z.J., and X.Z.; analysis, Y.L., Z.J., X.Z., Y.Z., and H.L. All authors participated in manuscript writing or editing. All authors have read and agreed to the published version of the manuscript.

This research was funded by the Ministry of Education in the China Profject of Humanities and Social Sciences, undner Grant No. 20YJC790176, and the Fundamental Research Funds for the Central Universities in China, under Grant No. 2242020S30024.

Data Availability Statement

Conflicts of interest.

The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Want to create or adapt books like this? Learn more about how Pressbooks supports open publishing practices.

4.2 Government Intervention in Market Prices: Price Floors and Price Ceilings

Learning objectives.

• Use the model of demand and supply to explain what happens when the government imposes price floors or price ceilings.
• Discuss the reasons why governments sometimes choose to control prices and the consequences of price control policies.

So far in this chapter and in the previous chapter, we have learned that markets tend to move toward their equilibrium prices and quantities. Surpluses and shortages of goods are short-lived as prices adjust to equate quantity demanded with quantity supplied.

In some markets, however, governments have been called on by groups of citizens to intervene to keep prices of certain items higher or lower than what would result from the market finding its own equilibrium price. In this section we will examine agricultural markets and apartment rental markets—two markets that have often been subject to price controls. Through these examples, we will identify the effects of controlling prices. In each case, we will look at reasons why governments have chosen to control prices in these markets and the consequences of these policies.

Agricultural Price Floors

Governments often seek to assist farmers by setting price floors in agricultural markets. A minimum allowable price set above the equilibrium price is a price floor . With a price floor, the government forbids a price below the minimum. (Notice that, if the price floor were for whatever reason set below the equilibrium price, it would be irrelevant to the determination of the price in the market since nothing would prohibit the price from rising to equilibrium.) A price floor that is set above the equilibrium price creates a surplus.

Figure 4.8 “Price Floors in Wheat Markets” shows the market for wheat. Suppose the government sets the price of wheat at P F . Notice that P F is above the equilibrium price of P E . At P F , we read over to the demand curve to find that the quantity of wheat that buyers will be willing and able to purchase is W 1 bushels. Reading over to the supply curve, we find that sellers will offer W 2 bushels of wheat at the price floor of P F . Because P F is above the equilibrium price, there is a surplus of wheat equal to ( W 2 − W 1 ) bushels. The surplus persists because the government does not allow the price to fall.

Figure 4.8 Price Floors in Wheat Markets

A price floor for wheat creates a surplus of wheat equal to ( W 2 – W 1 ) bushels.

Why have many governments around the world set price floors in agricultural markets? Farming has changed dramatically over the past two centuries. Technological improvements in the form of new equipment, fertilizers, pesticides, and new varieties of crops have led to dramatic increases in crop output per acre. Worldwide production capacity has expanded markedly. As we have learned, technological improvements cause the supply curve to shift to the right, reducing the price of food. While such price reductions have been celebrated in computer markets, farmers have successfully lobbied for government programs aimed at keeping their prices from falling.

While the supply curve for agricultural goods has shifted to the right, the demand has increased with rising population and with rising income. But as incomes rise, people spend a smaller and smaller fraction of their incomes on food. While the demand for food has increased, that increase has not been nearly as great as the increase in supply. Figure 4.9 “Supply and Demand Shifts for Agricultural Products” shows that the supply curve has shifted much farther to the right, from S 1 to S 2 , than the demand curve has, from D 1 to D 2 . As a result, equilibrium quantity has risen dramatically, from Q 1 to Q 2 , and equilibrium price has fallen, from P 1 to P 2 .

On top of this long-term historical trend in agriculture, agricultural prices are subject to wide swings over shorter periods. Droughts or freezes can sharply reduce supplies of particular crops, causing sudden increases in prices. Demand for agricultural goods of one country can suddenly dry up if the government of another country imposes trade restrictions against its products, and prices can fall. Such dramatic shifts in prices and quantities make incomes of farmers unstable.

Figure 4.9 Supply and Demand Shifts for Agricultural Products

A relatively large increase in the supply of agricultural products, accompanied by a relatively small increase in demand, has reduced the price received by farmers and increased the quantity of agricultural goods.

The Great Depression of the 1930s led to a major federal role in agriculture. The Depression affected the entire economy, but it hit farmers particularly hard. Prices received by farmers plunged nearly two-thirds from 1930 to 1933. Many farmers had a tough time keeping up mortgage payments. By 1932, more than half of all farm loans were in default.

Farm legislation passed during the Great Depression has been modified many times, but the federal government has continued its direct involvement in agricultural markets. This has meant a variety of government programs that guarantee a minimum price for some types of agricultural products. These programs have been accompanied by government purchases of any surplus, by requirements to restrict acreage in order to limit those surpluses, by crop or production restrictions, and the like.

To see how such policies work, look back at Figure 4.8 “Price Floors in Wheat Markets” . At P F , W 2 bushels of wheat will be supplied. With that much wheat on the market, there is market pressure on the price of wheat to fall. To prevent price from falling, the government buys the surplus of ( W 2 – W 1 ) bushels of wheat, so that only W 1 bushels are actually available to private consumers for purchase on the market. The government can store the surpluses or find special uses for them. For example, surpluses generated in the United States have been shipped to developing countries as grants-in-aid or distributed to local school lunch programs. As a variation on this program, the government can require farmers who want to participate in the price support program to reduce acreage in order to limit the size of the surpluses.

After 1973, the government stopped buying the surpluses (with some exceptions) and simply guaranteed farmers a “target price.” If the average market price for a crop fell below the crop’s target price, the government paid the difference. If, for example, a crop had a market price of $3 per unit and a target price of $4 per unit, the government would give farmers a payment of $1 for each unit sold. Farmers would thus receive the market price of $3 plus a government payment of $1 per unit. For farmers to receive these payments, they had to agree to remove acres from production and to comply with certain conservation provisions. These restrictions sought to reduce the size of the surplus generated by the target price, which acted as a kind of price floor.

What are the effects of such farm support programs? The intention is to boost and stabilize farm incomes. But, with price floors, consumers pay more for food than they would otherwise, and governments spend heavily to finance the programs. With the target price approach, consumers pay less, but government financing of the program continues. U.S. federal spending for agriculture averaged well over $22 billion per year between 2003 and 2007, roughly $70 per person.

Help to farmers has sometimes been justified on the grounds that it boosts incomes of “small” farmers. However, since farm aid has generally been allotted on the basis of how much farms produce rather than on a per-farm basis, most federal farm support has gone to the largest farms. If the goal is to eliminate poverty among farmers, farm aid could be redesigned to supplement the incomes of small or poor farmers rather than to undermine the functioning of agricultural markets.

In 1996, the U.S. Congress passed the Federal Agriculture Improvement and Reform Act of 1996, or FAIR. The thrust of the new legislation was to do away with the various programs of price support for most crops and hence provide incentives for farmers to respond to market price signals. To protect farmers through a transition period, the act provided for continued payments that were scheduled to decline over a seven-year period. However, with prices for many crops falling in 1998, the U.S. Congress passed an emergency aid package that increased payments to farmers. In 2008, as farm prices reached record highs, Congress passed a farm bill that increased subsidy payments to $40 billion. It did, however, for the first time limit payments to the wealthiest farmers. Individual farmers whose farm incomes exceed $750,000 (or $1.5 million for couples) would be ineligible for some subsidy programs.

Rental Price Ceilings

The purpose of rent control is to make rental units cheaper for tenants than they would otherwise be. Unlike agricultural price controls, rent control in the United States has been largely a local phenomenon, although there were national rent controls in effect during World War II. Currently, about 200 cities and counties have some type of rent control provisions, and about 10% of rental units in the United States are now subject to price controls. New York City’s rent control program, which began in 1943, is among the oldest in the country. Many other cities in the United States adopted some form of rent control in the 1970s. Rent controls have been pervasive in Europe since World War I, and many large cities in poorer countries have also adopted rent controls.

Rent controls in different cities differ in terms of their flexibility. Some cities allow rent increases for specified reasons, such as to make improvements in apartments or to allow rents to keep pace with price increases elsewhere in the economy. Often, rental housing constructed after the imposition of the rent control ordinances is exempted. Apartments that are vacated may also be decontrolled. For simplicity, the model presented here assumes that apartment rents are controlled at a price that does not change.

Figure 4.10 Effect of a Price Ceiling on the Market for Apartments

A price ceiling on apartment rents that is set below the equilibrium rent creates a shortage of apartments equal to ( A 2 − A 1 ) apartments.

Figure 4.10 “Effect of a Price Ceiling on the Market for Apartments” shows the market for rental apartments. Notice that the demand and supply curves are drawn to look like all the other demand and supply curves you have encountered so far in this text: the demand curve is downward-sloping and the supply curve is upward-sloping.

The demand curve shows that a higher price (rent) reduces the quantity of apartments demanded. For example, with higher rents, more young people will choose to live at home with their parents. With lower rents, more will choose to live in apartments. Higher rents may encourage more apartment sharing; lower rents would induce more people to live alone.

The supply curve is drawn to show that as rent increases, property owners will be encouraged to offer more apartments to rent. Even though an aerial photograph of a city would show apartments to be fixed at a point in time, owners of those properties will decide how many to rent depending on the amount of rent they anticipate. Higher rents may also induce some homeowners to rent out apartment space. In addition, renting out apartments implies a certain level of service to renters, so that low rents may lead some property owners to keep some apartments vacant.

Rent control is an example of a price ceiling , a maximum allowable price. With a price ceiling, the government forbids a price above the maximum. A price ceiling that is set below the equilibrium price creates a shortage that will persist.

Suppose the government sets the price of an apartment at P C in Figure 4.10 “Effect of a Price Ceiling on the Market for Apartments” . Notice that P C is below the equilibrium price of P E . At P C , we read over to the supply curve to find that sellers are willing to offer A 1 apartments. Reading over to the demand curve, we find that consumers would like to rent A 2 apartments at the price ceiling of P C . Because P C is below the equilibrium price, there is a shortage of apartments equal to ( A 2 – A 1 ). (Notice that if the price ceiling were set above the equilibrium price it would have no effect on the market since the law would not prohibit the price from settling at an equilibrium price that is lower than the price ceiling.)

Figure 4.11 The Unintended Consequences of Rent Control

Controlling apartment rents at P C creates a shortage of ( A 2 − A 1 ) apartments. For A 1 apartments, consumers are willing and able to pay P B , which leads to various “backdoor” payments to apartment owners.

If rent control creates a shortage of apartments, why do some citizens nonetheless clamor for rent control and why do governments often give in to the demands? The reason generally given for rent control is to keep apartments affordable for low- and middle-income tenants.

But the reduced quantity of apartments supplied must be rationed in some way, since, at the price ceiling, the quantity demanded would exceed the quantity supplied. Current occupants may be reluctant to leave their dwellings because finding other apartments will be difficult. As apartments do become available, there will be a line of potential renters waiting to fill them, any of whom is willing to pay the controlled price of P C or more. In fact, reading up to the demand curve in Figure 4.11 “The Unintended Consequences of Rent Control” from A 1 apartments, the quantity available at P C , you can see that for A 1 apartments, there are potential renters willing and able to pay P B . This often leads to various “backdoor” payments to apartment owners, such as large security deposits, payments for things renters may not want (such as furniture), so-called “key” payments (“The monthly rent is $500 and the key price is $3,000”), or simple bribes.

In the end, rent controls and other price ceilings often end up hurting some of the people they are intended to help. Many people will have trouble finding apartments to rent. Ironically, some of those who do find apartments may actually end up paying more than they would have paid in the absence of rent control. And many of the people that the rent controls do help (primarily current occupants, regardless of their income, and those lucky enough to find apartments) are not those they are intended to help (the poor). There are also costs in government administration and enforcement.

Because New York City has the longest history of rent controls of any city in the United States, its program has been widely studied. There is general agreement that the rent control program has reduced tenant mobility, led to a substantial gap between rents on controlled and uncontrolled units, and favored long-term residents at the expense of newcomers to the city (Arnott, R., 1995). These distortions have grown over time, another frequent consequence of price controls.

A more direct means of helping poor tenants, one that would avoid interfering with the functioning of the market, would be to subsidize their incomes. As with price floors, interfering with the market mechanism may solve one problem, but it creates many others at the same time.

Key Takeaways

• Price floors create surpluses by fixing the price above the equilibrium price. At the price set by the floor, the quantity supplied exceeds the quantity demanded.
• In agriculture, price floors have created persistent surpluses of a wide range of agricultural commodities. Governments typically purchase the amount of the surplus or impose production restrictions in an attempt to reduce the surplus.
• Price ceilings create shortages by setting the price below the equilibrium. At the ceiling price, the quantity demanded exceeds the quantity supplied.
• Rent controls are an example of a price ceiling, and thus they create shortages of rental housing.
• It is sometimes the case that rent controls create “backdoor” arrangements, ranging from requirements that tenants rent items that they do not want to outright bribes, that result in rents higher than would exist in the absence of the ceiling.

A minimum wage law is another example of a price floor. Draw demand and supply curves for unskilled labor. The horizontal axis will show the quantity of unskilled labor per period and the vertical axis will show the hourly wage rate for unskilled workers, which is the price of unskilled labor. Show and explain the effect of a minimum wage that is above the equilibrium wage.

Case in Point: Corn: It Is Not Just Food Any More

Figure 4.12

Herry Lawford – Stocks – CC BY 2.0.

Government support for corn dates back to the Agricultural Act of 1938 and, in one form or another, has been part of agricultural legislation ever since. Types of supports have ranged from government purchases of surpluses to target pricing, land set asides, and loan guarantees. According to one estimate, the U.S. government spent nearly $42 billion to support corn between 1995 and 2004.

Then, during the period of rising oil prices of the late 1970s and mounting concerns about dependence on foreign oil from volatile regions in the world, support for corn, not as a food, but rather as an input into the production of ethanol—an alternative to oil-based fuel—began. Ethanol tax credits were part of the Energy Act of 1978. Since 1980, a tariff of 50¢ per gallon against imported ethanol, even higher today, has served to protect domestic corn-based ethanol from imported ethanol, in particular from sugar-cane-based ethanol from Brazil.

The Energy Policy Act of 2005 was another milestone in ethanol legislation. Through loan guarantees, support for research and development, and tax credits, it mandated that 4 billion gallons of ethanol be used by 2006 and 7.5 billion gallons by 2012. Ethanol production had already reached 6.5 billion gallons by 2007, so new legislation in 2007 upped the ante to 15 billion gallons by 2015.

Beyond the increased amount the government is spending to support corn and corn-based ethanol, criticism of the policy has three major prongs:

• Corn-based ethanol does little to reduce U.S. dependence on foreign oil because the energy required to produce a gallon of corn-based ethanol is quite high. A 2006 National Academy of Sciences paper estimated that one gallon of ethanol is needed to bring 1.25 gallons of it to market. Other studies show an even less favorable ratio.
• Biofuels, such as corn-based ethanol, are having detrimental effects on the environment, with increased deforestation, stemming from more land being used to grow fuel inputs, contributing to global warming.
• The diversion of corn and other crops from food to fuel is contributing to rising food prices and an increase in world hunger. C. Ford Runge and Benjamin Senauer wrote in Foreign Affairs that even small increases in prices of food staples have severe consequences on the very poor of the world, and “Filling the 25-gallon tank of an SUV with pure ethanol requires over 450 pounds of corn—which contains enough calories to feed one person for a year.”

Some of these criticisms may be contested as exaggerated: Will the ratio of energy-in to energy-out improve as new technologies emerge for producing ethanol? Did not other factors, such as weather and rising food demand worldwide, contribute to higher grain prices? Nonetheless, it is clear that corn-based ethanol is no free lunch. It is also clear that the end of government support for corn is nowhere to be seen.

Sources: Alexei Barrionuevo, “Mountains of Corn and a Sea of Farm Subsidies,” New York Times , November 9, 2005, online version; David Freddoso, “Children of the Corn,” National Review Online, May 6, 2008; C. Ford Runge and Benjamin Senauer, “How Biofuels Could Starve the Poor,” Foreign Affairs , May/June 2007, online version; Michael Grunwald, “The Clean Energy Scam,” Time 171:14 (April 7, 2008): 40–45.

Answer to Try It! Problem

A minimum wage ( W min ) that is set above the equilibrium wage would create a surplus of unskilled labor equal to ( L 2 – L 1 ). That is, L 2 units of unskilled labor are offered at the minimum wage, but companies only want to use L 1 units at that wage. Because unskilled workers are a substitute for a skilled workers, forcing the price of unskilled workers higher would increase the demand for skilled labor and thus increase their wages.

Figure 4.13

Arnott, R., “Time for Revisionism on Rent Control,” Journal of Economic Perspectives 9(1) (Winter, 1995): 99–120.

Principles of Economics Copyright © 2016 by University of Minnesota is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License , except where otherwise noted.

Downloadable Content

Government Intervention and Free Markets: Case Studies in Voter Opinion

• Student Research
• Rhodes, Robert T.
• Guerrero, Mario
• Letters, Arts, & Social Sciences
• Political Science
• California State Polytechnic University, Pomona
• political science
• free market
• 2015-04-21T18:18:21Z
• http://hdl.handle.net/10211.3/138063
• All rights reserved.

Relationships

• 2015 SRC Oral Presentation Sessions
• Dr. Mario Guerrero

Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.

In: Economics

Case Study 1: Government Intervention In Germany in 2009 there was considerable debate about the extent...

• Case Study 1: Government Intervention

In Germany in 2009 there was considerable debate about the extent to which the government should be intervening in the economy. For example, its citizens were worried about the future of Opel, a German car brand that was part of the ailing General Motors. Some wanted the government to make sure jobs were saved no matter what. Others, however, were more hesitant and worried about becoming the government becoming too interventionist. Traditionally since the Second World War the German government has seen itself as a referee in market issues and has avoided trying to control parts of the economy. It would regulate anti-competitive behavior, for example, but not try to run many industries. However, in the recession of 2009 when the economy was shrinking the government was forced to spend more to stimulate demand and had to intervene heavily to save the banking sector from collapse. The government also had to offer aid to businesses to keep them alive.

1. What are the possible benefits of a government intervening in an economy?

3. What prompted greater intervention by the German government in 2009?

4. What would determine whether the German continued to intervene on this scale in

the future?

Expert Solution

1. Governments can intervene to provide a basic security net – unemployment benefit, minimum income for those who are sick and disabled. This increases net economic welfare and enables individuals to escape the worst poverty. This government intervention can also prevent social unrest from extremes of inequality. There are many advantages of government intervention such as even income distribution, no social injustice, secured public goods and services, property rights and welfare opportunities for those who cannot afford.

3. Germany was shaken by the financial crisis that began in mid-September 2008 on two levels: First, its banks were hit by the spreading financial panic, forcing the government to intervene in the overleveraged banking sector; second, its exports were hit by falling worldwide demand as credit dried up.

4. The value of currency and the economic power of Germany will determine such an intervention in future.

Related Solutions

Case-1 in 1920, there was considerable debate regarding how much the us government should intervene in..., from this case study "the 2007-2009 financial crisis" , whats the role of derivative securities, government,..., in the intervention section of your case study, you need to describe what the intervention is..., discuss the extent to which market failure provides sufficient justification for government intervention into your country's..., to what extent the government response in each of the three countries usa, uk, and germany was effective in containing and solving the crisis, what is so special about healthcare that justifies government intervention, to what extent should government be concerned about the prospects of deflation, 1. a study conducted in march 2009 found that about half of u.s. adults trusted the..., case study: after a considerable investment of both money and time, executives at leman healthcare were..., case study - warehouse case study - warehouse the following case about otter aerospace is a....

• Advanced Math
• Anatomy and Physiology
• Civil Engineering
• Computer Science
• Electrical Engineering
• Mechanical Engineering
• Operations Management
• Statistics and Probability
• In April 2020, the U.S. national debts has exceeded $23.8 Trillion. It is rising much faster...
• what al ghazali thinks about virtue?
• 1. what intersectionality is and why it might be important to psychologists 2. what is  the difference...
• Use the Black-Scholes model to find the value for a European put option that has an...
• Using C++ Write a program that reads a text from the keyboard until the user presses...
• Some economists claim that inflation is always a “monetary phenomenon.” What do they mean by this...
• How do you think recessions influence elections?

Lead in the Americas: A Call for Action (1996)

Chapter: session d: case studies of interventions, case studies of interventions, introduction: case studies of interventions.

D AVID R ALL *

The task of reducing lead exposures and preventing lead poisoning across the Americas seems a daunting task. This session highlights four case studies of interventions in different sectors. The first describes a successful voluntary industry initiative in Mexico to remove lead solder from the canning process. The second presents an example of government regulation—in this instance, the U.S. Clean Air Act—that dramatically reduced population lead levels in the United States. The third case study explores the role of international organized labor—in this case the U.S. Carpenter's Union—in educating and training workers about lead poisoning and describes the union's efforts to work with government agencies to ensure stricter protective policies for workers and their families. The final case study describes the role of community activism and education in empowering local communities in Mexico to design and implement focused public health programs to reduce lead exposure in their populations.

What these four case studies illustrate is that successful control and prevention strategies require the involvement of people and organizations at all levels of society, from the federal sector down to the grassroots or community. Lead poisoning is a problem that directly affects people at all levels of society. Its solutions must, therefore, also be shared.

A LFONSO DE L EÓN *

In 1992, the metallic containers industry in Mexico stopped producing tin cans with lead soldering as food containers, substituting instead a process that closes tin cans with electrical solder. Mexican public health authorities are now interested in determining whether a quantifiable reduction in population blood lead levels, especially in children, has occurred as a result of this voluntary industry change.

The National Chamber for Metallic Containers of Mexico represents more than 85 percent of steel tin can and 100 percent of aluminum tin can production, making it the leading manufacturer of metallic containers in Mexico. The process leading to the total elimination of lead soldering in food cans was begun as a voluntary initiative by industry, although the increased pace of change in the latter phases of conversion was dictated by external events. This process is described briefly below.

Metallic containers have been used to hold and conserve food for more than 180 years. In the beginning, tin was used in soldering, but it resulted in little flexibility and a fragile seam. Lead rapidly replaced tin in solder because it is a ductile material that easily adheres to the tin plate and can be mixed with the tin to produce a more flexible and less fragile soldering. At one point the solder commonly used contained 90 percent lead and 10 percent tin. Such soldering was universally adopted and, with it, many billions of cans were produced globally, without an understanding that lead in external soldering posed a public health hazard.

When medical and public health authorities began to acknowledge concerns about the effects of lead exposure on human health, attempts were made to identify the different sources of the metal. Leaded gasoline, paints, and ceramicware glazes and food cans containing leaded solder were rapidly identified as important sources.

Manufacturers of metallic containers had already began, by the late 1970s, to substitute lead soldering with electric soldering. This change was instituted for reasons apart from public health concerns about lead, which remained largely unknown at the time. The main motives for converting to electrical soldering were to have a cleaner production process, greater productivity, and a larger surface area on the cans for the purposes of advertising (leaded seams are broader than electrically soldered seams). Mexico adopted the electrical soldering technique, despite the many lead soldering production lines that still had many years of usable life and the $1.2-2.0 million investment that was required to refit production lines. This change was an unusual one for Mexico, which does not usually afford itself the luxury of getting rid of production equipment that is still considered useful.

At the beginning of 1991, a strong popular movement against lead in the environment was organized in Mexico City. The metallic container industries became involved in this movement, along with other lead industries. As an outcome, ecologists, united with health authorities, insisted that the electrical soldering substitution process that had been under way for over 10 years be abruptly accelerated. We cannot deny that moments in the ensuing deliberations were amusing.

The industry's basic argument in 1991 against an abrupt acceleration in the substitution process was based on the following: (1) in 1979 the industry voluntarily initiated a gradual substitution of production lines that utilized lead solder; (2) the cost for substituting each line was too high to warrant wholesale, rapid substitution; (3) until research proved that inward migration of lead from the external soldering of tin cans was contributing to the amount of lead in canned foods, accelerated action was premature; and (4) the recommended threshold for lead in foods was still subject to debate.

To support industry's argument, we noted U.S. FDA reports that indicated similar voluntary changes occurring at the same pace in the U.S. metallic container industry. According to these reports, in the early 1980s, 90 percent of tin cans produced in the United States had leaded solder; by the 1990s, the proportion had dropped to approximately 4 percent and was expected to decrease even more in the coming years. Clearly, industry in two major markets was voluntarily making changes, and the demand that the Mexican sector suddenly accelerate its rate of conversion—at considerable additional cost—was unfair.

The FDA , to my knowledge, has not yet been able to establish legislation totally prohibiting the use of lead soldering in tin cans for food. In 1994, for

example, 139,000 million tin cans were produced in the United States; of these, 20,000 million were three-piece tin cans used for food. Assuming that only 2 percent of these had lead soldering means that 400 million tin cans would have still contained lead soldering. It is our industry's belief that, in the United States, voluntary action by industry will eliminate lead soldering in the near future without the need for legislation.

Returning to the story in Mexico, none of the arguments listed above swayed the necessary Mexican decisionmakers. Health authorities, pressured by ecologists, continued to demand immediate conversion to electrical soldering. This activity culminated in the 5 July 1991 signing of the “Actions for the Integral Solution of the Problems Related to Lead Content in Products that Could Constitute a Risk for Health and Ecosystems” in Mexico City. The National Chamber for Metallic Containers agreed in this document to accelerate the transformation of all of its production lines in order to eliminate lead from the soldering process. We were given a very short time to complete the necessary changes. This posed special problems—for example, the machinery for producing electrical soldering was manufactured in Europe and required 10 to 12 months for delivery. Our organization was also required to close 15 new production lines. Recognizing the difficulties of implementing wholescale change in a short time frame, the Mexican government generously granted our industry 18 months to fulfill our task.

At that time, as an active member of the “Normalization of the Metallic Food and Drink Containers Subcommittee, ” I began work with colleagues on the elaboration of a preliminary program to regulate the types of solder that would be permitted on tin cans for food. On 2 June 1993, a text of the program was made available and accepted by both the private and public sectors. The health secretary in the Office of the General Management of Environmental Health visited our plants to verify that lead soldering had effectively been eliminated from the soldering process.

Finally, on 11 November 1993, the report of the subcommittee was published in the Diario Oficial (Mexico's largest newspaper). According to the law, 90 days were given to interested parties to express comments and concerns about the proposed project. On 8 February 1994, the secretary of health held a meeting to analyze the commentaries given. Revisions were made, and the final text revised. On 14 November 1994, the Norma Oficial Mexicana (Official Mexican Norm) NOM-002-SSAT-1993 “Metallic Containers for Food and Beverages: Seal Specifications and Sanitary Requirements,” was published in the Diario Oficial . The norm prohibits the use of

lead soldering in metallic containers for food. The norm is also applicable to imported products, serving as a way to prohibit lead-soldered cans from entering the country. Helped by a loan of $35 million, our industry had fulfilled its commitment to change to lead-free soldering in the 18 months accorded us.

One aspect of Mexico's experience that was not given the attention it deserved was the unfilled need for assistance to workers who had been poisoned through occupational exposure to the lead soldering process. On many of our old production lines, the containers of melted solder sat open, generating lead fumes. Systems of absorption and filtration existed for these emissions, but were rarely adequate. In addition, masks usually given to workers were frequently not used because the workers complained of hindered breathing. As a result, cases of lead poisoning were common, and many of these workers still require medical attention.

Our organization also believes that food containing lead before processing and canning should be given careful attention. Tuna, for example, because of its metabolism, has a tendency to absorb heavy metals such as lead, cadmium, mercury, and arsenic. The National Institute of the Consumer ( INCO ) of Mexico on one occasion reported large amounts of canned tuna containing lead above the threshold limit established at that time. The cans in which the tuna was stored were of the two-piece variety that did not contain solder of any kind.

R OBERT A. S CALA *

This paper will trace a successful story of government intervention to reduce the risk of airborne lead intake and possible lead poisoning. The story is told from the perspective of the U.S. Environmental Protection Agency ( EPA ), the main government agency involved.

Lead had been used as an additive in gasoline from the 1920s to boost octane and to provide lubrication for certain engine parts. Tetraethyl lead and tetramethyl lead, both high in octane value, lubricate intake and exhaust valves and help to reduce engine knock (EIA, 1992). Over time the use of additives became increasingly widespread, and the amount in fuel increased as octane demand increased. Although the limit for lead in gasoline was approximately 4 grams a gallon, a usage level of 2.5 g/gal was more typical. Because of increasing health and environmental concerns over atmospheric lead, lead reduction began with the 1970 Clean Air Act, which authorized restrictions on the use of lead in gasoline. In 1970 there were almost 90 million passenger cars registered in the United States, and motor gasoline consumption was 5.78 million barrels daily (approximately 915 million liters daily). Under the Clean Air Act, the U.S. EPA was authorized to regulate fuels and fuel additives. Unleaded fuels also appeared in the early 1970s. Auto manufacturers were required to design and build vehicles that could operate on unleaded fuels or low-lead fuels, and a schedule was set for the reduction of lead levels in leaded fuels.

Before the Clean Air Act, the EPA had strong concerns about the potentially harmful effects of lead, but was unable to persuade the scientific community or industry that airborne lead represented a health hazard outside the workplace. Under the Clean Air Act, the EPA was given the authority to control airborne lead attributable to motor vehicle emissions for reasons beyond potential health hazards. A 1980 National Research Council publication outlined a model for regulatory decisionmaking re-

garding potential health hazards of environmental agents (NRC, 1980). There were nine steps:

Identify sources of lead and pathways of environmental transfer;

Identify specific human populations with exposure to lead;

Estimate the level of exposure to lead by each environmental pathway for each specific population;

Establish the association between exposure to lead and the level of lead in the body for each specific population;

Establish the association between the level of lead in the body and biological change caused by lead for each specific population;

Estimate the upper limit of nondetrimental biological change for each specific population and the level of lead in the body associated with that degree of biological change;

Identify and describe alternative control strategies;

Apply risk-benefit, cost-benefit, and other considerations, compare alternatives for control, and decide what is an acceptable level of lead in the environment for each specific population; and

Evaluate the process and the decision.

Of greatest interest with respect to motor gasoline is its contribution to airborne lead. The amount of lead in the air appears to be related in large measure to the amount of lead in fuel. Leaded fuels generated 24,000 µg/m3 of lead at the tail pipe in the era before lead phasedown (NRC, 1993). Typical lead levels in urban environments in the 1970s were in the range of 0.5 to 10 µg/m3, and perhaps 90 percent of this is attributable to lead from gasoline. Most lead is emitted as halides and oxides, but virtually all of it is eventually converted to the sulfate. Most of the lead is deposited near the vehicular source. Particles with diameters in the range of 10 µm are deposited over a broad distance, and there is long-range transport of particles with a diameter of less than 0.1 µm for over a month (NRC, 1993). Lead is widely distributed in the body, with a preferential uptake by bone.

From the outset, EPA held that leaded gasoline was a source of air and dust lead that could be reduced readily and significantly in comparison with other sources. It also held that young children in the age range of 1 to 5

years should be regarded as a group sensitive to lead exposure (EPA, 1978). The third link in the EPA chain of logic is that contaminated dust and dirt from motor vehicle exhausts are the most important exposure routes for children (EPA, 1973). The health status of children was the principal driving force for the regulation of the lead content of fuels. As late as 1984, about 6 million children and 400,000 fetuses in the United States were exposed to lead at concentrations that placed them at risk of adverse health effects, defined as blood lead levels of at least 10 µg/dl (NRC, 1993).

The EPA put great emphasis on the work of Azar et al. (1975), which showed a ratio of lead in the air to lead in the blood of 1:1.8 at airborne levels of 1.5 µg lead/m3, where lead in blood was expressed in the usual terms of µg/dl (micrograms lead per deciliter of blood). The work was in adults, not children. the EPA holds that children have a greater net absorption and retention of lead than adults. The agency assumes that the air-lead to blood-lead relationship for this sensitive population exposed to lead in ambient air equals or exceeds the relationship for adults. The literature also suggests (ACGIH, 1991) that this relationship is nonlinear with concentration in air. Yankel et al. (1977), using children living near a smelter, found an average ratio of 1:1.9. Some of the nonlinearity in the relationship is explained by particle size changes with concentration, 24-hour vs. 8-hour exposures, and certain kinds of avoidance behavior by workers.

Preceding presentations have discussed the issues centering on clinical and subclinical effects of various body burdens of lead, expressed either in blood lead levels or tissue concentrations. the EPA emphasized minimizing lead burden. The agency position was that air lead contributed to general population lead exposure and that airborne lead levels below 2 µg/m3 affect blood lead levels. With the promulgation of a National Ambient Air Quality Standard for Lead in 1978 (EPA, 1978), the EPA stated that for the sensitive population previously defined (children ages 1–5 years), a blood lead level above 30 µg/dl was associated with an impairment in heme synthesis in cells, as indicated by an elevation in erythrocyte protoporphyrin. This finding was judged by the EPA to be adverse to the health of chronically exposed children. the EPA also declared that there were a number of other adverse health effects associated with blood lead

levels above 30 µg/dl in children, as well as in the general population. Three systems appear to be most sensitive to the effects of lead: the hematopoietic system, the nervous system, and the renal system. Inhibition of enzymes systems has a threshold as low as 10 µg/dl; at the other end of the scale, permanent, serious neurological damage or death have thresholds approaching or exceeding 80 to 100 µg/dl in children.

