Political economy analysis of health taxes (tobacco, alcohol drink and sugar-sweetened beverage): qualitative study of three provinces in Indonesia

Abstract

Objective

Efforts to implement health tax policies to control the consumption of harmful commodities and enhance public health outcomes have garnered substantial recognition globally. However, their successful adoption remains a complex endeavour. This investigates the challenges and opportunities surrounding health tax implementation, with a particular focus on subnational government in Indonesia, where the decentralisation context of health tax remains understudied.

Design

Employing a qualitative methodology using a problem-driven political economy analysis approach.

Setting

We are collecting data from a total of 12 focus group discussions (FGDs) conducted between July and September 2022 in three provinces—Lampung, Special Region of/Daerah Istimewa Yogyakarta and Bali, each chosen to represent a specific commodity: tobacco, sugar-sweetened beverages (SSBs) and alcoholic beverages—we explore the multifaceted dynamics of health tax policies.

Participant

These FGDs involved a mean of 10 participants in each FGD, representing governmental institutions, non-governmental organisations and consumers.

Results

Our findings reveal that health tax policies have the potential to contribute significantly to public health. Consumers understand tobacco’s health risks, and cultural factors influence both tobacco and alcohol consumption. For SSBs, the consumers lack awareness of long-term health risks is concerning. Finally, bureaucratic complexiting and decentralised government hinder implementation for all three commodities.

Conclusion

Furthermore, this study underscores the importance of effective policy communication. It highlights the importance of earmarking health tax revenues for public health initiatives. It also reinforces the need to see health taxes as one intervention as part of a comprehensive public health approach including complementary non-fiscal measures like advertising restrictions and standardised packaging. Addressing these challenges is critical for realising the full potential of health tax policies.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• This study addresses the research gap by exploring the political economy of health tax policy implementation at the subnational level.

• Focus on specific provinces may limit its external validity, even with attempts to diversify the sample across three provinces.

• Lacks insights from elected policy actors, such as governors or the provincial regional legislative council, limiting the understanding of the political economy of health taxes.

• Absence of data post implementation of health taxes on sugar-sweetened beverages, hindering a comprehensive analysis of its impact on this specific commodity.

• Potential selection bias arising from participants recruited through open online invitation and limiting the participants’ age.

Introduction

Health taxes are a crucial policy tool for controlling harmful commodity consumption and improving public health outcomes.¹,⁵ The WHO deems this approach a ‘best buy’ for addressing non-communicable diseases (NCDs).⁶,⁹ However, many low-income and middle-income countries (LMICs) struggle to adopt this policy,¹⁰ prompting global health leaders to emphasise the need for research on health tax policy processes.¹¹

The political economy of a health system shapes policy development and implementation in important ways.¹² Conflicts of interests, poor interagency and multisectoral coordination, as well as excessive bureaucracy were among the most common challenges to fully institutionalising evidence-informed health taxes policy recommendations.¹³,¹⁵ Public sector decentralisation may hinder the health policy adoption due to challenges including limited shared understanding of health tax objectives among local authorities, political complexities and financing issues at the subnational level, resulting in suboptimal implementation.^{14 16} Indonesia being one of the largest countries with escalating concerns in the healthcare sector, primarily due to rising the NCDs epidemic.¹⁷,¹⁹ The country faces challenges in health policy adoption, primarily attributed to political contestation, industry interference and a rigid bureaucratic system, exacerbated by decentralisation.²⁰,²³

Indonesia has been decentralising by increasing greater autonomy for local government and decentralising public finance.²⁴ The decentralisation of public finance encompassing various fiscal policy authorities, such as revenue generation, earmarking and spending mandates, has recently undergone reforms through Law No. 1/2022.^{25 26} Regarding health tax policy, both tobacco and alcoholic beverage taxes have been designed to involve subnational governments.^{27 28} Specifically for the tobacco tax collected through the Ministry of Finance (MoF), that is, excise tax, the national government also enacted legislation to establish a tobacco-tax sharing fund (Dana Bagi Hasil Cukai Hasil Tembakau/DBHCHT), with the purpose of allocation to subnational leaders based on their levels of tobacco farming production.²⁹

Table 1 provides a summary of the division of roles between the national and subnational governments in Indonesia. There are many types of taxes for commodities that fall under ‘unhealthy’ products category:³⁰ (1) Excise Tax, regulated under Law No. 39/2007, in conjunction with Law No. 7/2021; (2) Local Tax, which currently exclusively applies to tobacco products (as part of other five categories outlined in Law No. 28/2009, which have been subsequently replaced by Law No. 1/2022) and (3) Value-Added Tax. In Indonesian case, health tax is more in line with excise tax, as one of its primary objectives is consumption control—rather than solely revenue generation.

Table 1. The division of role in health tax management between national and subnational governments in Indonesia.

