Unpacking Informality: Expanding Retirement Savings Coverage for Informal Workers in Indonesia Ilsa Meidina & Tanya Adi Putri This work is a product of the staff of The World Bank. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions © 2024 The World Bank 1818 H Street NW, Washington DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org Some rights reserved The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. All queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@ worldbank.org. Attribution Please cite the work as follows: “Meidina, Ilsa; and Putri, Tanya Adi; 2024. Unpacking Informality: Expanding Retirement Savings Coverage for Informal Workers in Indonesia.” © World Bank.” Unpacking Informality: Expanding Retirement Savings Coverage for Informal Workers in Indonesia Ilsa Meidina & Tanya Adi Putri Acknowledgements The policy note Unpacking Informality: Expanding Retirement Savings Coverage for Informal Workers in Indonesia was prepared by the Social Protection and Jobs team, Indonesia and Timor Leste under the leadership of Ilsa Meidina and Tanya Adi Putri, with important inputs from Muhammad Noor Farid (Consultant) and Robert Palacios (Lead Economist, HEASP). The team is grateful for excellent comments and feedback from the peer reviewers Melis U. Guven (Lead Economist, HSPGE), Oleksiy A. Sluchynskyy (Senior Social Protection Specialist, HMNSP), and Nga Thi Nguyen (Social Protection Specialist, HEASP). The team thanks Yasser El-Gammal for his oversight and Kurniasih Susilawati for her diligent administrative support. The team extends special thanks to the Ministry of National Development Planning (Bappenas), in particular Directorate for Population and Social Security, for their support and partnership as well as the Coordinating Ministry of Human Development and Culture, Ministry of Manpower, the Fiscal Policy Agency, National Social Security Council, BP Jamsostek, TNP2K and Government of Australia’s Department of Foreign Affairs and Trade for feedback provided during the finalization of this study. Funding for this study was provided through the Australia World Bank Indonesia Partnership (ABIP) by the Government of Australia’s Department of Foreign Affairs and Trade. The team thanks the Australian Government for their generous support. 4 Contents Acknowledgements ............................................................................................................. 4 A. Introduction ............................................................................................................. 6 Informal workers in Indonesia: A Working Definition ..................................................... 7 Analytical Framework and Methodology ........................................................................ 8 B. Who are Indonesia’s Informal Workers? ..................................................................... 10 Demographic Characteristics .......................................................................................... 10 Employment and Socio-Economic Characteristics .......................................................... 11 Behavioral Characteristics ............................................................................................... 12 C. Informal workers’ retirement savings coverage .......................................................... 14 D. Understanding the challenges to retirement savings participation ............................... 18 E. Devising Coverage Expansion Strategy for Informal Workers ...................................... 20 References ........................................................................................................................... 26 Annex 1 - Descriptive Statistics by Sector ............................................................................ 28 Annex 2 - Sector Profiles ...................................................................................................... 33 5 A. Introduction In 2004, Indonesia launched the National Social Security System (Sistem Jaminan Sosial Nasional, SJSN) which, de jure, have made protection against health and employment-related risks available to all workers regardless of their employment status. The schemes include Health Insurance (JKN), Work Injury and Death Benefit scheme (JKK and JKM), a life insurance and protection against work-related injuries, and a defined- contribution old-age savings program (JHT). The salaried workers are also eligible for a defined-benefit pension program, Jaminan Pensiun (JP) which provides annuitized benefit upon retirement. The SJSN fully came into effect in 2015, after the establishment of the BPJS Kesehatan and BP Jamsostek (hereafter, BPJS) as administrators of health social security program and employment social security programs respectively.1 In 2021 the Government launched Jaminan Kehilangan Pekerjaan (JKP), the long-anticipated unemployment insurance which provides temporary income support and access to job counselling and training for salaried workers during unemployment spell. From a design perspective, the current set of social security programs in Indonesia is on the right track in providing income protection along the life-cycle but the remaining challenges include the limited coverage of the tax-financed programs for the poorest and most vulnerable and the low uptake of the contributory programs among informal workers. Expanding social security coverage particularly the retirement savings’ is a challenge faced by many countries with heightened urgency in countries where population ages rapidly and informality is persistently high. Indonesia is no exception. The share of elderly people in Indonesia accounts for 7 percent of the total population in 2023 and is expected to increase significantly over the coming years (UNDESA, 2022). This rapid ageing is unfortunately not followed by retirement savings coverage increase. In 2021, informal workers accounts for 60 percent of total workers in Indonesia and large share of them constitute the so called “missing middle”: those who are not poor enough to be receiving social assistance but is not rich enough to fully participate in voluntary social insurance programs. In real terms, many are also “missing” due to geographical and accessibility challenges. Eight years into its implementation, pension coverage in Indonesia remains low with less than one in ten working age population (15-64) actively contributing to the public retirement saving schemes. Among the working age population, only 7.5 percent of workers are contributing to an annuitized pension program2 while only 9.4 percent are enrolled in a public defined contribution pension (J aminan Ha ri Tu a, JHT). The coverage is especially low among workers in informal employment. Figure 1 Coverage of Jaminan Hari Tua (JHT) JHT Contributors and Coverage 20 9.5% 9.4% 18 16 9.0% 9.0% Coverage of working age 8.9% population (15-64) 14 JHT Contributors 12 8.6% 8.6% 8.5% Millions 10 8.2% 8 8.0% 6 7.8% 4 7.5% 2 0 7.0% 2016 2017 2018 2019 2020 2021 2022 JHT Active contributors JHT Coverage out of working-age populations (15-64) Source: World Bank Staff calculation using UN Population data and BPJS Ketenagakerjaan data 1 BPJS Kesehatan and BPJS Ketenagakerjaan (BP Jamsostek) were formed in 2011 through Law 24/2011. BPJS Kesehatan became fully operational in 2014 while BPJS Ketenagakerjaan in 2015. 2 Provided that they meet the 15 years of minimum contribution requirement at the time of retirement. 6 In 2021, the Presidential Instruction 2/2021 was issued mandating various ministries, government Institutions, and local governments to take necessary actions to increase uptake of the employment social security programs. The medium-term development plan (RPJMN) also set the target of covering more than 30 percent of workers in either formal or informal employment and a plan to subsidize 20 million workers’ contribution to the employment social security by 2024. Furthermore, the social security roadmap issued in June 2023 through Presidential Regulation 36/2023 also sets an ambitious target to cover all workers in employment social security programs by 2030. Despite these regulatory mandates, the increase in retirement savings among informal workers remain low. International evidence suggest that such low uptake is driven by a combination of factors.3 First, informal economy, as also proven through analysis presented in this policy note, are heterogenous which implies that one-size-fits all approach will not work. Secondly, informal workers’ level of income and the frequency of income receipts also varied. Most informal workers are geographically hard to reach, mobile, or are working in a low-skilled sector where digital literacy is low, making outreach efforts difficult. Pre-existing savings behavior, including through informal financial and risk sharing products, also exist which may influence workers’ preference and trust in new, government-sponsored programs. This policy note presents insights synthesized from the 2022 Indonesia’s Retirement Savings among Informal Workers Survey. It aims to unpack informality in Indonesia as it relates to retirement savings decision, following the premise that informality is heterogenous and that retirement savings decision is a function of different demographic, socio-economic, and behavioral characteristics. This note will first explain the analytical framework used to design the survey and the methodology used for data collection, followed by discussion on Indonesia’s informal workers’ characteristics and how such characteristics could inform the design and delivery of retirement savings for workers in the informal sector, using the social insurance delivery chain as the reference point. The analysis presented in this note is also expected to inform the government plans to subsidize the social insurance contribution for informal workers. Options considered include subsidizing the contribution of the short-term benefits program (JKK/JKM) or co-contribution to the defined-contribution retirements saving program (JHT). Focusing the discussion on JHT scheme is considered appropriate given that, to date, JHT is the only retirement scheme accessible to all workers irrespective of the formality of their work arrangements. Informal workers in Indonesia: A Working Definition Indonesia’s Statistical Bureau (BPS) uses primarily employment status to determine whether a worker is formal or informal. Using BPS’ simplified definition4, employer with paid permanent workers and people working as employee and labor are considered formal workers, while self-employed, employer with unpaid worker, and casual and unpaid workers are categorically informal. Under this definition, there are approximately 78 million people in Indonesia working as informal workers in 2021.5 While some choose to engage in the informal economy to avoid taxes, social security contribution, and labor regulations, others stay in the informal economy because of its easy entry, lack of requirement for education and skills, and low capital and technology. Interestingly, Indonesia’s Social Security Laws do not employ the term “informal workers”. The laws differentiate programs’ eligibility criteria based on the presence of employer who are mandated to deduct and co-contribute to social security programs. The term “salaried worker” and “non-salaried worker” are used instead. This implies that even employers of a small or micro enterprises are obliged to enroll, deduct, and pay contributions of their employees. Table 1 shows the de jure eligibility and coverage target of contributory social security in Indonesia. The survey and analysis presented in this note uses the expanded de jure definition of non-salaried workers to include the small/micro enterprise employees given their similarities with the informal, non-salaried workers. 