The EPA acknowledged that the lead exposure problem arose from a combination of sources, including food, water, air, leaded paint, and dust. The contribution of each source varies depending on the environment, bioavailability, and individual exposure and uptake. Reducing lead from all sources would improve health; the EPA , however, viewed lead in gasoline as a source of air and dust lead that could be readily and significantly reduced.

The first steps taken to reduce lead in gasoline in the United States resulted from the introduction of the platinum-based catalytic converter. This device, which reduced emissions of polycyclic aromatic hydrocarbons and other pollutants from gasoline, was damaged by lead. Starting in 1973, it was necessary for EPA to ensure a supply of lead-free fuels for new cars equipped with catalytic converters. In 1978, EPA moved to evaluate the public health benefits of removing lead from gasoline in order to reduce lead in the ambient environment. The initial regulations set quarterly limits on allowable amounts of lead used by refiners and permitted averaging this amount across all grades of gasoline produced. It was assumed that the natural replacement of older vehicles with cars requiring unleaded fuels would result in the programmed reduction of lead in gasoline, without further intervention by government. By 1982, however, two trends were recognized that prevented this natural reduction: first, a substantial number of motorists continued to use leaded gasoline because of its lower pump price (misfueling); and second, since the allowable content of gasoline was defined as the average of all fuels produced, the per gallon content of leaded gasoline could actually increase as the fraction of leaded/unleaded decreased. Because of these two phenomena, the EPA took further regulatory action to ensure the phasedown of lead in gasoline in 1983–1985. Simply by changing the basis of calculating allowable levels of lead in gasoline from the total fuel base to leaded gasoline only, EPA caused a substantial decrease in the amount of lead used in gasoline. In 1990, the U.S. Congress mandated the eventual phaseout of lead in gasoline in the United States.

There were some environmental consequences of the motor gasoline lead phasedown. One was misfueling, or using leaded fuel in a vehicle designed to use unleaded fuel. Misfueling was found in about 6 percent of vehicles in the inspection and maintenance programs.

Of interest is how much lead use was reduced by the phasedown of leaded fuels. With the introduction of the catalytic converter in 1973, the next 10 years showed an increase in unleaded gasoline use. By 1983, unleaded fuel was just over 50 percent of the total gasoline market in the United States, and by 1995 it represented almost the entire market. the EPA estimated a 34 percent reduction in lead use over the interval 1983 –1990, with a savings of almost 130,000 tons of lead (EPA, 1982c). Natural Resources Defense Council figures estimate more comprehensible figures. Lead used in motor gasoline in the United States totaled 243 thousand metric tons in 1971; 309 thousand metric tons in 1976; and 138 thousand metric tons in 1981. When the phasedown really took effect, lead use was reduced to 7.1 thousand metric tons in 1986 and 4.4 thousand in 1992 (NRDC, n.d.).

Companies that sold lead additives for gasoline were faced with the prospect of seeing their business virtually vanish over a period of just a few years. They were in the position of buggy whip makers at the beginning of the automotive era. The prudent companies diversified. There was also some reasonable resistance on the part of auto manufacturers. Lead in gasoline not only provided octane value but was also a lubricant for engine parts. The presence of lead prevented valve seats from being “beaten in” because the relatively soft lead deposits provided a form of cushion. Work on metallurgy had to be done as part of the lead phaseout. In the end, unleaded and low-lead gasolines were widely available on schedule.

The EPA examined the costs of lead phasedown in terms of reduced lead use (EPA, 1984). The agency estimated that the marginal cost of removing lead from gasoline was about U.S. 1¢ for each gram of lead. EPA also found that lead removal presented a positive cost-benefit. The benefit fell into three

categories: savings on vehicle maintenance, reduced misfueling, and reduced health care costs. For the period 1986–1988, the projected 0.1 gram per gallon for leaded gasoline would produce annual benefits over costs of greater than $1 billion ($1,000 million).

The findings of the National Health and Nutrition Examination Surveys II ( NHANES II ), which covered the period of 1976–1980 (NRC, 1993), used a stratified multistage probability cluster sample of U.S. households. Blood lead levels were measured for persons aged 6 months to 74 years, and almost 10,000 samples were used. As Figure 3-8 depicts, there was a striking

Figure 3-8. Lead used in gasoline production and average NHANES II blood lead (February 1976-February 1980). Source: Reprinted, with permission, from J.L Annest, “Trends in the bood lead levels of the U.S. population: The Second National Health and Nutrition Examination Survey ( NHANES II ) 1976-1980,” in Lead Versus Health, M. Rutter and R.R. Jones, eds., New York: John Wiley & Sons. ©1983, John Wiley & Sons, Ltd.

association between declining lead levels in gasoline in the period of 1976-1980 and the subsequent decline in blood lead levels in the U.S. population. Although the choice of scale for each of the vertical axes in Figure 3-8 permitted the dramatic overlay of the declines in amount of lead in gasoline and blood lead levels, one is hardpressed to deny that the association is not a correlation. For NHANES III (Pirkle et al., 1994), the survey covered the period 1988-1991, with over 12,000 blood samples from a population whose ages ranged upward from one year. The mean blood lead level of persons aged 1-74 years decreased 78 percent, from 12.8 to 2.8 µg/dl, a drop of 10 µg/dl. Comparable percentage decreases were seen for non-Hispanic white and non-Hispanic black children aged 1 to 5 years. There was a decrease in the prevalence of blood lead levels equal to or greater than 10 µg/dl, from 85 percent to 5.5 percent for non-Hispanic white children and from 97.7 percent to 20.6 percent for non-Hispanic black children. The authors attributed the majority of this decrease in blood lead levels to the virtual removal of lead from gasoline and the reduction of lead in soldered cans (Pirkle et al., 1994).

Government intervention in the form of control regulations can be successful in reducing the body burden of lead in the population as measured by blood lead levels. This reduction in burden surely represents a comparable reduction in the risk of lead-related diseases and dysfunctions.

J OHN D. R EPKO *

The United Brotherhood of Carpenters, with about 550,000 members in North America, has members who work in all aspects of the trade, including general construction, renovation, and repair, as well as working on bridges and other steel structures. Our members work in buildings and on steel structures known to contain lead-based paint. Piledrivers are frequently exposed to lead-based paint when applying acetylene torches to bridges undergoing renovation.

Lead intoxication is a continuum, in which the adverse effects are expressed at the cellular, organ, or whole organism level, depending on the dose (Silbergeld et al., 1991). There is no magic level; the toxic effects of lead are evident when increased absorption begins.

In addition to working in an occupation that exposes a worker to lead, the worker may come to the job with exposures to lead from the home or living environment. Whatever standards are ultimately established for workers, they must provide for this margin of prior exposure.

“Take-home” lead is also a concern. Recently the National Institute for Occupational Safety and Health ( NIOSH ) (Sussell, in press) reported 62 incidents worldwide of paraoccupational, or “take-home” lead exposure. Industries with reported paraoccupational lead exposures included lead smelting, battery manufacturing and recycling, radiator repair, electrical components manufacturing, pottery and ceramic production, and stained glass making.

The Carpenters' general approach to intervention is very much proactive. Generally, our approaches fall into four categories: (1) education and training, (2) collaboration with scientists involved in occupational health research, (3) support for occupational and environmental health standards and regulations, and (4) what we call, “dealing an honest hand.”

Health and safety education for workers about the occupational dangers of lead exposure is an important part of both apprenticeship education and the ongoing training of the journeyman. The U.S. Society for Occupational and Environmental Health, for example, developed a document that was used extensively by the EPA in the development of regulations governing required curriculums for lead-based paint worker and supervisor training (SOEH, 1993). This document and the training programs it has engendered have been effective among our membership in preventing work-related diseases and injuries, keeping workers up-to-date, and helping workers develop an attitude and willingness to practice prevention.

The Carpenters have collaborated on many research activities designed to identify health problems arising from workplace exposures to lead. In 1991, Dr. Irwin Selikoff, who headed Mount Sinai Environmental Sciences Laboratory, in collaboration with NIOSH and Harvard Medical School, conducted the first membershipwide screening of Carpenter workers. A subgroup underwent an even more detailed study, including questionnaires on lead exposure history, blood lead testing, and in vivo measurement of bone lead levels. The study found that age was the dominant predictor of both tibia and patella bone lead. Demolition, carpet laying, and alcohol ingestion were also significant predictors of bone lead (Watanabe et al., 1994). The authors concluded that the data reflect a subclinical effect of bone stores of lead on hematopoiesis and that this effect is the first epidemiologic evidence that bone lead may be an important biologic marker of ongoing chronic toxicity. The researchers also found that the differences in concentrations of bone lead between the tibia and patella are suggestive of ALAD -2-associated pharmacokientic effects. Further, they suggest that subclinical lead-associated kidney dysfunction is found with relatively low

current blood lead concentrations, and that the ALAD -2 genotype may be an additional modifier of this effect (Smith et al., 1995). The Carpenters have also worked with Dr. Selikoff's successors, a network of devoted scientists, by promoting continued study of the membership.

These studies demonstrate that information useful to workers' health can be obtained when workers and unions collaborate with their scientific and academic colleagues. By providing additional evidence of lead's toxicity at very low blood and bone levels, such studies can promote changes in occupational standards that better protect the health of workers.

Numerous standards have been set for lead in both the occupational and nonoccupational setting. Some standards set limits on allowable concentrations of lead in the ambient air or workplace air, as well as other environmental compartments, and some standards are based upon biological monitoring. In the workplace, both airborne lead and biomarkers are components of preventing lead poisoning in most national systems.

The U.S. occupational lead standard was promulgated in 1978, one of the first de novo standards developed by the Occupational Safety and Health Administration ( OSHA ), without reliance on earlier guidelines proposed by the American Council of Government and Industrial Hygienists ( ACGIH ). the OSHA lead standard was noteworthy in that it proposed both a limit on airborne lead in the workplace (50 microgram/m 3 ) and a mandatory program of biological monitoring in most work settings. In addition, the OSHA standard protected worker health and employment rights by establishing a medical removal program: workers whose blood lead levels exceeded the standards were temporarily shifted to jobs without lead exposure with no loss of pay, benefits, or seniority. This approach was intended to change the incentives in the labor management relationship, to encourage employers to reduce lead exposures, and to protect workers from job termination or loss of income.

In the United States, occupational standards cover most, but not all, workers exposed to lead. In 1993 the lead standard was finally extended to workers in the construction industry, who are often highly exposed to lead during repair and maintenance of steel structures (which may still be painted with lead-based paints); demolition workers; and workers involved in abating lead hazards in housing. Small workplaces are still imperfectly covered, and some of these, such as battery repair shops, may be sources of intense exposure that not only pose problems for workers, but

can also be sources of contamination to workers' homes and releases to the environment (Matte et al, 1989b; Nunez et al, 1993). In several countries of the Americas, informal “cottage industries,” such as recycling facilities, are almost wholly outside any regulatory surveillance, as demonstrated in a report from Tijuana, Mexico (Leung, 1988).

Historically, organized labor and the lead industries have not had an easy relationship. A number of incidents illustrate their lack of cooperation. The events surrounding the American Smelting and Refining Company ( ASARCO ) lead smelter emissions in El Paso, Texas, in the late 1960s (Repko et al., 1978) and the Bunker Hill smelter in Kellog, Idaho, in the 1970s (EDF, 1992) are two such examples. The lead industry also consistently attacks studies that demonstrate a causal relationship between lead exposure and adverse health effects in order to counter this increasing body of evidence. Another industry approach is to claim simply that lead is not a problem and that nothing, therefore, needs to be done about limiting exposures to the metal.

As a result, labor has had to consistently challenge the positions taken by the lead industry. NIOSH 's legal right of entry, for example, has been used to conduct health hazard evaluations in the workplace. Because of the active involvement and support of labor, however, the lead industries have never won a major victory over the EPA or OSHA , the government organization most responsible for worker health.

In response, there is a current effort by industry to shift its worksites into international markets where regulation is less restrictive or does not exist. Moreover, environmental advocacy groups or scientific communities may also be absent or small in a number of these areas. Workers all over the world are increasingly under attack by corporate and financial interests who use the globalization of products such as lead as a lever to restrict worker rights and to lower workplace standards. It is the responsibility of labor to fight against this trend.

Workers continue to be poisoned by lead on the job. Members of labor groups or health professionals can help reduce worker exposure in several ways:

Worker education is the first step to prevention. Workers have the right to know. An informed workforce increases its control over its exposure. Inspectors and government officials responsible for workplace health and safety must also be well-trained and informed about lead hazards.

International and national protective standards for workers should be widely adopted across the Americas, but these should not necessarily copy those of the United States. Labor and workers should have an opportunity to establish stringent standards for lead exposure consistent with new scientific evidence that was not available or not used when the United States first set its standards.

Existing environmental or workplace standards that limit workers' exposure to lead must be enforced. To adequately enforce regulations will require a coordinated effort by an educated workforce, competent inspectors, an effective judicial system, and policymakers who can support the courts and the inspectors.

Productive collaborations among unions, scientists, occupational health professionals, and academics should be fostered.

Union officials should be informed of the importance of occupational health and encouraged to raise the level of workers' health to the same level of concern in contract negotiations that is routinely given to economic issues.

Finally, persist, and don't give up. Whenever new ways of making a dollar, peso, real, other currency involve new hazards, those who work to protect worker health are playing “catch up.” Remember that it is always the workers who pay first and worst.

C HRISTINA VON G LASCOE , M.D., P H .D. *

If people in the developing world are to have radically improved lives, it is first of all necessary to teach them to be dissatisfied with the present situation and at the same time make them appreciate how they can work towards a better future.

Bidwell, 1988

Environmental health issues are of recent concern at the northern border of Mexico, but to date have not been addressed in a concerted fashion by either community interest groups or public health authorities. For example, public health authorities and two major nongovernmental organizations ( NGO s) working in primary care and family planning in Tijuana, Mexico, have yet to incorporate environmentally driven concerns such as lead poisoning into their regular health activities. Nevertheless, there are indications that lead poisoning and other environmental health concerns are gaining visibility at the local level in Mexico, as they are elsewhere in the Americas.

A major requirement in creating the necessary infraculture for community-based action is the need to increase popular knowledge and public awareness of environmental health hazards and the steps that can and should be taken to reduce these hazards. An example of such work is provided in this paper. Ways to incorporate such knowledge into the cultural horizon of a community under environmental health stress are highlighted.

Anthropologists interested in health recognize that community members practice a wide range of activities related to public health, with or without the direction of health authorities. In any community there coexist interpretations of an official and a traditional system of dealing with public health threats, and these do not necessarily influence each other. In order to deal

effectively on a grassroots level, it is necessary as a first step to define the “cultural horizon” of the community. Broadly defined, the cultural horizon is the wide range of public health activities practiced by community members, with or without the direction of the health authorities. Thus, the development of effective community strategies to reduce or prevent lead poisoning will depend not only on information on the sources of contamination and number and distribution of cases identified, but also on being able to convey that information in a manner that is reflective of the community's cultural horizon. That horizon can be defined by a number of factors, including:

the level of personal behavior engaged in health promotion (individual, family, and community/public);

the level of interaction and cooperation with local health authorities in matters of personal and public health.

Responsibility for health is transferred to individual households through the empowerment of individuals to act in ways consonant with that responsibility. Assumptions underlying this transition are:

People will act responsibly on their own behalf given sufficient information, resources, and interest, and in the context of a particular cultural horizon that sanctions this behavior;

Information can consist of information and resources, but interest must be stimulated from within the community. Information presentation in and of itself does not guarantee the “success” of community-based programs;

Health professionals need to determine what the focus of prevention activities should be;

The potential threat of a hazardous substance must be made visible to the community so that it may be communicated in ways that can be understood by community members. In transmitting this information, it is imperative to impart knowledge on specific steps that can and should be taken by community members to reduce their exposure, and that of their families, to the hazardous substance in question. Allowances should be made for community-driven modifications in these steps deriving from the culture or other values of the inhabitants.

Underlying these four assumptions is the recognition that in the past, the community has often been the missing link in addressing environmental and occupational health problems at the local level. The community has often been excluded because members were not considered knowledgeable enough to participate in problem solving. While this paternalistic belief continues to be held by many traditional health professionals, particularly in less-developed countries or in inner-city or rural regions of the more developed countries of the Americas, there is a growing movement toward including community members and other “stakeholders” in the development and implementation of prevention and control programs. Growing experience indicates that such individuals can be vital to the development of innovative and sustainable solutions. Thus, community empowerment should remain a cornerstone of prevention and control efforts.

The best community interventions expand the cultural health horizon of a community by adding culturally relevant alternatives to standard or “textbook” approaches to prevention. In order to do this, the goals and methods of agencies must be aligned with the goals and methods of the community; in other words, they must be made relevant to the community members and vice-versa. How can this be done?

The first step is to appraise the “health horizon" of the community—for example, by determining, in the case of lead, the level of personal behavior directed toward reducing industrial and family exposures. The current degree of information exchanged about the sources and nature of lead poisoning in the community and the interaction between the public health authorities and community members on health issues important to the community also need to be assessed.

As a general rule, we have observed that where the health authorities have a strong bureaucracy, community leadership is suppressed, and health professionals manage the public's health; conversely, where the public health bureaucracy is weak, alternative forms of representation naturally arise. For example, in underserved areas, health authorities are often helped by local grassroots organizations and other NGO s. As a consequence, some of the most innovative experimentation in community education and empowerment is occurring in the most economically disadvantaged populations or in rural areas. This experimentation should be carefully studied to determine the circumstances that foster increased

community autonomy and decisionmaking power that translates into improved community health and well-being.

To what extent can interested volunteers support and sustain the health of their communities in the face of an increasingly toxic environment? As a first step, there have to be sufficient health personnel, laboratories, and budget to respond to the threat of exposure to hazardous substances. Mexico provides an example where the direct participation of citizens is a real resource. In the small rural towns of Sonora, for example, there are voluntary groups called health committees. These health committees, which are organized by local health authorities but include community membership, have general oversight in matters pertaining to the running of the local health centers and their prevention programs. The committees, in general, have significant political clout with the regional health administrators; thus they can often exert sufficient pressure to obtain needed personnel and services.

In many of the northern border cities of Mexico, NGO s also depend heavily on community volunteers. Although few are currently addressing the issue of lead poisoning in the region, NGO s provide a potentially useful mechanism for dealing with environmental hazards in a manner that can be useful to communities. the NGO s are often staffed by volunteers who act on behalf of their own and their neighbors' best interests. The question then becomes one of how to motivate community members to participate.

Lead toxicity is a major problem in Mexico, as it is in every other country of the Americas. Whole neighborhoods in Tijuana, for example, are exposed to lead poisoning, and this has resulted in significantly lower IQ levels in grade school children living near the lead emission point sources (Guzmán, 1994). Of the children surveyed by Guzmán, 80 percent had blood lead levels that have been associated in cognitive studies with a seven-point performance IQ deficit, and the average blood lead of residents is higher than that associated with severe symptoms of lead poisoning.

To date there has been no consistent public health response to the crisis of lead poisoning in Tijuana, nor has any effective community action been undertaken, in a large part because of a lack of community awareness about the hazards. For many in Tijuana, lead is invisible; for others, who may be aware of the hazards of lead, there is little knowledge about common routes

of exposure, for example, lead-glazed ceramicware or local remedies for such maladies as stomach disorders that contain lead (Baer et al., 1989).

Our study of community perceptions of lead as a health risk in Tijuana produced seven distinct and contrasting perspectives (see Box 3-1 ). These seven views demonstrate increasing knowledge of the dangers, and the nature and sources of community lead exposure.

The preferred approach to developing strategies to reduce lead exposures is one that utilizes and is based on the capacities, skills, and assets of community members. Much of past public health experience demonstrates that significant advances in community health take place only when local community mentors are committed to investing themselves and their resources in the effort. Depending on its size and sophistication, a given community may have local institutions (such as businesses, schools, parks, libraries, hospitals, and colleges), citizens' associations (including churches, block clubs, and cultural groups), and gifted individuals (for example, artists, retired people, and young people) whose skills and knowledge can be used in the development of effective education and intervention.

Specific steps for accomplishing these ends include:

Discovering the existence of active and trusted organizations or other resources within a community. These may include formal organizations (churches, government offices, and the like); informal organizations and networks (such as family and friends); or capacities and assets of individuals, citizens' associations, and local institutions;

Undertaking a community-led needs assessment that will allow residents to identify problems and priorities that are most important to them; this step should involve the building of productive relationships among local and state health networks who share the common goal of preventing lead poisoning; and

Identifying resource needs and the mechanisms to meet them.

READ FREE ONLINE

Welcome to OpenBook!

You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

Do you want to take a quick tour of the OpenBook's features?

Show this book's table of contents , where you can jump to any chapter by name.

...or use these buttons to go back to the previous chapter or skip to the next one.

Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

Switch between the Original Pages , where you can read the report as it appeared in print, and Text Pages for the web version, where you can highlight and search the text.

To search the entire text of this book, type in your search term here and press Enter .

Share a link to this book page on your preferred social network or via email.

View our suggested citation for this chapter.

Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

Get Email Updates

Do you enjoy reading reports from the Academies online for free ? Sign up for email notifications and we'll let you know about new publications in your areas of interest when they're released.

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• My Account Login
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Open access
• Published: 09 November 2023

Government intervention, industrial structure, and energy eco-efficiency: an empirical research on new energy demonstration in cities

• Xiaoyi Zhang 1 ,
• Rui Zhang 1 ,
• Yue Wang 1 ,
• Meilin Zhao 1 &
• Xin Zhao 1  

Scientific Reports volume  13 , Article number:  19446 ( 2023 ) Cite this article

937 Accesses

Metrics details

• Energy efficiency
• Energy policy

This study investigates the relationships among government intervention, industrial structure, and energy eco-efficiency (EE). Energy eco-efficiency was measured based on a non-radial directional distance function for 236 cities in China from 2005 to 2019. Additionally, the difference-in-difference model (DID) method and spatial econometric models were used to analyse the impact of government intervention and industrial structure on energy eco-efficiency and their spatial spill-over effects. Government intervention includes fiscal expenditures and policy orientation for new energy demonstration construction. Our results indicate that: China’s EE has a fluctuating upward trend and increased 17.85% in the period, and its spatial distribution imbalance gradually developed into a regional distribution balance. Moreover, government intervention and adjustment of the industrial structure improved urban energy eco-efficiency by 7.43% and 0.92%, respectively, which also has spatial spill-over effects in neighbouring regions. Furthermore, economic development, technological innovation, and foreign direct investment enable EE. However, urbanisation hinders the improvement of energy eco-efficiency. Finally, heterogeneity analysis showed that the policy of the new energy demonstration city has better effects on eastern and western cities in promoting EE.

Similar content being viewed by others

How does the opening of high-speed rail drive energy restructuring? New micro evidence from China

Yanchao Feng, Juan Zhang, … Shuhai Niu

The impact of China’s energy saving and emission reduction demonstration city policy on urban green technology innovation

Changfei Nie, Ruyi Li, … Zhi Chen

Assessment of green technology innovation on energy-environmental efficiency in China under the influence of environmental regulation considering spatial effects

Wei Li, Xiaomin Xu, … Can Zhang

Introduction

Global warming poses an irreversible threat to human economic and social development. The consumption of fossil fuels and emission of industrial waste are the leading causes of the greenhouse effect. Using renewable energy can effectively alleviate the negative impact of fossil fuels, mitigate climate change, and improve energy security. In China, the renewable energy sector has developed rapidly owing to recent policies and incentives. From 2005 to 2006, China promulgated and implemented the Renewable Energy Law. A series of policies, including target planning, fiscal support, tariff control, and consumption guarantee, were introduced, which promoted the large-scale utilisation of renewable energy. Renewable energy consumption in China reached 454 million tons of standard coal in 2022 1 , accounting for 8.34% of the total primary energy consumption, representing an increase of 18.01% from 2021. This renewable energy development benefits China's energy transition and low-carbon economic development, substantially improves the energy structure, helps cope with climate change, and improves EE.

Energy eco-efficiency (EE) combines the economic and ecological benefits generated by energy consumption 2 , 3 , and can be interpreted as maximising socio-economic benefits with minimal energy consumption and ecological damage. This indicator comprehensively reflects the system efficiency of the "energy-economy-environment". In social development, a laisser-faire market economy that pursues economic interests often neglects damage to the environment, thus necessitating government intervention.

To achieve a clean energy structure and ecological civilisation construction strategy, China launched the construction of new energy demonstration cities, with the first batch of 81 cities (districts) being announced in January 2014. The construction of new energy demonstration cities allows China to explore an energy development path with Chinese characteristics, and substantially impacts EE. In this study, the construction of new energy demonstration cities was considered as the policy orientation of government intervention. Policy orientation refers to the policies and regulations issued by the government to encourage the development of new energy, including information resources, financial support, market resource allocation, development direction, talent attraction, and technological innovation.

Fiscal expenditure, a means of government intervention in the social market economy, can promote the economy and reduce emissions. Fiscal expenditure has a "multiplier effect", and fiscal input by the government can effectively promote national economic growth. However, problems with energy consumption and environmental pollution accompany economic development models. Moreover, the government's strengthening of financial investment and taxation measures in specific fields can also effectively promote technological innovation, and carbon reduction. However, it remains controversial as to whether the emission reduction effect of fiscal expenditure and the pollution caused by economic growth are balanced and coordinated, and whether this balanced effect will influence EE. The industrial structure is a major determinant that affects the intensity of resources and the degree of environmental friendliness in economic growth, and is closely related to EE 4 . EE improves when the pattern of structural change in economic development has a prominent effect on pollution. However, the change in the industrial structure caused by the simple accumulation of factors cannot improve the efficiency of resource utilisation, and may damage the environment 5 . Therefore, to effectively implement China's high-quality development strategy, it is necessary to elucidate the relationship between China's industrial structure changes and EE.

The potential contributions of this study are twofold. First, the impact of government intervention and industrial structure on EE was explored by constructing a non-radial directional distance function (NDDF) and a difference-in-difference model (DID). In the evaluation process, the EE of 236 cities in China was measured based on the NDDF. Additionally, the DID was used to identify the impact of government intervention and industrial structure on EE. Second, the spatial econometric model was used to analyse the spatial spill-over effects and regional heterogeneity of government intervention and industrial structure, and to provide policy recommendations for new energy demonstration cities.

This article is divided into seven sections. The next presents a literature review. The “ Models and variables ” section describes econometric models, selected sample data, and variables. The “ Empirical results and analysis ” section explains the analysis of the EE measurement results and benchmark regression. The “ Robustness tests ” section discusses parallel trend tests, superposition policy tests, and Propensity Score Matching and Difference-in-Difference (PSM-DID) tests. The sixth section, “ Spatial spill-over effect and heterogeneity analysis ”, presents the empirical results of the heterogeneity analysis and extended analysis of spatial spill-over effects. Finally, the conclusions and policy recommendations are discussed in the last section.

Literature review

EE evaluation methods include the super Slacks-Based Measure Model (Super-SBM) 2 , stochastic frontier model analysis (SFA) 6 , and super-efficiency models based on NDD (Super-NDDF) 7 . An analysis of the spatiotemporal characteristics and influencing factors of EE showed that, the EE of cities in China is generally at a low level; complex fluctuations characterise the time pattern, and the EE of each city has improved to varying degrees 6 . China’s EE has significant global and local spatial agglomeration characteristics, but the spatial distribution is uneven, and there are prominent spatial effects 8 . Additionally, Wang et al. 9 calculated and analysed the EE of the energy-enriched area of the Yellow River Basin using SFA and the Spatial Durbin Model (SDM), and found that the EE was relatively low and showed a downward trend. Innovation and industrial structure were found to be prominent factors enhancing EE in this region.

The academia has faced considerable controversy over the impact of government intervention on factor productivity. Some researchers have concluded that government intervention can achieve economic and environmental objectives by improving the impact of resource endowments on energy efficiency 10 , thus effectively improving the efficiency of the green economy 11 . However, some researchers have found that government intervention will disrupt the market by restricting private investment, reducing the efficiency of capital allocation, and inhibiting the increase in factor productivity 12 . The government intervenes in social development mainly through economic intervention and policy orientation, in which fiscal expenditure is the primary method of economic intervention. Researchers have studied the relationship between factor productivity and fiscal expenditure, finding that the emission of sulphur dioxide and other pollutants reduces with an increase in public financial expenditure 13 , which can significantly promote total factor energy efficiency 14 . Zhang et al. 15 found that government fiscal expenditure can improve EE; however, many researchers also concluded that fiscal expenditure will negatively affect factor productivity. The inhibitory effect on green total factor productivity increases with increasing fiscal expenditure 16 . Fang et al. 17 found that local governments may underestimate environmental protection in industrial diversification, while the decentralisation of fiscal expenditures will inhibit the improvement of EE.

By studying the impact of relevant policies and regulations with policy-oriented backgrounds on EE, researchers have found that it is difficult to meet the dual needs of improving life satisfaction and economic level by solving the conflict between the environment and energy utilisation through market mechanisms 18 . Chen et al. 7 measured the EE of 282 cities in China using Super-NDDF and found that environmental regulation has a spatial spill-over effect in improving ecological efficiency. Cui et al. 2 studied the impact of environmental regulation on EE based on Tobit and threshold regression models. They found that mandatory environmental regulations and market incentive environmental regulations have a more significant inhibitory effect on EE. In contrast, inhibiting voluntary environmental regulations has a time lag effect.

Regarding the research on industrial structure, many researchers have concluded that industrial structure adjustment positively affects EE. Guan and Xu 8 found that industrial structure is the most important factor affecting EE. Industrial transformation facilitates the effective allocation of resources through factor flow and professional division of labour to effectively improve EE 19 . Liu et al. 20 found that upgrading the regional industrial structure can considerably enhance eco-efficiency. However, some studies had different results. Meng and Zou 21 reported that the effect of industrial structure adjustment on EE is not significant. However, increasing the proportion of industrial output will increase regional social and economic benefits, and thus inflict more severe damage to the environment. The above research shows the need for a consistent conclusion on the impact of fiscal expenditure and policy-oriented government intervention and industrial structure on EE, especially in the form of budgetary expenditure.

Researchers have found that the government can effectively guide the public to improve energy efficiency through reasonable intervention. Implementing energy policies and financial subsidies can substantially improve energy efficiency. Many countries are actively committed to implementing various policies to promote energy efficiency and establish a sustainable energy mix. The policy orientation of constructing new energy demonstration cities facilitates the utilisation of renewable energy and can effectively alleviates pollutant emissions through government support, private financing, industrial structure optimisation, and resource allocation adjustment 22 , 23 . Yang et al. 24 concluded that policy orientation will encourage local governments to increase the intensity of environmental regulation for high-polluting industries and enterprises, and promote a certain amount of production factor resources transfer to policy-oriented new energy industries. The exploitation and utilisation of renewable energy can reduce dependence on traditional fossil fuel energy, enhance cities’ energy security, and promote sustainable economic, social, and environmental development 25 .

Although the spatiotemporal characteristics of EE and its impact have been studied from different perspectives, the impact of government intervention on EE and its spatial spill-over effects have yet to be explored from the perspective of new energy demonstration city policy orientation and financial expenditure. Furthermore, the influence of industrial structure adjustment on EE requires further verification. Therefore, in this study, we constructed a NDDF model considering the system of "energy-economy-environment", and then analysed the impact of policy orientation, fiscal expenditure, industrial structure, and spatial spill-over effects of government intervention on EE and its regional heterogeneity.

Models and variables

Econometric model.

Government intervention is measured by fiscal expenditure and the policy orientation of constructing new energy demonstration cities. Here, we studied the impact of government intervention and industrial structure on EE. As DID can solve endogenous problems commonly faced in the existing literature 26 , the construction of new energy demonstration cities can be viewed as a "natural experiment". In this study, fiscal expenditure, policy orientation, and industrial structure were considered as core explanatory variables and a DID model was constructed to estimate the effect of government intervention and fiscal expenditure on EE. The equation for this model is as follows:

where, \(i\) and \(t\) are cities and years, respectively; the explained variable \(Y_{it}\) is the annual EE of each city, \(du_{it}\) is a region dummy variable, and \(dt_{it}\) represents the time dummy variable. The interaction coefficient \(\beta_{3}\) reflects the net effect of policy orientation on EE. \(FE_{it}\) is fiscal expenditure, \(IND_{it}\) is industrial structure, \(X_{it}\) is the control variable matrix, comprising economic development level, urbanisation level, foreign investment and technology level. \(T_{t}\) is the time-fixed effect, \(\mu_{i}\) is the individual-fixed effect, and \(\varepsilon_{it}\) is the random disturbance term. Logarithmic processing was performed on the data of variables to reduce the influence of skewness and heteroscedasticity.

Spatial econometric model

A certain spatial correlation was observed because cities are not independent. Fiscal expenditure, policy orientation, and industrial structure affect regional EE and may also impact adjacent areas. Thus, in this study, spatial factors were incorporated into the model, taking government intervention and industrial structure as core explanatory variables. The spatial econometric model is as follows.

where \(W\) is the geospatial distance weight, \(WLnEE_{it}\) is the spatial lag of the explained variable, \(\rho\) and \(\xi\) are the space-effect coefficients. \(\nu_{it}\) is the error term of \(\varepsilon_{it}\) . When \(\rho { = }\xi { = 0}\) , it degenerates into a spatial error model (SEM). When \(\xi { = }\gamma { = 0}\) , it degenerates into a spatial lag model (SLM). When \(\gamma { = 0}\) , it is the SDM.