Product category	Regulation	Description	Responsibility
Tobacco	Law No. 39/2007 on Excise Tax (as has been replaced by Law No. 7/2021 on Tax Harmonisation)	Article 4: any tobacco product (produced locally or imported) is subject to an excise tax.	National government: determine the tax rate and tax collection.
		Article 66A: any excise tax collected from locally produced tobacco product is to be shared with the producing provinces through tobacco excise sharing fund (DBHCHT).	National government: disburse DBHCHT to the provinces based on the share of their production. Subnational government: manage and allocate the DBHCHT following the mandate of Ministry of Finance (MoF).
	Law No. 28/2009 on Regional Tax and Retribution (as has been repealed by Law No. 1/2022)	Article 33: 10% of the excise tax on tobacco is to be dedicated for local tobacco tax.	National government: collection and disbursement to the local government based on population size (MoF). Subnational government: manage and allocate the local tobacco tax following the law mandate.
Alcoholic beverages	Law No. 39/2007 on Excise Tax (as has been replaced by Law No. 7/2021 on Tax Harmonisation)	Article 4: ethyl alcohol and any beverages containing ethyl alcohol (produced locally or imported) is subject to an excise tax.	National government: determine the tax rate and tax collection.
Sugar-sweetened beverages	N/A	A regulatory bill has been proposed since 2019; delayed implementation.	N/A; current regulatory draft is under MoF.

DBHCHTDana Bagi Hasil Cukai Hasil TembakauN/Anot applicable

Currently, only tobacco and alcoholic beverages are subject to health tax. Although sugar-sweetened beverages (SSBs) were proposed in 2019, these have not yet been implemented.³¹ The national government primarily focuses on tax collection, while the subnational government plays a critical role in tax management and allocation. Additionally, regulations mandate that subnational governments allocate health taxes for the healthcare sector, encompassing both promotive and preventive measures.^{20 32}

With the current practice of health taxation, criticisms have emerged. The health tax initiative relies heavily on the national government with the MoF and Ministry of Health as policy champions.²³ Policy champions influence the political commitment and leadership on the design of the policy implementation, which may result in stronger and more sustained health tax endorsement.¹¹ Furthermore, including the health tax as part of a health-related policy serves as an opportunity in gaining public legitimation.¹⁵ However, the reliance on the national government may be hindered by a limited understanding of the objectives of health taxation and complicated politics implement, particularly among subnational governments.^{22 33} The current practice prioritises revenue generation over consumption control, especially for subnational leaders who aim to protect revenues.²⁰ Finally, decentralisation continues to face several issues, with one of the most significant being financing despite regulations mandating health tax allocation for this sector.^{18 34 35} The financing issues include budget insufficiency and suboptimal public insurance coverage due to disparities in revenue generation across region.³⁵

While decentralisation becomes central to policy implementation, understanding the subnational government behaviour in response to health tax remains important, yet understudied. This study fills this gap by exploring the political economy of health tax policy implementation at subnational level. This study aims to understand the political economy of health tax policy implementation at the subnational level in Indonesia. The study investigates the political economy of three taxable commodities: tobacco, alcoholic beverages and SSBs, focusing on their distinct attributes and characteristics in three provinces. Our study focuses on the subnational level of Indonesia’s political economy of health taxation due to abundant evidence at the national level.^{2021 23 36},³⁹

Material and methods

Study design and case selection

We employed a qualitative study approach as the information to be gathered requires in-depth insights and is highly contextual.⁴⁰ A qualitative study is also more suitable to answer the ‘why’ and ‘how’ questions which make them more explanatory and descriptive,⁴¹ aligning it with the objectives of this research.

The subnational context is documented at the provincial level. We choose the provincial level government because it plays an important role in communicating the central government health policy while simultaneously hold a discretion over districts’ budgeting process.⁴² Each province was selected based on their level of consumption for each commodity (see online supplemental appendix 1). Lampung, Special Region of/Daerah Istimewa Yogyakarta and Bali were chosen for tobacco, SSBs and alcoholic beverages, respectively.

Analytical approach and stakeholders’ identification

We used a problem-driven political economic analysis (PEA) adapted from Overseas Development Institute (ODI) framework^{14 15 43} to guide our analysis on specific policies across a wide spectrum of issues.^{44 45} Figure 1 illustrates our adoption of the problem-driven PEA framework modified from Carriedo et al.¹⁵ We adopted the framework by repeatedly aligning it with our problem statement and findings. We identified the current policy circumstances as a challenge for the health tax implementation at subnational level in Indonesia. The structural issues include the social context and institutional arrangements, while agency issues encompass stakeholders and incentives. In our analysis of both structural and agency issues, we concentrated on organisations and institutions, including their roles, relationships and interactions in the implementation of health taxation policies at the subnational government level. Ultimately, the analysis identifies opportunities and barriers, and proposes policy actions.

Figure 1. Problem-driven political economy analysis approach, adopted from overseas development institute.

We identified three relevant categories of participants: governmental institutions, non-governmental organisations and consumers. Institutional stakeholders were recruited through close invitations, while individual (consumer) participants were recruited through open online invitations. Governmental institutions provide in-depth information about governmental progress and cross-sectoral coordination. For each province, we recruited local government’s officer who were at the managerial level and in charge of: (1) public finance; (2) public health; (3) human capital development and (4) development planning. The non-governmental organisations (NGOs) or experts provide a counterbalance to the governmental institutions. For each province, the NGOs/experts recruited were from: (1) public health or economic experts from public/private higher education institution; (2) local public health or economic NGOs and (3) local faith-based or religious-based organisations. Finally, the consumers, above 18 years old, are the third category of FGD to understand the social context and confirm the local government’s effort in increasing consumer awareness. Exploring consumers’ opinions also helped us gain more information from their perspective.