3 See Guven, 2023 4 BPS has a more detailed classification that is based on employment status and position/type of work they do (KBJI – Klasifikasi Baku Jabatan Indonesia), but the simplified definition has been used in the numbers released by BPS. 5 Informal workers comprise of mainly self-employees, which accounts for 35 percent of total informal workers in Indonesia. Other types of informal workers are employer with temporary or unpaid worker (25%), unpaid workers (23%), and casual workers (17%). Agriculture, forestry, and fishing sector is dominated by informal workers, with nearly 90 percent of total workers in the sector being informal. Informal workers are also often found in restaurant and hotel sector (75%) and retail and trade sector (69%) (Labor Force Survey, 2021) 7 Table 1 Indonesia de jure Social Security Coverage by Programs and Workers’ Category Types of Protection Work Injury Defined Defined Health and Life Contribution Benefit Unemployment Insurance Insurance Pension Pension Insurance Administrator • Salaried in large- mandatory medium firm • Salaried in small firm mandatory optional Private sector • Salaried in micro firm mandatory optional BPJS Kesehatan and BP Jamsostek Types of workers • Construction mandatory not available • Non-salaried mandatory optional not available • Migrant worker mandatory optional not available BPJS Kesehatan and • Civil servants mandatory not available sector Public TASPEN BPJS Kesehatan and • Military/ Police mandatory not available ASABRI • Non-workers mandatory not available BPJS Kesehatan Analytical Framework and Methodology Early theories explaining retirement savings assume that individuals and households will accumulate and decumulate assets to maximize their lifetime utilities (Modigliani and Brumberg, 1954; Brumberg, 1957). Individuals begin by estimating, explicitly or implicitly, their life-cycle wealth then choose a spending path that will smooth their spending over the rest of their life (Statman, 2017). Follow on research underpinned by these early thinking shows that retirement savings will depend on economic and demographic factors such as level of income, assets, level of education, access to information, and financial literacy.6 Meanwhile, from behavioral economic point of view, decisions made today to smooth consumption in the future face the problems of self- control, mental accounting, and framing (Shefrin & Thaler, 1988). It is important to note that literatures on long-term savings in the context of informality remains limited. Figure 2 Retirement Savings Determinants Demographic Determinants Retirement Saving Decisions • Age • Intention to save • Education Socio-economic Determinants • Enrollment • Gender • Employment status • Continuous Contribution • Domicile Behavioral Determinants • Income • Financial Literacy • Welfare status • Social Norms • Perceptions Source: Adapted from Wiener & Doescher, 2008; Yusof & Sabri, 2017; Guven, 2023 6 Empirical evidence confirms the relevance of socioeconomic and demographic determinants. Capital area of residence and property ownership seems to affect pension plan participation. These are confirmed by analysis of tax returns in the Netherlands among 20-64 years old (Stinglhamber, Zachary, Wuyts, & Valenduc, 2007), analysis of Survey on Health and Retirement in Europe in Spain and Portugal (Ares, Lopez, & Bua, 2015), in the USA through survey of consumer finance of pre-retired household (DeVaney & Chiremba, 2005), as well as the analysis of Vanguard Group data of 401(k) participants (Huberman, Iyengar, & Jiang, 2007). These studies also assessed the demographic determinants of retirement savings. Level of education and age seem to correlate positively with participation in pension plan based on a study conducted in Spain and Portugal (Ares, Lopez, & Bua, 2015) and in the US (Lum & Lightfoot, 2003). A study using US Survey of Consumer data also revealed that race and gender also correlates with decision to own a pension plan where white male and black male are more likely to save for retirement compared to their hispanic peers (Lum & Lightfoot, 2003). In addition, people with high perceived financial knowledge and financial management capacity are more likely to save for retirement (Chou et al., 2014) while another study also confirms the importance of information and knowledge and conformity with social norms when it comes to enrolling for a retirement savings among US employees (Duflo & Saez, 2003). 8 A survey to investigate the characteristics and participation of informal sector workers in the old-age saving scheme was then designed based on the above framework. The survey asks questions related to demographic, socio-economic characteristics as well informal workers’ perception and preferences on retirement savings. The survey was carried out with the support of a local survey firm under the supervision of the World Bank Social Protection and Jobs team. The self-enumerated, online survey was conducted during the period of March – April 2022. Participants were recruited using purposive and snowball sampling method. The survey was divided into two phases. First, using participant database of World Bank previous surveys and in the second phase, using social media, Facebook. The survey mainly targeted workers in informal employment as well as workers in micro and small enterprises. Small/micro enterprises are characterized by highly informal working engagement and high turnover making the workers in this sector face comparable uncertainties in a level that is similar to that of informal workers. A total of 4525 clean sample were obtained from the country’s 34 provinces. Weighting protocol was applied to create nationally representative data. For this analysis, we excluded respondents who are unpaid workers and those who are outside the labor force (e.g. unemployed persons and family caretaker). A total of 3861 responses were then used in this analysis. Around 61 percent came from Java-Bali region and the rest came from other parts of Indonesia. Sixty nine percent were male workers, and 52 percent were between the age of 25-44. More than half of these workers attain lower-secondary school and below. Figure 3 Overview of survey respondents Regional Distribution Gender 39% Java-Bali 31% Male Outside Female 61% 69% Age-Group Education 30% 25% 27% 25% 20% 33% 17% 35% 15% 15% 19% 18% 10% 12% 0% 20% 40% 60% 80% 100% 5% Primary or below Lower secondary 0% Upper secondary Tertiary 15-24 25-34 35-44 45-54 55+ 9 B. Who are Indonesia’s Informal Workers? This first section will discuss this question by first putting forward the concept of distance to formality. This notion implies, based on Indonesia’s social security regime, that presence of employer and employment contracts are key indicators of how formal a job is. In addition, use of digital intermediation is established as a proxy to formality.7 Workers who perform their works using internet or digital intermediation are identifiable and reachable despite the absence of formal work contract or employer. Based on this premise, micro-small enterprise salaried workers with contract (15.48%), whose working arrangements are most similar to formal employment, are at one end of the spectrum. Salaried workers without contract (15.8%) and non-salaried digital workers (11.08%) are in the middle. Their working arrangements are somewhat informal, but because there is some form of formal intermediaries, these workers are identifiable and easier to reach. Salaried workers without contract are employed by formal enterprises while digital workers are independent workers including gig workers who use digital platforms to find jobs and those who are using websites, including e-commerce platforms, to perform their works or businesses. Finally, furthest from formal employment, sits non-digital non-salaried workers (57.6%). They are self-employees and other form of non- salaried workers who operates in traditional setting, without digital platform or internet’s intermediation. The following discussion will look at four categories of workers namely the salaried workers with contract, salaried workers without contract, digital workers, and non-digital workers. Demographic Characteristics The survey suggests that informal sector in Indonesia is mainly dominated by young, male workers. Approximately 64 percent of workers are below the age of 45 which implies majority of them could contribute to a pension scheme for at least 15 years8 before retirement. Digital workers and salaried workers without contract are the youngest with over 80 percent aged below 45 years old (Figure 4). While half of the workers are low-educated, with 50 percent having completed only primary or lower-secondary schooling, there are variations across sub-types of informal workers. Around 49 percent of digital workers are high-school graduates and 23 percent of them has tertiary degree. This level of education attainment is similar to their peers engaged as salaried workers in micro-small enterprises. Figure 4 Informal Workers by Age Group and Education Level Age Group Education Salaried, with contract 15% 32% 26% 15% 11% 8% 11% 49% 33% Salaried, without contract 20% 37% 25% 12% 6% 18% 24% 47% 11% Non-salaried, digital 19% 37% 25% 11% 7% 14% 14% 49% 23% Non-salaried, non digital 8% 18% 28% 22% 24% 48% 18% 25% 9% TOTAL 12% 25% 27% 19% 18% 33% 17% 35% 15% 0% 20% 40% 60% 80% 100% 0% 20% 40% 60% 80% 100% Primary or below Lower secondary 15-24 25-34 35-44 45-54 55+ Upper secondary Tertiary 7 Traditionally, people working in informal employment are difficult to reach as they tend to be unobservable and disorganized. Social security administrators do not usually have information about where they are located, the type of work they do, or their income. Furthermore, ensuring compliance is difficult as there is no registered employer to mandate and to track the workers’ income. Digital workers are different, as some of their characteristics resemble more to workers in formal employment. Despite not having an employer, a formal institution manages, organizes, and keeps track of the workers and their income. Furthermore, they must comply with the rules set by the institution regardless of their status as employee or independent worker. Therefore, these people are more reachable, and it is possible to mandate and enforce compliance similarly to what has been done in the formal sector. The mandate can be to cover the workers’ social security contribution or to facilitate and ensure the workers under the institution are contributing to social security. 