Samples and data

In this study, 3540 balanced panel observations of 236 cities in China from 2005 to 2019 were used to investigate the impact of fiscal expenditure, new energy demonstration city construction policy orientation, and industrial structure on EE. The investigation of the construction policy orientation of new energy demonstration cities only considered the first batch of demonstration cities, and data samples at the prefecture-level were used. To ensure the robustness of the conclusion, county-level cities, districts (autonomous prefectures), and industrial park cities were eliminated from the first batch of demonstration cities established in 2014. Subsequently, 47 and 189 cities were generated in the experimental and control groups, respectively. To investigate its regional characters, the 30 provinces in China were divided into eastern, central and western regional according to geographic location. Specifically, the eastern region includes 12 provinces and cities, namely Beijing, Hebei, Tianjin, Shandong, Jiangsu, Shanghai, Zhejiang, Fujian, Guangdong, Hainan, Liaoning and Guangxi; The central region includes the nine provinces of Shanxi, Henan, Hubei, Hunan, Anhui, Jiangxi, Inner Mongolia, Heilongjiang and Jilin; The western region covers nine provinces, namely Chongqing, Sichuan, Shaanxi, Yunnan, Guizhou, Gansu, Qinghai, Ningxia and Xinjiang. Above 236 cities were divided into three regions based on their belong provinces. The data used in this study were obtained from the statistical yearbook of cities at all levels, the China City Statistical Yearbook, and the Statistical Bulletin of Social and Economic Development. The missing values were uniformly filled by interpolation.

Variable description

Interpreted variables.

The direction distance function (DDF) proposed by Chung et al. 27 is extensively used in energy and environmental efficiency. However, a limitation of using the DDF is that all the input and output elements must change in the same direction. Zhou et al. 28 further proposed the NDDF, which can effectively solve the problem of input and output factors changing in the same direction. Therefore, in the present study, the NDDF was used to measure EE. The NDDF is defined as follows 28 :

where \(g = ( - g_{x} ,g_{y} , - g_{b} )\) represents the specified direction vector, \(W = (W_{x} ,W_{y} ,W_{b} )\) represents the weighting vector of each input–output element, and \(\beta = (\beta_{x} ,\beta_{y} ,\beta_{b} )^{T} \ge 0\) represents the variable proportion of each input and output factor.

Energy (E), capital (K), and labour (L) were taken as input factors. The gross domestic product (GDP) of each city (G) is taken as desirable output. Sulphur dioxide (S), smoke (dust) emissions (C), and wastewater discharge (P) were undesirable outputs in each city 2 . Using NDDF, the DEA model of EE of 236 cities was constructed as follows:

where the direction vector \(g = ( - K, - L, - E, + G, - C, - S, - P)\) , referring to the research of Liu et al. 29 , the weight matrix \(W^{T} = \left( {\frac{1}{9},\frac{1}{9},\frac{1}{9},\frac{1}{3},\frac{1}{9},\frac{1}{9},\frac{1}{9}} \right)\) is given, substituting into Eq. ( 3 ), and obtaining \(\beta_{j}^{ * } = (\beta_{jK}^{ * } ,\beta_{jL}^{ * } ,\beta_{jE}^{ * } ,\beta_{jG}^{ * } ,\beta_{jC}^{ * } ,\beta_{jS}^{ * } ,\beta_{jP}^{ * } )\) through linear programming, that is, the optimal solution of the slack variable of the \(j\) th city. The EE for each city in the corresponding year is calculated as follows:

In measuring EE, the labour force (L) of each city was measured based on the number of employees. Owing to the lack of city-scale energy consumption data in the statistical yearbook, provincial energy consumption data were retrieved for each city according to the light data value using a linear model without intercept 30 . The amount of capital investment was estimated using the perpetual inventory method 31 . The GDP of each city was used to express the desirable outputs. Sulphur dioxide (S), wastewater discharge (P), and smoke (dust) emissions (C) were estimated using data on industrial sulphur dioxide emissions, industrial wastewater discharges, and industrial smoke and dust emissions 2 . In this study, the impact of price factors on research was reduced by adjusting all price data in 2005.

Core explanatory variables

Policy orientation ( \(du \times dt\) ) The policy orientation investigated in this article is the new energy demonstration city pilot dummy variable, \(du \times dt\) , where \(du\) is the processing variable, indicating whether the city was selected as the first batch of demonstration cities in 2014; if selected, \(du{ = 1}\) , otherwise \(du{ = 0}\) . Additionally, \(dt\) is a time dummy variable; \(dt{ = 0}\) before the pilot city was selected, and \(dt{ = 1}\) was chosen after the selection.

Fiscal expenditure Fiscal expenditure is a government macroeconomic control measure that can intervene in economic development, pollution and carbon emission reductions, and resource allocation. By investing in new energy industry infrastructure and technological innovation, the government can continuously improve the technological innovation environment and technological service facilities, guiding the flow of innovative resources and attracting more enterprises and research institutions to increase their investment. Increasing investment in the new energy industry will accelerate the development and utilisation of new energy, reduce pollutant emissions, and promote the improvement of EE. In this study, the level of urban financial expenditure was measured as the proportion of urban general financial budget expenditure to GDP.

Industrial structure The continuous adjustment of the industrial structure can promote the rational allocation of resource elements in various industries and maintain a balance between input and output. The Thiel index is an essential indicator for measuring the reasonable allocation of industrial resource elements in cities. In this study, the Thiel index was used to measure the industrial structure. As described in Gao et al. 32 , the calculation is performed as follows:

where \(Y_{i}\) and \(Y\) represent the added value and GDP of the three industries, respectively; \(L_{i}\) and \(L\) are the employment and total employment of the three industries. The \(IND\) value reflects a reasonable degree of industrial structure. The larger the value, the more reasonable the allocation of resources and production factors among sectors.

Control variables

The economic development level (PGDP) was expressed in GDP per capita. Foreign direct investment (FDI) is measured as the proportion of FDI utilised by each city in regional GDP 33 . Urbanisation level (URBAN) was measured as the ratio of the urban population to the total population. Technological innovation (TI) is measured as the proportion of employees engaged in scientific research, technical services, and geological prospecting to those employed in the unit at the end of the year. To eliminate heteroscedasticity, logarithmic processing was performed for each variable. The descriptive statistical analysis of each variable is presented in Table 1 .

Empirical results and analysis

Analysis of ee calculation results.

Equations ( 3 ) and ( 4 ) were applied to measure the EE of 236 cities in China from 2005 to 2019, and the EE kernel density curves 34 were plotted, as shown in Fig.  1 . Figure  1 shows the dynamic evolution characteristics of the EE levels of the sampled cities in 2005, 2010, 2014, and 2019. Overall, the main peak of the nuclear density curve tends to shift to the right over time, indicating that the EE level in China is constantly improving. From the perspective of the distribution position, the EE kernel density curves for 2005, 2010 and 2019 show a "double peak", with a "1 main and 2 secondary peak" feature in 2014, where the main peak is "high" and to the left, and the secondary peak is "low" and to the right (the secondary peak in 2014 refers to the right side). The part with a lower EE level has a higher peak kernel density, whereas the part with a higher EE level has a lower kernel density. This means that the current level of EE in most cities in China still needs to increase, and only a few cities have a high level of EE. The value of the main peak in each region first increased and then decreased. The width initially decreased and then increased, indicating that the absolute difference in EE among cities in China first narrowed and then expanded. The small subpeak on the left side of the kernel density curve in 2014 may owe to the new energy demonstration city policy in 2014, which made some cities start to increase the construction of renewable energy infrastructures, following energy consumption increasement and the damage to the ecological environment, Hence is energy eco-efficiency is at a low level, appearing in the left side of the main peak of the small subpeak.

Kernel density curve of energy eco-efficiency (EE) in main years.

The vector diagram of the average value of EE in 2005–2013 before and after the implementation of the new energy demonstration cities policy was created using ArcGIS. Combined with the breakpoint method, the average value of EE is divided into three grades: low efficiency (below 0.6), medium efficiency (between 0.6 and 0.8), and high efficiency (above 0.8), as shown in Fig.  2 .

Average distribution map of energy eco-efficiency (EE). ( a ) Average distribution of EE before the policy, ( b ) Average distribution of EE after the policy. NEDC refers to the new energy demonstration city. Note: The map is produced based on the standard map with review number GS(2019)1822 downloaded from the standard map service website of the State Administration of Surveying, Mapping and Geographic Information of China, with no modifications to the base map.

Since 2005, China’s EE has generally shown an upward trend. By 2019, it had increased by 17.85%, but still in the transition stage from low to medium efficiency. The average value of EE before and after the implementation of the policy indicates that the number of cities with different levels of EE has changed significantly. The proportion of cities with high-efficiency levels increased from 5.93% before the implementation of the policy to 7.20%. The proportion of cities with medium-efficiency levels increased from 12.71 to 40.68%, while the proportion of low-efficiency cities decreased from 81.36 to 59.75%. As shown in Fig.  2 , the regional distribution of EE levels has changed. Before the implementation of the policy, the average EE in the eastern region (0.5508) was the highest, followed by that in the central (0.4887) and western (0.4828) regions. The level of EE in China has gradually increased from the western to the eastern coastal cities, and cities with medium and high-efficiency levels are mainly concentrated in the eastern region. After implementing the policy, the average EE value in all the cities increased. The region with the highest average value was still in the eastern region (0.5889), followed by the central region (0.5619), and the lowest value was observed in the western region (0.5606). Medium- and high-efficiency cities are developing towards the west and central regions, and the spatial distribution of EE has gradually achieved balanced development.

In general, medium- and high-efficiency cities have gradually developed from regional concentration to balanced regional distribution, from being concentrated in the eastern coastal areas to balanced development in the central and western regions. The western and central regions could benefit from their unique natural resources and natural gas, solar energy, clean resources such as hydroelectricity, actively cultivating green industries, and promoting win–win industry and ecology.

Regression results and analysis

Table 2 lists the estimated results of Eq. ( 1 ). Models (1) to (4) in Table 2 introduce policy orientation, fiscal expenditure, industrial structure, and control variables in sequence. The regression model was determined to be a fixed-effects model using the Hausman test. The goodness of fit in Table 2 ranges from 0.6694 to 0.7630, indicating that the selected explanatory variables are key variables affecting EE. Among them, the goodness of fit of the regression with core explanatory variables and control variables is better, indicating that the estimated results of Model (4) can better reflect the effect of government intervention and industrial structure on EE.

The policy orientation ( \(du \times dt\) ) and fiscal expenditure that indicate the effect of policy have a significant positive impact on EE; both have passed the 1% significance test, indicating that government intervention with policy orientation and fiscal expenditure affects EE. Specifically, the policy orientation, financial expenditure and industrial restructuring significantly increased urban energy eco-efficiency by about 2.52%, 4.91% and 0.92%, respectively. It can be inferred that the policy formed by policy orientation is conducive to improving and supplementing the infrastructure and industrial environment on which the development of the new energy industry depends. Policy orientation often integrate innovative resources and capital. Learning and sharing effects are conducive to upgrading and innovating new energy technologies. Moreover, the optimal combination of conventional, renewable, and new comprehensive energy utilisation models can increase energy utilisation efficiency and reduce pollutant emissions, which is conducive to improving urban EE. Additionally, fiscal expenditure has a "Keynesian multiplier" effect, which promotes economic growth. Local governments have focused on green energy and environmental protection in environmental protection accountability and performance appraisal. This would promote the construction of new energy projects by increasing financial investment and support for energy conservation and environmental protection industries, actively promoting the healthy and rapid development of energy conservation and environmental protection industries, and effectively reducing pollutant emissions.

The regression coefficient of the industrial structure is significantly positive, indicating that EE can be improved by adjusting the industrial structure. This is because the allocation efficiency of resource elements can be effectively improved, the total amount of pollutants discharged in the production process can be reduced, and healthy development of the urban ecosystem and industrial economy can be achieved by coordinating the balanced development of industries. Furthermore, technological innovation, economic development level, and foreign direct investment can effectively promote the improvement of EE, indicating that in the context of industrial restructuring and government intervention, EE can be improved by attracting FDI, technological innovation, and promoting economic development. However, urbanisation has hindered EE improvement. This could be attributed to the increasing population in the city, the energy consumed in production and living, and the increase in pollution emissions, which hinders the improvement of urban EE. Moreover, urbanisation has led to the construction and use of a large amount of housing, entertainment, education, medical care, and transportation infrastructure, which has led to a sharp increase in the demand for energy and other resources. Therefore, in the planning of urban modernisation development, the government should focus on emphasising quality over quantity.

Robustness test

Parallel trend test.

An effective assumption of the DID method is to satisfy the parallel trend 35 . Therefore, a parallel trend test was conducted using the method of Beck et al. 36 . Specifically, the regression was performed by replacing the dummy variable \(dt\) in regression Eq. ( 1 ) with the dummy variable \({D}^{T}\) for each year (in which 2013 is the base period), as shown in Eq. ( 5 ). The parallel-trend test results are shown in Fig.  3 . The regression coefficients of years before the implementation of the policy orientation in 2014 fluctuated around the 0 axis (the 90% confidence interval includes the 0 value). Therefore, the difference in EE between the testing and control groups is not apparent, indicating that the testing and control groups had a parallel trend before the policy was implemented. Meanwhile, the estimated coefficients pass the 10 per cent significance test from the third year of construction of the demonstration city, indicating that the impact of the pilot policy on energy eco-efficiency has a lag of three years and grows fluctuatingly thereafter. The reason could be inferred that the improvement of EE relies on industrial restructuring, and industrial restructuring has the characteristics of high investment and long cycle, coupled with the long cycle of renewable energy infrastructure investment and construction, and in the short term, it is not possible to get rid of the dependence on fossil energy consumption, and optimise the structure of energy consumption with a certain time lag, so that the impact of pilot policies on the improvement of EE has a lag.

The results of the parallel trend test.

Superposition policy inspection

Considering that many similar or related policies between regions are implemented simultaneously or cross-wise, there is a certain overlay policy effect. Therefore, to exclude the impact of other policies in the same period, this study controlled for the impact on EE of the low-carbon city pilot and carbon trading pilot policies implemented during the sample period. Specifically, this study adds the above two policy dummy variables (the interaction term of the policy grouping dummy variable and the policy time dummy variable) to the benchmark regression to examine the causality between policy orientation and EE after controlling for other policy disturbances. Table 3 presents the corresponding estimated results. Models 5 and 6 are regression results that only contain policy variables, and Models 7 and 8 are the estimated results after adding control variables. According to Table 3 , after adding other relevant policy variables, the estimated coefficient of \(du \times dt\) remains significantly positive, which is the same as the result of the benchmark regression. This shows that, after considering the impact of the above policies, government intervention and industrial structure still significantly affect EE.

PSM-DID test

In the present study, technological innovation, economic development, urbanisation level, foreign investment, fiscal expenditure, and industrial structure were used as matching variables, and the neighbour 1:1 method for matching, the matching effect is shown in Fig.  4 . After the abovematching treatment, the probability distribution of tendency scores in the experimental group and the control group has basically tended to be consistent. It shows no significant systematic error, which is meeting the hypothesis of the PSM model. Regression estimation was performed as described in Eq. ( 1 ), and Table 4 shows the regression estimation results of PSM-DID.

Kernel density distribution comparison of propensity score values before and after kernel matching.

Models (9) to (12) in Table 4 introduce policy orientation, fiscal expenditure, industrial structure and control variables orderly, respectively. The regression coefficient is still significant at the 10% level and the sign is positive, indicating that the improved PSM-DID model results of the DID regression model are further improved. Therefore, supporting the industrial structure and government intervention has positive significance in promoting the improvement of EE.

Spatial spill-over effect and heterogeneity analysis

Spatial spill-over effect.

The general and specific Moran's I tests showed that the explained variables had a positive spatial autocorrelation, indicating the spatial layout of high-high and low-low clustering. A robust Lagrangian multiplier test (R-LM) and Lagrange multiplier (LM) test were performed to obtain a better fitting effect. LM test results significantly reject the null hypothesis, but the R-LM (error) failed the test 37 , 38 , and LM (lag) > LM (error), R-LM (lag) > R-LM (error), indicating that the SLM should be selected 39 . Furthermore, the Hausman test supported the fixed effects model. According to the \(R^{2}\) and Log-likelihood values of the SDM, SAR, and SEM models in Table 5 , combined with the Hausman test results, it is more reasonable to choose the SLM under individual fixed effects for empirical testing.

According to the SLM model results in Table 5 , the regression coefficients of policy orientation, fiscal expenditure, and industrial structure on urban EE are all significantly positive, which shows that industrial structure and government intervention have significantly improved the EE effect. Moreover, the spatial autoregressive coefficient \(\rho\) is significantly positive, indicating that EE has a significant endogenous spatial interaction effect among cities. The improvement of energy efficiency in this region can drive the improvement of surrounding areas and form a high–high agglomeration state in space. Specifically, the industrial structure and government intervention behaviour in a region will directly affect the local EE and indirectly affect the EE of neighbouring areas and produce a feedback effect, which will eventually further change the industrial structure and the actual impact of government intervention behaviour on local EE. LeSage and Pace 40 proposed that the influence of independent variables on dependent variables can be divided into direct, indirect, and total effects. We conducted a decomposition analysis based on the spatial lag model under individual fixed effects 37 , as shown in Table 6 .

According to the decomposition results of the spatial lag model in Table 6 , the direct effect of policy orientation, fiscal expenditure, and industrial structure on the EE of the city is positive. It passed the significance test, confirming that government intervention and industrial structure promote EE in the region. Moreover, the spatial spill-over effects of government intervention and industrial structure were significantly positive, indicating that fiscal expenditure, policy orientation, and industrial structure improve the EE of the region as well as that of neighbouring regions. The construction of demonstration cities is based on the market mechanism. It is oriented by government intervention, which could encourage enterprises to implement technological innovation, rationally allocate resource elements in various industries, exert technological innovation and demonstration effects, and promote the rational use and mutual flow of high-quality resource elements among regions. Furthermore, through knowledge diffusion and technological spill-over, a good demonstration effect and space radiation effect on adjacent areas can promote the "learning effect" in adjacent areas, thereby reducing the difficulty of EE improvement.

Analysis of regional heterogeneity

The impact of government intervention and the industrial structure of cities in different geographical locations on EE may differ. Therefore, we divided the research sample into three regions in the east, middle, and west and examined the impact of fiscal expenditure, policy orientation, and industrial structure on urban EE under location heterogeneity. Table 7 presents the estimated results of the spatial-lag model.

During the study period, the policy orientation coefficients of the demonstration cities were all positive, which verified that policy orientation is conducive to promoting the improvement of the EE of the region. The policy orientation of western and eastern cities passed the significance test, showing direct effects. The spatial spill-over effects of western and eastern cities account for 36.23% and 39.73% of their total effects, respectively, while the driving effects of policy orientation of the central cities were not significant. Fiscal expenditure significantly impacted EE in eastern cities rather than the west and central regions. Additionally, although the industrial structure can substantially promote the EE of eastern cities, it had no significant effect on western cities and had an inhibitory effect on the EE of central cities.

This regional difference may be because the central region has a high endowment of coal resources and an incomplete industrial structure. The extensive economic growth model causes the secondary industry to account for a substantial proportion, and there are many high-energy-consuming and high-polluting enterprises. Financial expenditure and policy orientation make it challenging to adjust the energy structure dominated by traditional fossil fuel energy in the short term, which has an impact on EE. Additionally, the regional economic development model has not fully transformed from "resource factor promotion" to "green development driving" due to the weak local technology foundation, the long-term output cycle of R&D input and results, and the long-term mechanism of R&D support is not apparent. Factor productivity and technological levels are not increased as planned and even hinders the improvement of EE.

By relying on solar energy, wind energy, and natural gas resources, the western region actively supports the green industry. It promotes EE via the spatial transfer of capital, technology, and advanced concepts brought about by policy guidance. The policy benefits from the new energy demonstration construction pilots have attracted high-tech projects transferred from the central and eastern regions, and the technological spill-over effects of these projects have led to corresponding improvements in the management capabilities and technical levels of the western regions, thereby improving EE in western regions. The east is a relatively developed region with a reasonable industrial structure. It could take good use of the dominant position of a developed economy and a good industrial foundation, actively responding to policy orientation, consciously innovating green technologies, accelerating scientific and technological research and development, vigorously developing clean energy, and improving green production capacity.

Conclusions and recommendations

Studying the internal relationship between industrial structure, government intervention behaviour, and EE in China has important implications for developing new energy, reducing pollutant emissions, and realising sustainable development. In this study, we selected panel data from 236 cities in China from 2005 to 2019, constructs the NDDF model to calculate the EE of each city, and establishes a DID model to empirically verify the effect of policy orientation, fiscal expenditure, and industrial structure on EE. The spatial impact and regional heterogeneity of government intervention and industrial structure on EE were analysed using a spatial econometric model. The research results show that China's EE is at a low level and shows an increasing trend during the research period, with a spatial distribution grid of low–low and high–high concentrations. The policy orientation, financial expenditure and industrial restructuring adjustment significantly increased urban energy eco-efficiency by about 2.52%, 4.91% and 0.92%, respectively, and industrial structure and government intervention have a positive spatial spill-over effect on EE improvement. Its spatial spill-over effect is transmitted between regions through the "learning effect" and "demonstration effect", thus promoting the improvement of regional EE. Additionally, there is spatial heterogeneity in this spill-over effect, and the driving impact of policy orientation on eastern and western cities is stronger. Foreign direct investment, economic development, and technological innovation can help improve EE, while urbanisation can inhibit the improvement of EE.

Based on the findings of this study, the following policy recommendations are proposed to promote the advancement of China's EE and promote the coupled development of energy, economy, and environment:

Strengthen policy guidance and financial support for high-tech and environmental protection and energy-saving industries; actively exert government intervention and industrial structure effects to cultivate new economic growth; and promote the advancement of EE using foreign direct investment and technological innovation. The government should take policy as guidance and financial support to encourage the development of strategic emerging industries in an overall way, promote the sustainable development of resource elements to industries such as low energy consumption, low pollution, high value-added environmental protection and energy conservation, high-tech and other sectors, reasonably promote the optimisation and upgrading of industrial structure, and promote the sustainable development of the green economy. Moreover, the government should advance the coordinated development of the business environment, technological innovation, and sustainable urbanisation. The government should improve the business environment and compensation mechanism for technological innovation, strengthen the construction of public research and development institutions and experimental platforms, and make more efficient and standardised use of funds from various sources, such as foreign capital. Additionally, it promotes the innovation and application of green technology in urbanisation and achieves the high-quality, coordinated development of urbanisation and the environment.

Actively promote the construction of new energy demonstration cities, strengthen the policy orientation effect of government intervention, and provide greater policy guarantees for the country’s overall new energy development. On the one hand, give full play to the policy-oriented resource allocation function, continue to promote the construction of new energy demonstration cities, strengthen the spatial spill-over effect of the construction of demonstration cities, improve the radiation scope and influence of the development of new energy cities, and achieve the high-quality development of new energy. But on the other hand, they are continuously optimising demonstration policies, strengthening policy systems such as intellectual property protection, ensuring reasonable monopoly income from innovation achievements of innovation subjects, and laying a policy foundation for constructing a strong new energy city and consummating the innovation environment.

Optimise the industrial structure according to local conditions, take advantage of the spatial spill-over effect of high-energy-efficiency regions, strengthen technical exchange and resource sharing, accelerate the coordinated development of EE regions, and jointly improve EE. First, it is necessary to make good use of the radiation effect of EE, strengthen inter-regional technical exchanges and resource sharing, and promote benign interactions and coordinated development between cities. Second, owing to the different characteristics of the city itself, the influence of government intervention and industrial structure on EE will have regional heterogeneity. Therefore, give full play to the advantages of resource endowment and implement a demonstration city construction plan according to local conditions.

The limitations of this study are as follows: First, this study focussed on China, which may have more reference value for developing countries but is less applicable for developed countries. Second, city-level macro-data research was used but the enterprise-level micro-angle was not considered, which has certain limitations.

Data availability

The datasets used and/or analyzed during the current study available from the corresponding author on reasonable request.

Energy Institute. Statistical Review of World Energy 2023 [EB/OL] . https://www.energyinst.org/statistical-review . Accessed 22 Sep 2023.

Cui, S., Wang, Y., Zhu, Z., Zhu, Z. & Yu, C. The impact of heterogeneous environmental regulation on the energy eco-efficiency of China’s energy-mineral cities. J. Clean. Prod. 350 , 131553 (2022).

Article   Google Scholar  

Feng, S., Kong, Y., Liu, S. & Zhou, H. Spatiotemporal pattern and convergence test of energy eco-efficiency in the Yellow River Basin. Int. J. Environ. Res. Public Health. 20 (3), 1888 (2023).

Article   PubMed   PubMed Central   Google Scholar  

Zeng, D.-Z. & Zhao, L. Pollution havens and industrial agglomeration. J. Environ. Econ. Manage. 58 (2), 141–153 (2009).

Article   MATH   Google Scholar  

Wang, Q.-S., Su, C.-W., Hua, Y.-F. & Umar, M. Can fiscal decentralisation regulate the impact of industrial structure on energy efficiency?. Econ. Res. Istraz. 34 (1), 1727–1751 (2021).

Google Scholar  

Liu, W., Zhan, J., Zhao, F., Wei, X. & Zhang, F. Exploring the coupling relationship between urbanization and energy eco-efficiency: A case study of 281 prefecture-level cities in China. Sustain. Cities. Soc. 64 , 102563 (2021).

Chen, P., Xie, R., Lu, M. & Huang, Z. The impact of the spatio-temporal neighborhood effect on urban eco-efficiency in China. J. Clean. Prod. 285 , 124860 (2021).

Guan, W. & Xu, S. Study of spatial patterns and spatial effects of energy eco-efficiency in China. J. Geogr. Sci. 26 (9), 1362–1376 (2016).

Wang, L., Zhang, F., Fu, W., Tan, Q. & Chen, J. Analysis of temporal and spatial differences and influencing factors of energy eco-efficiency in energy-rich area of the Yellow River Basin. Phys. Chem. Earth. 121 , 102976 (2021).

Besco, L. & Kirk, E. A. Industry perceptions of government interventions: Generating an energy efficiency norm. J. Environ. Pol. Plan. 23 (1), 130–142 (2021).

Zhai, X., An, Y., Shi, X. & Liu, X. Measurement of green transition and its driving factors: Evidence from China. J. Clean. Prod. 335 , 130292 (2021).

Wei, X., Hu, Q., Shen, W. & Ma, J. Influence of the evolution of marine industry structure on the green total factor productivity of marine economy. Water. 13 (8), 1108 (2021).

Article   CAS   Google Scholar  

Jin, Y., Tang, Y. M., Chau, K. Y. & Abbas, M. How government expenditure mitigates emissions: A step towards sustainable green economy in belt and road initiatives project. J. Environ. Manage. 303 , 113967 (2022).

Article   CAS   PubMed   Google Scholar  

Tang, L. & He, G. How to improve total factor energy efficiency? An empirical analysis of the Yangtze River economic belt of China. Energy. 235 , 121375 (2021).

Zhang, R., Hu, Y.-Y. & Qie, X.-T. Research on dynamic response of energy eco-efficiency of China’s logistics industry to influencing factors. Econ. Probl. 08 , 9–17 (2021) ( in Chinese ).

Wang, K.-L., He, S. & Zhang, F.-Q. Relationship between FDI, fiscal expenditure and green total-factor productivity in China: From the perspective of spatial spillover. PLoS ONE. 16 (4), 0250798 (2021).

Fang, X.-C., Tian, S.-Y. & Wang, X.-L. Analysis on the spatial effects of urban ecological efficiency, fiscal decentralization and industrial agglomeration in China: Empirical research based on spatial Durbin panel model. Urban Probl. 4 , 62–71 (2020) ( in Chinese ).

Bao, J. & Guo, B.-Q. Influence of heterogeneous environmental regulation on regional ecological efficiency. J. Arid Land Resourc. Environ. 36 (2), 25–30 (2022) ( in Chinese ).

Zhu, B. & Shan, H. Impacts of industrial structures reconstructing on carbon emission and energy consumption: A case of Beijing. J. Clean. Prod. 245 , 118916 (2020).

Liu, Z., Zhang, H., Zhang, Y.-J. & Zhu, T.-T. How does industrial policy affect the eco-efficiency of industrial sector? Evidence from China. Appl. Energ. 272 , 115206 (2020).

Meng, F.-S. & Zou, Y. Dynamic evaluation of energy eco-efficiency based on PP-SFA: Take the 30 autonomous regions of China as an example. Syst. Eng. 36 (05), 47–56 (2018) ( in Chinese ).

Cheng, Z., Yu, X. & Zhang, Y. Is the construction of new energy demonstration cities conducive to improvements in energy efficiency?. Energy. 263 , 125517 (2023).

Zhang, Q., Huang, X., Xu, Y., Bhuiyan, M. A. & Liu, P. New energy demonstration city pilot and green energy consumption: Evidences from China. Energy. Rep. 8 , 7735–7750 (2022).

Yang, X., Zhang, J., Ren, S. & Ran, Q. Can the new energy demonstration city policy reduce environmental pollution? Evidence from a quasi-natural experiment in China. J. Clean. Prod. 287 , 125015 (2021).

Wang, Q. & Yi, H. New energy demonstration program and China’s urban green economic growth: Do regional characteristics make a difference?. Energ. Policy. 151 , 112161 (2021).

Ye, P., Cai, W. & Zhou, Y. Can green industrial policy promote the total factor productivity of manufacturing enterprises?. Environ. Sci. Pollut. Res. 29 (58), 88041–88054 (2022).

Chung, Y. H., Färe, R. & Grosskopf, S. Productivity and undesirable outputs: A directional distance function approach. J. Environ. Manage. 51 (3), 229–240 (1997).

Zhou, P., Ang, B. W. & Wang, H. Energy and CO 2 emission performance in electricity generation: A non-radial directional distance function approach. Eur. J. Oper. Res. 221 (3), 625–635 (2012).

Liu, K., Lu, S. & Zhang, G. Regional difference in global unified efficiency of China: Evidence from city-level data. Sci. Total. Environ. 713 , 136355 (2020).

Article   ADS   CAS   Google Scholar  

Shi, D. & Li, S.-L. Emissions trading system and energy use efficiency: Measurements and empirical evidence for cities at and above the prefecture level. China Ind. Econ. 09 , 5–23 (2020) ( in Chinese ).

Shan, H. Reestimating the capital stock of China:1952–2006. J. Quant. Tech. Econ. 25 (10), 17–31 (2008) ( in Chinese ).

Gao, L., Tian, Q. & Meng, F. The impact of green finance on industrial reasonability in China: Empirical research based on the spatial panel Durbin model. Environ. Sci. Pollut. Res. https://doi.org/10.1007/s11356-022-18732-y (2022).

Namahoro, J. P., Wu, Q., Zhou, N. & Xue, S. Impact of energy intensity, renewable energy, and economic growth on CO 2 emissions: Evidence from Africa across regions and income levels. Renew. Sustain. Energ. Rev. 147 , 111233 (2021).

Niu, S., Luo, X., Yang, T., Lin, G. & Li, C. Does the low-carbon city pilot policy improve the urban land green use efficiency? Investigation based on multi-period difference-in-differences model. Int. J. Environ. Res. Public Health. 20 (3), 2704 (2023).

Chen, Z. & Wang, H. Research on the impact of the Belt and Road Initiative on the sustainability of the resource-based economy of participating countries. Environ. Sci. Pollut. Res. 29 (60), 91139–91154 (2022).

Beck, T., Levine, R. & Levkov, A. Big bad banks? The winners and losers from bank deregulation in the United States. JoF 65 (5), 1637–1667 (2010).

Xu, Y.-Z. & Chen, Y. Spatial spillover effects of carbon lock-in among provinces in China: An empirical study based on spatial auto-regressive model. J. South Chin. Norm. Univ. 02 , 126–134 (2018) ( in Chinese ).

Song, C., Chen, Y., Yin, G. & Hou, Y. Spatial correlation and influencing factors of industrial agglomeration and pollution discharges: A case study of 284 cities in China. Environ. Sci. Pollut. Res. 30 , 434–450 (2023).

Huang, Y.-P., Zhou, J.-J. & Shang, X.-T. Analysis of the impact of China’s housing price rise on the per capita income gap of the Province. Econ. Geogr. 38 (10), 29–35 (2018) ( in Chinese ).

Dolores Ugarte, M. Introduction to spatial econometrics. J. R. Stat. Soc. A. Stat. 174 (2), 513–514 (2011).

Download references

This research was funded by the Fundamental Research Funds for the Central Universities, China [No. 2022YJSGL13], and the Beijing Natural Science Foundation [No. 9222026].

Author information

Authors and affiliations.

School of Management, China University of Mining and Technology (Beijing), Beijing, 100083, China

Xiaoyi Zhang, Rui Zhang, Yue Wang, Meilin Zhao & Xin Zhao

You can also search for this author in PubMed   Google Scholar

Contributions

X.-Y.Z: Conceptualization, Methodology, Investigation, Visualization, Writing-original draft. R.Z: Resources, Methodology, Supervision, Funding acquisition. Y.W: Data curation, Writing-review and editing. M.-L.Z: Data curation, Software. X.Z: Writing-review and editing. All authors have read and agreed to the published version of the manuscript.

Corresponding author

Correspondence to Rui Zhang .

Ethics declarations

Competing interests.