Data collection

The study used semistructured (see online supplemental appendix 2 for guiding questions) focus group discussions (FGD) to explore a range of views, while observing the social interaction between participants.⁴⁶ Four separate categories of FGDs were conducted in each province for each category of participants (see table 2) which in total involve 12 FGDs between July and September 2022. Specifically for consumers, we organised FGDs for those aged 30 and above separately from those aged 18–30 to understand different behaviours among different age groups. Each FGD lasted approximately 120 min, accommodated a maximum of 15 participants and arranged in a U-shaped room setting. To ensure that participants understood the study’s context, each FGD began with a presentation by the researcher. All FGD activities were audio recorded and complemented with researchers’ notes. We reached data saturation at the last FGD and no additional data were collected.

Table 2. Summary of focus group discussions (FGDs) conducted.

Topic	Participant category	Total number of FGDs conducted (n=12)
Tobacco (Lampung)	Governmental institutions	1
	NGOs and experts	1
	Consumers	2
SSBs (DI Yogyakarta)	Governmental institutions	1
	NGOs and experts	1
	Consumers	2
Alcoholic beverages (Bali)	Governmental institutions	1
	NGOs and experts	1
	Consumers	2

NGOsnon-governmental organisationsSSBssugar-sweetened beverages

Data analysis

The study used transcribed verbatim to conduct a framework analysis.^{47 48} A matrix was created to display the participants’ arguments against the PEA framework, from which we derived keywords to align with the themes and subthemes outlined in the PEA for each commodity category (see online supplemental appendix 3), adopted from Rodríguez et al.¹⁴ Cross-member checking was performed, and the matrix was independently developed by at least two researchers. The matrix appropriateness for each theme, subtheme and categories was iteratively reviewed by all team members. The insights from the findings were extensively contextualised, reflected and confirmed with our main author’s (AA) expertise in the health tax policy context. Regular research meetings were attended to critically assess the findings and resolve disagreements.

Patient and public involvement

The study’s design did not involve general public. Public was involved during the data collection process by participating in the FGD. All participants were informed of the research objectives and were handed consent before participating in the study, to ensure voluntary participation. No input from patients was sought in interpreting or writing up the results. Finally, the research is planned to be disseminated to relevant participating stakeholders.

Results

First, we provide an analysis of the context surrounding the health tax policy. This analysis encompasses both structural and agency-related factors. Second, we explore the barriers and challenges that are anticipated in the implementation of the policy. Finally, we present the recommendations proposed by stakeholders to tackle these identified issues, comparing the characteristics of each commodity.

Structural diagnosis

Structural diagnosis concerns the structural factors that shapes the stakeholders’ behaviour.⁴⁵ We identified two subthemes under the domain of the structural diagnosis in our PEA, that is, social context and institutional arrangement.

Social context

The social context influences health systems, and our case selection focuses on high consumption levels for each commodity in each province to understand the underlying factors influencing consumption and inform necessary policy interventions. The study has identified three recurring keywords or categories. First, cultural factors emerged as a significant determinant. Our FGDs with consumers and NGOs/experts revealed that high consumption rates are predominantly influenced by cultural factors, in addition to price-related factors. Remarkably, this pattern holds true across all provinces, despite variations in the specific commodity category. For example, in Lampung, tobacco consumption is deeply ingrained in local traditions as ‘Tobacco is a cultural item in Lampung during wedding ceremonies’ (NGOs 1, Lampung). This also applies to alcoholic beverages where alcohol consumption is common during celebrations, ceremonies, or religious activities:

Most Balinese people have arak (distilled spirit, traditional alcoholic drinks commonly found in Bali) at home for ceremonies, and there are numerous traditional alcoholic beverage producers in Bali. (NGOs 1, Bali)

Finally, in DI Yogyakarta, like other provinces, the consumption of sweetened foods and beverages appears to be commonplace due to the inclusion of sugary ingredients in many traditional food and beverage items. All three products are easily accessible at nearby stores and come at a affordable price, starting as low as IDR 500 (=US$0.03):

A sachet beverage can be as cheap as Rp 500 per pack, relatively inexpensive compared to daily expenses. (Consumer 1, DI Yogyakarta)

A second aspect of the social context is consumer awareness. Our findings indicate that there may be varying levels of consumer awareness regarding regulated and unregulated commodities. In the case of regulated items such as tobacco and alcoholic beverages, despite high consumption rates, consumers are generally aware of the harm associated with these products, although many remain unaware of the specific purpose of taxing them. Conversely, in the case of SSBs, which are still unregulated, some consumers lack awareness of the long-term disadvantages of SSB consumption. As a common case in Asia, this is especially true for certain types of SSBs marketed as ‘healthy’, such as dairy products or probiotic drinks, which mislead consumers into believing in their health benefits while remaining unaware of the sugar content:

In my house, we usually stock up ready-to-drink tea and coffee and consume them up to three times a week. For probiotic drinks, it can be as much as three times a day (Consumer 2, DI Yogyakarta)

Consumers argued that due to current packaging, it is difficult to easily understand the nutritional content. Moreover, as there is no age limit for SSBs consumption, unlike tobacco and alcoholic beverages, parents make consumption decisions for children in their households, contributing to a relatively high SSBs consumption rate among individuals aged three and above. This common misconception among parents, exacerbates SSBs consumption as it can begin at an early age and lead to developing unhealthy dietary habits. Additionally, in the case of SSBs, advertising and packaging are recognised as two significant factors driving their purchase:

The packaging is attractive, leading to curiosity. As in case for probiotic drinks, [I usually consume] for health (Consumer 3, DI Yogyakarta)

The third aspect is interlinked with the second factor in that consumer awareness shapes the social perception of a commodity Regulated products like tobacco and alcohol are negatively viewed, leading to a stronger desire to quit or reduce consumption. In contrast, SSBs, which have never been marketed negatively, are still seen as relatively normal and harmless commodities, making health taxes implementation more challenging for SSBs compared with tobacco and alcoholic beverages.

Institutional arrangement

The discussion focuses on the institutional arrangement of governmental and non-governmental institutions, their power dynamics, and shared understanding in adopting national government policy, identifying two categories: ideology on health tax objectives and regulation and programmes.

First, the ideological perspective on the health tax objectives focuses on shared policy objectives among institutions, with most agreeing that fiscal measures like health taxes are needed for consumption control, especially among minors and lower-income individuals. Nevertheless, some institutions with a vested interest in industry and trade protection expressed reservations about the concept of health taxes, particularly in the case of alcoholic beverages and SSBs, due to their potential disadvantages to local businesses operating at a micro-to-small scale. This concern is particularly relevant in the case of alcoholic beverages and SSBs, given that two provinces also host local businesses engaged in the production of these commodities:

The sugary food and beverage industry is among the most popular in DI Yogyakarta. When Small and Microenterprises (SMEs) are asked to change their products, they will clearly refuse. Therefore, consumers are encouraged to control their intake. (Dinkes, DI Yogyakarta)

The second aspect of the institutional arrangement pertains to regulations and programmes. Subnational level institutions argued that the national government has the authority to implement fiscal measures to control NCDs, with subnational governments focusing on non-fiscal measures, such as tobacco taxes. This is particularly evident in the implementation of tobacco taxes, where the national government clearly defines the non-fiscal measures to be adopted by subnational governments in accordance with Law No. 17/2023 on Health.⁴⁹ In tobacco and alcoholic beverage cases, these measures include smoke-free area (specific for tobacco) and limit or ban advertising.

In the case of alcoholic beverages, although advertisements are more prominently visible in tourism areas, the sale of arak by micro enterprises remains prevalent. To address the potential impact of arak production, the Governor of Bali issued Governor Regulation No. 1/2020 on the Management of Balinese Fermented and/or Distilled Beverages.⁵⁰ However, the orientation of this regulation is towards industry development, rather than protecting minors from alcohol consumption:

There is Governor Regulation No. 1 of 2020, which aims to be a solution for marginalized Balinese alcoholic beverages [arak]. Many people in Bali used to consume imported beverages. Meanwhile, in Bali, we have our own distinctive Balinese alcoholic beverages, but due to regulations, they seem to be illegal, causing traditional drinks to be undermined. [As a result] they are consumed less by our local community, which is unfair. Balinese beverages, distinct to Bali, have become illegal, while imported ones are considered legal. (DinkopUKM, Bali)

In the case of tobacco, even though Lampung has already established local regulations for Smoke-Free Areas (Kawasan Tanpa Rokok/KTR), tobacco advertisements can still be prominently observed along the main roads. This is compounded by the fact that microenterprises continue to sell tobacco to minors. Furthermore, the implementation of KTR itself remains notably weak:

Many people continue to smoke in places where KTR have been designated. A significant portion of the population remains unaware of the KTR policy, and there are still people who smoke in hospital/health center areas. (NGOs 1, Lampung)

Regarding SSBs, while the fiscal measure has not yet been implemented, the non-fiscal measures that the subnational government can take have not been clearly outlined. The initiative to raise issues related to limiting SSB consumption primarily came from public health government officials or NGOs/experts responsible for public health. Even in the governmental institutions responsible for the education sector or women and child empowerment, these initiatives have not been specifically adopted:

Educational outreach to schools, especially in high schools and vocational schools, regarding health (the dangers of NCDs) has not specifically addressed SSBs. (Disdik, DI Yogyakarta)

Agency diagnosis

Agency diagnosis outlines power, incentives and behaviour.⁴³ These three aspects are closely related to the stakeholders or actors who shape policy implementation. In our case, we are interested in discussing the main stakeholder’s power surrounding the implementation of health taxes at the subnational level, as well as the factors influencing their behaviour (ie, motives) and their interactions. While subnational governments do not have a role in tax collection itself, they play a strategic role in tax allocation and in assisting national government agencies in their areas (eg, the Customs and Excise Office or Kantor Perwakilan Bea dan Cukai/KPBC). We have identified two main aspects of agency diagnosis in our case, namely stakeholders and incentives. See figure 2. for a summary of the stakeholders, their decision-making power and their motives that we have determined from our analysis.