8 The contribution period needed to receive annuitized payment for the public defined contribution scheme Jaminan Pensiun (JP). 10 Employment and Socio-Economic Characteristics Informal workers tend to receive lower and less predictable income. Informal workers earn approximately only half of what their formal peers earn. According to Sakernas 2021, the average income of informal self-employee and casual workers is IDR 1.5 million/month and IDR 1.4 million/month respectively while the average income of formal employees is IDR 2.7 million/month. However, as with the level of education, our survey shows that there are great variations in the level of income as well as its predictability among informal workers. Figure 6 shows the average income and income predictability of different types of informal workers. Salaried workers generally have better income predictability, especially among those with contract. Around 88 percent and 42 percent of salaried workers with contract and without contract respectively receive their income on monthly basis, compared to only 18 percent and 11 percent of digital and non-digital salaried workers. Non-salaried digital workers earn the highest with the average of IDR 5.07 million/month. However, since their income is based on output (or sales) and they do not have secure employment arrangement, this group face uncertainty in maintaining their income. Salaried workers with contract earn slightly more than those without contract, that is IDR 3.27 million and IDR 2.46 million respectively. Traditional non-salaried workers earn the lowest, equivalent to only about one third of their digital counterparts. Figure 5 Average Income and Income Regularity Mean Individual Income Income Payment Salaried, with contract 3.27 8% 49% 2% 2% Salaried, without contract 2.46 20% 29% 42% 9% Non-salaried, digital 5.07 20% 9% 18% 52% Non-salaried, non digital 1.76 28% 21% 11% 40% TOTAL 2.47 21% 19% 32% 28% 0 1 2 3 4 millions IDR 0% 20% 40% 60% 80% 100% Daily Weekly Monthly Output-based Furthermore, informal workers have high financial responsibility and limited government income supports. The survey reveals that 82 percent of respondents are part of sandwich generation who are financially responsible not only for their family but also for their parents. Although most informal workers have health insurance, about a third of them are estimated to be in the bottom two deciles, receiving the Family Hope Program assistant (PKH) and Non-Cash Food Assistance (BPNT) or other food assistance programs. The share of cash transfer recipients is highest among non-salaried-non-digital workers, followed by salaried workers without contract. Figure 6 Social Assistance Receipts Social Assistance Recipient 70% 60% 59% 52% 53% 52% 50% 46% 39% 37% 40% 35% 35% 36% 33% 30% 28% 21% 20% 16% 16% 10% 0% TOTAL Non-salaried, Non-salaried, Salaried, Salaried, non digital digital without contract with contract PKH and BNPT PKH, BNPT, or COVID-19 SA PBI-JKN 11 Behavioral Characteristics Financial capability of informal workers, comprising of financial literacy and financial capacity, also differs. With regards to financial literacy whereby ownership and usage of bank account are often used as main indicators, the analysis shows that although more than half of informal workers own and actively use bank accounts, about the same proportion of workers do not actually understand basic concepts in retirement savings, case on point, compound interest, as assessed by asking a simple question on expected return of a hypothetical savings. It is also interesting to note that despite low objective financial literacy, most workers are confident about their ability to handle financial problems, signaling some degree of overconfidence bias. Further, the level of objective financial literacy seems to differ among different types of informal workers with those engaged in digital works showing better results. Figure 7 Objective and Subjective Financial Literacy Objective Financial Literacy Subjective literacy: Handling financial problems TOTAL 51% 49% 29% 22% 49% Non-salaried, digital 37% 63% 12% 17% 71% Non-salaried, non digital 57% 43% 38% 20% 42% Salaried, with contract 41% 59% 18% 23% 59% Salaried, without contract 50% 50% 21% 29% 50% 0% 20% 40% 60% 80% 100% 0% 20% 40% 60% 80% 100% Incorrect answer Correct answer not confident confident very confident Confidence bias is also observed when assessing financial capacity to save, measured by calculating the difference between household income and household spending.9 Based on the survey, 59 percent of informal workers claimed that given their current income and spending pattern, they are not able to save. However, analysis on the actual household income and main expenses suggest that 77 percent of workers still have some capacity to save and among those with real ability to save, they can set aside on average up to 25 percent of their household income. Although one cannot assume that this saving space is completely unallocated, this mismatch highlights that ‘perceived inability to save’ is potentially a prevalent mental barrier that should be addressed accordingly, especially among those who are engaged in traditional type of informal works. Equally pertinent is perception on the importance of retirement savings and old age responsibilities. Although majority of informal workers agree that retirement savings is very important, most still think that their children and spouse will take care of them when they are old. This entrenched filial piety10 is more prominent among agriculture and construction workers while those working in more formal sector has better sense of self sufficiency during old age. When data is disaggregated based on types of work, again, those using digital technology have better sense of self-sufficiency during old age and tend to agree that they themselves are responsible for their well-being during old age. Meanwhile, analysis using logistic regression shows that the likelihood of having retirement savings, both in general form and specifically in BPJS-JHT, is strongly associated with the type of employment, family welfare, and financial literacy.11 Workers with higher degree of formality, i.e. salaried workers with contract and non-salaried-digital workers are more likely to have retirement savings than other types of workers.12 9 Household income includes spouse’s income, money earned from investment and any government cash transfer. Household expenditure consists of two components, which are food expenditure (such as rice, meat, eggs, vegetables fruits, eating-out, etc., including cigarettes) and non-food expenditure (such as rent, electricity, fuel, child education, healthcare, clothings, phone credit, etc.). 10 Filial piety is a Confucian concept derived from Chinese culture, which advocates a set of moral norms, values, and practices of respect and caring for one’s parents. This concept is prevalent and widely observed in Asian culture. 11 The model includes demographic, employment, financial literacy, welfare and asset ownership, view on retirement savings, and knowledge of social security variables. 12 In the model workers are classified into 4 categories: (1) salaried worker with contract; (2) salaried worker without contract; (3) non-salaried worker who are not utilizing digital platform to perform their work; and (4) non-salaried worker who are utilizing digital platform. 12 Having higher and regular income, bank account and health insurance are also strong determinants for retirement savings ownership. Finally, workers who belong to households receiving social assistance are less likely to have retirement savings. Savings for retirement in general form – not necessarily in BPJS scheme – is also affected by perception on the importance of retirement savings and savings habit in general. Workers with perceived ability to save and those who have other forms of savings (such as general savings) are more likely to save for retirement. Similarly, strong perception on the importance of retirement savings is associated with retirement savings ownership. Interestingly, worker’s actual savings space (real ability to save) does not seem to affect the likelihood of one contributing to retirement savings. Finally, when it comes to retirement savings specifically in BPJS scheme, knowledge of BPJS and knowing someone who is a member of BPJS are associated with savings for retirement in BPJS-JHT. Figure 8 Perceived vs Real Saving Ability Perceived Saving Ability Real Saving Ability Salaried, with contract 46% 54% 23% 77% Salaried, without contract 59% 41% 27% 73% Non-salaried, digital 40% 60% 19% 81% Non-salaried, non digital 66% 34% 23% 77% TOTAL 59% 41% 23% 77% 0% 20% 40% 60% 80% 100% 0% 20% 40% 60% 80% 100% No Yes No Yes 13 C. Informal workers’ retirement savings coverage Coverage of retirement savings among informal workers in Indonesia is low and varies across different types of jobs. On average, 43 percent of workers have some form of retirement savings or assets intended to support during retirement, but most workers are still saving for their retirement in general savings account and only around 17 percent of workers are contributing to a retirement saving scheme either in one managed by BPJS or other private retirement savings plan. Fairly high retirement savings coverage is found among salaried workers with contract and digital non-salaried workers. Coverage is highest among salaried workers in micro-small enterprises who work under a contract. Approximately 74 percent of these workers have some form of retirement savings and 50 percent have retirement savings under BPJS. Among digital non-salaried workers, retirement savings coverage is equally high, at 64 percent, however, most are not saving with BPJS retirement savings scheme. Coverage is lowest among those engage in non-digital, non-salaried works. Figure 9 Share of Workers with Retirement Saving and Saving in BPJS-Managed Scheme by type of jobs Workers with Retirement Savings by Type of Jobs 100% 90% 80% 74% 70% 64% 60% 50% 50% 45% 43% 40% 31% 25% 30% 18% 17% 20% 10% 6% 0% Non-salaried, Non-salaried, Salaried, Salaried, TOTAL non digital