The authors declare no competing interests.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Zhang, X., Zhang, R., Wang, Y. et al. Government intervention, industrial structure, and energy eco-efficiency: an empirical research on new energy demonstration in cities. Sci Rep 13 , 19446 (2023). https://doi.org/10.1038/s41598-023-46799-1

Download citation

Received : 25 March 2023

Accepted : 05 November 2023

Published : 09 November 2023

DOI : https://doi.org/10.1038/s41598-023-46799-1

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

By submitting a comment you agree to abide by our Terms and Community Guidelines . If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Government Intervention in Intermodal Transportation Case Study

Private companies have dominated the United States freight transportation system for a long time. They also control passenger transportation systems, but the control is shared with the government, unlike freight transport. Through fees, tools, and taxes, government transport departments can safeguard the public best interests in the passenger transportation industry. In contrast, government departments do not have a means to directly control or foster innovation in intermodal freight transport partly because of the resistance by trackers. In their book, Konings et al. (2008) suggest that the US government should be the catalyst for innovations in intermodal transportation. This argument can be justified by looking at a historical example of how a government-supported initiative has outperformed the free market-driven industry.

Why the US Government Should Be a Catalyst for Innovations in Intermodal Transport Systems

I agree with the author that the government should act as a catalyst for intermodal transport. The government brings tremendous benefits in transport and regulates limitations posed by the free market economy. This type of economy brings a problem of lack of interconnectivity and interoperability. This challenge is evidenced in intermodal road-rail transport because different parties use different resources. Private-owned intermodal transportation systems are different because they lack a centralized control unit that regulates the designs. Transportation is an essential aspect of the lives of Americans, and the time wasted when interconnecting problems occur leads to massive losses. For effectively centralized control to be achieved in such centers, the US government needs to step in and assist in innovation, ensuring uniform standards and compatible systems are deployed everywhere.

Additionally, I agree with the writer that the government needs to step in because it is able to pass information better. Intermodal transport involves several stakeholders and components, thus requiring an intensive flow of information (Kine, Gebresenbet, Tavasszy, & Ljungberg, 2022). An institution like a government capable of reaching all should be in control. In social-technical systems, evidence shows that governments outperform privatized institutions. The state plays 13 different roles in the governance of socio-technical systems: Observer, mitigator, opportunist, facilitator, lead user, enabler of societal engagement, promoter, moderator, gatekeeper, initiator, and watchdog (Borrás & Edler, 2020). Incorporating these in the transport industry would save a lot of wasted time and hassle.

The US government has intelligence tools at its disposal, and it can do better work in observing than private individuals. It is also respected and can-do mitigation in case of intermodal transport conflict. It can be relied on to facilitate innovation in the system because of the ease of additional funding. The government also far outperforms the private industry in promoting, moderating, and initiating intermodal transportation projects. With access to security officials, no organization can beat the government in the enforcement of intermodal transport principles.

Justification

The days of the US Department of Defense (DOD) being the global leader in defense technology are over. This is despite being the most funded defense department in the world. The declining dominance is attributed to the department’s public-private partnerships to procure technological resources. On the other hand, the nuclear power production industry, which for unavoidable reasons, is left in the control of the government, continues to prosper. The state’s role in the mode of governance of the nuclear power socio-technical system is very central to its success (Borrás & Edler, 2020). It offers transformative, innovative, and responsible leadership in a conflict-filled field. One of the conditions for the development of intermodal transport is better access to intermodal logistics networks offering appropriate line and point infrastructure (Przybylska & Dohn, 2019). This can only be achieved by the government as private investors seek to maximize profit at the expense of productivity, as observed in the case of the DOD.

Borrás, S., & Edler, J. (2020). The roles of the state in the governance of Socio-Technical Systems’ transformation. Research Policy, 49 (5), 103971.

Kine, H. Z., Gebresenbet, G., Tavasszy, L., & Ljungberg, D. (2022). Digitalization and automation in intermodal freight transport and their potential application for low-income countries. Future Transportation, 2 (1), 41-54.

Konings, R., Priemus, H., & Nijkamp, P. (2008). The role of government in fostering intermodal transport innovations: Perceived lessons and obstacles in the United States. In J. W. Konings (Eds.), The future of intermodal freight transport: Operations, Design and Policy (pp. 302-322). Cheltenham: Edward Elgar.

Przybylska, E., & Dohn, K. (2019). Analysis of Intermodal Freight Transport: Stakeholders in a Selected Cross-Border Area (Doctoral dissertation, Silesian University of Technology).

Use these guidelines if the customer asks for appendices. The first paragraph of the appendix should be flush with the left margin. Additional paragraphs should be indented.

Begin each appendix on a new page with the word “Appendix” at the top center. Use an identifying capital letter (e.g., Appendix A, Appendix B, etc.) if you have more than one appendix. If you are referring to more than one appendix in your text, use the plural appendices (APA only).

Label tables and figures in the appendix as you would in the text of your manuscript, using the letter A before the number to clarify that the table or figure belongs to the appendix.

Demographic Information for Cummings et al. (2002)’s Review

If an appendix consists entirely of a table or figure, the title of the table or figure should serve as the title of the appendix.

• Chicago (A-D)
• Chicago (N-B)

IvyPanda. (2023, January 6). Government Intervention in Intermodal Transportation. https://ivypanda.com/essays/government-intervention-in-intermodal-transportation/

"Government Intervention in Intermodal Transportation." IvyPanda , 6 Jan. 2023, ivypanda.com/essays/government-intervention-in-intermodal-transportation/.

IvyPanda . (2023) 'Government Intervention in Intermodal Transportation'. 6 January.

IvyPanda . 2023. "Government Intervention in Intermodal Transportation." January 6, 2023. https://ivypanda.com/essays/government-intervention-in-intermodal-transportation/.

1. IvyPanda . "Government Intervention in Intermodal Transportation." January 6, 2023. https://ivypanda.com/essays/government-intervention-in-intermodal-transportation/.

Bibliography

IvyPanda . "Government Intervention in Intermodal Transportation." January 6, 2023. https://ivypanda.com/essays/government-intervention-in-intermodal-transportation/.

• Intermodal Transport: Comparison Between USA and Europe
• Regulation and Its Impact on Intermodal Transportation
• Intermodal Transportation Impacts on Environment
• Containerization of Intermodal Transport Service
• Intermodal Transport and Drayage Operations
• Intermodal Transportation Systems and Communication Technologies
• Sociotechnical System in an Organization
• Intermodal Transportation: Overview
• The Advantages of Intermodal Transportation in the Global Economy
• Intermodal Transportation in the European Union and the United States
• New Technology Pros and Cons
• Key Performance Indicators in Inventory Management in Logistics
• The Matters Regulated by the Hague-Visby Rules and the Rotterdam Rules: Air Cargo
• The Houston Galveston Regional Transportation Plan
• The Causes of Congestion in Transportation

Impact of government interventions on the stock market during COVID-19: a case study in Indonesia

• Original Article
• Open access
• Published: 17 August 2022
• Volume 2 , article number  136 , ( 2022 )

Cite this article

You have full access to this open access article

• Josua Sinaga 1 ,
• Ting Wu 1 &
• Yu-wang Chen   ORCID: orcid.org/0000-0002-2007-1821 1  

6961 Accesses

2 Citations

Explore all metrics

This paper aims to examine the short-term impact of government interventions on 11 industrial sectors in the Indonesian Stock Exchange (IDX) during the COVID-19 pandemic. Whereas earlier studies have widely investigated the impact of government interventions on the financial markets during the pandemic, there is lack of research on analysing the financial impacts of various interventions in different industrial sectors, particularly in Indonesia. In this research, five key types of government interventions are selected amid the pandemic from March 2020 to July 2021, including economic stimulus packages, jobs creation law, Jakarta lockdowns, Ramadan travel restrictions, and free vaccination campaign. Based on an event study methodology, the research reveals that the first economic stimulus package was critical in reviving most sectors following the announcement of the first COVID-19 case in Indonesia. Jakarta lockdowns impacted stock returns negatively in most sectors, but the impacts were relatively insignificant in comparison to other countries in the region. The recurrence of lockdowns in Jakarta had a minor detrimental impact, showing that the market had acclimated to the new normal caused by the COVID-19 pandemic. Additionally, Ramadan travel restrictions caused minor negative impacts on the stock market. Furthermore, the second Ramadan travel restrictions generated a significant reaction from the technology sector. Finally, while free vaccination campaign and job creation law did not significantly boost the stock market, both are believed to result in a positive long-term effect on the country’s economy if appropriately executed. The findings are critical for investors, private companies, and governments to build on recovery action plans for major industrial sectors, allowing the stock market to bounce back quickly and efficiently. As this study limits its analysis to the short-term impact of individual interventions, future studies can examine long-term and combined effects of interventions which could also help policy makers to form effective portfolios of interventions in the event of a pandemic.

Similar content being viewed by others

Reaction of stock market returns to COVID-19 pandemic and lockdown policy: evidence from Nigerian firms stock returns

Isiaka Akande Raifu, Terver Theophilus Kumeka & Alarudeen Aminu

Covid-19 pandemic and stock returns in India

Munusamy Dharani, M. Kabir Hassan, … Mohammad Zoynul Abedin

The Impact of Covid-19 Pandemic on the Financial Performance Indicators for Palestinian Firms Listed at PEX

Avoid common mistakes on your manuscript.

Introduction

A contagious disease that started in December 2019, so-called COVID-19, had infected more than 197 million people globally, with 4.2 million deaths as of 31 July 2021 (WHO 2021 ). The virus attacks the human respiratory system, which leads to a more severe effect on people with underlying medical problems. There have been a few disease outbreaks since the beginning of the twenty-first century when a deadly Severe Acute Respiratory System (SARS) virus tore East Asian countries in 2002 with over 8000 cases. In addition, in early 2009, the United States discovered a new influenza outbreak called Swine Flu (H1N1), which lasted around 19 months and was estimated to cause between 105,000 and 395,000 deaths in over 214 countries. The COVID-19 pandemic is still incomparable to these outbreaks as it utterly changed the way of living, which appeared to cause a domino effect to the world’s economy. The pandemic led to an economic downturn indicated by most countries declaring a recession in the third quarter of 2020. In addition, the stock markets worldwide had significantly lost investor’s confidence due to the uncertainty during the COVID-19 pandemic, which was supported by past experiences (Fan 2003 ; Bloom et al. 2005 ). However, although an increase in new cases has shown a negative return in the stock market, each country has faced different impacts from the COVID-19 pandemic. Thus, despite the relatively low mortality rates, this pandemic has created a more significant financial crisis than any other extraordinary event in this century.

This study centres its discussion on the COVID-19 situation in Indonesia, which is selected to reach wider audience from relevant countries in terms of geographic and economic conditions. Geographically, Indonesia is the largest tropical archipelago in the world and it is located in the most populous continent (Cribb and Ford 2009 ; United Nations 2021 ). Furthermore, from an economic perspective, Indonesia is also the largest economy in Southeast Asia and the world's seventh largest economy by purchasing power (World Bank 2018 ). The fourth most populated country declared its first case on 2 March 2020 when the president announced that foreign citizens had transmitted the virus to a local citizen in Jakarta. The COVID-19 cases in Indonesia have proliferated for the past 16 months since March 2020, with 3.4 million cases that caused over 90,000 (WHO 2021 ). By the end of July 2021, Indonesia has the most active cases in the Southeast Asia region, with over 40,000 daily new cases as shown in Fig.  1 .

The daily COVID-19 cases in Indonesia from March 2020 to July 2021 (WHO 2021 )

This study uses the local stock market as the instrument to quantify the impact of the COVID-19 pandemic to the Indonesia’s economy. A stock market is used as it could be utilised as another vital economic indicator beside the country’s GDP. Masoud ( 2013 ) supported that the stock market has played a significant role in emerging countries like Indonesia. Hall ( 2020 ) also found that the market volatility trend was correlated negatively and significantly with real per capita GDP growth. Therefore, it is believed that any significant impact from government interventions could be detected from the stock market’s short-term movements.

Indonesia is selected also due to its relatively stable financial performance over any major indices in the Southeast Asia region as illustrated in Fig.  2 . Despite its rapid growth in the number of cases, there is an indication that the government interventions might have played a significant role to recover the stock market’s fall. Although it experienced a significant loss of around 20% in March 2020, the Indonesian Stock Exchange (IDX) has rebounded and performed relatively well at least until July 2021. Hence, the Indonesian Government’s approach during the COVID-19 pandemic could provide an exciting insight into the literature.

Monthly returns of Southeast Asia’s major indices from July 2019 to July 2021 (Reuters 2021 )

The challenging situation has led the Indonesian Government to implement a series of non-pharmaceutical interventions, such as social distancing, travel restriction, and lockdown. In addition, several economic policies, including debt relief, electricity subsidy, tax exemption, and microbusiness support, have also been implemented as it is critical to keep the investors’ positive sentiment, as the country targets an optimistic 5.6–6.2% average GDP growth by 2024 (Reuters 2020 ). Furthermore, the president had also used political interventions by restructuring his cabinet to strengthen the health and economy ministers to gain more trust from the public.

To understand the effectiveness of the Indonesian government interventions, this study uses an event study methodology to quantify the possibility of short-term indirect impact on each individual sector in the Indonesian Stock Exchange (IDX). Additionally, an event study can analyse numerous similar events to forecast how stock values in a certain sector normally react to a particular event. The methodology has also been applied extensively on both firm specific and industry level in analysing the short-term impact of corporate-related news, financial crisis, marketing strategy, natural disasters and disease outbreaks.

The study contributes to the literature by providing insights into the short-term impact of government interventions during a pandemic. The findings are expected to help investors, regulators, and government understand the short-term impact of government interventions on each industry sector. The insights could also help them create better policies and decisions to avoid a severe economic impact on any future pandemics. The remainder of the study is organised as follows: Sect.  2 reviews the literature related to the research topic, while Sect.  3 discusses the methodology and data collection process for event study analysis. Section  4 presents the results and findings from the analysis. Last, Sect.  5 concludes the study.

Literature review

Impact of pandemics on the economy.

Many studies have investigated the economic impact of pandemics on regional and worldwide levels. Most of the literature indicated that a pandemic would have a short-term impact on the global economy, with a longer duration for countries identified as the pandemic's epicentres. SARS, the first pandemic of the twenty-first century, began in Hong Kong and infected over 8000 individuals worldwide in 18 months, with a 10% fatality rate. Mackellar ( 2007 ) calculated that a relatively small outbreak triggered a 1% drop in China’s GDP and a 0.5% drop in Southeast Asia’s GDP, resulting in a substantial loss of US$ 30 billion. Although Sánchez and Liborio ( 2012 ) determined that a relatively insignificant decrease in GDP would not immediately cause a dramatic increase in unemployment rates, they argued that a recession could be incurred with a likelihood of 33%. H1N1 was initially reported in the United States in the spring of 2009, with over 400, 000 confirmed cases by the end of the year (CDC 2019a ). The economic impact was difficult to quantify as the economy was still dealing with the effects of the 2008 Global Financial Crisis. However, Barua ( 2020 ) suggested that a clear enlightenment might be gained from the tourism business, and they found that Asia’s tourism sector was the quickest to recover from the pandemic by reviving tourist arrivals in the second half of 2009. In addition, Rassy and Smith ( 2013 ) analysed the effect on Mexico, which was one of the pandemic’s epicentres, estimated US$ 2.8 billion losses from the tourism industry given the intensity of the H1N1 Kim et al. ( 2012 ) estimated that the indirect cost from a pandemic could be 5–10 times more than the direct cost. The 2014 Ebola pandemic was 11 times larger than the previous Ebola outbreaks combined, with over 11,000 deaths by the end of 2016 (CDC 2019b ). The outbreak was centred in the West African region, affecting the following three nations in particular: Guinea, Liberia, and Sierra Leone. According to the World Bank report ( 2016 ), the three countries lost a total of US$ 2.8 billion. Furthermore, the report also stated that Sierra Leone’s private sector had lost half of its workers, while the unemployment rate in Liberia had climbed by 40%. Interestingly, Ighobor ( 2014 ) found that neighbouring countries with no confirmed cases also faced a substantial decline of their GDP by at least 1%, suggesting that tourist and transportation sectors had spillover effects. According to Omoleke et al. ( 2016 ), high death rates have resulted in an unprecedented number of persons being pulled from the labour market, resulting in a drastic fall in public consumption.

In comparison to the above-mentioned pandemics, the COVID-19 pandemic has a greater impact. The IMF (International Monetary Fund 2020 ) reported that the cumulative GDP loss could be around US$ 9 trillion in 2020, with US$ 4 trillion solely contributed from the tourism industry. Additionally, emerging economies were significantly damaged as many declared recessions. Moreover, Jawaid and Garrido ( 2020 ) discovered that in the third quarter of 2020, 31 developed economies with a GDP over US$ 200 billion were in a recession. Rose ( 2021 ) also predicted that the economic impact of the COVID-19 pandemic would likely to be 30–50 times greater than the 9/11 tragedy.

Stock market reaction to the COVID-19 pandemic

The adverse reaction in the global stock markets started when WHO declared COVID-19 as a global pandemic, and many governments started to declare its first cases (more details in Appendix A). Most agreed that stock markets underreacted as many showed positive returns within the first few days but suddenly plummeted to their lowest level in the decade. Additionally, firms were showing different reactions to the COVID-19 as SMEs showed significant negative returns while large firms still survived with slight positive returns in the United States (Harjoto Rossi and Paglia 2020). The stock markets’ delayed response might have two possible explanations as follows: different market efficiencies in different countries and the investors’ confidence in the government to deal with the pandemic (Khatatbeh et al. 2020 ).

The countries with more complex COVID-19 situations had proven to have higher volatility in the stock market. Harjoto et al. ( 2020a , b ) found that emerging markets tend to be more volatile than the developed market with 1% increase in daily cases and deaths caused a 2.37% and 14.94% volatility in the stock market. Fu et al. ( 2021 ) examined the global stock markets and concluded that South America was highly exposed to the contagion economic risk, while Asia experienced the most negligible severe risks in the same period. Their study also indicated that the panic caused by the uncertainties led to a severe drop in investor confidence, resulted in a nose-dive response in the global stock market. Gupta et al. ( 2021 ) strengthened the claim that the COVID-19 pandemic has negatively impacted all major stock markets in the short term. According to Reuters ( 2021 ), the monthly returns for major indices, including Dow Jones, FTSE 100, Euro Stoxx 50, Shanghai and ASX 200 in Fig.  3 showed a relatively stable return before the COVID-19 outbreak but declined sharply from February 2020 to April 2020. Australia’s ASX-200 was one of the most affected indices at the beginning of the pandemic, with a negative 20% return in March 2020. On the other hand, China’s Shanghai index showed a relatively stable return as the country successfully controlled the disease’s transmission rate.

The monthly returns of global indices from July 2019 to July 2021 (Reuters 2021 )

It was discussed that the increase in investor herding behaviour during the COVID-19 epidemic might have contributed to stock market volatility. In general, herding behaviour aggravates fluctuations, resulting in inefficiencies in the stock markets (Blasco et al. 2012 ). In research of 49 countries around the world, Bouri et al. ( 2021 ) discovered a strong connection between herding behaviour and stock market uncertainty, and they discussed that emerging markets exhibit more herding behaviour as a result of more volatility, which is consistent with prior research. Chong et al. ( 2016 ) concluded that companies with a high turnover ratio and systematic risk would be swiftly exposed to herding behaviour in their study of China's stock market. In addition, analyst recommendation is one of the factors that played a vital role in causing herding behaviour.

Government interventions during the COVID-19 pandemic

Disease outbreaks, including the rapid spread of COVID-19, have caused severe human and economic costs to any country globally. Governments were forced to implement strict measures to limit these costs in the short and longer-term. Several studies showed the importance of enforcing the interventions that could change people’s behaviour, which eventually would help reduce the COVID-19 transmission rate, as mentioned in Cowling et al. ( 2020 ) study in Hong Kong. Non-pharmaceutical measures, according to many studies, effectively reduced the number of cases reproduced; however, pharmaceutical interventions, such as immunisation, were critical in controlling the COVID-19 pandemic.

We reviewed the impact of the four widely implemented non-pharmaceutical interventions, including contact tracing, social distancing, lockdown, and border restriction in Appendix B. Koh, Naing and Wong (2020) stated that these interventions were implemented by over 142 countries before the 100th COVID-19 cases identified in each country due to their effectiveness in limiting people’s movement and hence reducing the transmission rate of the infectious disease. Governments implemented lockdown as one of the earliest interventions. Lockdown is widely regarded as an effective tool to restrict the transmission rate of infectious diseases, though any lockdown over 120 days is considered ineffective (Koh et al. 2020 ). Goldstein et al. ( 2021 ) suggested that countries should not impose a blind national lockdown which could place the underprivileged people at high risk of unemployment. Instead, a short and strict lockdown would be ideal for emerging countries with a data-driven decision on hospital system situations. When combined with quarantine policies, contact tracking proved to be effective. However, the implementation could be costly for emerging countries, while data protection could be a complicated challenge to overcome by Western countries.

Furthermore, Adekunle et al. ( 2020 ) argued that the impact of border restrictions was largely reliant on the underlying situation of the country. Steyn et al. ( 2021 ) suggested that Australia and New Zealand had the strongest border restrictions, which were seen to be crucial during the early epidemic, delaying the pandemic by four weeks and giving the government more time to prepare. On the other hand, African countries that enforced border restrictions witnessed an increase in new cases due to a lack of official support to ensure the lives of the people (Emeto et al. 2021 ). In fact, social distancing was found to be the most effective non-pharmaceutical intervention by significantly reducing the transmission rate by 25% (Li et al. 2020 ).

Impact of government interventions on the stock markets

Non-pharmaceutical measures were also applied to reduce COVID-19 transmission. Governments implemented a variety of economic recovery interventions. In doing so, governments previously had enforced direct and indirect stock market interventions. During a crisis, direct interventions had varying results. During the Asian Financial Crisis of 1998, Hong Kong allocated US$15 billion to acquire the Hang Seng Index’s 33 stocks. A positive abnormal return for at least 30 days was found to restore investor confidence successfully, with a short-term spillover impact to other stocks (Su et al. 2001 ). During the 2008 Financial Crisis, Russia injected the banking system with US$ 150 billion. As a result, the market overreacted on the intervention day, resulting in a significant negative return. Therefore, many governments selected the safer approach with indirect interventions, which had succeeded in many developed countries (Swaine 2008 ; Murphy 2008 ). Many studies found that social distancing and lockdown had a short-term negative influence on markets.

We reviewed the impact of three non-pharmaceutical interventions, including lockdown, gathering restrictions, and economic support on the stock markets in Appendix C. Stock markets across the world had plummeted as a result of the lockdown’s implementation. When local governments announced the lockdown, both developed and emerging markets overreacted. Lockdown also had a spillover impact on interconnected countries, as several economies experienced a brief downturn when their neighbours went into lockdown (Eleftheriou and Patsoulis 2020 ). The cancellation of public events was proven to be the most impactful restriction in causing excessive volatility in the global stock markets (Zaremba et al. 2020 ). Gathering restrictions were also found to cause high volatility in the global stock markets, with the cancellation of public events believed to be the most impactful restriction (Zaremba et al. 2020 ). The restriction also significantly reduced the illiquidity situation in America, Europe, and the Middle East emerging economies. However, Asian emerging economies showed no impact on the market’s liquidity (Haroon and Rizvi 2020 ). Last, economic supports were insignificant in helping the recovery of stock markets. However, several studies concluded that these interventions directly targeted households and not corporations, which generated a relatively small indirect impact. Besides, monetary policies and fiscal policies were impactful in helping stock markets to rebound in all continents. Asian emerging markets were found to have a spillover impact from developed countries’ quantitative easing policies, contributing to an 8% surge on average (Beirne et al. 2021 ).

Data and methodology

Data collection, stock market.

The Indonesia Stock Exchange (IDX) was established in 2007 and has shown significant development during the past decade. As of 31 July 2021, the stock market has 746 listed companies, an approximately 50% increase since 2014. However, during the COVID-19 pandemic, the Jakarta Composite Index (JKSE) has reached its 7-year low in March 2020. Nevertheless, JKSE recovered towards the end of 2020 with only a 4.8% loss and outperformed other major indexes in the Southeast Asia region (Maulia 2021 ).

As a sample for the analysis, we selected the industry sector leaders based on the market capitalisation before the COVID-19 pandemic began. Indonesia, as an emerging market, has a different characteristic from the developed market. As illustrated in Table 1 , the financial sector still dominates and has become most of the market in Indonesia, where the technology sector has been on the top list for decades in developed countries like the United States. The stock market data used in this study is considered as a secondary dataset obtained from the Yahoo Finance database from 1 January 2020 to 31 July 2021 to analyse the impact of government interventions selected for this study.

• Government interventions

This study selected nine events between March 2020 and July 2021, including two events of economic stimulus packages, one event of jobs creation law, three events of Jakarta lockdowns, two events of Ramadan travel restrictions, and one instance of a free vaccination campaign. The timing for each intervention's announcement is highlighted in Fig.  4 . The selection of the nine events is justified in the following sub-sections.

The timeline of five Indonesian government interventions for the analysis

Economic stimulus packages

The Indonesian president unveiled the economic stimulus package on 24 March 2020, and it was the first economic policy responded to the COVID-19 pandemic. According to his statement, the economic stimulus package was designed to assist corporate firms in surviving and maintaining people’s purchasing power. The package includes tax breaks for all industries, credit relief for small businesses, and an increase in the amount accessible to Staple Food Card recipients. A total of nine incentives were implemented, resulting in a rise of IDR 405 trillion in the state budget, which is equivalent to US$ 27 billion (Gorbiano and Akhlas, 2020 ).

This intervention was chosen for various reasons, including the fact that it was the Indonesian government’s first and largest economic intervention during the COVID-19 pandemic. The economic stimulus package also resulted in a substantial increase to the state budget for 2020, which was in place until December 2020. It was believed that the nine incentives provided possibly had an indirect impact on various business sectors, which might be the turning point for stock market downturns. The government had also decided to extend several incentives from the 2020 economic stimulus package into 2021, which was announced by the president on 4 January 2021. The 2021 package focused more on small businesses, impoverished families, and unemployed citizens with a total of IDR 110 trillion allocated, which is reduced than the previous year.

Jobs creation law

The global unemployment rates were soaring at the beginning of the COVID-19 pandemic caused by cost-optimisation by most businesses. Indonesia has also experienced a sudden increase within a few months after the pandemic. Figure  5 shows the sharp movement between the first and second quarters of 2020, indicating that at least three million people lost their jobs throughout the period.

The unemployment rates in Indonesia from February 2018 to February 2021 (BPS Statistics Indonesia 2020 )

During the challenging times, the Indonesian parliament decided to authorise a new law to simplify investment regulations called Omnibus Law (Jennings 2020 ). The complexity of foreign investments in Indonesia has been a long-standing issue, as The World Bank ( 2020 ) highlighted the country’s rigid investment regulations as a factor limiting growth. It is also reflected from the 2020 World Bank’s Ease of Business Index that put Indonesia in 73rd place among 190 countries, far behind other Southeast Asia countries. Oxford Business Group ( 2020 ) also suggested that the law could provide jobs for six million people who have been left unemployed during the COVID-19 pandemic.

This intervention was selected as one of the most vital and controversial interventions during the COVID-19 pandemic. This is because the international world looked at the law as a brighter future for the Indonesian industries, while the domestic workforce strongly opposed the law due to a weakening in job security. Nevertheless, the Omnibus Law was a concrete step for Indonesia’s investments regulations, which could positively impact the country’s economy during the COVID-19 pandemic.

Jakarta lockdowns

Jakarta is the Indonesian capital, categorised as a unique province among 33 other provinces in the country. The capital is the most populated city in the Southeast Asia region, with over 10.57 million population at the beginning of 2020 (BPS Statistics Indonesia 2020 ). In addition, Jakarta is also the headquarters of big companies and where the Indonesia Stock Exchange sits.

Jakarta’s population density is extremely high, over one hundred times the country’s average, with 16,704 people per Km 2 in 2020 (BPS Statistics Jakarta 2020 ). Due to the high population density, the COVID-19 cases’ growth in Jakarta was the fastest among other provinces. Figure  6 illustrates the significant difference in the number of COVID-19 cases in Jakarta compared to the five most populated provinces. The six provinces had around 63% of the total cases in Indonesia at the end of 2020, implying that high-population provinces were the pandemic’s epicentres in Indonesia.

The number of COVID-19 cases in six provinces in Indonesia (Indonesia COVID- 19 Response Acceleration Task Force 2021 )

This has forced the Indonesian government to impose a regional-level lockdown instead of a national lockdown like other high-population countries like China and Brazil. Jakarta was the first province to impose the lockdown when the governor announced 14 days of large-scale social restrictions on 7 April 2020 when the number of daily new cases surpassed one hundred for two consecutive days (BPS Statistics Jakarta 2020 ). This was followed by two other lockdowns in September 2020 and July 2021 mainly to avoid public health collapse due to the increase in the new cases.

This intervention was selected for several reasons, considering that Jakarta is the most populated city in Indonesia and headquarter of most businesses in the country. Although Jakarta is the smallest province geographically, it represents the biggest contributor to the country’s GDP with 17.67% in the second quarter of 2020. This indicates that any extreme measurements in Jakarta could impact the country’s economic condition.

Ramadan travel restrictions

Indonesia has the largest Muslim population, accounting for approximately 12% of the global Muslim population. The country’s biggest annual event occurs during Ramadan month when tens of millions of people travel to their hometown; a tradition locally called Mudik, like Christmas in western countries or the Chinese New Year. This event generally impacts the country’s economy, as people tend to spend more due to the compulsory Ramadan bonus granted to all employees. Moreover, Muslims must pay alms during Ramadan, which creates a massive surge in the money circulation during the season. Hence, the consumer sector typically significantly impacts Ramadan, with an approximately 30% increase in sales (Halimatussadiah 2015 ).

However, Ramadan travel during the COVID-19 pandemic could result in a massive surge in the new cases as millions of people would travel in any transportation mode. For example, 23 million people travelled domestically during Ramadan in 2019 (Wight 2020 ). To prevent the disaster, the Indonesian government temporarily banned domestic flights, busses, and ferries for at least 14 days before and after Ramadan Day. These restrictions were imposed in both 2020 and 2021, which have proven to reduce the transmission rate. However, the economic impact was unclear to the country’s economic condition. Hence, this intervention was selected as it would be insightful to explore the economic impact created by the restrictions.

Free vaccination campaign

The Indonesian journey with vaccination finally started when the first batch of Sinovac vaccines arrived in Jakarta on 7 December 2020. A week later, President Widodo announced to provide free Covid-19 vaccines, which plays a vital role in other countries to recover their economic condition quickly. For example, the vaccination policy and people’s willingness to get vaccinated gave strong sentiments from the US Stock Market, which is a good indicator of economic healing (US Bank Asset Management Group 2021 ). In other words, vaccines could help the government to have significant economic growth as people’s movement would be less limited.

Several studies have also found a reasonable hope that vaccines should help the world stop the pandemic. For example, Powell ( 2021 ) concluded that vaccines are created to establish herd immunity, which can be achieved when 50–60% population is vaccinated regardless of the virus mutations, as it would still have the same structure. Therefore, this intervention was selected as it might be the booster for the stock market to gain more trust from the investors and started to grow strongly towards 2021.

Methodology

The study used the event study methodology, which has been widely utilised to assess the valuation impacts of extraordinary corporate actions (more details in Appendix D). In addition, event study has revealed vital information about how an industrial sector is likely to react in a short-term to a given extraordinary event, such as natural disaster, disease outbreaks, and geopolitical issues. A short-term analysis is critical for the stock market since an extraordinary event might alter investor behaviour and the entire market environment.

The approach examines the stock price’s response around the announcement by looking at the stock returns first introduced by Dolley ( 1933 ). Moreover, event study with known event dates has a relatively statistical solid power to support the result, which is required to understand the short-term impact during unprecedented events like the COVID-19 pandemic. Although Dyckman et al. ( 1984 ) pointed out several problems of daily return analysis, such as nonsynchronous trading and biased estimation, Brown and Warner ( 1985 ) concluded that the potential problems with daily returns are unimportant easily corrected in the standard event study. Figure  7 illustrates the process taken by this study to perform the event study analysis, and the key steps are described in the following subsections.

The process diagram for event study analysis

Three event study models were commonly used in previous studies, namely constant mean return, market model, and capital asset pricing model (CAPM). The main difference among these models lies in the way of calculating the expected returns and abnormal returns. The constant mean return model has the simplest from by subtracting the stock return with the simple mean return, but Brown and Warner ( 1985 ) criticised that the method does not consider the abnormal return in reflection to the stock market condition. The market model and CAPM are the most popular yet similar methods in practice. The difference is that the CAPM imposes an additional restriction (e.g., intercept equals the risk-free rate). Due to the added restriction, the variance of error terms in CAPM is generally more significant than the market model (MacKinlay 1997). Consequently, a significant variance of error leads to a less powerful test for the result than the market model. This study uses the seminal market model introduced by Scholes and Williams ( 1977 ).