Figure 2. Key actors relationship in supporting health tax implementation at subnational level in Indonesia. Bappeda, Badan Perencanaan Pembangunan Daerah; BPKAD, Badan Perencanaan Keuangan dan Anggaran Daerah; Dinkes, Dinas Kesehatan; Disdik, Dinas Pendidikan; DinkopUKM, Dinas Koperasi dan UMKM; DP3A, Dinas Pemberdayaan Perempuan dan Perlindungan Anak; DPRD, Dewan Perwakilan Rakyat Daerah; Disperindag, Dinas Perdagangan dan Perindustrian; Setda, Sekretariat Daerah.

Stakeholders

This subtheme identifies key stakeholders in the implementation of subnational health taxes, their roles and their interactions (see figure 2). Two topics have been derived from this subtheme, that is, key actors at the subnational level and the power and relationships between actors at the subnational level.

The first aspect comprises actors responsible for decision-making at the subnational level. Based on our FGD findings, we have identified three categories (see figure 2, the colour of each institution differs based on their roles) of actors at the subnational level who can influence the adoption of health tax policies. While they may not be exactly the same across all provinces, they typically share similar sets of institutions/agencies. The first category (yellow coloured) includes actors responsible for strategic policy development and priority setting. At the provincial level, this includes the governor, the provincial regional legislative council (Dewan Perwakilan Rakyat Daerah/DPRD), the regional secretariat (Sekretariat Daerah/Setda), the regional agency of development planning (Badan Perencanaan Pembangunan Daerah/Bappeda) and the regional agency of finance and budgeting (Badan Perencanaan Keuangan dan Anggaran Daerah/BPKAD).

The second category comprises actors concerned with human capital development (orange coloured), including the regional department of health (Dinas Kesehatan/Dinkes), the department of education (Dinas Pendidikan/Disdik) and the department of child protection and women empowerment (Dinas Pemberdayaan Perempuan dan Perlindungan Anak/DP3A). Finally, the third category comprises actors in the economic sector (purple coloured), including the regional department of trade and industry (Dinas Perdagangan dan Perindustrian/Disperindag) and the department for cooperative and micro, small and medium enterprises (Dinas Koperasi dan UMKM/DinkopUKM). In addition to these subnational policy makers, there are actors without decision-making roles who interact with them: the local agency of the Customs and Excise Office—the representative of the MoF in each region—assumes the responsibility for revenue collection and supervision at the local level, and the local agency of the Food and Drug Authority (Badan Pengawas Obat dan Makanan/BPOM), which exercises authority over nutritional value and safety in consumed products.

The second aspect is the power and relationships between the identified stakeholders. In this aspect, we identified four categories (see figure 2). The four categories ranked from highest position in the strategic decision-making include the policy agenda and priority setting, policy coordination and planning, policy budgeting, and policy implementation, respectively.

The first category of actors holds higher positions in terms of regional decision-making compared with the other two categories. This is because they have the authority to set policy targets and priorities, which are then translated into policy planning and budget allocation by Bappeda in collaboration with BPKAD. Additionally, Setda and Bappeda coordinate and supervise intersectoral policy implementation. Actors in the human capital development sector and the economic sector mostly implement policies but can propose programmes for prioritisation before the budgeting process. They must provide evidence for why certain programmes should be prioritised during the annual budgeting and planning process. Finally, the local KPBC does not hold a vertical position with other policymakers. However, in terms of implementing its supervisory role (ie, eradicating illicit trade), it can collaborate with the subnational government.

Motives

This subtheme pertains to the factors that shape the behaviour of stakeholders. We have identified three main keywords that emerged from our analysis in this domain: the actors’ orientation/position towards public health and their incentive to support the health tax policy.

First is the actors’ orientation/position towards public health. To illustrate each institution’s objectives, we have created a spectrum of the actors’ orientation, ranging from public health motives to economic motives (see figure 2). At the implementor level, Dinkes is the main actor with a public health orientation, along with other actors in human capital development. However, they maintain a relatively neutral stance towards the health tax policy issue since it is not their core institutional motive, but they actively engage in public health advocacy:

There are provincial and district programs for child-friendly areas with indicators that include a smoking ban. There are child forums throughout Indonesia that have actively campaigned against smoking with active outreach in schools. DP3A has also conducted anti-smoking campaigns at the neighborhood level, and there are regional regulations supporting smoke-free child-friendly areas. Disdik needs to be involved in raising awareness about the dangers of smoking. (DP3A, Lampung).

While at the right hindside, for Disperindag and DinkopUKM, the orientation is more towards economy, in this case industry protection:

As supporters of MSMEs, we agree because this excise tax applies to large manufacturers. Therefore, it can enable UMKMs to compete with large manufacturers. (DinkopUKM, DI Yogyakarta)

As for the coordinating bodies, BPKAD’s position leans more towards the economic aspect since their motive is to collect revenue for the regional government. We position Bappeda and Setda in a more neutral role since they both house units for human capital policy and economic policy development and primarily function as coordinators:

This tax serves as an instrument to reduce consumption or circulation. I agree that this tax should increase, especially if there is revenue-sharing. The revenue-sharing from the tax can be used for SMEs in the alcoholic beverage industry and for awareness campaigns to improve the understanding that alcoholic beverages are detrimental to health. (Bappeda, Bali)

Finally, political actors like the governor and DPRD are also in a neutral position due to their political motives, which can change with each election year.