digital without contract with contract Have retirement savings Savings specifically at BPJS Retirement savings ownership also vary greatly across sectors. Analysis shows that in agriculture, only 30 percent of workers have some form of retirement savings while, in contrast, 67 percent in hospitality and 66 professional service sectors do. Of those claiming to save for retirement, only 8 percent of agriculture workers contribute to BPJS-managed scheme, while the highest coverage can be found among workers in the manufacturing sector. Figure 10 Workers with Retirement Savings by Sector Workers with Retirement Savings by Sector 80% 70% 66% 67% 60% 60% 54% 52% 50% 44% 40% 43% 40% 39% 30% 32% 32% 32% 30% 28% 20% 18% 16% 19% 17% 10% 8% 8% 0% Agriculture Transportation Retail and Manufacturing Restaurant and logistic trade and hotel Construction Other services Energy and Professional All-sector extractive industry services Have retirement savings Savings specifically at BPJS 14 Analysis also suggests that the varying level of coverage are consistent with different demographic and socio- economic characteristics, the workers’ financial capability, as well as perceptions and values related to old age and retirement. Agriculture and construction are dominated by elderly workers with the median age of 43 and 42 respectively. A third of them are 55 years old which means that they will not yield significant retirement income if they start contributing to a pension scheme today. In addition, agriculture and construction workers tend to have low level of education. Half of the workers in these sectors only completed elementary school. This is a contrast when compared with informal workers who work in professional services sector which tend to have higher education whereby around 67 percent have tertiary degree. Agriculture and construction can be seen as the most challenging sectors in expanding retirement savings coverage. Workers in these sectors have the lowest score on capability to accumulate savings, reachability, financial literacy, and knowledge of social security compared to all other sectors. In the agriculture sector 65 percent of workers are considered low-income and 64 percent are in poor and vulnerable households13 with 88 percent not having predictable income. Reaching this sector is difficult as 75 percent of the workers are in rural area. Furthermore, bank account use is also low among agriculture workers. About 70 percent of them either do not have bank account or have an account but have not used it in the past year and 72 percent perceived that they are unable to save regularly. The same share of workers are also without other forms of savings. Interestingly, 70 percent of workers in agriculture have heard of BPJS, but 63 percent of them have never heard of JHT program. Finally, 77 percent of them do not know anyone who is participating in BPJS. Similarly, construction workers also tend to earn low and unpredictable income. Around 60 percent of the workers are also in poor/vulnerable households. When it comes to financial literacy, 59 percent of the workers do not use bank account and 76 percent do not have other types of savings. Their knowledge of BPJS is about the same as the national average but when it comes to JHT, 64 percent of workers in the sector have never heard of the program. In Agriculture sector, where financial literacy, capability to accumulate savings, and knowledge of social security is low, social pension for older workers and subsidized contribution for those who still have enough time to accumulate savings can be considered as one of the interventions to improve coverage. Construction sector face similar challenges to agriculture – although not as difficult. In addition to demographic and socio-economic challenges, these workers typically have relatively short-term employment and despite the mandatory social security coverage for construction worker in work-injury and life insurance program, these workers cannot participate in the retirement savings scheme even voluntarily or out of their own expenses. To improve retirement savings coverage among construction workers access to retirement savings program should be considered, either as part of mandatory protection or as a top-up by worker with subsidized contribution as incentive. Figure 11 Construction and Agriculture Workers’ Informality Workers without network of BPJS participant Knowledge of Workers without knowledge of JHT Social Security Workers without knowledge of BPJS Workers who perceive retirement savings low-moderately important Financial Workers with perceived inability to save literacy Workers without any other types of savings Workers with low bank account ownership and use Workers with low level of formality Reachability Workers in rural area Workers with unpredictable income Capability Workers in the bottom 40% for savings accumulation Low-income workers (below 1.8 mln) Older workers (45+) 0.0 0.2 0.4 0.6 0.8 1.0 Construction Agriculture 13 In this analysis workers are considered low income when they earn less than IDR 1.8 million/month, which was the lowest provincial minimum in Indonesia in 2022, and are considered poor and vulnerable if they received subsidized health insurance, which implies that they are in the bottom 40. 15 Meanwhile, analysis shows that workers in digital economy have more consistent relation between perception, intention, and the actual participation in JHT. This signifies that those engaged in digital economy is a low-hanging fruit when it comes to expanding retirement savings coverage (See Box A for discussion on gig workers retirement savings preference). To date, there is yet a legal mandate to facilitate social security registration, contribution deduction, or co-contribution for digital gig work platform or e-commerce operating in Indonesia. Efforts are on voluntary basis, depending on the platforms’ willingness, and are often limited to registration per se. Box A – Expanding Retirement Savings to Gig Workers The survey estimated that around 8% of informal workers in Indonesia are full-time gig workers. The term gig works refers to short-term, non-permanent type of work that involves delivering task or ‘gig’ that is facilitated by digital platform. The study also reveals that most of gig workers in Indonesia are providing location-based services mainly in urban settings (63%). Common tasks include transporting personal items (reported by 44% of gig workers), transporting people (35%), running errands such as grocery shopping service (29%), and logistic services (18%). Meanwhile, a small percentage of gig workers seemed to be engaged in non-location-based works providing administration and data input (10%), creative and multimedia (8%), and other professional services (5%) as indicated by the platforms they use which includes Freelancer.com and Sampingan.co.id. Majority of the gig workers cited that they engage in platform works for its flexibility but at the same time, most of them work for more than 40 hours a week. Many also choose to do gig works as side job and to overcome the income shock during COVID19. Indonesian gig workers have distinct characteristics when compared with their “traditional” self- employed peers. Gig workers in Indonesia are younger and better-educated with most of them completing at least upper secondary education. Their financial capability is relatively high compared to non-gig informal workers as marked by ownership and usage of bank accounts (62 percent), understanding of financial and investment concepts (41 percent), and expressed confidence in performing financial tasks (64 percent). When it comes to income, gig workers earn 57 percent more than their non-gig, self-employed peers. In terms of savings, gig workers are able to set aside money for monthly savings, on average, twice as much compared to their self-employed peers. Most do so by savings in a general savings scheme. Despite higher financial literacy, income, and propensity to save, gig workers in Indonesia are categorically vulnerable as most of them do not receive social assistance and are uncovered for employment-related risks. The study reveals that only 34 percent of gig workers have precautionary savings and around 60 percent of them are struggling to meet their financial obligation (e.g. mortgage, debt). Only 17 percent of gig workers receive main social assistance programs although the coverage is higher for the subsidized health insurance. Meanwhile, participation in employment social security programs among gig workers is low despite the fact that the programs are de jure available for all workers without exception. Categorized as “non-salaried workers”, gig workers are eligible for three contributory social security programs: life insurance, work injury, and old-age savings but in absence of employers, the workers have to register and pay their contributions themselves. High financial literacy and awareness of the importance of retirement savings does not seem to translate to higher participation in contributory programs. Only around 33 percent of gig workers are enrolled in any social security scheme and the coverage is even lower for retirement savings at merely 17 percent. To extend social protection coverage to gig workers, awareness-raising and innovative design could be pursued. Lack of knowledge on program benefits and eligibility coupled with perceived inability to pay the contribution are two most-cited reasons for non-participation in social security schemes. The existing information gaps need to be urgently addressed and platforms can be engaged to help disseminate information on social security programs, the eligibility criteria, and the benefit. Some location-based platforms in Indonesia like Gojek and Grab are already facilitating participation in social security schemes by mandating deduction for work accident and death benefits from the workers’ e-wallet. 16 Participation in the retirement savings scheme, however, are still entirely voluntary. Simple choice experiment embedded in the survey suggests that subsidization (either in the form of matching- contribution or direct contribution subsidy) and allowing more frequent payment of contribution in smaller amounts would make the retirement scheme more appealing to the gig workers and to informal workers in general. Digital platforms can play a more active role in encouraging participation of gig workers in social security programs. Offering some level of organization to the otherwise unorganized sector (Sluchynskyy, 2021), the digital platforms have the technological capacity to conduct massive outreach activities – even individually-tailored framing and messaging – to encourage enrollment and contribution payment. Similarly, digital platforms could also help roll out a gig workers-friendly social security design which could allow, for example, automatic enrollment, small yet frequent contribution deduction, and payment reminders. In the long run, the digital platforms, capitalizing on their fin-tech innovation capacity, could also set up micro-pension scheme not only for gig workers but also for informal workers in general. Inevitably, these changes will require regulatory adjustments which necessitates a strong concerted efforts from all relevant stakeholders involved in the design and implementation of social security in Indonesia. 