There are two crucial parameters that determine the analysis’s outcome in the event study, namely the estimation window and the event window. This study used 42 days of the estimation window, equivalent to 2 months of trading days, as the events analysed happen within a short period of 16 months. Furthermore, we used an event window of 7 days, which centres symmetrically around the event day, as shown in Fig.  8 , and a similar approach has been taken by previous literature (Bash and Alsaifi 2019 ; Buigut and Kapar 2020 ). A short event window is also applied to prevent overlapping event window periods, as some events in the analysis are only a few weeks away. Further, the small period within the event window, the anticipation window and the adjustment window are used to capture the short-term abnormal returns before and after the event day. In contrast, a long event window could reduce the statistical power of analysis but suffer from essential limitations (Brown and Warner 1985 ).

Illustration of the event window timeline used in the paper

Estimation of expected return

The market model assumes that the asset returns are given by the following:

where \({R}_{i,t}\) represents the return for each company i on day t, which belongs to the estimation window, while the expected return is established as follows:

The \({R}_{m,t}\) represents the market portfolio’s return, and the linear specification of the model arises from the assumed joint normality of returns. The market portfolio used is the Jakarta Composite Index (JKSE), the composite index for IDX. The market model also assumes that \({\varepsilon }_{i,t}\) changes related to the return on the market portfolio iare removed as follows:

Estimation of abnormal return

This study used Buy and Hold Abnormal Returns (BHAR), which employs a geometric method to calculate abnormal returns. The BHAR approach was used as several economists such as Ritter ( 1991 ) and Lyons ( 1999 ) have argued that CAR is not appealing from the economic perspective. The CAR approach could lead to biases due to the continuous compound rate of appreciation. The calculation of BHAR is established as follows:

where i represents each company in the analysis, t represents the event window start date, and k represents the duration of the event window. Furthermore, the calculation of Abnormal Return (AR) is as follows:

Test procedure

The parametric test is the only approach to test the null hypothesis for the event study that analyses multiple individual events’ impact (Boehmer 1991 ). We used a statistical test to determine whether enough evidence exists to reject a hypothesis about the process. The following hypothesis testing is adapted to the parametric test:

The statistical test is formulated as follows:

The \({t}_{\mathrm{BHAR}}\) represents the t -score of the BHAR for each company at different event windows and \({\sigma }_{BHAR}\) is the standard deviation of the BHAR for the estimation window.

Implementation

This paper used an analytical approach to answer the research questions, mainly for the event study analysis. Open-source packages in Python are used to automate the data extraction process, calculate the daily returns, estimate the abnormal returns, and validate the results.

Results and discussion

Preliminary analysis.

To demonstrate the unique circumstances surrounding the COVID-19 pandemic, descriptive statistics for the years 2019–2021 are presented for comparison. A majority of industries experienced a fall in the average stock price, as demonstrated by the Jakarta Composite Index (JKSE), which experienced a roughly 16% decline in the average index price in 2020 (Tables 2 and 3 ). However, two of the eleven sectors, basic material and property, experienced a significant increase in the average stock price, nearly doubling the previous year.

Table 4 summarises the statistics data for 2021, which is considerably different from the previous year. The market has generally recovered, with an average index price increase of 16% until July 2021. Ten of the eleven industries also had strong growth, except for the property sector, which experienced a substantial decline. Although there are still a few months remaining before the end of 2021, these figures indicate that the market has begun to recover, despite Indonesia continuing to have the most cases in Southeast Asia as of 31 July 2021.

The range is a crude measure of the spread in stock prices, and it indicates that the range for the majority of sector leaders increased significantly in 2020 compared to 2019. The range, however, is subject to outliers that occur under an extreme COVID-19 situation. In comparison, the standard deviation is a more accurate tool for detecting outliers in a normal distribution. This method showed a substantial increase from 2019, indicating that the stock market saw much higher volatility returns as the dispersion of company prices compared to their average increased dramatically. Additionally, it implies that the stock market became a riskier investment during the COVID-19 pandemic in 2020.

On the other hand, both measures have been significantly reduced for the first half of 2021. The composite index’s range and standard deviation have decreased by 72% compared to the previous year. Additionally, nine of eleven industry sectors followed the movement of the composite index. This could indicate that the market has grown less risky and has rebounded from its early 2020 collapse. Furthermore, low volatility attracts additional investors, signalling the prospect of significant growth during the pandemic.

Hale et al. ( 2021 ) developed a government intervention tracer that covers 23 different types of government interventions, including containment and closure policies, economic policies, health system policies, and vaccination policies. In this research, we focussed primarily on the following two indices: overall government response and stringency. We compared and analysed the daily global average government response index to the Indonesian daily index in Fig.  9 . At the start of the pandemic, the Indonesian government responded more adequately than the global average, implementing at least three critical interventions. However, this condition did not last long, as it continued below the global average for the remainder of 2020. In contrast, the Indonesian government has responded more positively from the beginning of 2021, whereas the global average has declined since May 2021.

The comparison of Government Overall Index between Indonesia and the Global average

The stringency index provided a similar trend to the government response index in Fig.  10 . Global average stringency increased rapidly from March to May 2020 but then stabilised until early 2021. Henceforth, the worldwide average has been close to or below the index level of 60. In comparison, Indonesia showed its most stringent condition in May 2020, when the government planned to suspend all modes of public transportation for more than 14 days, with an index was above 80. Although the trend had been downward for several months, it started to rise again in September due to Jakarta's second lockdown. Thus, throughout the COVID-19 pandemic, it is safe to say that Indonesia had implemented tighter restrictions than the global average.

The comparison of Stringency Index between Indonesia and the Global average

Event study analysis

The results for each intervention type are presented and discussed in the following sub-sections, with the individual parametric test results presented in Appendix E.

The impact made by the first economic stimulus package was believed to be highly significant for several sectors (Table 5 ). It is evident that, on the announcement day, the stock prices for financial, consumer non-cyclical, and consumer cyclical sectors were significantly improved. At the same time, infrastructure and healthcare were also increased to a certain extent. In contrast, significant adverse reactions were shown by the stock prices of basic material and property sectors.

It is believed that the financial sector’s reaction was due to the inclusion of financial system stability policies in the package. The policies grant authority to five vital government bodies to establish steps on handling financial stability matters by formulating government support such as short-term liquidity loans and financing on the sharia principle to all financial institutions (Molina and Ramadhan 2020 ). The stability policies were implemented to avoid a serious banking crisis in 1998 when half of the private banks collapsed due to the government’s unpreparedness in handling a crisis (Fane and McLeod 2002 ).

Moreover, additional funds for staple food card beneficiaries and pre-employment card holders significantly impacted both consumer sectors’ stock prices. According to Fang (2021), additional funds could stimulate the subsidy by the government to generate more consumption. Liu et al. ( 2020 ) strengthened the theory by finding that a consumption coupon of RMB 1 can drive excess spending of RMB 3.4 to RMB 5.8 at the beginning of the pandemic in China. Thus, these findings could be an essential variable that increased the market’s confidence in both consumers sectors on the announcement day.

Additionally, the property sector signalled an under-reaction but rebounded with a positive 16.50% abnormal return during the adjustment window. The under-reaction to stock-related news could be due to an anchoring bias or slow information diffusion (Lansdorp and Jellema 2013 ). On the other hanfd, basic materials, industrials, and technology significantly suffered. We believed that the corporate tax reduction for 2020 and 2021 is the only policy that directly impacts these sectors.

The impact made by the second economic stimulus package on stock markets was not significant compared to the first one (Table 6 ). The announcement day has resulted in diverse reactions, where six sectors reacted negatively and five sectors reacted positively. The second economy package was different as it was more focused on income support and debt relief. Ashraf ( 2020 ) found that the insignificant impact caused by the package was only directed to households and did not directly impact corporations. Additionally, the results showed that consumer non-cyclical overreacted on the announcement day, followed by a significant negative abnormal return of 3.10% in the subsequent days, while basic material showed an under-reaction with a significant increase of 4.95% to the stock prices in the same period. Besides, basic material showed an underreaction with a significant increase of 4.95% to the stock prices in the same period. Although the impact was not significant, we believed that the announcement was still made at the right time. The market could lead to high volatility if the government did not announce the continuation of several economic stimulus at the beginning of 2021.

In addition, we also believed that the Indonesian government had also successfully maximised its capabilities. Table 7 shows the percentage of total COVID-19 economic support to the respective country’s GDP in 2020. Indonesia only spent 2.6% of its GDP to support the economy during the COVID-19 pandemic, which was only higher than China and Taiwan. From the relatively low budget, the Indonesian government successfully targeted the support’s recipients, which reflected by the stock market reaction. Moreover, a low cost of living could also contribute heavily to the economic stimulus package. As an illustration, the average meal price in Jakarta is US$2.51 compared to Singapore and Hong Kong, with US$9.84 and US$7.70, respectively (Numbeo 2021 ). Therefore, Singapore and Hong Kong must have provided direct household support over US$1000 a month for vulnerable groups. In comparison, the Indonesian government only provided additional support of less than US$100 a month.

The Authorisation of the Omnibus Law elicited negative sentiments from most Indonesia’s sectors, with eight out of eleven sectors reacting negatively on the day of the announcement (Table 8 ). By comparison, the infrastructure and financial sectors generated strong positive abnormal returns of 4.17% and 1.92%, respectively. Moreover, under-reaction was detected as the stock price for the transportation sector rose significantly by 19.31% in the subsequent days after the announcement. Overall, most sectors suffered a short-term negative impact during the event window, where the healthcare sector plunged significantly by 9.07% and transportation rose strongly by 17.15%.

The early reaction of the financial and infrastructure sectors was expected, given our belief that the Omnibus Law would directly influence those sectors. For instance, the financial sector might save high operational costs because of worker protections. On the other hand, enterprises involved in telecommunications infrastructure may benefit from the law as it enables them to share their infrastructure, generating enormous synergy and accelerating the sector’s growth.

Despite successfully lowering unemployment rates in the hope of accelerating economic growth during COVID-19, the stock market reacted unexpectedly. Before the event, various Non-Governmental Organisations (NGOs) publicly criticised the law on social media platforms and on national televisions. Valle-Cruz et al. ( 2022 ) proved that social media transmission via Twitter directly affected the indices' behaviour, particularly in Indonesia, which has the fourth-largest Twitter user base in July 2021 with 15.7 million active users. Additionally, they determined that the drop in market values during the COVID-19 pandemic was more severe than during the H1N1 pandemic, owing to the abundance of speculation and rumours about the virus. Furthermore, thousands of people protested in October 2020 to express their dissatisfaction with the new law, which could temper market enthusiasm, as happened in the United States when the government’s proposal was encountered with widespread scepticism on its implementation (Randall 2021 ).

While the Omnibus Law was insignificant in the short term, we believed that the new laws would positively affect several sectors over time. The property sector should be a clear winner, as the Omnibus Law simplified the land permission process, which had been a major obstacle for decades. The simplified foreign property ownership restrictions would also entice many international investors to invest in the Indonesian market. Additionally, the energy and basic material sectors would greatly benefit from a 0% royalty on value-added to raw materials. The industrial sector may also benefit from simplified foreign direct investment (FDI), which could boost the sector’s long-term growth. To summarise, the Omnibus Law represents a promising start for Indonesia’s future investment, while implementation would be critical.

Indonesia was the last major country in Southeast Asia to impose a lockdown. The announcement of Jakarta’s first lockdown appeared to cause adverse reactions from most sectors in Indonesia (Table 9 ). The stock market showed relatively normal returns on the announcement day of the first lockdown, with only the financial sector, reacting significantly positive. Financials and consumer non-cyclical sectors were underreacted on the announcement day as both plummeted significantly during the adjustment window. At the same time, properties continued to plunge significantly with a negative 14.45% of abnormal return. Overall, eight sectors suffered negative abnormal returns during the event window, with properties significantly suffered while basic materials and transportations were positively affected.

The first lockdown had such a minor detrimental impact that could be due to the Indonesian government’s delay in enforcing one. According to Ozili and Arun ( 2020 ), the first regional restriction had a more significant impact than the first national restriction. In other words, when the Philippines announced its first lockdown, the stock market in Indonesia may have already experienced a spillover effect. Furthermore, an introduction of lockdown measures has also been linked to undesirable reactions known as overreaction. Within a few days after the release, most industries showed overreaction, suggesting a delay in absorbing unusual news. Liew and Puah ( 2020 ) also verified that various industries experienced varied levels of lockdown based on their market conditions and business nature.

The property sector was also predicted to suffer substantial losses during the initial lockdown. This suggests that the debt-relief scheme included in the economic stimulus package was insufficient to convince the market. Furthermore, transportation sectors reacted negatively to the statement, which was expected given that logistic companies were permitted to operate regularly during the lockdown. Instead, the sector should have received more demands due to the drastic increase in transactions from online marketplaces, as offline stores were forced to halt its operations during lockdown.

The second lockdown in Jakarta elicited a range of responses from most sectors, with less severe adverse effects (Table 10 ). Though six sectors saw abnormal returns on the announcement, the impact was substantially negative for consumer non-cyclical and basic materials. Additionally, the stock market experienced under-reaction, as basic materials and technology stocks rose 5.75% and 21.89%, respectively, in the days following the announcement. The technology sector’s response was expected, as most enterprises, organisations, and schools must rely extensively on technology companies because of the lockdown restrictions. Additionally, the basic materials sector has recovered globally and was expected to be immune to COVID-19 by the second half of 2020 due to a resurgence in demand from China as the largest global importer (Barman 2020 ). Overall, the announcement resulted in five of eleven sectors suffering negative abnormal returns, while the remaining nine experienced positive abnormal returns.

The repetition of the same interventions appeared to have had a stabilising effect on the stock market, particularly proven by the third lockdown in Jakarta (Table 11 ). Our results aligned with Scherf et al. ( 2022 ), who concluded that multiple lockdown restrictions generally caused smaller negative returns than the first one. On the announcement day, consumer non-cyclical and property substantially impacted stock prices, gaining 5.78% and 23.79%, respectively. Additionally, a substantial under-reaction was detected in the consumer cyclical sector, which increased by 18.85% during the adjustment window. Six sectors, in aggregate, responded negatively to the announcement, with the infrastructure sector suffering the most, with an abnormally negative return of 8.43%. In comparison, five sectors experienced positive reactions, with consumer cyclical and property experienced considerable increases in stock prices of 12.42% and 35.87%, respectively.

China enforced its first nationwide travel restriction during the 2020 Lunar New Year. Huo and Qiu ( 2020 ) discovered a significant negative impact on the Chinese stock market during the period, with 22 out of 28 sectors experiencing negative abnormal results. Similarly, Indonesia enforced four-week nationwide travel restrictions during Ramadan in 2020 and 2021, but it was found to have a minor impact than China.

On the announcement day, only four sectors suffered a negative abnormal return, while the remainder experienced a positive abnormal return (Table 12 ). Additionally, throughout the anticipation window, a similar trend was seen with no substantial abnormal returns. However, the consumer cyclical sector underreacted on the announcement day, as indicated by a significant increase of 8.64% within the adjustment window, while properties underreacted by 11.43%. In general, the announcement had no discernible effect on any sector in Indonesia. Most sectors demonstrated insignificantly positive reaction, apart from the financial, energy, property, and transportation sectors, which suffer severe abnormal returns throughout the event window. Additionally, the overall positive returns demonstrated by both consumer sectors indicated that Ramadan spending was resilient to the COVID-19 pandemic.

We believed that information leakage was critical in preventing a short-term stock market collapse. For instance, a few days before the announcement, government officials talked about the possibility of travel restrictions in an open public forum. Journalists spread the rumours over multiple media resulting in a higher anticipated reaction from the stock market. Brunnermeier ( 2005 ) supported the notion that information leakage makes price processes more informative in the short run, indicating that information leakage frequently helped investors in managing expectations, hence stabilising the stock price in the short term. However, the study also discovered that information leakage might eventually diminish information efficiency. Additionally, the relatively positive responses from most sectors may reflect the market's confidence in the government's commitment to reduce the COVID-19 transmission rate.

A less severe impact because of subsequent measures is also demonstrated by the second nationwide travel ban in 2021 (Table 13 ). Only the technology sector reacted positively on the announcement, with an 11.64% gain in its stock price. Additionally, there were no substantial reactions before the announcement day since all sectors experienced divergent returns, which is believed to have been exacerbated by information leaks a few days before the announcement. However, the properties sector overreacted, with the stock price plunging by 11.01% following a slight positive abnormal return on the day of the announcement. Most sectors responded positively, with the technology sector yielding a remarkable 24.54% throughout the event window. On the other hand, the financial, energy, and transportation sectors all suffered declines, with the property sector suffering the most, with a decline of 22.01%.

Most sectors expressed support for the free vaccination campaign (Table 14 ). The reaction on the announcement day was moderate, with an expected significant increase in the healthcare sector’s stock price. Additionally, only the basic material and consumer cyclical sectors showed significant abnormal returns during the anticipation window, at − 6.23% and 7.03%, respectively. Moreover, a signal of underreaction was observed in the technology sector, with the stock price increasing significantly by 29.66% inside the adjustment window. In comparison, we concluded that the considerable reduction in basic material was unrelated to the intervention since the intervention was irrelevant to the sector. Overall, eight out of eleven sectors responded positively to the announcement, with the healthcare and technology sectors bearing the brunt of the impact.

The Indonesian stock market remained unaffected by the free vaccination campaign. The reaction was entirely contradictory for that in the United Kingdom and the United States, where the FTSE100 and Dow Jones both increased by around 5% on the day of the announcement. Additionally, in some cases, both countries have seen a substantial improvement in the stock prices of airlines, hotels, and energy companies, in some cases by more than 40% (Jack 2020 ).

Furthermore, Rouatbi et al. ( 2021 ) indicated that free vaccination campaigns had decreased global stock market volatility. According to the study, a 10% rise in vaccination might result in a 0.245% positive reaction in the stock market. However, adoption in emerging economies was regarded to be a hurdle to gaining investor confidence. A significant divide between emerging and developed countries could hamper economic growth, as The Economist ( 2021 ) estimated that emerging countries would not have widespread access to vaccines until 2023. Emerging economies such as Indonesia and India only have fully vaccinated rates of 11.2% and 9.2%, respectively. In contrast, developed economies such as the United Kingdom and the United States have already surpassed the 50% mark. As a result, the vaccination campaign is a critical factor in a country's economic recovery from the uncertainty during the COVID-19 pandemic.

Summary and conclusions

This paper examined the short-term impact of government interventions on the industrial sectors of the Indonesian stock market during the COVID-19 pandemic, and the analysis focused on the following five types of interventions: economic stimulus packages, job creation law, Jakarta lockdowns, Ramadan travel restrictions, and free vaccination campaign. The results from the proposed event study analysis indicated that the initial economic stimulus package was critical in reviving the stock market following its collapse to a 7-years low in March 2020. It was also observed that the combination of household and corporate support was the most powerful economic stimulus package. In contrast, the enactment of ****the jobs creation law ushered in a new era of hope for the Indonesian bureaucracy. Although the authorisation had a minor influence on several sectors, it was anticipated that the law would benefit the country’s economy in the future. Furthermore, the Jakarta lockdowns had no noticeable impact on any industrial sector in Indonesia. Indonesia was the last major country in the Southeast Asia region to impose lockdown, which indicated that the Indonesian stock market had experienced spillover impact from other countries’ announcements. Additionally, the September and July lockdowns were less severe than the initial one, showing that the COVID-19 situation affected investors’ behaviour, likely resulting in market resistance during the COVID-19 pandemic. Whereas Ramadan travel restrictions had historically resulted in significant negative sentiment in global stock markets, it was not the case in Indonesia and the announcement had no discernible impact on any industrial sector amid the pandemic. Additionally, the stock market was unaffected negatively by the recurrence of Ramadan travel restrictions in 2021. It was also anticipated that a free vaccination campaign would benefit the Indonesian healthcare sector in the short run. Additionally, the announcement was well received by most sectors. However, it was believed that emerging countries’ lack of access to vaccines could impede gaining market trust. As a result, the intervention’s long-term impact is likely to rely on the government’s commitment to distribute vaccines.

In conclusion, the Indonesian government took a relatively conservative strategy by enforcing a series of government interventions to reduce COVID-19 transmission rates while also stabilising stock market volatility. This was reflected in the stock market’s rapid recovery with a monthly return of 6.53% in December 2020, which was a higher return than other major countries in Southeast Asia, including Singapore (1.13%), Malaysia (4,13%), Thailand (2.91%), and Philippines (5.13%) (Yahoo Finance 2021 ). Presumably the government had gained knowledge from prior disease outbreaks such as SARS and H1N1, in which Indonesia was involved in the fight against the virus. In addition, Indonesia had also experienced two financial crises in the past 25 years, including the 1997 Asian Financial Crisis and the 2008 Global Financial Crisis, which helped the current government take steps to prevent another crisis.

The event study analysis used in the paper limited the focus to immediate and short-term analysis. However, it would be interesting to examine the longer-term impact of the interventions implemented by the government and further extend the market model to include investor behaviour and political factors in quantifying the short-term impact of recurrence events. Furthermore, due to many interventions implemented in a tight timeframe, an analysis of the impact of combined interventions could also help policy makers to gain better understanding on the formation of effective portfolios of interventions in the event of a pandemic.

Availability of data and material (data transparency)

Publicly accessible data used in the research.

Code availability

Available on request.

Adekunle A, Meehan M, Rojas-Alvarez D, Trauer J, McBryde E (2020) Delaying the COVID-19 epidemic in Australia: evaluating the effectiveness of international travel bans. Aust N Z J Public Health 44(4):257–259

Article   Google Scholar  

Ashraf BN (2020) Stock markets’ reaction to COVID-19: Cases or fatalities? Res Int Bus Financ 54(1):101249

Barman A (2020) 5 Basic materials stocks to play the sector’s strong rebound. Nasdaq. https://www.nasdaq.com/articles/5-basic-materials-stocks-to-play-the-sectors-strong-rebound-2020-09-24

Barua S (2020) Understanding Coronanomics: the economic implications of the coronavirus (COVID-19) Pandemic. Available at SSRN 3566477

Bash A, Alsaifi K (2019) Fear from uncertainty: an event study of Khashoggi and stock market returns. J Behav Exp Financ 23(2019):54–58

Beirne J, Renzhi N, Sugandi E, Volz U (2021) COVID -19, asset markets and capital flows. Pac Econ Rev 2021:1–41

Google Scholar  

Blasco N, Corredor P, Ferreruela S (2012) Market sentiment: a key factor of investors’ imitative behaviour. Account Finance 52(3):663–689

Bloom E, de Wit V and Carangal - San Jose MJ (2005) Potential economic impact of an avian flu pandemic on Asia. Asian Development Bank. https://www.adb.org/sites/default/files/publication/28082/pb042.pdf

Boehmer E (1991) Event-study methodology under conditions of event-induced variance. J Financ Econ 30(2):253–272

Bouri E, Demirer R, Gupta R, Nel J (2021) COVID-19 pandemic and investor herding in international stock markets. Risks 9(9):168

BPS Statistics Indonesia (2020) Statistical Yearbook of Indonesia 2020. BPS Statistics Indonesia. Jakarta: BPS Statistics Indonesia. https://www.bps.go.id/publication/2020/04/29/e9011b3155d45d70823c141f/statistik-indonesia-2020.html

BPS Statistics Jakarta (2020) DKI Jakarta province in figures. BPS Statistics Jakarta. Jakarta: BPS Statistics Jakarta. https://jakarta.bps.go.id/publication/2020/04/27/20f5a58abcb80a0ad2a88725/provinsi-dki-jakarta-dalam-angka-2020.html .

Brown SJ, Warner JB (1985) Using daily stock returns. J Financ Econ 14(1):3–31

Brunnermeier MK (2005) Information leakage and market efficiency. Rev Financ Stud 18(2):417–457

Buigut S, Kapar B (2020) Effect of Qatar diplomatic and economic isolation on GCC stock markets: an event study approach. Financ Res Lett 37(2020):101352

CDC (2019a). 2009 H1N1 Pandemic (H1N1pdm09 virus). Centers for Disease Control and Prevention. Available at: https://www.cdc.gov/flu/pandemic-resources/2009-h1n1-pandemic.html

CDC (2019b) Cost of the ebola epidemic. [online] CDC. https://www.cdc.gov/vhf/ebola/history/2014-2016-outbreak/cost-of-ebola.html

Chong TT-L, Liu X, Zhu C (2016) What explains herd behavior in the chinese stock market? SSRN Electron J 18(4):448–456

Cowling BJ, Ali ST, Ng TWY, Tsang TK, Li JCM, Fong MW, Liao Q, Kwan MY, Lee SL, Chiu SS, Wu JT, Wu P, Leung GM (2020) Impact assessment of non-pharmaceutical interventions against coronavirus disease 2019 and influenza in Hong Kong: an observational study. Lancet Public Health 5(5):e279–e288

Cribb RB, Ford M (eds) (2009) Indonesia beyond the water’s edge: managing an archipelagic state. Institute of Southeast Asian Studies, Singapore

Dolley JC (1933) Open market buying as a stimulant for the bond market. J Polit Econ 41(4):513–529

Dyckman T, Philbrick D, Stephan J (1984) A comparison of event study methodologies using daily stock returns: a simulation approach. J Account Res 22:1–30

Eleftheriou K and Patsoulis P (2020) COVID-19 lockdown intensity and stock market returns: a spatial econometrics approach. MPRA Paper, 100662. https://mpra.ub.uni-muenchen.de/100662/

Emeto TI, Alele FO, Ilesanmi OS (2021) Evaluation of the effect of border closure on COVID-19 incidence rates across nine African countries: an interrupted time series study. Trans R Soc Tropic Med Hygiene 115(10):1174–1183

Fan E (2003) SARS: Economic Impacts and Implications. ERD Policy Brief by Asian Development Bank. https://www.adb.org/publications/sars-economic-impacts-and-implications

Fane G, McLeod RH (2002) Banking Collapse and Restructuring in Indonesia, 1997–2001. 1997–2001. Cato J 22:277

Fu S, Liu C, Wei X (2021) Contagion in Global Stock Markets during the COVID-19 Crisis. Global Chall 2021(2000130):1–10

Goldstein P, Yeyati EL and Sartorio L (2021) Lockdown fatigue: the declining effectiveness of lockdowns. VoxEU.org. https://voxeu.org/article/declining-effectiveness-lockdowns

Gorbiano M and Akhlas A (2020) Indonesia announces Rp 405 trillion COVID-19 budget, anticipates 5% deficit in historic move. The Jakarta Post. https://www.thejakartapost.com/news/2020/03/31/indonesia-announces-rp-405-trillion-covid-19-budget-anticipates-5-deficit-in-historic-move.html . ( Accessed 2 Jul. 2021 )

Gupta H, Chaudhary R, Gupta S (2021) COVID-19 impact on major stock markets. FIIB Bus Rev 1(11):231971452199451

Hale T, Angrist N, Goldszmidt R, Kira B, Petherick A, Phillips T, Webster S, Cameron-Blake E, Hallas L, Majumdar S, Tatlow H (2021) A global panel database of pandemic policies (Oxford COVID-19 Government Response Tracker). Nat Hum Behav 5(4):529–538

Halimatussadiah A (2015) Why Ramadan is a special economic season in Indonesia. The Conversation. https://theconversation.com/why-ramadan-is-a-special-economic-season-in-indonesia-43399 . ( Accessed 2 Jul 2021 )

Hall M (2020) How the stock market affects GDP. Investopedia. https://www.investopedia.com/ask/answers/033015/how-does-stock-market-affect-gross-domestic-product-gdp.asp

Harjoto MA, Rossi F, Lee R, Sergi BS (2020a) How do equity markets react to COVID-19? Evidence from emerging and developed countries. J Econ Bus 1(1):105966

Harjoto MA, Rossi F, Paglia J (2020b) COVID-19: stock market reactions to the shock and the stimulus. SSRN Electron J 28(10):795–801

Haroon O, Rizvi SAR (2020) Flatten the curve and stock market liquidity—an inquiry into emerging economies. Emerg Mark Financ Trade 56(10):2151–2161

Huo X and Qiu Z (2020) How does China’s stock market react to the announcement of the COVID-19 pandemic lockdown? Economic and Political Studies, pp 1–26

Ighobor K (2014) Ebola threatens economic gains in affected countries | Africa Renewal. Un.org. https://www.un.org/africarenewal/magazine/december-2014/ebola-threatens-economic-gains-affected-countries

International Monetary Fund (IMF) (2020) World Economic Outlook: the great lockdown. International Monetary Fund, Washington

Indonesia COVID-19 Response Acceleration Task Force (2021) Daily cases in Indonesia. https://covid19.go.id/peta-sebaran

Indonesia Stock Exchange (2021) IDX Index fact sheet. PT Bursa Efek Indonesia. https://www.idx.co.id/en-us/market-data/statistical-reports/index-fact-sheet/

Jack S (2020) Covid-19: Global stock markets rocket on vaccine hopes. BBC News. 9 Nov. https://www.bbc.co.uk/news/business-54874108 . ( Accessed 10 Dec. 2020 )

James A, Plank MJ, Hendy S, Binny R, Lustig A, Steyn N, Nesdale A, Verrall A (2021) Successful contact tracing systems for COVID-19 rely on effective quarantine and isolation. PLoS One 16(6):e0252499

Jawaid A and Garrido F (2020) COVID-19 recession touches all 5 continents as recovery looks distant. www.spglobal.com . https://www.spglobal.com/marketintelligence/en/news-insights/latest-news-headlines/covid-19-recession-touches-all-5-continents-as-recovery-looks-distant-60168984 . ( Accessed 24 Jun. 2021 )

Jennings R (2020) Indonesia’s pandemic response: a law to create millions of jobs | Voice of America - English. VOA News. https://www.voanews.com/economy-business/indonesias-pandemic-response-law-create-millions-jobs . ( Accessed 2 Jul. 2021 )

Khatatbeh H, Abu-Alfoul (2020) The impact of COVID-19 pandemic on global stock markets: an event study. Int J Econ Bus Adm 8(4):505–514

Kim Y-W, Yoon S-J, Oh I-H (2012) The economic burden of the 2009 pandemic H1N1 influenza in Korea. Scand J Infect Dis 45(5):390–396

Koh WC, Naing L, Wong J (2020) Estimating the impact of physical distancing measures in containing COVID-19: An empirical analysis. Int J Infect Dis 100:42–49. https://doi.org/10.1016/j.ijid.2020.08.026

Lansdorp S and Jellema W (2013) Momentum from Underreaction. https://www.robeco.com/media/6/5/5/65562514d3c17a4f69cf50f5963b2a0b_robeco-whitepaper-momentum-from-underreaction-2013-10_tcm20-11257.pdf

Li ML, Bouardi HT, Lami OS, Trikalinos TA, Trichakis NK and Bertsimas D (2020) Forecasting COVID-19 and analyzing the effect of government interventions. MedRxiv, pp 2020–2006

Liew V and Puah C-H (2020) Chinese stock market performance in the time of novel coronavirus pandemic. MPRA (100414). https://mpra.ub.uni-muenchen.de/100414/

Liu H, Manzoor A, Wang C, Zhang L, Manzoor Z (2020) The COVID-19 outbreak and affected countries stock markets response. Int J Environ Res Public Health 17(8):2800

Lyons C (1999) Definiteness. Cambridge University Press

Book   Google Scholar  

MacKellar L (2007) Pandemic influenza: a review. Popul Dev Rev 33(3):429–451

Masoud N (2013) The impact of stock market performance upon economic growth. Int J Econ Financ Issues 3(4):788–798

Maulia E (2021) Indonesia bourse launches tech classification to lure investors. Nikkei Asia. https://asia.nikkei.com/Business/Markets/Indonesia-bourse-launches-tech-classification-to-lure-investors . ( Accessed 8 Jul. 2021 )

Molina K and Ramadhan IF (2020) COVID-19 response of the indonesian government on funding and financial support measures. White & Case LLP. https://www.whitecase.com/publications/alert/covid-19-indonesian-government-financial-assistance-measures

Murphy K (2008) Swan’s $4bn injection for small lenders. The Sydney Morning Herald. https://www.smh.com.au/national/swans-4bn-injection-for-small-lenders-20080926-4ovd.html . ( Accessed 15 Aug. 2021 )

Numbeo (2021) Cost of living. Numbeo.com. https://www.numbeo.com/cost-of-living/in/Singapore

Omoleke SA, Mohammed I, Saidu Y (2016) Ebola viral disease in West Africa: a threat to global health, economy and political stability. J Public Health Afr. https://doi.org/10.4081/jphia.2016.534

Oxford Business Group (2020) Can Indonesia’s Omnibus Law drive its Covid-19 economic recovery? Oxford Business Group. https://oxfordbusinessgroup.com/news/can-indonesia-s-omnibus-law-drive-its-covid-19-economic-recovery . ( Accessed 5 Jul 2021 )

Oxford Economics (2020) The economic impact of Covid19 on Asia Pacific. Oxford Economics, p 14. https://www.oxfordeconomics.com/resource/the-economic-impact-of-covid-19-on-asia-pacific/

Ozili PK and Arun T (2020) Spillover of COVID-19: impact on the Global Economy. Available at SSRN 3562570

Powell A (2021) Vaccines should end the pandemic, despite the variants, say experts. Harvard Gazette. https://news.harvard.edu/gazette/story/2021/02/vaccines-should-end-the-pandemic-despite-the-variants-say-experts/

Randall LK, David (2021) Infrastructure bill could fuel stocks, but investors temper enthusiasm. Reuters. 25 Mar. https://www.reuters.com/article/us-usa-stocks-infrastructure-analysis-idUSKBN2BH0G1 . ( Accessed 20 Aug 2021 )

Rassy D, Smith RD (2013) The economic impact of H1N1 on Mexico’s tourist and pork sectors. Health Econ 22(7):824–834

Reuters (2020) Indonesia targets 5.6–6.2% average GDP growth to 2024. Reuters. https://www.reuters.com/article/us-indonesia-economy-idUSKBN1Z516O . ( Accessed 6 Jul 2021 )

Reuters (2021) Stock market headlines | breaking stock market news. Reuters. https://www.reuters.com/markets/stocks/asia-pacific/

Ritter JR (1991) The long-run performance of initial public offerings. J Financ 46(1):3–27

Rose A (2021) COVID-19 economic impacts in perspective: a comparison to recent U.S. disasters. Int J Disaster Risk Reduct 60:102317

Rouatbi W, Demir E, Kizys R, Zaremba A (2021) Immunizing markets against the pandemic: COVID–19 vaccinations and stock volatility around the world. SSRN Electron J. https://doi.org/10.2139/ssrn.3815000

Sánchez JM and Liborio CS (2012) The relationships among changes in GDP, employment, and unemployment: this time, it’s different. research.stlouisfed.org. https://research.stlouisfed.org/publications/economic-synopses/2012/05/18/the-relationships-among-changes-in-gdp-employment-and-unemployment-this-time-its-different

Scherf M, Matschke X, Rieger MO (2022) Stock market reactions to COVID-19 lockdown: a global analysis. Fin Res Lett 45:102245

Scholes M, Williams J (1977) Estimating betas from nonsynchronous data. J Financ Econ 5(3):309–327

Steyn N, Plank MJ, James A, Binny RN, Hendy SC, Lustig A (2021) Managing the risk of a COVID-19 outbreak from border arrivals. J Royal Soc Interface 18(177):20210063

Su Y, Yip Y, Wong RW (2001) The impact of government intervention on stock returns evidence from Hong Kong. Int Rev Econ Financ 11(2002):277–297

Swaine J (2008) Bank bailout: Alistair Darling unveils ?500billion rescue package. Telegraph.co.uk. https://www.telegraph.co.uk/finance/financialcrisis/3156711/Bank-bailout-Alistair-Darling-unveils-500billion-rescue-package.html

The Economist (2021) More than 85 poor countries will not have widespread access to coronavirus vaccines before 2023. Economist Intelligence Unit. https://www.eiu.com/n/85-poor-countries-will-not-have-access-to-coronavirus-vaccines/

The World Bank (2016) The economic impact of the 2014 ebola epidemic. https://openknowledge.worldbank.org/bitstream/handle/10986/20592/9781464804380.pdf . ( Accessed 12 Aug 2021 )

The World Bank (2020) Reforms in Indonesia’s financial sector open new economic opportunities. World Bank. https://www.worldbank.org/en/news/press-release/2020b/03/23/reforms-in-indonesias-financial-sector-open-new-economic-opportunities

United Nations (2021) World population prospects—population division—United Nations. un.org. https://population.un.org/wpp/

US Bank Asset Management Group (2021) How the COVID vaccine affects the economy | U.S. Bank. www.usbank.com . https://www.usbank.com/investing/financial-perspectives/market-news/how-does-the-covid-vaccine-affect-the-economy.html . Accessed 2 Jul 2021

Valle-Cruz D, Fernandez-Cortez V, López-Chau A, Sandoval-Almazán R (2022) Does twitter affect stock market decisions? financial sentiment analysis during pandemics: a comparative study of the h1n1 and the covid-19 periods. Cogn Comput 14(1):372–387

WHO (2021) WHO coronavirus disease (COVID-19) dashboard. Who.int. https://covid19.who.int/table

Wight A (2020) Will the coronavirus follow when 20 million indonesians go home for Mudik? Forbes. https://www.forbes.com/sites/andrewwight/2020/04/08/will-the-coronavirus-follow-when-20-million-indonesians-go-home-for-mudik/?sh=d1eb562fc13c . ( Accessed 2 Jul. 2021 )

World Bank (2018) Indonesian overview. World Bank. https://www.worldbank.org/en/country/indonesia/overview

Yahoo Finance (2021) Historical stock market data. Accessed through Yahoo Finance API

Zaremba A, Kizys R, Aharon DY, Demir E (2020) Infected markets: novel coronavirus, government interventions, and stock return volatility around the globe. Financ Res Lett 1(1):101597

Zaremba A, Aharon DY, Demir E, Kizys R, Zawadka D (2021) COVID-19, government policy responses, and stock market liquidity around the world: a note. Res Int Bus Financ 56(2021):101359

Download references

Not applicable.