Considering each actor’s motive, we can also identify their incentives for supporting and prioritising the health tax policy. For Dinkes, their main performance indicator focuses on enhancing public health status, making them staunch supporters of adopting the health tax. In addition to the potential reduction in unhealthy consumption, this move offers substantial public health benefits. Furthermore, the public health sector anticipates receiving allocations from the tobacco tax and expects similar outcomes for other types of health taxes:

I agree with the health tax and revenue-sharing that can be used to improve the quality and the healthcare sector. (Dinkes, Bali)

As for DP3A and Disdik, their support is somewhat conditional, hinging on the expectation that health tax revenue will be allocated to programmes that bolster the implementation of health taxes. These programmes could include initiatives to raise public awareness about unhealthy consumption in educational institutions or to aid women entrepreneurs in developing healthier product alternatives. These conditions are especially relevant in the case of SSBs, where there are currently no regulations in place to curb consumption, and public awareness is lacking:

For health and for the education of children, women, families in rural areas, and SMEs. Because many of the supported families also produce juices and beverages, so they can have better awareness [on the harm of SSBs]. (DP3A, DI Yogyakarta)

Conversely, Disperindag and DinkopUKM are primarily driven by the motive of safeguarding the local industry. Consequently, their support for health taxes hinges on whether these measures will protect local businesses. For instance, if the tax structure is tailored to the size of the industry, if it contributes to industrial development with an export focus, or if, in the case of SSBs, it serves to educate local industries about adopting healthier alternatives, these factors can influence their support:

It [health tax] can be used for education and development of MSMEs [in the sector]. (DinkopUKM, DI Yogyakarta)

Regarding BPKAD, their support for the health tax is more pronounced when the tax revenue can be earmarked for local government revenue, mirroring the structure of the tobacco tax. In the case of Bappeda, they maintain a more neutral stance towards the incentive, as policy priorities are subject to a bottom-up discussion, beginning with sector-specific priorities and later being communicated and decided on by higher-ranking government entities such as Setda, the governor and the DPRD.

Way forward

In this subsection, we discuss the main barriers and opportunities associated with the implementation of health tax policies at the subnational level. Recommendations for addressing these barriers and optimising opportunities will be discussed in the subsequent section.

Barriers

After conducting the structural and agency diagnosis, we can now identify potential barriers that may impede the implementation of health tax policies at the subnational level. Three main barriers have been identified: a lack of awareness, distrust towards governmental institutions and the existence of product substitution.

First and foremost is the issue of awareness. As discussed previously in the social context, there remains a significant lack of awareness among consumers regarding the health risks associated with the consumption of certain products. This lack of awareness is particularly pronounced in the cases of tobacco and SSBs. Concerning tobacco, the early age at which consumers often initiate smoking, combined with the addictive nature of tobacco, presents a formidable challenge in curbing consumption through the imposition of health taxes. While alcoholic beverages are also considered addictive substances, we do not observe the same patterns of consumption:

It is somewhat impossible that Bali has the highest alcohol consumption rate. Most of it is related to traditional and religious purposes (it should be emphasized whether the factors driving this high alcohol consumption are for traditional and religious ceremonies or for recreational purposes and regular consumption). (NGOs 2, Bali)

Regarding SSBs consumption, the public still lacks a negative perception of these products, unlike the clear aversion to tobacco and alcoholic beverages. Furthermore, consumers currently struggle with understanding how to interpret the nutritional information on packaging and determining a safe sugar intake limit. Consequently, shifting their perspective on these products and raising awareness about limiting their consumption poses a significant challenge.

Second, there is an issue of trust in governmental institutions. Both consumers and institutions share a lack of trust in the government’s intent behind imposing health taxes. Presently, health taxes are predominantly viewed as revenue-generation tools rather than instruments for controlling consumption:

I want this tax to be clearly used for a specific purpose. Let’s not allow it to be used solely for new political targets just to increase revenue. (Consumer 3, DI Yogyakarta)

Hence, consumers and institutions driven by the motive of safeguarding local industries still harbour reservations about the necessity of implementing health taxes, fearing potential harm to the local economy. The absence of transparent policy communication at the subnational level, particularly to consumers, regarding the public health benefits of health tax implementation exacerbates this barrier.