17 D. Understanding the challenges to retirement savings participation This section will discuss the challenges to retirement savings as experienced and perceived by the informal workers in Indonesia. From the survey, unavailability of funds is the most-cited reasons for not saving for retirement in general (expressed by 84% of respondents) while non-participation in JHT among informal workers are due to factors related to the design of the program. Reasons for not participating in JHT scheme includes unfamiliarity with JHT program (mentioned by 32% of respondents), perceived ineligibility (24%), low understanding of the program (20%), concerns around ability to pay in the future (14%), and concerns over irregular income (12%). While lack of knowledge can be addressed by designing fit-for-purpose communication and outreach strategy, the remaining concerns will require substantial rejigging of the existing JHT parameters which includes revisiting the scheme’s prerequisites, contribution rate, and contribution payment frequency and modalities. A complex set up of terms and conditions for participating in retirement savings scheme may have also, inadvertently, contributed to the low uptake. The current regulation requires informal workers to save for JKK/ JKM programs first before they can participate in the JHT program. The JKK/JKM contribution rate also follows a schedule that increases progressively based on declared income which may discourage workers from applying to the JKK/JKM schemes which in turns affect their decision to enroll in JHT scheme or the amount they are willing to save for their retirement. Put simply, an informal worker will have to pay more for short-term risks protection if they want to save more for their retirement. The survey also suggests that there is non-negligible cognitive bias prevailing among informal workers with regards to responsibility for old age care. This bias is most likely due to cultural norms whereby children are expected to take care of their parents when they are old. The survey also reveals perception – intention – action gap as shown by the high understanding of the importance of retirement savings which do not translate to corresponding rate of participation in retirement savings in general and in JHT scheme in particular. Such gaps are due to low level of knowledge on JHT, concerns or worries that one will not be able to contribute continuously or regularly to the scheme, as well as the general perceived inability to save. Meanwhile, a regression analysis based on the survey data also revealed four main challenges hampering retirement savings uptake, namely capability for savings accumulation, reachability, financial literacy, and knowledge of social security. Table 3 summarizes the prevalence of these challenges as faced by workers in different sectors. As can be inferred from the table, encouraging workers in agriculture and construction workers to save for retirement savings will require significant efforts. Workers in this sector are poor, and most of them are categorically in the bottom 40% (as indicated by social assistance recipient status) and most of them are older workers which signifies shorter remaining period of contribution. On the other end of the spectrum, informal workers offering professional services or those working in manufacturing sector experience relatively low challenges for long term savings. When existing one-size-fits-all strategy does not yield significant increase in coverage, it is timely to employ a sectoral approach where possible, which will also be in support of the implementation of Presidential Instruction No. 2/2021. 18 Table 2 Challenges in Retirement Savings by Sector14 Energy and extractive industry Transportation and logistic Restaurant and hotel Professional services All-sector average Retail and trade Manufacturing Other services Construction Agriculture Challenges Capability for savings accumulation Older workers 0.34 0.43 0.20 0.28 0.40 0.26 0.10 0.30 0.18 0.38 Low-income workers 0.50 0.65 0.38 0.24 0.40 0.43 0.36 0.43 0.38 0.52 Workers in the bottom 40% 0.53 0.64 0.42 0.43 0.60 0.50 0.39 0.46 0.42 0.46 Workers with unpredictable income 0.75 0.88 0.49 0.53 0.85 0.82 0.51 0.74 0.26 0.74 Reachability Workers in rural area 0.47 0.75 0.45 0.27 0.42 0.35 0.28 0.36 0.39 0.33 Workers with low level of formality 0.61 0.82 0.35 0.25 0.67 0.58 0.32 0.41 0.30 0.63 Financial literacy Workers with low bank account ownership 0.51 0.70 0.50 0.31 0.59 0.42 0.25 0.39 0.25 0.47 and use Workers without any other types of savings 0.63 0.72 0.72 0.47 0.76 0.50 0.37 0.66 0.48 0.70 Workers with perceived inability to save 0.60 0.72 0.61 0.49 0.67 0.49 0.36 0.60 0.48 0.60 Workers who perceive retirement savings 0.20 0.22 0.23 0.15 0.22 0.16 0.17 0.27 0.15 0.21 low-moderately important Knowledge of Social Security Workers without knowledge of BPJS 0.22 0.30 0.22 0.19 0.22 0.16 0.19 0.21 0.08 0.20 Workers without knowledge of JHT 0.50 0.63 0.38 0.37 0.64 0.40 0.35 0.48 0.35 0.46 Workers without network of BPJS participant 0.64 0.77 0.55 0.54 0.71 0.55 0.44 0.57 0.48 0.62 The numbers show the share of workers within respective sector who show certain characteristics (e.g., 43 percent of workers in the 14 agriculture sector are older workers). Indicators above benchmark or above 50% are shaded in red, indicating that the challenge is faced by more than 50% of the informal workers in respective sector. The benchmarks are national all-sector average. 19 E. Devising Coverage Expansion Strategy for Informal Workers Based on the analysis presented earlier and drawing from international experiences, the following policy actions can be pursued to accelerate the retirement savings’ coverage expansion efforts among informal workers in Indonesia. The recommendations are divided into two parts: ones that concern the design issue which in turn will require revision of higher-level regulation and ones that concern the delivery systems which would relatively be easier to implement. Overcoming the design issues 1 Conduct reforms of the existing parameters of the JHT program. Flexibility in program design is essential for informal workers’ scheme. Intended to bundle social security programs together, Presidential Regulation 109/2013 mandates participation in JKK/JKM programs before workers can participate in JHT. While bundling programs together is a good initiative and can be attractive when communicated well, workers with limited financial space might opt-out altogether from participating in JHT when participation in the other two programs requires them to pay a higher amount of contribution. In combination with the bundling requirement, how contribution for non-salaried workers is designed also compromises the adequacy of the savings. Under Government Regulation 44/2015 and 46/2015, contribution for non-salaried workers is decided based on a contribution schedule with a minimum monthly income set at IDR 1,000,000. Work injury insurance and old-age savings (JHT) contributions are at around 1 and 2 percent respectively, while life insurance premium is a flat rate of IDR 6,800. As the programs are tied together, a worker who would like to save more in JHT are required to pay more for work-injury insurance. This could potentially discourage worker to commit more for their savings, as saving more also means spending more on the insurance part. To improve JHT coverage and adequacy, the Government could reconsider the bundling requirement and providing workers with the freedom to choose social security program that best serve their individual needs. As an alternative, keeping work-injury contribution at the minimum level or flat rate without tying it to JHT contribution is an option that can be considered. Providing more flexibility in payment frequency and contribution amount might also be desirable for informal workers whose income and availability of jobs are often irregular. Box B summarizes findings from the survey-based experiment to elicit preference of informal workers when it comes to retirement savings. Box B. Choice Experiment: Which Scheme Attribute is Preferred? The choice experiment study aimed to identify potential incentives and delivery system improvement to increase workers’ participation in the old-age savings program. We presented respondents with two hypothetical old-age savings scheme packages with three attributes: the amount of monthly contribution, payment frequency, and the incentive they would receive while contributing to the scheme. Each of these attributes had equal opportunity to be selected. We asked respondents to choose their preferred scheme. They may also choose not to participate in either of the savings scheme. How did people respond to the offer? The experiment showed that 65 percent of the respondents would take up the scheme with preference on one or more attributes (contribution amount, payment frequency, and/or incentive), 16 percent would take up the scheme without any preference on these attributes, while 19 percent chose not to participate in the scheme. 20 . A B Attributes • Contribution amount 20 000 40 000 • Payment frequency Flexible Fixed • Incentive Savings bonus Work-injury and life insurance Which one would you choose? Savings A Savings B Neither Among those who based their choice on certain attribute(s), what matters? We found that 35 percent chose based on a combination of two or three attributes while 65 percent chose based solely on one specific attribute, which is one of the following: contribution amount (26%), the type of incentive (22%), and payment flexibility (17%). Lower amount of contribution is important, especially for those who showed strong preference on this attribute (i.e., people who based their choice only on contribution amount). Around 83 percent of respondents with strong preference chose schemes with lower amount of contribution, while 17 percent chose schemes with higher amount. When compared with respondents who based their choice on a combination of other attributes, only 63 percent of people in this group chose the lower contribution while 37 percent chose higher amount. Similarly, flexible payment is also important. Around 80 percent of respondents with strong preference chose flexible payment over fixed monthly payment (as compared with 60 percent among respondents with mixed attribute preference). Finally, work-injury and life insurance are generally preferred as the type of “bonus” respondents would like to have. However, this preference is not as strong compared with the other two attributes. 