Author information

Authors and affiliations.

Alliance Manchester Business School, The University of Manchester, Manchester, M15 6PB, UK

Josua Sinaga, Ting Wu & Yu-wang Chen

You can also search for this author in PubMed   Google Scholar

Corresponding author

Correspondence to Yu-wang Chen .

Ethics declarations

Conflict of interest.

The authors declare that they have no conflict of interest.

Supplementary Information

Below is the link to the electronic supplementary material.

Supplementary file1 (DOCX 79 kb)

Rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Sinaga, J., Wu, T. & Chen, Yw. Impact of government interventions on the stock market during COVID-19: a case study in Indonesia. SN Bus Econ 2 , 136 (2022). https://doi.org/10.1007/s43546-022-00312-4

Download citation

Received : 24 December 2021

Accepted : 26 July 2022

Published : 17 August 2022

DOI : https://doi.org/10.1007/s43546-022-00312-4

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Financial markets
• Emerging country
• Stock return
• Find a journal
• Publish with us
• Track your research

• Open access
• Published: 02 April 2024

Towards universal health coverage in Vietnam: a mixed-method case study of enrolling people with tuberculosis into social health insurance

• Rachel Forse   ORCID: orcid.org/0000-0002-0716-3342 1 , 2 ,
• Clara Akie Yoshino 2 ,
• Thanh Thi Nguyen 1 ,
• Thi Hoang Yen Phan 3 ,
• Luan N. Q. Vo 1 , 2 ,
• Andrew J. Codlin 1 , 2 ,
• Lan Nguyen 4 ,
• Chi Hoang 4 ,
• Lopa Basu 5 ,
• Minh Pham 5 ,
• Hoa Binh Nguyen 6 ,
• Luong Van Dinh 6 ,
• Maxine Caws 7 , 8 ,
• Tom Wingfield 2 , 7 ,
• Knut Lönnroth 2 &
• Kristi Sidney-Annerstedt 2  

Health Research Policy and Systems volume  22 , Article number:  40 ( 2024 ) Cite this article

146 Accesses

4 Altmetric

Metrics details

Vietnam’s primary mechanism of achieving sustainable funding for universal health coverage (UHC) and financial protection has been through its social health insurance (SHI) scheme. Steady progress towards access has been made and by 2020, over 90% of the population were enrolled in SHI. In 2022, as part of a larger transition towards the increased domestic financing of healthcare, tuberculosis (TB) services were integrated into SHI. This change required people with TB to use SHI for treatment at district-level facilities or to pay out of pocket for services. This study was conducted in preparation for this transition. It aimed to understand more about uninsured people with TB, assess the feasibility of enrolling them into SHI, and identify the barriers they faced in this process.

A mixed-method case study was conducted using a convergent parallel design between November 2018 and January 2022 in ten districts of Hanoi and Ho Chi Minh City, Vietnam. Quantitative data were collected through a pilot intervention that aimed to facilitate SHI enrollment for uninsured individuals with TB. Descriptive statistics were calculated. Qualitative interviews were conducted with 34 participants, who were purposively sampled for maximum variation. Qualitative data were analyzed through an inductive approach and themes were identified through framework analysis. Quantitative and qualitative data sources were triangulated.

We attempted to enroll 115 uninsured people with TB into SHI; 76.5% were able to enroll. On average, it took 34.5 days to obtain a SHI card and it cost USD 66 per household. The themes indicated that a lack of knowledge, high costs for annual premiums, and the household-based registration requirement were barriers to SHI enrollment. Participants indicated that alternative enrolment mechanisms and greater procedural flexibility, particularly for undocumented people, is required to achieve full population coverage with SHI in urban centers.

Conclusions

Significant addressable barriers to SHI enrolment for people affected by TB were identified. A quarter of individuals remained unable to enroll after receiving enhanced support due to lack of required documentation. The experience gained during this health financing transition is relevant for other middle-income countries as they address the provision of financial protection for the treatment of infectious diseases.

Peer Review reports

Contributing to universal health coverage (UHC) by improving access to fair and sustainable health financing, of which one mechanism is health insurance, has become a priority among low- and middle-income countries [ 1 , 2 ]. Many countries in the Asia Pacific region have made steady progress towards UHC coverage through sustained political commitments and fiscal policy aligned with their commitment [ 3 ]. By 2020, 27 countries had implemented a social health insurance (SHI) financing mechanism, which typically includes open enrollment for the full population along with partial or full subsidization of healthcare costs for vulnerable groups [ 4 ].

Vietnam’s first SHI scheme was piloted in 1989 and grew through successive pilots and expansions. In 2009 the national-level Health Insurance Law (HIL) went into effect, uniting the existing health insurance programs and schemes for the poor [ 5 ]. Amendments to the HIL effective in 2015 made SHI compulsory for all and pooled risk by re-structuring registration around the household unit [ 4 ]. A household in Vietnam is defined by inclusion in the ‘family book ’, the national system of family and address registration [ 6 ].

Access to SHI in Vietnam increased rapidly, principally through subsidization of premiums. Specific groups were enrolled automatically with full subsidy, including vulnerable populations (e.g., households classified as ‘poor’, children aged < 6, people aged > 80), pensioners and meritorious groups (e.g., veterans). Partial premium subsidization was also available for students, households classified as ‘near-poor’ and some farmers [ 7 ]. More than half of SHI members are entitled to 80% coverage with a 20% co-payment for services [ 8 ]. However, co-payments are reduced to 5% or are eliminated for subsidized groups (e.g., households classified as ‘poor’ and ‘near-poor’, children < 6) [ 4 ].

By 2020, Vietnam recorded a 91% national SHI coverage rate [ 7 ]. Those remaining uninsured mainly consisted of informally employed individuals [ 7 ]. Enrollment rates were highest among low- and high-income groups, leaving the so-called “missing middle” of uninsured [ 5 ].

Vietnam continues to transition to domestic financing of healthcare from donor financing by expanding the breadth of the national SHI. The Ministry of Health and Vietnam Social Security (VSS) have begun to close service gaps and integrate vertical health programs (e.g., those with stand-alone budget allocations and/or direct donor financing) into SHI financing [ 7 ]. The costs for antiretroviral therapy (ART) were transitioned from donor funding to SHI in 2019 [ 9 ], COVID-19 treatments were covered by SHI in 2020, and financing for tuberculosis (TB) care was fully transitioned to SHI in 2022 [ 7 ].

Until this financing transition, anti-TB medications and consultations were provided free of charge in the public sector, funded by a mixture of domestic and international funding [ 10 ]. While first-line TB medications were included in the SHI-reimbursable list of essential medicines, the government network of District TB Units (DTUs) were ineligible for registration with VSS, or reimbursement for services provided. Since July 2022, TB health facilities that met certain conditions could register with VSS and receive reimbursements for TB consultations, diagnostics and anti-TB medications [ 11 ]. The financing for drug-resistant (DR-)TB tests and medications remains largely unchanged, co-financed by the Global Fund and domestic budgets [ 12 ].

This transition of the TB financing model in Vietnam is a large undertaking as the country has the world’s 10th highest TB burden and the SHI benefits package is already considered to be generous, and the sustainability of the SHI fund is a concern [ 4 , 13 ] An estimated 169,000 individuals developed TB in 2021, and the disease killed approximately 14,200 [ 14 ]. A national costing survey of TB-affected households showed that 63% experienced catastrophic costs, spending ≥ 20% of their annual income on TB [ 10 ]. Many face food insecurity and cope with TB-related costs by taking loans, dissavings and informally borrowing money [ 10 , 15 , 16 ].

As Vietnam continues to expand SHI financing for the TB program, it is now vital for people with TB to have SHI. Those without SHI coverage will need to finance their care out of pocket (OOP) or purchase SHI and make co-payments for their care to be subsidized. For these reasons, it is important to understand why certain people with TB are uninsured, the feasibility of enrolling them in insurance when they begin treatment, and the challenges they may face with enrolling in SHI.

We conducted a convergent parallel mixed-method case study [ 17 ]. A case study was selected because it is well-suited to describe a complex issue in a real-life setting [ 18 ]. We used a naturalistic design with theoretical sampling of uninsured persons with TB using an interpretivist approach [ 19 ]. Mixed methods were selected to facilitate comparisons between quantitative and qualitative data and interpretation of the findings. An intervention, assisting TB-affected households to enroll in SHI, was conducted between November 2019 and January 2022, prior to the integration of the TB program into the SHI financing scheme. Quantitative data collection sought to answer questions regarding enrollment success rate, time to enrollment and cost of SHI enrollment for uninsured TB-affected households upon TB treatment initiation. The qualitative data explored barriers to SHI enrollment to explain and contextualize the quantitative findings. The quantitative and qualitative data were weighted equally [ 17 ].

Intervention description

A pilot intervention was conducted to facilitate SHI enrollment for people with TB in ten districts of Ha Noi and Ho Chi Minh City (HCMC). The standard process for first-time enrollment into SHI was mapped and costed from a household’s perspective (Additional file 1 ). Uninsured individuals were identified from the TB treatment register when they were enrolled in drug-susceptible (DS-)TB treatment at DTUs [ 20 ]. Study staff then attempted to facilitate enrollment of the person with TB and up to three household members into SHI.

SHI enrollment support included home visits by study staff to provide detailed information and counseling about the process of SHI enrollment, assistance with SHI application preparation including obtaining photocopies of all required documents, follow-up to obtain missing documentation within the household, accompaniment to the SHI office for application submission, and direct payment of the annual SHI premium for the household. For people who did not have the paperwork certifying temporary residence in Hanoi or Ho Chi Minh City, staff visited the local government office to obtain the information about the process for individual cases to obtain residency certificates and support participants with navigation of the bureaucracy. TB-affected people and their household members were also provided with a hotline number to call and receive support during working hours from the social workers who were employed by the study. Study staff attempted to facilitate the SHI enrollment process throughout the entire 6-month duration of DS-TB treatment. After a TB treatment outcome was recorded by the DTU, study staff stopped assisting with SHI enrollment and participants were recorded as ‘not enrolled in SHI’ in the study’s evaluation.

Quantitative methods

Case-level TB treatment notification data and SHI status were exported from VITIMES, the government-implemented electronic TB register for Vietnam, for all individuals who started TB treatment during the intervention period. The pilot intervention recruited participants from two TB treatment support projects (Project 1, n  = 59 and Project 2, n  = 56) [ 21 , 22 ] and tracked study forms housed in ONA.io. The sample size was determined by the availability of funding provided by the donor for treatment support service delivery, rather than to measure a specific end point of SHI enrollment. Descriptive statistics summarizing the enrollment cascade and turnaround time of enrollment were calculated using Stata v17 (Stata17 Corp, College Station, USA). To obtain the mean costs for household SHI enrollment, total direct costs for purchasing SHI were summed and divided by the total number of participants. Costs were captured in Vietnamese Dong (VND) and converted to United States Dollars (USD) using the exchange rate from the mid-point of the pilot intervention (1 June 2020) from OANDA.com.

Qualitative methods

Individuals were purposively sampled for maximum variation to ensure representation of all implementation areas and provide gender balance [ 23 ].The concept of information power guided the sample size [ 24 ]. Given the well-defined study aim, high quality in-depth responses from the participants and the authors’ expertise in the subject area, the sample size of 19 individual interviews and three focus group discussions was deemed appropriate. These were conducted in Ha Noi and HCMC. A total of 34 individuals participated in the interviews (Table  1 ).

They included 14 people enrolled in the pilot intervention, five community members who were non-beneficiaries of the treatment support intervention, 13 TB program staff from the national-, provincial- and district-levels and two study staff. Interviews were conducted at two time points: June 2019 and 2020. SHI enrollment barriers were collected as part of a qualitative study on the acceptability of providing cash transfers and SHI enrollment to adults with TB [ 25 ]. During the second round of interviews in 2020, study staff were included due to their in-depth knowledge of the challenges faced by TB-affected households when attempting to enroll in SHI and their ability to suggest programmatic-level solutions to these challenges. These interviews were conducted one-on-one, after the other interviews and focus groups had been conducted to reduce bias. The interviews were conducted at the National Lung Hospital, HCMC Provincial Lung Hospital, study office or DTUs. All interviews were conducted and transcribed in Vietnamese, translated into English, checked and finalized by a lead translator.

The interviews were analyzed through an inductive approach and themes were drawn through a framework analysis [ 26 ] to identify barriers to enrolling in SHI using Dedoose Version 7.0.23 (SocioCultural Research Consultants, Los Angeles, USA).

Data triangulation

Quantitative and qualitative data were collected in parallel. Triangulation of quantitative and qualitative data was conducted to synthesize findings and assess the level of agreement, convergence, and divergence from the findings generated by the different methods [ 17 ].

During the study, 5887 individuals were treated for DS-TB across the 10 intervention districts (Table  2 ). TB registers indicated that 2846 (48.3%) individuals were uninsured upon treatment initiation, or their SHI enrollment status was not recorded. Among 115 uninsured study participants, 88 (76.5%) were successfully enrolled in SHI before the end of their TB treatment. Among those, the household had an average of two members, resulting in a total of 206 individuals living in TB-affected households receiving SHI coverage through the pilot intervention.

The median time between DS-TB treatment initiation and SHI card issuance was 34.5 days (IQR 24–68): 11 days (IQR 5–23) between treatment initiation and pilot enrollment, 7 days (IQR 1–19.5) for SHI application preparation and submission, and 12 days (IQR 9–20) for application processing and SHI card provision.

The qualitative data showed that participants across all participant groups broadly understood that SHI is a system designed to prevent catastrophic OOP medical expenditure. As shown in Table  3 , National and provincial-level TB staff described SHI as a human right and spoke about achieving UHC as a nation; no other participant groups discussed SHI in this way. However, district-level doctors and intervention beneficiaries spoke in greater details about coverage and service gaps, and the practicalities of utilizing SHI. These participant groups expressed that when individuals purchase SHI only after a negative health event, such as a TB diagnosis, then the social safety net is unavailable to provide support until SHI coverage begins. Drawn from these views, the first theme indicated that the optimal time to purchase SHI is prior to a TB diagnosis.

One DTU staff member described how the standard processing time, or delays in processing SHI applications led to periods of high OOP expenditure:

“Unfortunately, claims are not immediately paid upon [SHI registration] submission. They may be handled in about 2 or 3 weeks, or even one month. That is why the insurance is not available at the time that they want to go for an examination and treat their condition using insurance.” (Female, District-level TB staff)

A complementary theme was that perceived lack of knowledge about SHI enrollment procedures prevents or delays enrollment. District-level TB doctors and program staff identified a lack of understanding and knowledge of the SHI enrollment process as a main contributor to lack of insurance or delays in obtaining coverage.

“Actually, for some people [with TB] who do not clearly understand the [enrollment] procedures… it will take a lot of time [to obtain SHI]. It also depends on the staff who handle the files at the commune; some staff are very enthusiastic and they help patients complete forms. There are cases [...] where they [people with TB] are required to fill in all information and write specific codes of each insurance card [from other family members] on a form. Meanwhile some people in their family work far from home and cannot send their insurance cards home in a timely manner.” (Female, program staff)

Participants tended to believe that individuals who lacked information about SHI made up the small minority of uninsured people in Vietnamese society. The above quote illustrated that the complicated administrative process prohibits enrollment; however, a factor potentially facilitating SHI enrollment may be the helpfulness of the person processing the SHI application.

The average cost per household to obtain SHI enrollment for one year (Table  2 ) was VND 1,503,313 (USD 65.52). (For detailed information on the costs of SHI enrollment, see Additional file 1 ). A third theme contextualized this finding and showed that SHI enrollment costs were perceived as prohibitively high for some. Cost was a greater challenge for lower income families, who did not meet the government’s criterion of households classified as ‘poor’ or ‘near-poor’, and were therefore ineligible for premium subsidies and SHI registration with lower co-payment rates. One DTU doctor reported that:

“We think that it is simple to buy health insurance cards, but that is only true for those who have sustainable income - when our income is much higher than the fee for buying health insurance. For some people, buying health insurance is a luxury.” (Male, District-level TB staff)

Twenty-seven people with TB (23.5%) were unable to obtain SHI coverage. The primary reason (70.4%) was missing documentation. In four instances (14.8%) a household member other than the person with TB refused to enroll in SHI. One individual (3.7%) died during the enrollment process. Three individuals (11.1%) did not enroll for other reasons.

SHI refusal by household members was not identified as a barrier to SHI enrollment in the qualitative data. However, a fourth theme confirmed the primary reason for non-enrollment by showing that some individuals do not possess the required documentation to obtain SHI, such as their identity card or ‘family book.’ [See Supplementary File] Even with six months of support from study staff, some TB-affected households were unable to gather the required documents for enrollment. The following quotation by an undocumented, elderly woman with TB illustrates the prolonged challenges she faced with obtaining formal employment, access to government services and SHI:

“I have had problems with my personal papers for a few decades and I cannot adjust my papers because I don’t have the money. […] I searched for my Identity Card and found out that I had lost it. Then I came back there [my hometown] to get the family book, to reissue my ID and to get my CV certified so I could join a company. I was very young at that time, just a little bit more than thirty years old, and I learned that I was cut from the family book.” (Female, pilot beneficiary)

To address challenges with documentation, one DTU officer in HCMC suggested that individuals who had never been insured required a change to the SHI registration requirements to ensure that everyone in Vietnam can access SHI:

“I think we should be flexible with these cases or we can find another way. Normally, the people who really need the support and the insurance or cash support, they are the people who have less information. […] We cannot have the same requirements for these people as for other people. Actually, for those who have [met] all conditions, they already have health insurance cards.” (Male, District-level TB staff)

Participants expressed that the uninsured had often not purchased SHI for a reason, and alternative registration procedures were needed to make SHI accessible for all. A fifth theme was identified indicating that current SHI enrollment procedures may prevent full population coverage.

Beyond the undocumented, some participants reported the enrollment mandate for the entire household (made under the Amendment to the HIL) for first-time enrollees was viewed as prohibitive of SHI coverage.

“Because in the old days, health insurance was sold individually for each person, but now it is sold to households, and many households do not have as good economic [situation]… so they can only afford to buy it for 50% or 60% of the household. Unskilled labor or low-income labor cannot afford to buy it for the whole family. That is to say, it is easier to buy it for each individual and it is difficult to buy for the whole family.” (Male, community member)

Though individual registration would make SHI more accessible to individuals with TB due to lower annual costs, household members with high vulnerability to TB would not be covered if policy promoted individual enrollment solely for TB.

This mixed-methods case study showed that by providing full subsidy and registration assistance, most uninsured people with TB could access SHI. However, the median time to insurance coverage meant that approximately 20% of a person’s DS-TB treatment duration remained uncovered by SHI despite successful enrollment. A substantial number of participants were unable to enroll in SHI and are likely to be perpetually locked out of SHI due to lack of personal documentation. Additional barriers to SHI enrollment were found to be lack of knowledge, the cost of obtaining coverage, and the household-based registration requirement.

The pilot intervention had dedicated staff who facilitated SHI application development and submission, yet it still took a median of 34.5 days for SHI coverage to take effect. In a context where this level of support is not available to all people with TB, it is likely that the turnaround time for SHI coverage is longer due to the complicated bureaucracy involved. This poses a major challenge, as TB-affected households incur the highest cost during the first two months of treatment [ 15 ]. One cost avoidance/mitigation strategy that people with a TB diagnosis may employ following the health financing transition is delaying TB treatment initiation until SHI coverage commences. This will likely lead to worse outcomes and sustained community transmission. The time between diagnosis and treatment should be rigorously monitored to ensure that this coping strategy is not employed, and alternative support should be made available to ensure that people diagnosed with TB are able to receive immediate treatment.

With the TB health financing transition, the uninsured will be asked to pay OOP for TB treatment and most insured individuals must co-pay for TB services which were previously provided free of cost. A national patient cost survey in 2018 found that 63% of TB-affected households experienced catastrophic costs under the previous health financing model [ 10 ]. There is a risk that the proportion of TB-affected households experiencing catastrophic costs could increase with the introduction of fees. This was not found to be the case for people living with HIV (PLHIV) when the costs of ART transitioned to SHI in Vietnam, but a new nationally representative TB costing survey is needed to assess this risk [ 9 ]. Several domestic solutions could ameliorate these challenges. As suggested for the Indian context, domestic revenues allocated by the Ministry of Finance to VSS could be increased to better support TB care [ 27 ]. VSS could also reclassify the category of TB disease and thus ensure that SHI paid for all diagnostics and drugs associated with TB treatment, without the need for a co-payment. A mid-term review of the Global Fund program in Vietnam has also called for a SHI package specifically designed to cover the OOP medical costs of TB care [ 28 ]. There are several potential mechanisms to prevent costs from falling on TB-affected households. A deeper investigation is needed to understand the fiscal space available within the Vietnamese government to cover such costs.

This case study showed that 23.5% of the uninsured people with TB were never able to enroll for the duration of their treatment, primarily due to lack of documentation. Specific provisions need to be made for the undocumented to receive free TB diagnosis, consultations, and medications through routine practice of the TB program. Multi- and bi-lateral funding mechanisms can also play a role in filling gaps by paying for TB tests for the uninsured, purchasing SHI for those diagnosed with TB, subsidizing or reimbursing OOP expenditure in the period before SHI coverage takes effect, and fully financing TB care for the undocumented. Furthermore, longer-term health system strengthening initiatives, such as creating a legal mechanism for the undocumented to obtain SHI, are likely needed to address the challenges faced by the 9% of the general population that remain uninsured. The ILO has called for “determining new strategies, which may include extension of state budget-funded subsidies to further support the participation of workers in the informal economy [ 7 ].” These forms of inclusive initiatives would solve the TB-specific challenges identified in this study and have a large positive impact on society.

We found that addressing the cost of SHI premiums and knowledge gaps in the enrollment procedures may improve SHI coverage. These findings mirror those following the transition of HIV financing to SHI in 2017. A study among PLHIV identified burdensome processes, lack of information about SHI registration procedures, and high SHI premium costs for a household as key barriers to SHI coverage [ 29 ]. However, a cluster randomized control trial which provided education, a 25% premium subsidy, or both to uninsured households found that these interventions had limited effects on SHI enrollment. Yet, “less healthy” individuals had higher SHI enrollment rates [ 30 ]. This suggests that people who have just received a TB diagnosis could be more receptive to interventions promoting SHI enrollment through premium subsidization and education. Vietnam’s National TB Program (NTP) has established a fund to subsidize SHI enrollment costs for TB-affected individuals. The size of the fund could be increased with additional support while access to the fund and the procedures for receiving support could be optimized [ 31 ]. Given the SHI transition, the NTP should also consider providing educational materials about the SHI enrollment process through the DTU network to uninsured persons with TB.

TB registers indicated that 52% of people starting TB treatment in the urban intervention districts had recorded SHI coverage. This rate is lower than other recent SHI coverage reports. A 2018–2022 DS-TB costing survey reported a SHI coverage of 70% [ 32 ], while in a DR-TB costing survey (2020–2022) it was 85% [ 16 ]. All available data sources indicate that SHI coverage among people with TB is lower than the general population, which is indicative of their socioeconomic vulnerability [ 33 ]. However, this large SHI coverage rate discrepancy may be explained by people with TB not revealing they had SHI coverage, or DTU staff could have also inconsistently recorded an individual’s SHI status in the paper TB registers since these data did not have much clinical relevance for TB treatment at the time. Now that DTUs receive financial reimbursements for the TB services from VSS, SHI coverage rates in treatment registers are likely to increase. Further research should be conducted to understand the national SHI coverage rate for people receiving TB treatment, along with the risk factors associated with being uninsured.

Limitations

This case study was conducted in the two largest cities of Vietnam and findings may not be representative of the entire country. Quantitative data were collected in a programmatic setting, and SHI coverage data for all individuals initiating TB treatment in the intervention areas appear to be underreported for reasons described above. Lastly, we were unable to collect SHI enrollment data from a control population, either prospectively during the pilot intervention or retrospectively during the pilot evaluation. As a result, we do not have information on the enrollment status or time to obtain SHI coverage among a population that did not receive assistance from the pilot intervention. However, given the substantial additional support provided by study staff for the enrollment process, we believe it is safe to assume that if left alone, TB-affected households would be slower in the enrollment process and likely enroll in lower rates.

Vietnam is viewed as a leader among Southeast Asian nations in its commitment and progress towards UHC. This mixed-methods case study illustrated the progress that Vietnam has made in its path to greater domestic financing of healthcare through SHI. This study is one of the first to examine the integration of TB services into SHI in Vietnam and define the challenges that people with TB face while attempting to gain access to financial protection after receiving a TB diagnosis. In order to make strides towards UHC in Vietnam and to close population coverage gaps, initiatives are required to specifically address the barriers faced by the uninsured. This study found that the majority of the uninsured were able to gain access to SHI through full subsidization of premiums, enrollment assistance and education. However, initiating TB care and SHI enrollment concomitantly left a significant portion of the 6-month TB treatment duration without financial protection. Additionally, a quarter of the uninsured with TB were unable to gain access to SHI during treatment, primarily due to a lack of documentation. There is great need for official mechanisms to be in place that enable those without sufficient state documents to access the TB program and to address the time-sensitive nature of providing effective financial protection during treatment of an infectious disease. These findings are relevant for other high TB burden, middle-income countries who are on a similar pathway for transitioning away from donor-financed TB programs to ones supported with a higher proportion of domestic resources.

Availability of data and materials

The quantitative dataset used and analyzed during the current study are available from the corresponding author on reasonable request. Seven anonymized transcripts of interviews with the people enrolled in the pilot intervention and non-beneficiaries have been uploaded to the following URL: https://doi.org/ https://doi.org/10.5281/zenodo.7736220 .

Abbreviations

Anti antiretroviral therapy

Drug resistant tuberculosis

Drug susceptible tuberculosis

District TB Unit

Ho Chi Minh City

Health Insurance Law

Human immunodeficiency virus

International Labour Organization

Interquartile range

National Tuberculosis Program

Out of pocket

People Living with HIV

Social Health Insurance

• Tuberculosis

Universal Health Coverage

United States Dollar

Vietnamese Dong

Vietnam Social Security

World health statistics 2023: monitoring health for the SDGs, sustainable development goals. https://www.who.int/publications-detail-redirect/9789240074323 . Accessed 31 May 2023.

SDG Target 3.8 | Achieve universal health coverage, including financial risk protection, access to quality essential health-care services and access to safe, effective, quality and affordable essential medicines and vaccines for all. https://www.who.int/data/gho/data/themes/topics/indicator-groups/indicator-group-details/GHO/sdg-target-3.8-achieve-universal-health-coverage-(uhc)-including-financial-risk-protection . Accessed 29 May 2023.

Hondo D, Kim Y. Achieving universal health coverage in developing Asia and the Pacific. Asian Development Bank Institute; 2023. Report No.: 2023-15 (November). https://doi.org/10.56506/VUWA5043 . Accessed 1 Feb 2024.

Le QN, Blizzard L, Si L, Giang LT, Neil AL. The evolution of social health insurance in Vietnam and its role towards achieving universal health coverage. Health Policy OPEN. 2020;100011. http://www.sciencedirect.com/science/article/pii/S2590229620300095 . Accessed 10 Sep 2023.

Somanathan A, Tandon A, Dao HL, Hurt KL, Fuenzalida-Puelma HL. Moving toward universal coverage of social health insurance in Vietnam: assessments and options. Washington, DC: World Bank; 2014, p. 177. (Directions in Development). https://doi.org/10.1596/978-1-4648-0261-4

Hardy A. Rules and resources: negotiating the household registration system in Vietnam under reform. Sojourn J Soc Issues Southeast Asia. 2001;16(2):187–212.

Article   CAS   Google Scholar  

Bales S, Phe Goursat M, Phuong NK. Extending social health protection in Viet Nam: accelerating progress towards universal health coverage. International Labour Organization; 2021, p. 1–14. (Social Protection in action: building social protection floors for all.). https://www.social-protection.org/gimi/gess/ShowRessource.action;jsessionid=iFmxgS-hSrmJjf_1uG90JAnijskVEc8XJD9-4lubAqbFD4-JPYOm!341651707?id=57668 . Accessed 26 Aug 2022.

Bao Hiem Xa Hoi [Vietnam Social Security]. 2019. Phát triển bền vững BHYT toàn dân [Sustainable development of universal health insurance]. https://baohiemxahoi.gov.vn/congkhai/Pages/so-lieu-thong-ke-nganh.aspx?CateID=130&ItemID=12916 . Accessed 26 Aug 2022.

Vu LTH, Johns B, Bui QTT, Duong Thuy A, Nguyen Thi D, Nguyen Thi H, et al. Moving to social health insurance financing and payment for HIV/AIDS treatment in Vietnam. Health Serves Insights. 2021;14:1178632920988843. https://doi.org/10.1177/1178632920988843 .

Article   Google Scholar  

Nhung NV, Hoa NB, Anh NT, Anh LTN, Siroka A, Lonnroth K, et al. Measuring catastrophic costs due to tuberculosis in Viet Nam. Int J Tuberc Lung Dis. 2018;22(9):983–90.