Finally, there is the challenge of product substitution and non-commercial consumption. Consumer substitution towards illicit alternatives represents a potential loss for both public health and revenue generation:

I disagree because if the tax is high, it can trigger illicit tobacco due to low supervision in Indonesia. So, it shouldn't be increased continuously; its upper limit should be assessed. The funds for law enforcement need to be increased to improve supervision. (Bappeda, Lampung)

However, an additional challenge arises from the fact that health taxes are levied on the end product rather than on the main ingredients of those products. In contrast to alcohol, tobacco is only taxed at the final product stage. Consequently, we can still observe instances where consumers purchase tobacco and hand-roll their own cigarettes, a practice known as ‘linting dewe’. In this scenario, the government has little control over the trade of these self-rolled cigarettes, and there are no limits on the quantity of tobacco that can be used, in contrast to commercially marketed products where such limits exist, potentially posing a health risk to the public:

I don’t agree [with the increase in tobacco tax]. If it’s raised, I will still buy, maybe reduce a bit, I can do ‘linting dewe’ or choose cheaper cigarettes, but still stick to stick my smoking habits. (Consumer 3, Lampung)

This is also a concern raised by consumers and NGOs/experts regarding the implementation of taxes on SSBs. Many small-scale SSB manufacturers are unaware of the sugar content in their products. This lack of awareness extends to homemade SSBs as well, where consumers, lacking an understanding of the potential health risks associated with excessive sugar consumption, like homemade soy milk, may add as much sugar as they desire without recognising the potential health drawbacks:

Homemade SSBs are actually more risky because there are no [limit] for the sugar doses. (Consumer 6, DI Yogyakarta)

As a result of health tax implementation, which can lead to higher product prices, consumers are increasingly inclined to consider more affordable alternatives, including homemade options.

Opportunities

Regarding opportunities, we have identified two primary avenues to bolster the implementation of health taxes at the subnational level. The first opportunity lies in garnering public support. While the lack of consumer awareness presents a challenge, there are individuals with a heightened awareness of the adverse health effects associated with tobacco, alcohol and SSBs who wholeheartedly endorse the health tax policy, even if they are consumers themselves. Consumers tend to rally behind policies that promise health benefits rather than revenue generation motives:

I agree, because it relates to the non-communicable disease (NCD) prevention program. Because not all of the population is educated yet. If, for example, the price increases, it will compete with healthier food and drinks that are relatively expensive. But it’s also necessary to raise the base ingredients of sweetened beverages. It should be supported by education. Previous policies should also be maintained to support the tax. (Consumer 7, DI Yogyakarta)

The second opportunity centres around engagement with non-governmental organisations (NGOs). At times, subnational governments lack the initiative to enhance public awareness, potentially undermining the effectiveness of health tax implementation. Local NGOs and academic institutions exhibit greater proactiveness in advocacy efforts. This presents an opportunity because focusing solely on engaging NGOs and academics at the national level may limit outreach to broader segments of society, especially beyond the Java region.

Discussion

Indonesia has encountered challenges in exercising local government autonomy, particularly in implementing fiscal autonomy (revenue generation) and delays in adopting national government regulations.^{20 22 51} Recognising the complex nature of health tax implementation not only horizontally among national actors but also vertically among local actors, we built on our previous work, specifically focusing on national-level policy debates on health taxes.⁵² This is especially important as limited fiscal resources lead subnational governments to prioritise curative measures over preventive or promotive ones.⁵³ The study suggests that promoting health taxes as a funding source for preventive and promotive actions in institutions is likely to gain more support from subnational governments. This insight is especially important as political work to-date on subnational health taxes is limited to a small set of research on SSBs taxes in the USA.⁵⁴ Our findings also align with previous studies emphasising the importance of earmarking health taxes for public health rather than framing them as revenue-generation tools.⁵⁵,⁵⁸ Inspite of that, detailing the well intention of the health tax generation invites counteractive arguments from antitax opposition.⁵⁴

The same principle applies to policy communication. In Indonesia, policies driven by economic motives seem less likely to garner public support, whereas those driven by public health motives have a better chance of gaining popularity. More mixed supports suggested that emphasising the revenue streams for wider purposes, such as education initiatives, appears to be more appealing to voters.⁵⁴ While recent cross-country work in LMICs has painted a rather mixed view about which arguments resonate in which settings, our findings are consistent with previous studies, suggesting that communicating health taxes as an effective measure to combat NCDs is likely to be publicly acceptable.⁵⁹,⁶² However, without providing a convincing moral case for health taxes that resonates with diverse sets of interests, these measures might face challenges in the face of competing arguments from opponents.^{63 64} Therefore, generating evidence to counter arguments from industry proponents should be a part of the policy agenda, as public acceptability is likely to be higher when strong evidence, provided by a credible source, supports a compelling claim.⁶⁴,⁶⁶

In addition to allocating funds for promotive and preventive activities, health tax implementation can be complemented by other fiscal measures that are more applicable for subnational governments. This could include subsidies for healthier alternatives, incentives for cessation⁶⁷,⁶⁹ and regulatory and fiscal support for industries producing healthy products.⁷⁰ Furthermore, health taxes often lead industries to reformulate their products to avoid the tax,^{71 72} which could be an important tax design consideration.⁷³ Therefore, health tax revenue can also be allocated to enable subnational governments to promote and ensure proper product reformulation for healthier alternatives by engaging local enterprises. This is consistent with our finding that institutions with industry protection motives support health taxes because they can benefit local enterprises, particularly for promoting healthier alternatives.