2 Expand social pension for the poorest and provides incentives to those who are better off. International experience suggests that subsidization of premium for pension help increase the coverage among poor and informal workers. For example, in addition to the employer annuity scheme, China and Colombia pay the entire premium for the poorest and more than half of the contribution for the non-poor. In the late 2000s, the Korean government paid half the contribution of farmers resulting in participation of the target group of more than 70 percent. Similarly, Thailand has been able to attract contributions from almost 20 percent of their informal sector workers while most informal workers in China are now contributing to their voluntary scheme. There is a clear relationship between the size of the subsidy and the take up rate, although other factors also play a role including trust and transaction costs. The amount of government subsidy for employment social security programs in Indonesia is still limited. To date, the government only subsidizes the training benefit of the unemployment insurance program while incentives to improve uptake of retirement savings program is virtually nonexistent. The Investing in People Report (World Bank, 2020) provides costs estimates of subsidizing contribution in select social protection schemes for the bottom 40% and for all adults as summarized in Table 3 below. 21 Table 3 Incremental Costs of Expanding Social Coverage through Contribution Subsidy Government contribution/ transfer offered to Current cost of Bottom 40% of all All adults Government adults contribution/ Schemes transfer Work Accident (JKK) and Death No government 0.03% 0.12% Benefit (JKM) contributions given to individuals currently enrolled Pension No subsidy given 0.18% 0.71% to those currently enrolled Health 0.2% 0.20% 0.49% Social assistance for elderly (65+) 0.02% 0.07% 0.17% Total cost 0.22% 0.58% 1.49% Note: Adults in our calculations include all individuals aged 20-64, i.e., 159 million as of 2019. JKK and JKM calculations assume govt. contribution equal to 0.54 percent of average monthly wage of IDR 2 million paid to all adult/bottom 40 percent of adults. Note that under current law JKK/ JKM/JP are mandated to cover only a subset of all adults. Health coverage increase would involve covering the remaining 80 million individuals. Pension contribution by the government is equal to 3 percent of IDR 2 million monthly and this contribution today would give future retiree a pension equivalent to IDR 200,000 monthly. Social assistance for elderly cost in third column is for 70 percent of those aged 65 and above as discussed in Chapter 2. 3 Introduce regulation to allow efficient identification of coverage gaps among different social insurance programs. Previously done specifically for Job-Loss Guarantee (JKP) program,15 data matching can be mandated for all social insurance programs. With the interoperable social security database, BPJS Employment should be able to identify members of BPJS Health who have not taken part in the employment programs and allows a more targeted approach. An integrated registration and contribution collection platform should also be created to streamline the delivery process. For non-salaried workers, the load of doing registration and making contribution payment is the responsibility of each individual. It is a burdensome process and, often, time consuming that it is inefficient to do separate process for each health insurance and employment social security. An integrated platform could simplify this process while at the same time allowing people to learn about the other programs that they might not be aware of previously. The unified social security database should ideally be interoperable with national ID and social registry as are the case in several countries. For example, Turkey created a unified information system of three different social security institutions. By using only citizenship identification number and date of birth, members can gain quick access to information on pension and health insurance status, registration, and premiums, without the need to make burdensome trips to the social security office or accessing different platforms. Updates are made in the system automatically every time there is a change in a member’s entitlements (payment of contributions, change in insurance status, benefit payments, etc.). Another good example of this exercise includes The National Pension Service of the Republic of Korea’s introduction of the Social Insurance Information Centre in 2003, which allows citizens to access all four social security schemes (pensions, health, employment and workers’ compensation) by visiting in person or virtually any of the three administering organizations. The integration of information systems simplifies administrative procedures for the insured persons as they can easily make inquiries and change, request or submit claims. At the same time, the administration has made significant efficiency gains by simplifying and streamlining various application forms and documents. Not to mention, a unified view of social protection programs also allows identification of individuals who are eligible for specific interventions, such as contribution subsidy and matching contribution, should the government introduce them in the future. The data matching is mandated through Government Regulation 37/2021 15 22 Tackling the delivery challenges 4 Adjust the social insurance delivery chain to attune to informal workers’ characteristic and preference. Improvements along social protection delivery chain can be pursued to make the system more accessible for informal sector workers. This means addressing challenges identified during the decision-making process, the enrollment, and the process of paying subsequent contribution. Figure 12 Journey Towards a Persistent Savings: Aer Pathway (Adapted from Guven, 2023) Decision to engage with saving product • Knowledge of the product • Forming intention Consistent Contribution • Overcoming bias and trust issue • Habit formation Enrollment and First Contribution • Identifying how to enrol • Choosing amount to commit Decision to engage in retirement saving product. As identified in the analysis, the decision to engage with the saving product is affected by; 1) lack of knowledge of the program and its benefits, and 2) perception and bias on the importance of retirement savings and capacity to save. Informal workers are not aware that they are eligible to participate in JHT despite having the intention to save for retirement. To close the apparent perception – intention – action gaps, in addition to improving the quality and modality of outreach activities, “framing” addressing the prevalent beliefs could play an important role. For example, BPJS as administrators could reframe the affordability of the contribution by comparing the amount needed to contribute to JHT to costs of enjoyments. Meanwhile, the strong prevalence of filial piety can be moderated by using outreach messages that is framed around the importance of preparing for the future for the ‘sandwich generation’. Information on JHT should be presented in the simplest way. For example, it would be appropriate to start framing JHT as a retirement savings with life insurance and work accident protection as additional benefits vis a vis positioning JKK/JKM contribution as prerequisite to JHT participation. Further, as the likelihood of someone savings for JHT increases when knowing other people contributing to the scheme, administrator could also design an informational material that can be easily forwarded with certain incentives system. International experience also suggests that proactive outreach can be implemented to reduce the barriers to information. Madagascar introduced mobile office to improve access to information for people living in rural and remote areas. This approach is very useful as it allows social security agency to provide tailored information, services, and recommendations that are suitable to the needs of the people who are otherwise left behind. Cabo Verde implemented targeted outreach to “easy-to reach” workers such as independent workers in urban areas and other informal workers in concentrated areas such as traders and fish sellers. Collaboration with workers’ associations was also established to facilitate contact with informal workers. For a quick win, short-term focus should be shifted to young workers in urban area and to those engaged in digital economy. The digital platforms provide some organization to the otherwise unorganized sector.16 The platforms have the ability to reach out to its captive audience and could disseminate information on social security programs including JHT in a concerted manner. There are currently more than 20 million workers who depend on digital platforms in Indonesia17, making them the low-hanging fruit for coverage expansion efforts. Granted that collaboration with the digital platforms would be required. Shluchynsky, 2022 16 “20 Million Workers Depend on Digital Platforms” https://www.kompas.id/baca/english/2023/07/25/en-20-juta-pekerja-bergantung-pada- 17 platform-digital, accessed on May 22nd 2024 23 It is also important to note that certain segments of informal workers are hard to reach. They are those who have low education background, low financial literacy, low income and inability to save where any outreach strategy would yield merely marginal success. In this context, contribution subsidy can be considered to incentivize savings for retirement in addition to expanding social pension for the poorest and most vulnerable. China, for example, provides subsidy to encourage savings among informal workers both in urban and rural area. In the case of Indonesia, agriculture and constructions are sectors where most informal workers are, and they are the hardest to reach. As such, subsidization of their pension contribution either fully or partially could be considered. For informal workers who are better off, such subsidization can be framed as enrollment bonus which can be claimed upon meeting minimum contribution duration. This type of incentive will also help form the saving habit. Ultimately, decision to engage in a retirement savings scheme should also be nurtured among future workers. Education on the importance of retirement savings and informal workers’ eligibility for JHT should be conducted not only for workers but to those who are about to enter the workforce, in schools and universities. As an example, Uruguay has integrated social security education in the curriculum starting from a very young age. Social