Article   CAS   PubMed   Google Scholar  

Tran VT. Circular No. 36/2021/TT-BYT dated December 31, 2021 on Medical examination and treatment, and payment for medical examination and treatment costs covered by health insurance relating to tuberculosis examination and treatment. 36/2021/TT-BYT Dec 31, 2021. https://hethongphapluat.com/circular-no-36-2021-tt-byt-dated-december-31-2021-on-medical-examination-and-treatment-and-payment-for-medical-examination-and-treatment-costs-covered-by-health-insurance-relating-to-tuberculosis-examination-and-treatment.html .

The Global Fund. Country Profiles Tuberculosis- 2022- Viet Nam. The Global Fund; 2023. https://data.theglobalfund.org/location/VNM/documents . Accessed 30 May 2023.

Nguyen Thi Thu Cúc. Socialist Republic of Viet Nam: strengthening the policy and institutional framework of social health insurance. Hanoi, Vietnam: Asian Development Bank; 2021, p. 1–27. Report No.: 50139–002. https://www.adb.org/projects/documents/vie-50139-002-tacr-4 . Accessed 26 Aug 2022.

Global tuberculosis report 2022. Geneva: World Health Organization; 2022. Licence: CC BY-NC-SA 3.0 IGO. https://www.who.int/publications-detail-redirect/9789240061729 .

Vo LNQ, Forse RJ, Codlin AJ, Dang HM, Van Truong V, Nguyen LH, et al. Socio-protective effects of active case finding on catastrophic costs from tuberculosis in Ho Chi Minh City, Viet Nam: a longitudinal patient cost survey. BMC Health Serv Res. 2021;21(1):1051. https://doi.org/10.1186/s12913-021-06984-2 .

Article   PubMed   PubMed Central   Google Scholar  

Pham TAM, Forse R, Codlin AJ, Phan THY, Nguyen TT, Nguyen N, et al. Determinants of catastrophic costs among households affected by multi-drug resistant tuberculosis in Ho Chi Minh City, Viet Nam: a prospective cohort study. BMC Public Health. 2023;23(1):2372. https://doi.org/10.1186/s12889-023-17078-5 .

Creswell JW, Clark Plano VL. Designing and conducting mixed methods research. 3rd ed. Thousand Oaks, California: SAGE Publications; 2017. p. 520.

Google Scholar  

George AL, Bennett A. Case studies and theory development in the social sciences. Cambridge, MA: MIT Press; 2005.

Crowe S, Cresswell K, Robertson A, Huby G, Avery A, Sheikh A. The case study approach. BMC Med Res Methodol. 2011;11(1):100. https://doi.org/10.1186/1471-2288-11-100 .

Thai LH, Nhat LM, Shah N, Lyss S, Ackers M. Sensitivity, completeness and agreement of the tuberculosis electronic system in Ho Chi Minh City. Viet Nam Public Health Act. 2017;7(4):294–8.

IMPACT TB Overview | IMPACT TB. https://impacttbproject.org/impact-tb-overview/ . Accessed 2 Jun 2023.

Fact Sheet: USAID Erase TB Project 2020. 2020. https://www.usaid.gov/vietnam/documents/fact-sheet-erase-tb-project . Accessed 24 Sep 2024.

Palinkas LA, Horwitz SM, Green CA, Wisdom JP, Duan N, Hoagwood K. Purposeful sampling for qualitative data collection and analysis in mixed method implementation research. Adm Policy Ment Health. 2015;42(5):533–44. https://doi.org/10.1007/s10488-013-0528-y .

Malterud K, Siersma VD, Guassora AD. Sample size in qualitative interview studies: guided by information power. Qual Health Res. 2016;26(13):1753–60. https://doi.org/10.1177/1049732315617444 .

Article   PubMed   Google Scholar  

Forse R, Nguyen TT, Dam T, Vo LNQ, Codlin AJ, Caws M, et al. A qualitative assessment on the acceptability of providing cash transfers and social health insurance for tuberculosis-affected families in Ho Chi Minh City, Vietnam. PLOS Glob Public Health. 2023;3(12): e0002439.

Ritchie J, Lewis J, Nicholls CM, Ormston R. Qualitative research practice: a guide for social science students and researchers. Second edn. SAGE Publications Ltd; 2013. p. 456.

Behera DK, Dash U, Sahu SK. Exploring the possible sources of fiscal space for health in India: insights from political regimes. Health Res Policy Syst. 2022;20(1):32. https://doi.org/10.1186/s12961-022-00831-4 .

Draft Report: Key Messages from the TB National Strategic Plan Mid Term Review (MTR) for Vietnam, between 8th to 18th January 2023. Hanoi, Vietnam: Vietnam’s National TB Program; 2023 Jan p. 1.

Matsumoto S, Nguyen HDT, Nguyen DT, Van Tran G, Tanuma J, Mizushima D, et al. The patient voice: a survey of worries and anxieties during health system transition in HIV services in Vietnam. BMC Int Health Hum Rights. 2020;20:1.

Wagstaff A, Nguyen HTH, Dao H, Bales S. Encouraging health insurance for the informal sector: a cluster randomized experiment in Vietnam. Health Econ. 2016;25(6):663–74. https://doi.org/10.1002/hec.3293 .

Hoa NB, Nhung NV. National tuberculosis patients cost survey: research findings lead to change in policy and practice. Viet Nam Public Health Act. 2019;9(2):50–2.

Wiemers A, Sidney-Annerstedt K, Forse R, Phan THY, Codlin A, Vo LNQ, et al. EP-13-729 COVID-19 lockdowns resulting in household income decline for people with TB in Ho Chi Minh City: interrupted time series analysis of household income changes. In: Proceedings of the union world conference on lung health 2022. Virtual event, 8–11 November 2022: the international journal of tuberculosis and lung disease (IJTLD); 2022. p. 191.

Lönnroth K, Jaramillo E, Williams BG, Dye C, Raviglione M. Drivers of tuberculosis epidemics: the role of risk factors and social determinants. Soc Sci Med. 2009;68(12):2240–6.

Download references

Acknowledgements

The authors would like to acknowledge the contributions of Hoang Thi My Linh, Chu Thi Hoang Anh, Nguyen Khac Cuong, Nham Thi Yen Ngoc and Tran Thai Hiep for conducting qualitative interviews and assisting with SHI enrollment activities. Special thanks to Dr. Kerri Viney for providing insightful comments on an early draft of this manuscript; they greatly strengthened the final version. This work was graciously supported by the staff of Vietnam’s National TB Program, the Hanoi Lung Hospital, Pham Ngoc Thach Provincial TB Hospital and 10 District TB Units. Lastly, we would like to thank the interview participants who shared their time and insights.

Open access funding provided by Karolinska Institute. The European Commission's Horizon 2020 program supported the provision of SHI and all data collection in 2019 through the IMPACT-TB study under grant agreement number 733174. For the period of 2020–2022, support to implement the pilot and conduct the evaluation was made possible by the generous support of the American people through the USAID under award number 72044020FA00001. TW was supported by grants from: the Wellcome Trust, UK ( Seed Award, grant number 209075/Z/17/Z); the Department of Health and Social Care (DHSC), the Foreign, Commonwealth & Development Office (FCDO), the Medical Research Council (MRC) and Wellcome, UK (Joint Global Health Trials, MR/V004832/1); the Medical Research Council (Public Health Intervention Development Award “PHIND”, APP2293); and the Medical Research Foundation (Dorothy Temple Cross International Collaboration Research Grant, MRF-131–0006-RG-KHOS-C0942). KSA was supported by the ASPECT Trial funded the Swedish Research Council (2022-00727). The contents of this study are the responsibility of the listed authors, and do not necessarily reflect the views of USAID or the United States Government.

Author information

Authors and affiliations.

Friends for International TB Relief, Hanoi, Vietnam

Rachel Forse, Thanh Thi Nguyen, Luan N. Q. Vo & Andrew J. Codlin

Department of Global Public Health, WHO Collaboration Centre on Tuberculosis and Social Medicine, Karolinska Institutet, Stockholm, Sweden

Rachel Forse, Clara Akie Yoshino, Luan N. Q. Vo, Andrew J. Codlin, Tom Wingfield, Knut Lönnroth & Kristi Sidney-Annerstedt

Centre for Development of Community Health Initiatives, Hanoi, Vietnam

Thi Hoang Yen Phan

IRD VN, Ho Chi Minh City, Vietnam

Lan Nguyen & Chi Hoang

USAID Vietnam, Hanoi, Vietnam

Lopa Basu & Minh Pham

National Lung Hospital, Hanoi, Vietnam

Hoa Binh Nguyen & Luong Van Dinh

Centre for TB Research, Department of Clinical Sciences, Liverpool School of Tropical Medicine, Liverpool, UK

Maxine Caws & Tom Wingfield

Birat Nepal Medical Trust, Kathmandu, Nepal

Maxine Caws

You can also search for this author in PubMed   Google Scholar

Contributions

This study was conceived of by RF, KSA, TTN, THYP, CAY, AJC, LNQV. The study was administered by RF, YP, TTN, AJC. Support from Vietnam’s National TB program was provided by HBN and LVD. The methodology was developed by RJ, CAY, KV, KL, KSA. The analysis was carried out by RF, CAY, TTN, and THYP. LNQV, AJC, TW, LN, CH, LB, MP, HBN, LVD, MC, KV, KL, and KSA supported the interpretation of findings. The first manuscript was written by RF. All co-authors reviewed and commented on the initial manuscript. The final manuscript was approved and reviewed by all authors.

Corresponding author

Correspondence to Rachel Forse .

Ethics declarations

Ethics approval and consent to participate.

All study procedures were conducted in strict adherence to the Declaration of Helsinki. Ethical approvals were granted by the National Lung Hospital Institutional Review Board (114/19/CT-HĐKH-ĐĐ), the Pham Ngoc Thach Hospital Institutional Review Board (1225/PNT-HĐĐĐ) and Ha Noi University of Public Health Institutional Review Board (300/2020/YTCC-HD3). All participants provided written informed consent and individual-level data were pseudonymized prior to analysis.

Consent for publication

Informed written consent was obtained for all individuals who the study attempted to enroll in SHI, as part of the pilot intervention. It was also obtained for all individuals who participated in the qualitative interviews.

Competing interests

Ten of the authors received salary support from one of the funding agencies to implement the pilot interventions and their evaluation. Two of the authors were employed by United States Agency for International Development (USAID), which funded one of the two pilot interventions. They played no role in the design or implementation of the pilot interventions or their evaluation, but during the development of the manuscript, they provided their insights about the context of the results and Vietnam’s health financing transition as experts in the field.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1..

Mapping of procedures and costs for first-time enrollment into Vietnam's social health insurance scheme.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ . The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Cite this article.

Forse, R., Yoshino, C.A., Nguyen, T.T. et al. Towards universal health coverage in Vietnam: a mixed-method case study of enrolling people with tuberculosis into social health insurance. Health Res Policy Sys 22 , 40 (2024). https://doi.org/10.1186/s12961-024-01132-8

Download citation

Received : 11 August 2023

Accepted : 13 March 2024

Published : 02 April 2024

DOI : https://doi.org/10.1186/s12961-024-01132-8

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Social health insurance
• Universal health coverage
• Health financing transition
• Financial protection

Health Research Policy and Systems

ISSN: 1478-4505

• Submission enquiries: Access here and click Contact Us
• General enquiries: [email protected]

• Open access
• Published: 09 April 2024

Creating culturally-informed protocols for a stunting intervention using a situated values-based approach ( WeValue InSitu ): a double case study in Indonesia and Senegal

• Annabel J. Chapman 1 ,
• Chike C. Ebido 2 , 3 ,
• Rahel Neh Tening 2 ,
• Yanyan Huang 2 ,
• Ndèye Marème Sougou 4 ,
• Risatianti Kolopaking 5 , 6 ,
• Amadou H. Diallo 7 ,
• Rita Anggorowati 6 , 8 ,
• Fatou B. Dial 9 ,
• Jessica Massonnié 10 , 11 ,
• Mahsa Firoozmand 1 ,
• Cheikh El Hadji Abdoulaye Niang 9 &
• Marie K. Harder 1 , 2  

BMC Public Health volume  24 , Article number:  987 ( 2024 ) Cite this article

Metrics details

International development work involves external partners bringing expertise, resources, and management for local interventions in LMICs, but there is often a gap in understandings of relevant local shared values. There is a widespread need to better design interventions which accommodate relevant elements of local culture, as emphasised by recent discussions in global health research regarding neo-colonialism. One recent innovation is the concept of producing ‘cultural protocols’ to precede and guide community engagement or intervention design, but without suggestions for generating them. This study explores and demonstrates the potential of an approach taken from another field, named WeValue InSitu , to generate local culturally-informed protocols. WeValue InSitu engages stakeholder groups in meaning-making processes which ‘crystallize’ their envelope of local shared values, making them communicable to outsiders.

Our research context is understanding and reducing child stunting, including developing interventions, carried out at the Senegal and Indonesia sites of the UKRI GCRF Action Against Stunting Hub. Each national research team involves eight health disciplines from micro-nutrition to epigenetics, and extensive collection of samples and questionnaires. Local culturally-informed protocols would be generally valuable to pre-inform engagement and intervention designs. Here we explore generating them by immediately following the group WeValue InSitu crystallization process with specialised focus group discussions exploring: what local life practices potentially have significant influence on the environments affecting child stunting, and which cultural elements do they highlight as relevant. The discussions will be framed by the shared values, and reveal linkages to them. In this study, stakeholder groups like fathers, mothers, teachers, market traders, administrators, farmers and health workers were recruited, totalling 83 participants across 20 groups. Themes found relevant for a culturally-informed protocol for locally-acceptable food interventions included: specific gender roles; social hierarchies; health service access challenges; traditional beliefs around malnutrition; and attitudes to accepting outside help. The concept of a grounded culturally-informed protocol, and the use of WeValue InSitu to generate it, has thus been demonstrated here. Future work to scope out the advantages and limitations compared to deductive culture studies, and to using other formative research methods would now be useful.

Peer Review reports

Although progress has been made towards the SDG of ‘Zero Hunger by 2025’, the global rates of malnutrition and stunting are still high [ 1 ]. Over the past 20 years, researchers have implemented interventions to reduce undernutrition, specifically focussing on the first 1000 days of life, from conception to 24 months [ 2 ]. However, due to both differing determinants between countries [ 3 , 4 ] as well as varying contextual factors, it is clear that no single fixed approach or combination of approaches can be relied on when implementing stunting interventions [ 5 , 6 , 7 ]. Furthermore, when external researchers design interventions for local areas in Low- and Middle-Income Countries (LMICs) they can often overlook relevant local cultural factors that consequently act as barriers to intervention uptake and reduce their effectiveness, such as geographical factors and the levels of migration in certain populations [ 8 , 9 ], or social norms or perceptions relating to accepting outside help, and power dynamics related to gender [ 10 , 11 , 12 ]. The inclusion of cultural level factors in behaviour change interventions has been proposed as a requirement for effective interventions [ 13 ]. However, despite the breadth of literature highlighting the negative impacts from failing to do this, the lack of integration or even regard of local culture remains a persistent problem in Global Health Research [ 14 ], possibly hindering progress towards the SDGs. Thus, there is a need for approaches to integrate local cultural elements into intervention design.

This lack of understanding of relevant local culture, social norms and shared values also has ethical implications. The field of Global Health Ethics was predominantly developed in the Global North, in High Income Countries (HICs), embedding values common in those countries such as the prominence of individual autonomy [ 15 , 16 ]. Researchers from HICs carrying out research in LMICs may wrongly assume that values held in the Global North are universal [ 14 ] and disregard some local values, such as those related to family and collective decision making, which are core to many communities in LMICs. It is therefore important for outside researchers to have an understanding of relevant local values, culture and social norms before conducting research in LMICs so as not to impose values that do not align with local culture and inadvertently cause harm or offence [ 16 , 17 ]. The importance of this is compounded by the colonial history that is often present in relationships between research communities in HICs and LMICs, and the fact that the majority of the funding and leading institutions are still located in the Global North [ 18 , 19 ]. Thus, conscious steps must be taken to avoid neo-colonialism in Global Health Research [ 20 ]. From a health-equity perspective, it is essential to ensure that those in vulnerable communities are not hindered from involvement in interventions to improve nutrition. Encouraging uptake by such communities could be provided if salient local shared values, norms and culture were taken into account [ 21 ].

In a recent paper, Memon et al., (2021) highlight the usefulness of first creating a cultural protocol that can precede and guide subsequent stages of community engagement or intervention design to ensure that salient local values are known to external researchers coming into the community [ 16 ]. We adopt the use of the concept of a cultural protocol, referring to locally-generated guidance about key values, norms, behaviours and customs relevant to working with the local community. However, we prefer the term, ‘culturally-informed protocol’ since this relates to only cultural elements deemed salient by the researchers, and locally, rather than any comprehensive notion of culture, nor extending beyond the research context.

Memon et al. (2021), point out links between the creation of such a protocol and existing codes of practice that have already been created for some cultures such as the Te Ara Tika, a Guideline for Māori Research Ethics [ 22 ]. Currently, research and interventions in Global Health can be informed by a stage of formative research involving one-to-one interviews, focus groups or direct observations, which can sometimes be ethnographic in nature such as within Focussed Ethnographic Studies or Rapid Assessment Procedures [ 23 , 24 , 25 ]. Although these methods can be effective to inform intervention designs, they have disadvantages like: can take long periods to complete [ 26 ], can be resource intensive [ 26 ] and can lack cultural acceptability [ 27 ]. These limitations may account for the frequent neglect of their use generally, highlighted by Aubel and Chibanda (2022) [ 14 ]. Additionally, none of these methods work towards making explicit local values, or towards the creation of a culturally-informed protocol. In brief, the literature suggests a need to develop alternative methods of Formative Research for understanding locally relevant cultural elements, that are less time-consuming and can generate data that is more easily translatable to intervention design. In addition, these approaches must be applicable in different cultures. Additionally, the protocols produced must be actionable and practical not only for guiding interactions between research teams but also for guiding the initial stages of intervention design.

The work presented here aims to address several of these needs. It includes an exploration of the usefulness of the WeValue InSitu ( WVIS ) approach because that has previously been shown, in environmental management domains, to offer a way to gather in-depth values-based perspectives from a target population [ 28 , 29 ] It was first created through action research, and co-designed to enable civil society organisations to better understand and measure the values-based aspects of their work [ 30 ]. The core WeValue InSitu process (detailed in Table 1 ) involves the crystallization of shared values, with a facilitator guiding a group of participants with shared experiences, through cycles of tacit meaning-making (using a stage of photo-elicitation and triggering) [ 31 ], until they can articulate more explicitly their shared values, in concise and precise statements. These statements are then linked together in a framework by the participants. In an example case in Nigeria, the results of the WVIS approach hinted at the creation of a culturally-informed protocol through an analysis of the shared values frameworks to find cultural themes for the creation of an indicator tool that was used to evaluate several development scenarios based on their social acceptability [ 29 ].

Furthermore, it has been found that if a group of WVIS participants take part in a specialised focus group discussion (FGD), named Perspectives EXploration (PEX:FGD) immediately afterward the main workshop, then they easily and articulately express their perspectives on the topics raised for discussion - and with allusions to the shared values they had crystallised just prior. In an example from Shanghai, the PEX:FGDs focussed on eliciting perspectives on climate change, which were shown to be closely linked with the cultural themes existing within the shared values frameworks produced immediately prior [ 32 ]. In that case, the PEX:FGDs allowed the cultural themes generated during the main WVIS workshop to be linked more closely to the research question. Those results suggested that the WVIS plus PEX:FGD approach could be used to create a specialised culturally-informed protocol for improved intervention design.

In the study presented here, the WVIS approach was explored for the purpose of creating culturally-informed protocols to inform the planning of interventions within two localities of the UKRI GCRF Action Against Stunting Hub [ 33 ]. The work was carried out in two parts. Firstly, the WVIS main workshop was used to elicit cultural themes within the target communities, indicating key elements to consider to ensure ethical engagement. Secondly, the PEX focus group discussions focussed on life practices related to stunting which we explored for the purpose of tailoring the culturally-informed protocols to the specific purpose of improving the design of an example intervention. The Action Against Stunting Hub works across three sites where stunting is highly prevalent but via different determinants: East Lombok in Indonesia (estimated 36% of under-fives stunted), Kaffrine in Senegal (estimated 16% of under-fives stunted) and Hyderabad in India (estimated 48% of under-fives stunted) [ 34 ]. We propose that, the information about local shared values in a given site could be used to inform the design of several interventions, but for our specific exploration the focus here is a proposed ‘egg intervention’, in which pregnant women would be provided with an egg three times per week as supplement to their diet. This study proposes that identifying shared values within a community, alongside information about local life practices, provides critical cultural information on the potential acceptability and uptake of this intervention which can be used to generate culturally-informed protocols consisting of recommendations for improved intervention design.

In this paper we aim to explore the use of the WVIS approach to create culturally-informed protocols to guide engagement and inform the design of localised egg interventions to alleviate stunting in East Lombok, Indonesia and Kaffrine, Senegal. We do this by analysing data about local shared values that are crystallized using the WeValue InSitu ( WVIS ) process to provide clear articulation of local values, followed by an analysis of life practices discussed during PEX:FGD to tailor the culturally-informed protocols for the specific intervention design.

Study setting

This research was exploratory rather than explanatory in nature. The emphasis was on demonstrating the usefulness of the WeValue InSitu ( WVIS ) approach to develop culturally-informed protocols of practical use in intervention design, in different cultural sites. This study was set within a broader shared-values workstream within the UKRI GCRF Action Against Stunting Hub project [ 33 ]. The Hub project, which was co-designed and co-researched by researchers from UK, Indonesia, Senegal and India, involves cohorts of 500 women and their babies in each site through pregnancy to 24 months old, using cross-disciplinary studies across gut health, nutrition, food systems, micro-nutrition, home environment, WASH, epigenetics and child development to develop a typology of stunting. Alongside these health studies are studies of the shared values of the communities, obtained via the WVIS approach described here, to understand the cultural contexts of that diverse health data. In this study the data from East Lombok, Indonesia and Kaffrine, Senegal were used: India’s data were not yet ready, and these two countries were deemed sufficient for this exploratory investigation.

The WVIS approach

The WVIS approach is a grounded scaffolding process which facilitates groups of people to make explicit their shared values in their own vocabulary and within their own frames (details in Fig. 1 and activities in Table 1 ). The first stage of the WVIS is Contextualisation, whereby the group identifies themselves and set the context of their shared experiences, for example, as ‘mothers in East Lombok, Indonesia’. Subsequently, there is a stage of Photo Elicitation, in which the group are first asked to consider what is important, meaningful or worthwhile to them about their context (e.g., ‘being mothers in East Lombok, Indonesia’) and then asked to choose photos from a localised set that they can use as props to help describe their answer to the group [ 29 ]. After this, a localised Trigger List is used. This Trigger List consists of 109 values statements that act as prompts for the group. Examples of these values statements are included below but all the statements begin with “it is important to me/us that…”. The group are asked to choose which statements within the trigger list resonate with them, and those are taken forward for group intersubjective discussion. After a topic of their shared values has been explored, the group begin to articulate and write down their own unique statements of them. These also all begin with “It is important to me/us that…”. After discussing all pressing topics, the group links the written statements on the table into a unique Framework, and one member provides a narrative to communicate it to ‘outsiders’. The WVIS provides a lens of each group’s local shared values, and it is through this lens that they view the topics in the focus group discussions which immediately follow, termed Perspectives EXplorations (PEX:FGDs).

Schematic of the macro-level activities carried out during the WeValue InSitu ( WVIS ) main workshop session

This results in very grounded perspectives being offered, of a different nature to those obtained in questionnaires or using external frameworks [ 31 ]. The specific PEX:FGD topics are chosen as pertinent to stunting contextual issues, including eating habits, food systems and environments, early educational environments, and perceptions of stunting. The local researchers ensured that all topics were handled sensitively, with none that could cause distress to the participants. The data for this study were collected over 2 weeks within December 2019–January 2020 in workshops in East Lombok, Indonesia, and 2 weeks within December 2020 in Kaffrine, Senegal.

The PEX:FGDs were kept open-ended so that participants could dictate the direction of the discussion, which allowed for topics that may not have been pre-considered by the facilitators to arise. Sessions were facilitated by local indigenous researchers, guided in process by researchers more experienced in the approach, and were carried out in the local languages, Bahasa in East Lombok, Indonesia and French or Wolof in Kaffrine, Senegal.

Development of localised WVIS materials

Important to the WVIS approach is the development of localised materials (Table 1 ). The main trigger list has been found applicable in globalised places where English is the first language, but otherwise the trigger lists are locally generated in the local language, incorporating local vocabulary and ways of thinking. To generate these, 5–8 specific interviews are taken with local community members, by indigenous university researchers, eliciting local phrases and ways of thinking. This is a necessary step because shared tacit values cannot be easily accessed without using local language. Examples of localised Trigger Statements produced this way are given below: (they all start with: “It is important to me/us that…”):

…there is solidarity and mutual aid between the people

…I can still be in communication with my children, even if far away

…husbands are responsible for the care of their wives and family

…the town council fulfils its responsibility to meet our needs

…people are not afraid of hard, and even manual work

Study participants

The group participants targeted for recruitment, were selected by local country Hub co-researchers to meet two sets of requirements. For suitability for the WVIS approach they should be between 3 and 12 in number; belong to naturally existing groups that have some history of shared experiences; are over 18 years old; do not include members holding significantly more power than others; and speak the same native language. For suitability in the PEX:FGD to offer life practices with relevance to the research topic of stunting, the groups were chosen to represent stakeholders with connections to the food or learning environment of children (which the Action Against Stunting Hub refer to as the Whole Child approach) [ 33 ]. The university researchers specialising in shared values from the UK, and Senegal and Indonesia respectively, discussed together which stakeholder groups might be appropriate to recruit. The local researchers made the final decisions. Each group was taken through both a WVIS workshop and the immediately-subsequent PEX:FGD.

Data collection and analysis

Standard data output from the WeValue session includes i) the jointly-negotiated bespoke Statements of shared values, linked together in their unique Framework, and ii) an oral recording of a descriptive Narrative of it, given by the group. These were digitized to produce a single presentation for each group as in Fig. 2 . It represents the synthesised culmination of the crystallisation process: a portrait of what was ‘important’ to each stakeholder group. Separately, statements from the group about the authenticity/ownership of the statements are collected.

An illustrative example of one digitized Shared Values Framework and accompanying Narrative from a teacher’s group in East Lombok, Indonesia. The “…” refers to each statement being preceded by “It is important to us that…”

When these Frameworks of ‘Statements of Shared Values’ are viewed across all the groups from one locality (Locality Shared Values Statements), they provide portraits of ‘what is important’ to people living there, often in intimate detail and language. They can be used to communicate to ‘outsiders’ what the general cultural shared values are. In this work the researchers thematically coded them using Charmaz constructionist grounded theory coding [ 35 ] to find broad Major Cultural Themes within each separate locality.

The second area of data collection was in the post- WVIS event: the PEX:FGD for each group. A translator/interpreter provided a running commentary during these discussions, which was audio recorded and then transcribed. The specific topics raised for each group to discuss varied depending on their local expertise. This required completely separate workstreams of coding of the dataset with respect to each topic. This was carried out independently by two researchers: one from UK (using NVivo software (Release 1.3.1)) and one from the local country, who resolved any small differences. All the transcripts were then collated and inductively, interpretively analysed to draw out insights that should be relayed back to the Action Against Stunting Hub teams as contextual material.

The extracts of discussion which were identified as relevant within a particular Hub theme (e.g. hygiene) were then meta-ethnographically synthesised [ 36 ] into ‘Hub Theme Statements’ on each topic, which became the core data for later communication and interrogation by other researchers within the Action Against Stunting Hub. These statements are interpretations of participants’ intended meanings, and links from each of them to data quotes were maintained, enabling future interpretations to refer to them for consistency checks between received and intended meaning.

In this investigation, those Hub Theme Statements (derived from PEX:FGD transcripts) were then deductively coded with respect to any topics with potential implications of the egg intervention. Literature regarding barriers and facilitators to nutrition interventions indicated the following topics could be relevant: attitudes to accepting help; community interactions; cooking and eating habits; traditional beliefs about malnutrition; sharing; social hierarchies [ 12 , 37 , 38 ] to which we added anything related to pregnancy or eggs. This analysis produced our Egg Intervention Themes from the data.

The Major Cultural Themes and Egg Intervention Themes were then used to create a set of culture-based recommendations and intervention specific recommendations respectively for each locality. These recommendations were then combined to form specialized culturally-informed protocols for the egg intervention in each locality: East Lombok, Indonesia and Kaffrine, Senegal. The process is displayed schematically in Fig.  3 .

Schematic representation of the method of production of the culturally-informed protocol for each locality

The preparation of the localised WVIS materials at each site took 6 hours of interview field work, and 40 person hours for analysis. The 10 workshops and data summaries were concluded within 10 workdays by two people (80 person hours). The analysis of the PEX:FGD data took a further 80 person hours. Thus, the total research time was approximately 200 person hours.

The stakeholder group types are summarised in Table 2 . The data is presented in three parts. Firstly, the Major Cultural Themes found in East Lombok, Indonesia and in Kaffrine, Senegal are described – the ones most heavily emphasised by participants. Then, the Egg Intervention Themes and finally, the combined set of Recommendations to comprise a culturally-informed protocol for intervention design for each location. Quotations are labelled INDO or SEN for East Lombok, Indonesia and Kaffrine, Senegal, respectively.

Major cultural themes from frameworks and narratives

These were derived from the Locality Shared Values Statements produced in the WVIS .

East Lombok, Indonesia

Religious values.

Islamic values were crucially important for participants from East Lombok, Indonesia and to their way of life. Through living by the Quran, participating in Islamic community practices, and teaching Islamic values to their children, participants felt they develop their spirituality and guarantee a better afterlife for themselves and their children. Participants stated the Quran tells them to breastfeed their children for 2 years, so they do. Despite no explicit religious official curriculum in Kindergarten, the teachers stated that it was important to incorporate religious teaching.

“East Lombok people always uphold the religious values of all aspects of social life.”

“It is important for me to still teach religious values even though they are not clearly stated in the curriculum.” – Workshop 1 INDO (teachers).

“In Quran for instance, we are told to breastfeed our kids for 2 years. We can even learn about that ” – Workshop 3 INDO (mothers).

Related to this was the importance of teaching manners to children and preventing them from saying harsh words. Teachers stated that it was important to create a happy environment for the children and to ensure that they are polite and well-behaved. Similarly, mothers emphasised the need to teach their children good religious values to ensure they will be polite and helpful to their elders.

“Children don’t speak harsh words.”

“My children can help me like what I did to my parents”.

– Workshop 8 INDO (mothers).

Togetherness within families and the community

The Locality Shared Values Frameworks stressed the importance of togetherness, both within family and community. Comments mentioned it being important that people rely heavily on their family and come together in times of need to support each other and provide motivation. This was also important more broadly, in that people in society should support each other, and that children grow up to contribute to society. This was also reflected in comments around roles within the family. Despite women being primary care givers, and men working to finance the family, participants stated that they follow a process of consultation to make decisions, and when facing hardships.

“that we have the sense of kinship throughout our society”.

“We have togetherness as mothers”.

“For the family side, whatever happens we need to be able to be united as a whole family. We need to have the [sense of] forgiveness for the sake of the children” – Workshop 2 INDO (mothers).

Attitudes about extra-marital pregnancy

In East Lombok, Indonesia, it was essential to both mothers and fathers that pregnancy happened within a marriage, this was to ensure that the honour of the family was upheld and that the lineage of the child was clear. The potential danger to health that early pregnancies can cause was also acknowledged.

“If they don’t listen to parents’ advice, there will be the possibility of pre-marital pregnancy happening, which will affect the family [so much].

The affect is going to be ruining the good name, honour and family dignity. When the children [are] born outside [of] marriage, she or he will have many difficulties like getting a birth certificate [and] having a hard time when registering to school or family” - Workshop 4 INDO (mothers).

“ To make sure that our children avoid getting married at a very young age and moreover [avoid] having free sex so that they will not get pregnant before the marriage” - Workshop 9 INDO (fathers).

Kaffrine, Senegal

The Major Cultural Themes which emerged from the Kaffrine data are described below. As these are grounded themes, they are different than those seen in East Lombok, Indonesia.

Access to healthcare

A recurring theme amongst the groups in Kaffrine were aspirations of affordable and easy-to-access healthcare. Community health workers stated the importance of encouraging women to give birth in hospitals and spoke of the importance of preventing early pregnancy which result from early marriages. Giving birth in hospitals was also a concern for Public Office Administrators who highlighted that this leads to subsequent issues with registering children for school. Mothers and fathers stated the importance of being able to afford health insurance and access healthcare so that they could take care of themselves.

“That the women give birth in the hospital” – Workshop 11 SEN (CHWS).

“To have affordable health insurance ” – Workshop 10 SEN (mothers).

“To have access to health care ” – Workshop 3 SEN (fathers).

“It is important that women give birth in the hospital in order to be able to have a certificate that allows us to establish the civil status” – Workshop 9 SEN (administrators).

Additionally, Community health workers spoke of their aspiration to have enough supplements to provide to their community so as to avoid frustration at the lack of supply, and mothers spoke of their desire to be provided with supplements.