In addition to fiscal measures, non-fiscal measures are equally important. While the majority of past studies have shown a negative relationship between health taxes and unhealthy consumption or NCD prevalence,⁶⁹⁷⁴,⁷⁹ anticipating consumers’ substitution with cheaper but equally harmful alternatives is important.^{67 80} Therefore, increasing consumer awareness to consciously limit their consumption of unhealthy products should be a priority. This can be achieved through extensive efforts to promote a healthy lifestyle through mass campaigns, civil society engagement and women’s empowerment.^{5960 81},⁸³ Our findings support the importance of optimising the engagement of local NGOs and academics, as well as raising awareness among institutions interested in women’s empowerment and children,⁶² such as Disdik and DP3A.

Another critical non-fiscal measure that should be strongly enforced is a complete ban or strict limitations on advertising. While regulatory frameworks for tobacco advertising are in place, lack of local government commitment and weak surveillance systems hinder adoption at the subnational level.^{22 84} While we found no issues with alcohol advertising, our findings suggest potential disadvantages of advertising and packaging information for SSBs. Since there are currently no specific interventions for SSBs, past studies have suggested Front-of-Pack (FOP) or ‘traffic-light’ nutritional labelling.⁸⁵,⁸⁷ This should be accompanied by extensive consumer education to increase their understanding of nutritional labelling, as previous studies have noted a lack of comprehension among consumers.⁸⁸ Furthermore, in the realm of law enforcement, the health tax can also be leveraged to impose more stringent penalties, enhance monitoring efforts and provide counselling for sellers found illegally selling tobacco and alcohol to minors.⁸⁹,⁹¹

Finally, given the cross-sectoral nature of health tax implementation, subnational and national governments must share a common understanding of the policy objectives to avoid policy polarisation. With health financing problems being one of the decentralisation issues facing the healthcare sector in Indonesia,³⁵ framing health taxes as a means of health financing will be crucial to avoid conflicting interests in tax allocation. This should be a clear priority during the planning process to ensure budget alignment and avoid implementation delays.^{14 16 92}

We propose three crucial recommendations for advancing the implementation of health taxes, applicable to both subnational and national governments.

First and foremost is the aspect of policy communication, which holds paramount importance in gaining the trust and support of the public. When communicating the health tax policy, advocates should consider a wider range of arguments in support of health taxes, such as ones based on health impacts, as opposed to prioritising just the revenue-generation potential of these taxes.

Second, the health tax earmarking should be a key consideration. Even though subnational governments may not directly engage in health tax collection, they can actively participate in the allocation of health tax revenues. Although hard earmarking is subject to a debate, a soft earmarking which allows a more flexible funding priorities and aligns with decentralisation, can garner greater public and subnational government support for health taxes.

The study recommends funding for the health sector, including public awareness campaigns, export-oriented industrial development (especially for alcoholic beverages, which may have unique local specialties suitable for export), encouraging healthier product alternatives, subsidies for lower-income households and strengthening law enforcement. It also suggests promoting healthier alternatives through product reformulation, offering subsidies for healthier alternatives, and eradicating the illicit trade of cigarettes and imposing fines on minors selling alcoholic beverages and tobacco.

Finally, non-fiscal measures are equally critical alongside fiscal ones, including advertising bans, reinforcing regulatory frameworks, enforcing standardised packaging, mandatory FoP labelling and extensive promotional campaigns. These measures should extend to SSBs, ensuring that they are marketed with pictorial health warnings and easily recognisable. The government should also consider implementing mandatory FoP labelling to reduce consumer confusion. Similar efforts should be made to raise awareness about health risks associated with SSB consumption.

This study is subject to certain limitations, despite its strengths in analysing the political economy of health taxes for three different commodities (tobacco, alcoholic beverages and SSBs) at the subnational level in Indonesia: its focus on specific provinces and the lack of involvement of elected policy actors like governors or DPRD members. To mitigate this, the sample was expanded to include three provinces and multiple stakeholders from various sectors. The method of recruiting participants through an open online invitation may also be subject to selection bias due to limited demographic representation. Additionally, some potential participants who are unaware of the ongoing research might not have the same probability of being recruited. The study’s scope is limited in terms of SSBs knowledge, as the tax on SSBs had not been implemented at the time of data collection. Future research should focus on SSBs after relevant regulations are implemented to provide a more comprehensive analysis of health tax impact on this commodity.

Conclusion

This study aims to investigate the challenges and opportunities surrounding health tax implementation at the subnational level in Indonesia, a country with significant decentralisation. By exploring the political economy of health taxes for tobacco, alcohol and SSBs, the research aimed to identify lessons applicable to countries facing similar governance structures.

The findings highlight the importance of effective policy communication to garner public support. This includes emphasising the health benefits of health taxes and earmarking the generated revenue for preventive public health initiatives, aligned with the principles of decentralisation. Additionally, addressing concerns from local businesses and promoting the development of healthier alternatives are crucial considerations.

Furthermore, this study underscores the need for a comprehensive public health approach that incorporates health taxes alongside non-fiscal measures like advertising restrictions and standardised packaging. Recognising the limitations of this research, future studies should delve deeper into the political economy of health taxes by comparing other regions in Indonesia or other countries with different social and political context to fully understand their potential for improving public health outcomes in decentralised settings.

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information.