security institution also organizes activities to inform and educate people on the role of the system in society as well as their rights and responsibilities. Meanwhile, Rwanda’s pension scheme allows parents to open an account for children under 16 years old, which then allows children to learn about the scheme early while at the same time encouraging parents to contribute regularly. Enrollment Decision. With regards to decision to enroll, which also entails making decision about the amount to save, the analysis reveals two main barriers namely predictability of income and the perceived inability to save. To respond to these challenges, registration and contribution payment should be made easy and flexible and could include behavioral economics components. More specifically, existing registration process could be made simpler by integrating the registration interface with other government services, most notably the health insurance (JKN). Similar to what has been implemented in Peru for digital workers, automatic enrolment is also an option that can be pursued whereby participation in JKN will automatically activate one’s membership in relevant employment social security programs. That said, the existing incompatibilities whereby JKN employs registration on household basis while employment social security uses individual registration should be addressed first. In the long run, full integration of JKN and employment social security programs’ registration process should be pursued. Equally important, database interoperability will also help a more systematic identification of coverage gaps among the different programs offered under SJSN. In addition, informal workers’ need to constantly have access to some kind of emergency savings should also be acknowledged. For example, Ghana’s retirement savings scheme features two-account system in which contribution is divided equally between the long-term and short-term account. The short-term account can then be withdrawn after 5 months of membership. Rwanda’s retirement savings scheme allows members to have access to 25% of their retirement savings account for liquidity and allow 40% of the total savings to be used as a collateral to obtain a loan. In addition, participants are provided with life and funeral insurance. Indonesia’s recently-passed Omnibus Law on Financial Sector Deepening has paved a way for similar approach to be implemented in Indonesia but the proportion of savings for each account needs to be determined carefully to ensure that adequacy for old age remains preserved. Forming a habit. The informal workers’ irregularity of income amount and receipts, as also shown in the analysis, and the absence of employers who can deduct and make contribution on the workers’ behalf will make continuous contribution a challenge. In the case of Indonesia’s JHT, this is also observed in the high number of inactive participants due to irregular contribution payment. In 2023, inactive participant accounts for 61 percent of total non-salaried worker registered in the scheme18. A more flexible contribution collection and payments arrangement can be introduced. Introduction of commitment devices such as automatic deduction for digital platform workers, digital business owners, and those with mobile money may overcome the issue of irregular contribution payments. Automatic deductions also help close the perception-intention-action gaps. Where appropriate, incentives in the form of matching BPJS Ketenagakerjaan Report to Financial Service Authority, as of December 2023 18 24 contribution can also be introduced to first get people enroll in the system and to incentivize more savings. In the long-run, introducing micro-pension feature enabling irregular amount and frequency of payments could also be considered. Innovations can be introduced to encourage contribution even for workers without access to digital or mobile technology. For example, Philippines introduced collection box, installed around workplace to allow workers to save small amounts in an individual slot when they have money available, allowing for daily or weekly deposits of small amounts. On a monthly basis, the amounts are counted using computer-generated payment transaction reports. In Indonesia, similar approach can be pursued by leveraging the existing agent/ aggregator model (PERISAI) where workers are allowed to pay their contribution in variable amount through the agents who will also be responsible in generating periodic report. 5 Clarify the social security mandate of digital economy platforms. To date, there is yet a legal mandate to facilitate social security registration, contribution deduction, or co- contribution for digital gig work platform or e-commerce operating in Indonesia. Efforts are on voluntary basis, depending on the platforms’ willingness, and are often limited to registration per se. Granted that digital platform workers are categorically non-salaried workers but given the size of the sector, its contribution to the economy, and its high reachability, it is timely for the Government to start tapping into the digital platforms in delivering the social security and pension programs not only to workers affiliated with their platform but also informal workers in general. Finally, it is important to acknowledge that reforms in the design and delivery of the retirement savings schemes will require a concerted effort from various stakeholders and may have to be implemented gradually. Table 4 below summarizes the recommendation and the potential timeline of its implementation. Table 4 Summary of Recommendations Recommendations Short-term Medium Term Long term Introduce more flexible package of programs (choices or more affordable contribution rate for JKK/JKm). Design Expand social pension for the poorest and provides incentives to those who are better off. Enforce data interoperability among different social insurance programs to identify coverage gaps. Employ framing addressing bias and behavioral barriers in outreach efforts to improve uptake and registration. For a quick win, target massive coverage expansion efforts to young, urban, digital workers. Delivery Introduce behavioral tools to establish savings habit (e.g. automatic deduction, small and irregular contribution payments) Develop educational materials/curriculum to instill importance of social security among future workers (middle and high school curriculum). Clarify the social security mandate of digital economy platforms 25 References Anderson, S., & Baland, J.-M. (2002). The Economics of Roscas and Intrahousehold Resource Allocation. Beshears, J., Choi, J. J., Laibson, D., Madrian, B. C., & Milkman, K. L. (2010). The Effect of Providing Peer Information on Retirement Savings Decisions. RAND Corporation. https://www.rand.org/pubs/working_papers/WR800. html Bikhchandani, S., Hirshleifer, D., & Welch, I. (1998). Learning from the Behavior of Others: Conformity, Fads, and Informational Cascades. Journal of Economic Perspectives, 12(3), 151–170. https://doi.org/10.1257/ jep.12.3.151 Carpena, F., Cole, S., Shapiro, J., & Zia, B. (2011). Unpacking the causal chain of financial literacy (Policy Research Working Paper Series 5798). The World Bank. https://econpapers.repec.org/paper/wbkwbrwps/5798.htm Choi, J. J., Laibson, D., Madrian, B. C., & Metrick, A. (2002). Defined Contribution Pensions: Plan Rules, Participant Choices, and the Path of Least Resistance. Tax Policy and the Economy, Volume 16, 67–114. Cole, S., Sampson, T., & Zia, B. (2011). Prices or Knowledge? What Drives Demand for Financial Services in Emerging Markets? The Journal of Finance, 66(6), 1933–1967. https://doi.org/10.1111/j.1540-6261.2011.01696.x DeVaney, S. A., & Chiremba, S. T. (2005). Comparing the Retirement Savings of the Baby Boomers and Other Cohorts. US Bureau of Labor Statistics, 12. Duflo, E., & Saez, E. (2002). The Role of Information and Social Interactions in Retirement Plan Decisions: Evidence from a Randomized Experiment. NBER Working Paper, 31 Fernandes, D., Lynch, J., & Netemeyer, R. (2014). Financial Literacy, Financial Education, and Downstream Financial Behaviors. Management Science. https://doi.org/10.1287/mnsc.2013.1849 Guven, M., Jain, H., & Joubert, C. (2021). Social Protection for the Informal Economy: Operational Lessons for Developing Countries in Africa and Beyond. World Bank. https://doi.org/10.1596/36584 Guven, M., and Jain, H. (2023). The Promise of Ejo Heza: A Brighter Future for All Rwandans. ©World Bank. License: CC BY 3.0 IGO Hong, H., Kubik, J. D., & Stein, J. C. (2004). Social Interaction and Stock-Market Participation. The Journal of Finance, 54(1), 27. IADB, I.-A. D. (2016). Saving for Development: How Latin America and the Caribbean Can Save More and Better. Springer. ILO. (2015). The Transition from the Informal to the Formal Economy. ILC 104/2015, Report V (2B). ILO Publications. ILO. (2018). Women and men in the informal economy: a statistical picture (third edition) / International Labour Office – Geneva: ILO, 2018 ILO. (2019). Extending social security coverage to workers in the informal economy: Lessons from international experience. International Labour Office Geneva. Karlan, D., McConnell, M., Mullainathan, S., & Zinman, J. (2016). Getting to the Top of Mind: How Reminders Increase Saving. Management Science, 62(12), 3393–3411. https://doi.org/10.1287/mnsc.2015.2296 Karlan, D., Ratan, A. L., & Zinman, J. (2014). Savings by and for the Poor: A Research Review and Agenda. Review of Income and Wealth, 60(1), 36–78. https://doi.org/10.1111/roiw.12101 Packard, T., Gentilini, U., Grosh, M., O’Keefe, P., Palacios, R., Robalino, D., & Santos, I. (2019). Protecting All: Risk Sharing for a Diverse and Diversifying World of Work. The World Bank. https://doi.org/10.1596/978-1-4648- 1427-3 Prina, S. (2015). Banking the poor via savings accounts: Evidence from a field experiment. Journal of Development Economics, 115, 16–31. https://doi.org/10.1016/j.jdeveco.2015.01.004 26 Schaner, S. (2018). The Persistent Power of Behavioral Change: Long-Run Impacts of Temporary Savings Subsidies for the Poor. American Economic Journal: Applied Economics, 10(3), 67–100. https://doi.org/10.1257/ app.20170453 Shefrin, H. M., & Thaler, R. H. (1988). The Behavioral Life-Cycle Hypothesis. Economic Inquiry, 26(4), 609–643. https://doi.org/10.1111/j.1465-7295.1988.tb01520.x Sluchynskyy, O. Approaches to Extending Social Protection Coverage to Digital Platform Workers. Presentation at APEC’s Workshop on Social Protection for Digital Economy, August 2021 Song, C. (2020). Financial Illiteracy and Pension Contributions: A Field Experiment on Compound Interest in China. The Review of Financial Studies, 33(2), 916–949. https://doi.org/10.1093/rfs/hhz074 United Nations, Department of Economic and Social Affairs, & Population Division. (2020). World population ageing, 2019 highlights. 