“To have dietary supplements in large quantities to give them to all those who need them, so as not to create frustration” – Workshop 11 SEN (CHWS).

Another aspect of access to healthcare, was mistrust between fathers and community health workers. Community health workers explained that sometimes men can blame them when things go wrong in a pregnancy or consider their ideas to be too progressive. Thus, to these community health workers the quality of endurance was very important.

“Endurance (Sometimes men can accuse us of influencing their wives when they have difficulties in conceiving)” – Workshop 5 SEN (CHWs).

Another recurring theme was the importance of having secure employment and a means to support themselves; that there were also jobs available for young people, and that women had opportunities to make money to help support the family. This included preventing early marriages so girls could stay in school. Having jobs was stated as essential for survival and important to enable being useful to the community and society.

“To have more means of survival (subsistence) to be able to feed our families”.

“To have a regular and permanent job”.

“We assure a good training and education for our children so that they will become useful to us and the community”.

“ Our women should have access to activities that will support us and lessen our burden” – Workshop 3 SEN (fathers).

It was considered very important to have a religious education and respect for religious elders. Moreover, living by, and teaching, religious values such as being hard working, humble and offering mutual aid to others, was significant for people in Kaffrine.

“Have an education in the Islamic Culture (Education that aligns with the culture of Islam)”.

“Respect toward religious leaders” – Workshop 3 SEN (fathers).

“ To organize religious discussions to develop our knowledge about Islam ” - Workshop 10 SEN (mothers).

“ Have belief and be prayerful and give good counselling to people ” - Workshop 4 SEN (grandmothers).

Egg intervention themes from each country from perspectives EXplorations focus group discussion data

Below are results of analyses of comments made during the PEX:FGDs in East Lombok, Indonesia and Kaffrine, Senegal. The following codes were used deductively: attitudes to accepting outside help, traditional gender roles, food sharing, traditional beliefs, social hierarchies and understanding of stunting and Other. These topics were spoken about during open discussion and were not the subject of direct questions. For example, topics relating to traditional gender roles came up in East Lombok, during conversations around the daily routine. Thus, in order to more accurately reflect the intended meaning of the participants, these were labelled food practices, under the “Other” theme. If any of the themes were not present in the discussion, they are not shown below.

Attitudes to accepting outside help

Few mentions were made that focussed on participants attitudes to accepting outside help, but participants were sure that they would not make changes to their menus based on the advice of outside experts. Additionally, teachers mentioned that they are used to accepting help from local organisations that could to help them to identify under-developed children.

“ We don’t believe that [the outsiders are] going to change our eating habits or our various menus ” – Workshop 3 INDO (Mothers).

Traditional gender roles

In East Lombok, mothers spoke about how their husbands go to work and then provide them with daily money to buy the food for the day. However, this was discussed in relation to why food is bought daily and is thus discussed below in the topics Other – Food practices.

Food sharing

In East Lombok, Indonesia, in times when they have extra food, they share it with neighbours, in the hope that when they face times of hardship, their neighbours will share with them. Within the household, they mentioned sharing food from their plate with infants and encouraging children to share. Some mothers mentioned the importance of weekly meetings with other mothers to share food and sharing food during celebrations.

“ Sometimes we share our food with our family. So, when we cook extra food, we will probably send over the food to our neighbour, to our families. So, sometimes, with the hope that when we don’t have anything to eat, our neighbour will pay for it and will [share with] us.” – Workshop 3 INDO (Mothers).

“Even they serve food for the kids who come along to the house. So, they teach the kids to share with their friends. They provide some food. So, whenever they play [at their] house, they will [eat] the same.” – Workshop 2 INDO (Mothers).

Understanding of stunting

The teachers in East Lombok were aware of child stunting through Children’s Development Cards provided by local healthcare organizations. They stated that they recognise children with nutrition problems as having no patience period, no expression, no energy for activities and less desire to socialise and play with other children. The teachers said that stunted children do not develop the same as other children and are not as independent as children who are the proper height and weight for their development. They also stated that they recognise stunted children by their posture, pale faces and bloated stomachs. They explained how they usually use the same teaching methods for stunting children, but will sometimes allow them to do some activities, like singing, later, once the other children are leaving.

“ They have no patience period, don’t have any energy to do any of the activities. No expression, only sitting down and not mingling around with the kids. They are different way to learn. They are much slower than the other kids .” – Workshop 1 INDO (teachers).

“ When they are passive in singing, they will do it later when everyone else is leaving, they just do it [by] themselves ” – Workshop 1 INDO (teachers).

Specific views on eggs

In East Lombok, Indonesia, there were no superstitions or traditional beliefs around the consumption of eggs. When asked specifically on their views of eggs, and if they would like to be provided with eggs, women in East Lombok said that they would be happy to accept eggs. They also mentioned that eggs were a food they commonly eat, feed to children and use for convenience. Eggs were considered healthy and were common in their house.

“ We choose eggs instead. If we don’t have time, we just probably do some omelettes or sunny side up. So, it happens, actually when we get up late, we don’t have much time to be able to escort our kids to the school, then we fry the eggs or cook the instant noodles. And it happens to all mothers. So, if my kids are being cranky, that’s what happens, I’m not going to cook proper meals so, probably just eggs and instant noodles.” – Workshop 3 INDO (Mothers).

Other important topics – food practices

Some detailed themes about food practices were heard in East Lombok, Indonesia. The women were responsible for buying and preparing the food, which they purchased daily mainly due to the cost (their husbands were paid daily and so provided them with a daily allowance) and lack of storage facilities. They also bought from mobile vendors who came to the street, because they could buy very small amounts and get occasional credit. The mother decided the menu for the family and cooked once per day in the morning: the family then took from this dish throughout the day. Mothers always washed their fruits and vegetables and tried to include protein in their meals when funds allowed: either meat, eggs, tofu or tempeh.

“ One meal a day. They [the mothers] cook one time and they [the children] can eat it all day long. Yes, they can take it all day long. They find that they like [to take the food], because they tend to feel hungry.” – Workshop 6 INDO (Mothers).

“ They shop every day because they don’t have any storage in their house and the other factor is because the husband has a daily wage. They don’t have monthly wage. In the morning, the husband gives the ladies the money and the ladies go to the shop for the food. ” – Workshop 4 INDO (Mothers).

In Kaffrine, the following themes emerged relating to an egg intervention: they were different in content and emphasis to Lombok and contained uniquely local cultural emphases.

Mothers were welcoming of eggs as a supplement to improve their health during pregnancy and acknowledged the importance of good nutrition during pregnancy. However, they also mentioned that their husbands can sometimes be resistant to accepting outside help and provided an example of a vaccination programme in which fathers were hesitant to participate. However, participants stated that the Government should be the source of assistance to them (but currently was not perceived to be so).

“But if these eggs are brought by external bodies, we will hesitate to take it. For example, concerning vaccination some fathers hesitate to vaccinate their children even if they are locals who are doing it. So, educating the fathers to accept this is really a challenge” – Workshop 11 SEN (CHWs).

Some traditional gender roles were found to be strong. The participants emphasised that men are considered the head of the household, as expected in Islam, with the mother as primary caregiver for children. This is reflected in the comments from participants regarding the importance of Islam and living their religious values. The men thus made the family decisions and would need to be informed and agree to any family participation in any intervention – regardless of the education level of the mother. The paternal grandmother also played a very important role in the family and may also make decisions for the family in the place of the father. Community Health Workers emphasised that educating paternal grandmothers was essential to improve access to healthcare for women.

“There are people who are not flexible with their wives and need to be informed. Sometimes the mother-in-law can decide the place of the husband. But still, the husband’s [permission] is still necessary.” – Workshop 1 SEN (CHWs).

“[We recommend] communication with mothers-in-law and the community. Raise awareness through information, emphasizing the well-being of women and children.” – Workshop 1 SEN (CHWs).

“The [grand]mothers take care of the children so that the daughters in-law will take care of them in return So it’s very bad for a daughter in law not to take care of her mother in-law. Society does not like people who distance themselves from children.” – Workshop 4 SEN (grandmothers).

Social hierarchies

In addition to hierarchies relating to gender/position in the family such as grandmothers have decision making power, there was some mention of social hierarchies in Kaffrine, Senegal. For example, during times of food stress it was said that political groups distribute food and elected officials who choose the neighbourhoods in which the food will be distributed. Neighbourhood leaders then decide to whom the food is distributed, meaning there is a feeling that some people are being left out.

“ It’s political groups that come to distribute food or for political purposes…organizations that often come to distribute food aid, but in general it is always subject to a selection on the part of elected officials, in particular the neighbourhood leaders, who select the people they like and who leave the others ” – Workshop 11 SEN (CHWs).

Participants explained that during mealtimes, the family will share food from one large plate from which the father will eat first as a sign of respect and courtesy. Sometimes, children would also eat in their neighbour’s house to encourage them to eat.

“ Yes, it happens that we use that strategy so that children can eat. Note that children like to imitate so that’s why we [send them to the neighbour’s house]” – Workshop 11 SEN (CHWs)”.

Traditional beliefs about malnutrition

In Kaffrine, Senegal, some participants spoke of traditional beliefs relating to malnutrition, which are believed by fewer people these days. For example, uncovered food might attract bad spirits, and any person who eats it will become ill. There were a number of food taboos spoken of which were thought to have negative consequences for the baby, for example watermelon and grilled meat which were though to lead to birth complications and bleeding. Furthermore, cold water was thought to negatively impact the baby. Groups spoke of a tradition known as “bathie” in which traditional healers wash stunted children with smoke.

“ There are traditional practices called (Bathie) which are practiced by traditional healers. Parents are flexible about the practice of Bathie ” – Workshop 1 SEN (CHWs).

Causes of malnutrition and stunting were thought to be a lack of a balanced diet, lack of vitamin A, disease, intestinal worms, poor hygiene, socio-cultural issues such as non-compliance with food taboos, non-compliance with exclusive breastfeeding and close pregnancies. Malnutrition was also thought by some to be hereditary. Numerous signs of malnutrition were well known amongst the groups in Kaffrine. For example, signs of malnutrition were thought to be a big bloated belly, diarrhoea, oedema of the feet, anaemia, small limbs and hair loss as well as other symptoms such as red hair and a pale complexion. Despite this, malnutrition was thought to be hard to identify in Kaffrine as not all children will visit health centres, but mothers do try to take their babies heights and weights monthly. The groups were aware of the effect of poverty on the likelihood of stunting as impoverished parents cannot afford food. Furthermore, the groups mentioned that there is some stigma towards stunted children, and they can face mockery from other children although most local people feel pity and compassion towards them. Malnourished children are referred to as Khiibon or Lonpogne in the local language of Wolof.

“ It is poverty that is at the root of malnutrition, because parents do not have enough money [and] will have difficulty feeding their families well, so it is the situation of poverty that is the first explanatory factor of malnutrition here in Kaffrine” – Workshop 9 SEN (administrators).

“It can happen that some children are the victim of jokes for example of mockery from children of their same age, but not from adults and older ” – Workshop 9 SEN (administrators).

Pregnancy beliefs

In Kaffrine, Senegal, there were concerns around close pregnancies, and pregnancies in women who were too young, and for home births. Within the communities there was a stigma around close pregnancies, which prevented them from attending antenatal appointments. Similarly, there were superstitions around revealing early pregnancies, which again delayed attendance at health centres.

Groups acknowledged the role of good nutrition, and mentioned some forbidden foods such as salty foods, watermelon and grilled meat (which sometimes related back to a traditional belief that negative impacts would be felt in the pregnancy such as birth complications and bleeding). Similarly, drinking cold water was thought to negatively affect the baby. Beneficial foods mentioned included vegetables and meat, during pregnancy.

“ Often when a woman has close pregnancies, she can be ashamed, and this particularly delays the time of consultation” – Workshop 5 SEN (CHWs).

“Yes, there are things that are prohibited for pregnant women like salty foods” – Workshop 11 SEN (CHWs).

In Kaffrine, Senegal, some participants spoke of a traditional belief that if a pregnant woman consumes eggs then her baby might be overweight, or have problems learning how to talk. Despite this, mothers in Kaffrine said that they would be happy to accept eggs as a supplement, although if supplements are provided that require preparation (such as powdered supplements), they would be less likely to accept them.

“These restrictions are traditional, and more women no longer believe that eggs will cause a problem to the child. But if these eggs are brought by external bodies, we will hesitate to take it.” – Workshop 11 SEN (CHWs).

“They don’t eat eggs before the child starts speaking (the child only eats eggs when he starts talking). This is because it’s very heavy and can cause bloating and may also lead to intestinal problems.” – Workshop 4 SEN (grandmothers).

Other important topics – access to health services

For the participants in Kaffrine, Senegal, accessing health services was problematic, particularly for pre- and post-natal appointments, which faced frequent delays. Some women had access due to poor roads and chose to give birth at home. Access issues were further compounded by poverty and social factors, as procedures in hospitals can be costly, and women with close pregnancies (soon after an earlier one) can feel shame from society and hide their pregnancy.

“Women really have problems of lack of finances. There are social services in the hospital; but those services rarely attend to women without finances. Even when a child dies at birth they will require money to do the necessary procedure ” – Workshop 11 SEN (CHWs).

Creation of the culturally-informed protocols

Recommendations that comprise a culturally-informed protocol for intervention design in each locality are given in Table 3 .

The Major Cultural Themes, and specific Egg Intervention Themes drawn out from only 9–11 carefully planned group sessions in each country provided a rich set of recommendations towards a culturally-informed protocol for the localised design of a proposed Egg Intervention for both East Lombok, Indonesia and Kaffrine, Senegal. A culturally-informed protocol designed in this way comprises cultural insights which are worthy of consideration in local intervention design and should guide future stages of engagement and provide a platform from which good rapport and trust can be built between researchers and the community [ 16 ]. For example, in Kaffrine, Senegal, the early involvement of husbands and grandmothers is crucial, which reflects values around shared decision making within families that are noted to be more prevalent in LMICs, in contrast to individualistic values in HICs [ 16 , 39 ]. Similarly, due to strong religious values in both East Lombok, Indonesia and Kaffrine, Senegal, partnerships with Islamic leaders is likely to improve engagement. Past studies show the crucial role that religious leaders can play in determining social acceptability of interventions, particularly around taboo topics such as birth spacing [ 40 ].

The WVIS plus PEX:FGD method demonstrated here produced both broad cultural themes from shared values, which were in a concise and easy-to-understand format which could be readily communicated with the wider Action Against Stunting Hub, as well as life practices relevant to stunting in Kaffrine, Senegal and in East Lombok, Indonesia. Discussions of shared values during the WVIS main workshop provided useful cultural background within each community. PEX:FGD discussion uncovered numerous cultural factors within local life practices that could influence on the Egg Intervention engagement and acceptability. Combining themes from the WVIS workshop and PEX:FGDs allowed for specific recommendations to be made towards a culturally-informed protocol for the design of an Egg Intervention that included both broad cultural themes and specific Intervention insights (Table 3 ). For example, in Kaffrine, Senegal, to know that the husband’s authoritative family decision-making for health care (specific) is rooted in Islamic foundations (wider cultural) points to an Intervention Recommendation within the protocol, involving consultations with Islamic Leaders to lead community awareness targeting fathers. Similarly, in East Lombok, Indonesia the (specific) behaviour of breastfeeding for 2 years was underpinned by (wider cultural) shared values of living in Islam. This understanding of local values could prevent the imposition of culturally misaligned values, which Bernal and Adames (2017) caution against [ 17 ].

There are a number of interesting overlaps between values seen in the WVIS Frameworks and Narratives and the categories of Schwartz (1992) and The World Values Survey (2023) [ 41 , 42 ]. For example, in both Kaffrine, Senegal and East Lombok, Indonesia, strong religious values were found, and the groups spoke of the importance of practicing their religion with daily habits. This would align with traditional and conservation values [ 41 , 43 ]. Furthermore, in Kaffrine, Senegal participants often mentioned the importance of mutual aid within the community, and similar values of togetherness and respect in the community were found in East Lombok, Indonesia. These would seem to align with traditional, survival and conservation values [ 41 , 43 ]. However, the values mentioned by the groups in the WVIS workshops are far more specific, and it is possible that through asking what is most worthwhile, valuable and meaningful about their context, the participants are able to prioritise which aspects of their values are most salient to their daily lives. Grounded shared values such as these are generally neglected in Global Health Research, and values predominant in the Global North are often assumed to be universal [ 14 ]. Thus, by excluding the use of a predefined external framework, we minimized the risk of imposing our own ideas of values in the community, and increased the relevance, significance and local validity of the elicited information [ 28 ].

Participatory methods of engagement are an essential step in conducting Global Health Research but there is currently a paucity of specific guidance for implementing participatory methods in vulnerable communities [ 16 , 44 ]. In addition, there is acknowledgement in the literature that it is necessary to come into communities in LMICs without assumptions about their held values, and to use bottom-up participatory approaches to better understand local values [ 14 , 16 ]. The WVIS plus PEX:FGD methodology highlighted here exemplifies a method that is replicable in multiple country contexts [ 28 , 32 ] and can be used to crystallize local In Situ Shared Values which can be easily communicated to external researchers. Coupled with the specialised FGD (PEX:FGD), values-based perceptions of specific topics (in this case stunting) can be elicited leading to the creation of specific Culture-based recommendations. This therefore takes steps to answer the call by Memon and colleagues (2021) for the creation of cultural protocols ahead of conducting research in order to foster ethical research relationships [ 16 ]. We believe that the potential usefulness of the WVIS approach to guide engagement and inform intervention design is effectively demonstrated in this study and WVIS offers a method of making explicit local values in a novel and valuable way.

However, we acknowledge that our approach has several limitations. It has relied heavily on the local university researchers to debate and decide which participant stakeholder groups should be chosen, and although they did this in the context of the Whole Child approach, it would have been advantageous to have involved cultural researchers with a deeper understanding of cultural structures, to ensure sufficient opportunities for key cultural elements to emerge. This would have in particular strengthened the intervention design derived from the PEX:FGD data. For example, we retrospectively realised that our study could have been improved if grandmothers had been engaged in East Lombok. Understanding this limitation leads to suggestion for further work: to specifically investigate the overlap of this approach with disciplinary studies of culture, where social interactions and structures are taken into account via formal frameworks.

There are more minor limitations to note. For example, the WVIS approach can only be led by a trained and experienced facilitator: not all researchers can do this. A training programme is currently under development that could be made more widely available through online videos and a Handbook. Secondly, although the groups recruited do not need to be representative of the local population, the number recruited should be increased until theoretical saturation is achieved of the themes which emerge, which was not carried out in this study as we focussed on demonstrating the feasibility of the tool. Thirdly, there is a limit to the number of topics that can be explored in the PEX:FGDs within the timeframe of one focus group (depending on the stamina of the participants), and so if a wider range of topics need formative research, then more workshops are needed. Lastly, this work took place in a large, highly collaborative project involving expert researchers from local countries as well as international experts in WVIS : other teams may not have these resources. However, local researchers who train in WVIS could lead on their own (and in this Hub project such training was available).

The need for better understanding, acknowledgement and integration of local culture and shared values is increasing as the field of Global Health Research develops. This study demonstrates that the WVIS plus PEX:FGD shared values approach provides an efficient approach to contextualise and localise interventions, through eliciting and making communicable shared values and local life practices which can be used towards the formation of a culturally-informed protocols. Were this method to be used for intervention design in future, it is possible that more focus should be given to existing social structures and support systems and a greater variety of stakeholders should be engaged. This study thus contributes to the literature on methods to culturally adapt interventions. This could have significant implications for improving the uptake of nutrition interventions to reduce malnutrition through improved social acceptability, which could help progression towards the goal of Zero Hunger set within the SDGs. The transferability and generalisability of the WVIS plus PEX:FGD approach should now be investigated further in more diverse cultures and for providing formative research information for a wider range of research themes. Future studies could also focus on establishing its scaling and pragmatic usefulness as a route to conceptualising mechanisms of social acceptability, for example a mechanism may be that in communities with strong traditional religious values, social hierarchies involving religious leaders and fathers exist and their buy-in to the intervention is crucial to its social acceptability. Studies could also focus on the comparison or combination of WVIS plus PEX:FGD with other qualitative methods used for intervention design and implementation.

Availability of data and materials

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request [email protected], Orcid number 0000–0002–1811-4597. These include deidentified Frameworks of Shared Values and Accompanying Narrative from each Group; deidentified Hub Insight Statements of relevant themes.

World Health Organization and Food and Agriculture Organization of the United Nations. Driving commitment for nutrition within the UN Decade of Action on Nutrition: policy brief. Geneva: World Health Organization; 2018.

Google Scholar  

Victora CG, Christian P, Vidaletti LP, Gatica-Domínguez G, Menon P, Black RE. Revisiting maternal and child undernutrition in low-income and middle-income countries: variable progress towards an unfinished agenda. Lancet. 2021;397(10282):1388–99.

Article   CAS   PubMed   PubMed Central   Google Scholar  

Aguayo VM, Nair R, Badgaiyan N, Krishna V. Determinants of stunting and poor linear growth in children under 2 years of age in India: an in-depth analysis of Maharashtra's comprehensive nutrition survey. Matern Child Nutr. 2016;12 Suppl 1(Suppl 1):121–40.

Article   PubMed   Google Scholar  

Beal T, Tumilowicz A, Sutrisna A, Izwardy D, Neufeld LM. A review of child stunting determinants in Indonesia. Matern Child Nutr. 2018;14(4):e12617.

Article   PubMed   PubMed Central   Google Scholar  

Hossain M, Choudhury N, Adib Binte Abdullah K, Mondal P, Jackson AA, Walson J, et al. Evidence-based approaches to childhood stunting in low and middle income countries: a systematic review. Arch Dis Child. 2017;102(10):903–9.

Brar S, Akseer N, Sall M, Conway K, Diouf I, Everett K, et al. Drivers of stunting reduction in Senegal: a country case study. Am J Clin Nutr. 2020;112(Suppl 2):860s–74s.

Heidkamp RA, Piwoz E, Gillespie S, Keats EC, D'Alimonte MR, Menon P, et al. Mobilising evidence, data, and resources to achieve global maternal and child undernutrition targets and the sustainable development goals: an agenda for action. Lancet. 2021;397(10282):1400–18.

Article   CAS   PubMed   Google Scholar  

Goudet SM, Bogin BA, Madise NJ, Griffiths PL. Nutritional interventions for preventing stunting in children (birth to 59 months) living in urban slums in low- and middle-income countries (LMIC). Cochrane Database Syst Rev. 2019;6(6):Cd011695.

PubMed   Google Scholar  

Desai S, Misra M, Das A, Singh RJ, Sehgal M, Gram L, et al. Community interventions with women's groups to improve women's and children's health in India: a mixed-methods systematic review of effects, enablers and barriers. BMJ Glob Health. 2020;5(12)

Muraya KW, Jones C, Berkley JA, Molyneux S. Perceptions of childhood undernutrition among rural households on the Kenyan coast – a qualitative study. BMC Public Health. 2016;16(1):693.

Isler J, Sawadogo NH, Harling G, Bärnighausen T, Adam M, Sié A, et al. 'If he sees it with his own eyes, he will understand': how gender informed the content and delivery of a maternal nutrition intervention in Burkina Faso. Health Policy Plan. 2020;35(5):536–45.

Zaidi S, Das JK, Khan GN, Najmi R, Shah MM, Soofi SB. Food supplements to reduce stunting in Pakistan: a process evaluation of community dynamics shaping uptake. BMC Public Health. 2020;20(1):1046.

Wight D, Plummer M, Ross D. The need to promote behaviour change at the cultural level: one factor explaining the limited impact of the MEMA kwa Vijana adolescent sexual health intervention in rural Tanzania. A process evaluation. BMC Public Health. 2012;12(1):788.

Aubel J, Chibanda D. The neglect of culture in global health research and practice. BMJ Glob Health. 2022;7(9):e009914.

Article   PubMed Central   Google Scholar  

Myser C. Defining "global health ethics": offering a research agenda for more bioethics and multidisciplinary contributions-from the global south and beyond the health sciences-to enrich global health and global health ethics initiatives. J Bioeth Inq. 2015;12(1):5–10.

Memon R, Asif M, Khoso AB, Tofique S, Kiran T, Chaudhry N, et al. Recognising values and engaging communities across cultures: towards developing a cultural protocol for researchers. BMC Med Ethics. 2021;22(1):47.

Bernal G, Adames C. Cultural adaptations: conceptual, ethical, contextual, and methodological issues for working with Ethnocultural and majority-world populations. Prev Sci. 2017;18(6):681–8.

Glickman SW, McHutchison JG, Peterson ED, Cairns CB, Harrington RA, Califf RM, et al. Ethical and scientific implications of the globalization of clinical research. N Engl J Med. 2009;360(8):816–23.

King KF, Kolopack P, Merritt MW, Lavery JV. Community engagement and the human infrastructure of global health research. BMC Med Ethics. 2014;15(1):84.

Horton R. Offline: is global health neocolonialist? Lancet. 2013;382(9906):1690.

Article   Google Scholar  

Woodward EN, Matthieu MM, Uchendu US, Rogal S, Kirchner JE. The health equity implementation framework: proposal and preliminary study of hepatitis C virus treatment. Implement Sci. 2019;14(1):26.

The Pūtaiora Writing Group. Te Ara Tika Guidelines for Māori research ethics: A framework for researchers and ethics committee members. Health Research Council of New Zealand. (no date). Online, Accessed 5 th May 2023. Available from: http://www.hrc.govt.nz/sites/default/files/2019-06/Resource%20Library%20PDF%20-%20Te%20Ara%20Tika%20Guidelines%20for%20Maori%20Research%20Ethics.pdf

Pelto GH, Armar-Klemesu M, Siekmann J, Schofield D. The focused ethnographic study 'assessing the behavioral and local market environment for improving the diets of infants and young children 6 to 23 months old' and its use in three countries. Matern Child Nutr. 2013;9 Suppl 1(Suppl 1):35–46.

Bentley ME, Johnson SL, Wasser H, Creed-Kanashiro H, Shroff M, Fernandez Rao S, et al. Formative research methods for designing culturally appropriate, integrated child nutrition and development interventions: an overview. Ann N Y Acad Sci. 2014;1308:54–67.

Kodish S, Aburto N, Dibari F, Brieger W, Agostinho SP, Gittelsohn J. Informing a behavior change communication strategy: formative research findings from the scaling up nutrition movement in Mozambique. Food Nutr Bull. 2015;36(3):354–70.

Mattes RD, Rowe SB, Ohlhorst SD, Brown AW, Hoffman DJ, Liska DJ, et al. Valuing the diversity of research methods to advance nutrition science. Adv Nutr. 2022;13(4):1324–93.

Robert RC, Bartolini RM, Creed-Kanashiro HM, Verney SA. Using formative research to design context-specific animal source food and multiple micronutrient powder interventions to improve the consumption of micronutrients by infants and young children in Tanzania, Kenya, Bangladesh and Pakistan. Matern Child Nutr. 2021;17(2):e13084.

Sethamo OA, Masika RJ, Harder MK. Understanding the role of crystallizing local shared values in fostering effective community engagement in adaptation planning in Botswana. Clim Dev. 2020;12(5):448–56.

Odii EC, Ebido CC, Harder MK. A values-based approach for generating localized social indicators for use in sustainability assessment and decision-making: test case of brownfield soft reuse in Nigeria. Sci Total Environ. 2020;711:135045.

Podger D, Piggot G, Zahradnik M, Janoušková S, Velasco I, Hak T, et al. The earth charter and the ESDinds initiative: developing indicators and assessment tools for civil society organisations to examine the values dimensions of sustainability projects. J Educ Sustain Dev. 2010;4(2):297–305.

Odii BC, Huang Y, Bouvrie N, Harder MK. Cycles of meaning-making crystallization in the WeValue InSitu process as clear contributions towards transformative learning. J Clean Prod. 2021;304:127024.

Huang Y, Wu W, Xue Y, Harder MK. Perceptions of climate change impacts on city life in Shanghai: through the lens of shared values. Cleaner Prod Lett. 2022;3:100018.

Action Against Stunting Hub. UKRI GCRF action against stunting: alleviating child undernutrition, globally. 2020. URL: https://actionagainststunting.org/ [Accessed 15/08/2022].

UNICEF. Nutrition country profiles . 2010. URL: https://data.unicef.org/resources/nutrition-country-profiles/ [Accessed 10/05/2023].

Charmaz K. Constructing grounded theory. 2nd ed. London: SAGE; 2014.

Noblit GW, Hare RD, Hare RD. Meta-ethnography: synthesizing qualitative studies: sage; 1988.

Stelle I, McDonagh LK, Hossain I, Kalea AZ, Pereira DIA. Nutrients. 2021;13(4):1140.

Muraya KW, Jones C, Berkley JA, Molyneux S. "If it's issues to do with nutrition…I can decide…": gendered decision-making in joining community-based child nutrition interventions within rural coastal Kenya. Health Policy Plan. 2017;32(suppl_5):v31–v9.

Aubel J. Grandmothers — a neglected family resource for saving newborn lives. BMJ Glob Health. 2021;6(2):e003808.

Egeh AA, Dugsieh O, Erlandsson K, Osman F. The views of Somali religious leaders on birth spacing - a qualitative study. Sex Reprod Healthc. 2019;20:27–31.

Schwartz SH. Universals in the content and structure of values: Theoretical advances and empirical tests in 20 countries. Advances in experimental social psychology. 25: Elsevier; 1992. p. 1–65.

World Values Survey Association. Findings and insights, 2020 [Available from: https://www.worldvaluessurvey.org/WVSContents.jsp .]

Inglehart R, Welzel C. Value change and the persistence of cultural traditions. Modernization, cultural change, and democracy: the human development sequence. Cambridge: Cambridge University Press; 2005. p. 48–76.

Glandon D, Paina L, Alonge O, Peters DH, Bennett S. 10 best resources for community engagement in implementation research. Health Policy Plan. 2017;32(10):1457–65.

Download references

Acknowledgements

We thank the Hub PI, Claire Heffernan, for feedback on a late draft of the manuscript.

The Action Against Stunting Hub is funded by the Medical Research Council through the UK Research and Innovation (UKRI) Global Challenges Research Fund (GCRF), Grant No.: MR/S01313X/1.

Author information

Authors and affiliations.

Values & Sustainability Research Group, School of Architecture, Technology and Engineering, University of Brighton, Brighton, UK

Annabel J. Chapman, Mahsa Firoozmand & Marie K. Harder

Department of Environmental Science and Engineering, Fudan University, Shanghai, People’s Republic of China

Chike C. Ebido, Rahel Neh Tening, Yanyan Huang & Marie K. Harder

Department of Zoology and Environmental Biology, University of Nigeria, Nsukka, Nigeria

Chike C. Ebido

Preventive Medicine and Public Health, Université Cheikh Anta Diop (UCAD), Dakar, Senegal

Ndèye Marème Sougou

Faculty of Psychology, Universitas Islam Negeri Syarif Hidayatullah, Jakarta, Indonesia

Risatianti Kolopaking

Southeast Asian Ministers of Education Organization Regional Centre for Food and Nutrition (SEAMEO RECFON) Universitas Indonesia, Jakarta, Indonesia

Risatianti Kolopaking & Rita Anggorowati

International Research Laboratory (IRL 3189) Environnement santé et sociétés/CNRS/UCAD, Dakar, Senegal

Amadou H. Diallo

Department of Medical Records and Health Information, Faculty of Health and Technology, Universitas Bandung, Bandung, Indonesia

Rita Anggorowati

Laboratory of Cultural Anthropology, IFAN, Université Cheikh Anta Diop (UCAD), Dakar, Senegal

Fatou B. Dial & Cheikh El Hadji Abdoulaye Niang

School of Education, Languages and Linguistics, Faculty of Humanities and Social Sciences, University of Portsmouth, Portsmouth, UK

Jessica Massonnié

Department of Learning and Leadership, IOE, UCL’s Faculty of Education and Society, University College London, London, UK

You can also search for this author in PubMed   Google Scholar

Contributions

MKH formulated the initial research question and study design. AJC developed the specific research question. Data collection in Senegal involved CCE, NMS, AHD, FBD, RNT, CEHAN and JM. Data collection in Indonesia involved RA, RK, YH and MKH. Cultural interpretation in Senegal Involved AHD, FBD, NMS, RNT and JM. Analysis involved AJC and MF. AJC and MKH wrote the paper.

Corresponding author

Correspondence to Marie K. Harder .

Ethics declarations

Ethics approval and consent to participate.

The research was conducted in accordance with the Declaration of Helsinki and has been approved by the Ethics Review Board of the University of Brighton, and national ethics committees for research in Indonesia and Senegal. Informed consent was obtained in the local vernacular language, Bahasa, French or Wolof. Participants retained a copy of the informed consent document for reference.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ . The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Cite this article.

Chapman, A.J., Ebido, C.C., Tening, R.N. et al. Creating culturally-informed protocols for a stunting intervention using a situated values-based approach ( WeValue InSitu ): a double case study in Indonesia and Senegal. BMC Public Health 24 , 987 (2024). https://doi.org/10.1186/s12889-024-18485-y

Download citation

Received : 27 September 2023

Accepted : 29 March 2024

Published : 09 April 2024

DOI : https://doi.org/10.1186/s12889-024-18485-y

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Health intervention
• Shared values
• Culturally-informed protocol
• Cultural factors
• WeValue InSitu

BMC Public Health

ISSN: 1471-2458

• Submission enquiries: [email protected]
• General enquiries: [email protected]