27 Annex 1 Descriptive Statistics by Sector Demographic Characteristics Share of Informal Workers by Sector 35 30.2 30 25 20.3 20 18.5 15 10 8.1 6.7 5.9 5.9 5 2.3 2.1 0 Agriculture Other services Energy and Retail Professional extractive industry and trade services Construction Transportation and logistic Manufacturing Restaurant and hotel Location All-sector 53% 47% Professional services 61% 39% Restaurant and hotel 72% 28% Retail and trade 65% 35% Manufacturing 73% 27% Energy and extractive industry 55% 45% Transportation and logistic 64% 36% Other services 67% 33% Construction 58% 42% Agriculture 25% 75% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Urban Rural 28 Education Level by Sector 100% 4% 5% 15% 14% 12% 13% 14% 14% 90% 26% 22% 80% 30% 70% 34% 35% 67% 20% 46% 46% 44% 60% 18% 61% 50% 55% 16% 17% 40% 15% 20% 30% 25% 53% 48% 20% 11% 24% 35% 34% 28% 28% 12% 10% 28% 28% 5% 28% 28% 0% Agriculture Other services Energy and Retail Professional extractive industry and trade services Construction Transportation Manufacturing Restaurant and hotel All sector and logistic Primary Lower secondary Upper secondary Tertiary Average age of workers by sector and “distance to retirement age” 60% 43 42 45 40 40 40 38 37 40 50% 36 35 34 35 Average age of workers 40% 30 32% 32% 25 30% 26% 24% 20 20% 18% 17% 17% 15 12% 12% 10 10% 8% 5 0% Professional 0 Agriculture Other services Energy and Retail extractive industry and trade services Construction Transportation Manufacturing Restaurant TOTAL and logistic and hotel >30 years 16-30 years 6-15 years 5 years or less Average age 29 Socio Economic Characteristics Workers’ Welfare Status by Sector 70% 60% 50% 40% 30% 64% 60% 50% 53% 41% 46% 46% 42% 43% 42% 20% 34% 39% 25% 26% 30% 28% 10% 24% 18% 18% 6% 0% Agriculture Other services Energy and Retail and Professional extractive industry trade services Construction Transportation Manufacturing Restaurant All-sector and logistic and hotel Social Assistance Recipient Heath Insurance PBI Income by Sector 70% 65% 60% 60% 53% 52% 51% 50% 49% 47% 50% 43% 45% 40% 41% 38% 38% 40% 35% 37% 30% 27% 24% 23% 38% 20% 20% 16% 36% 15% 13% 11% 12% 10% 8% 7% 9% 0% Agriculture Other services Energy and Retail and Professional services extractive industry trade Construction Transportation Manufacturing Restaurant and hotel TOTAL and logistic Less than 1.8 million 1.8 million - 4.59 million 4.6 million+ Asset ownership general savings land or house 8% Agriculture 34% 12% Construction 34% 14% Other services 34% 18% Transportation and logistic 24% Energy and extractive 16% industry 31% 24% Manufacturing 31% 25% Retail and trade 34% 33% Restaurant and hotel 33% 27% Professional services 39% 16% All-sector 34% 45% 40% 35% 30% 25% 20% 15% 10% 5% 0% 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 30 Behavioral Characteristics and Preferences Expectation at retirement: who will take care of me financially? 70% 60% 62% 57% 56% 50% 53% 48% 50% 47% 47% 47% 40% 44% 44% 40% 41% 41% 35% 37% 30% 34% 31% 28% 28% 20% 10% 0% Agriculture Other services Energy and Retail and Professional extractive industry trade services Construction Transportation Manufacturing Restaurant All-sector and logistic and hotel Self Children Desirable Financial Behavior Spending within income Save for emergency Contributing to pension Calculating retirement Compare prices for big needs purchase 0% 10% 20% 30% 40% 50% 60% Agriculture 46% 12% 1% 2% 12% Construction 15% 40% 2% 2% 13% Other services 38% 18% 1% 1% 19% Transportation and logistic 19% 49% 5% 2% 17% Energy and extractive industry 18% 44% 2% 4% 18% Manufacturing 42% 31% 15% 4% 17% Retail and trade 31% 48% 5% 4% 25% Restaurant and hotel 36% 45% 14% 9% 22% Professional services 31% 50% 13% 4% 21% All-sector 44% 21% 4% 3% 17% 0% 10% 20% 30% 40% 50% 60% 31 Knowledge and Participation in BPJS Employment Have heard of BPJS Have heard of JHT Member of BPJS Contributing to JHT Employment 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 70% 37% Agriculture 10% 6% 78% 36% Construction 11% 7% 80% 54% Other services 19% 13% 79% Transportation and logistic 52% 25% 17% 78% 62% Energy and extractive industry 42% 37% 81% Manufacturing 63% 42% 37% 84% 60% Retail and trade 21% 16% 81% 65% Restaurant and hotel 38% 26% 92% 65% Professional services 37% 27% 78% 50% All-sector 20% 14% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 32 Annex 2 Sector Profiles Agriculture • Average age of workers is 43 years old. • 65% has income lower than the lowest regional minimum wage in Indonesia. • 41% are social assistance recipients. • 57% thinks that their children are responsible for their old-age. Monthly Income Social Assistance Recipient Distance to Retirement 12% >30 years 27% 1.8 million - Yes 4.59 million 65% 41% No < 1.8 million 8% 4.6 million+ 59% 32% 16-30 years Expected Financial Responsibility Do you save for at Retirement retirement 25% 6-15 years 60% 50% Yes 40% 57% 27% 32% 5 years or less 30% 20% 35% children No 10% self 73% 0% Retirement Reasons for not saving for 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% retirement Relying on others - 5% Do not have enough money - 85% Do not know how - 3% Have not thought about retirement - 10% Forms of retirement savings General savings - 42% BPJS Ketenagakerjaan - 14% Retirement savings plan - 7% Buying propert/land - 44% BPJS Knowledge and Participation Have heard of BPJS Employment - 70% Have heard of JHT - 37% Member of BPJS - 10% Contributing to JHT - 6% 33 Constructions • Average age of workers is 42 years old. • 40% has income lower than the lowest regional minimum wage in Indonesia. • 34% are social assistance recipients. • 48% thinks that their children are responsible for their old-age. Monthly Income Social Assistance Recipient Distance to Retirement 6% >30 years 53% 1.8 million - 4.59 million Yes No 34% 66% 7% 4.6 million+ 41% 16-30 years 40% < 1.8 million Expected Financial Responsibility Do you save for at Retirement retirement 20% 6-15 years 50% 45% Yes 40% 48% 31% 32% 5 years or less 35% 44% children 30% self No 25% 69% 20% Retirement 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Reasons for not saving for retirement Relying on others - 1% Do not have enough money - 88% Do not know how - 4% Have not thought about retirement - 11% Forms of retirement savings General savings - 59% BPJS Ketenagakerjaan - 13% Retirement savings plan - 14% Buying propert/land - 24% BPJS Knowledge and Participation Have heard of BPJS Employment - 78% Have heard of JHT - 36% Member of BPJS - 11% Contributing to JHT - 7% 34 Transport and Logistics • Average age of workers is 40 years old. • 43% has income lower than the lowest regional minimum wage in Indonesia. • 26% are social assistance recipients. • 40% thinks that their children are responsible for their old-age. Monthly Income Social Assistance Recipient Distance to Retirement 12% >30 years 49% Yes 1.8 million - 4.59 million 26% No 9% 4.6 million+ 43% 74% < 1.8 million 41% 16-30 years Expected Financial Responsibility Do you save for at Retirement retirement 60% 29% 6-15 years 50% Yes 40% 34% 30% 53% self 40% No 18% 5 years or less 20% children 10% 66% 0% Retirement 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Reasons for not saving for retirement Relying on others - 0% Do not have enough money - 90% Do not know how - 3% Have not thought about retirement - 11% Forms of retirement savings General savings - 61% BPJS Ketenagakerjaan - 25% Retirement savings plan - 6% Buying propert/land - 21% BPJS Knowledge and Participation Have heard of BPJS Employment - 79% Have heard of JHT - 52% Member of BPJS - 25% Contributing to JHT - 17% 35 Trade • Average age of workers is 37 years old. • 43% has income lower than the lowest regional minimum wage in Indonesia. • 18% are social assistance recipients. • 44% thinks that their children are responsible for their old-age. Monthly Income Social Assistance Recipient Distance to Retirement 37% Yes 22% 1.8 million - >30 years 4.59 million 20% 18% 4.6 million+ No 43% 82% < 1.8 million 43% 16-30 years Expected Financial Responsibility Do you save for at Retirement retirement 50% 48% 18% 6-15 years Yes 46% 47% 44% 42% No 17% 5 years or less 44% 40% 41% children 53% 38% self Retirement 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Reasons for not saving for retirement Relying on others - 2% Do not have enough money - 81% Do not know how - 6% Have not thought about retirement - 17% Forms of retirement savings General savings - 57% BPJS Ketenagakerjaan - 19% Retirement savings plan - 7% Buying propert/land - 38% BPJS Knowledge and Participation Have heard of BPJS Employment - 84% Have heard of JHT - 60% Member of BPJS - 21% Contributing to JHT - 16% 36 Tourism and Hospitality • Average age of workers is 34 years old. • 36% has income lower than the lowest regional minimum wage in Indonesia. • 24% are social assistance recipients. • 31% thinks that their children are responsible for their old-age. Monthly Income Social Assistance Recipient Distance to Retirement 41% Yes 1.8 million - 4.59 million 25% >30 years 24% 23% No 36% 4.6 million+ 76% < 1.8 million 48% 16-30 years Expected Financial Responsibility Do you save for at Retirement retirement 60% 50% 40% No 20% 6-15 years 30% 56% 36% self Yes 20% 31% 64% 8% 5 years or less 10% children 0% Retirement 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Reasons for not saving for retirement Relying on others - 3% Do not have enough money - 85% Do not know how - 3% Have not thought about retirement - 22% Forms of retirement savings General savings - 73% BPJS Ketenagakerjaan - 34% Retirement savings plan - 9% Buying propert/land - 25% BPJS Knowledge and Participation Have heard of BPJS Employment - 81% Have heard of JHT - 65% Member of BPJS - 38% Contributing to JHT - 26% 37 Manufacturing • Average age of workers is 38 years old. • 24% has income lower than the lowest regional minimum wage in Indonesia. • 18% are social assistance recipients. • 34% thinks that their children are responsible for their old-age. Monthly Income Social Assistance Recipient Distance to Retirement Yes 60% 18% 18% >30 years 1.8 million - 4.59 million 16% 4.6 million+ No 24% 82% < 1.8 million 41% 16-30 years Expected Financial Responsibility Do you save for at Retirement retirement 60% 23% 6-15 years 50% 40% Yes No 30% 47% 49% 51% 20% 34% 17% 5 years or less self 10% children 0% Retirement 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Reasons for not saving for retirement Relying on others - 2% Do not have enough money - 83% Do not know how - 5% Have not thought about retirement - 15% Forms of retirement savings General savings - 55% BPJS Ketenagakerjaan - 43% Retirement savings plan - 5% Buying propert/land - 21% BPJS Knowledge and Participation Have heard of BPJS Employment - 81% Have heard of JHT - 63% Member of BPJS - 42% Contributing to JHT - 37% 38 39