Social Protection & Jobs Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: Insights and Recommendations Himanshi Jain, Nga Thi Nguyen, Agastya Yeachuri, Pham Bao Ha, Koustav De Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: Insights and Recommendations Himanshi Jain, Nga Thi Nguyen, Agastya Yeachuri, Pham Bao Ha, Koustav De Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS TABLE OF CONTENTS LIST OF ABBREVIATIONS8 ACKNOWLEDGEMENTS........................................................................................................................ 9 EXECUTIVE SUMMARY.......................................................................................................................10 I. OVERVIEW OF THE VOLUNTARY SOCIAL INSURANCE SCHEME ................................................15 II. INSTITUTIONAL ARRANGEMENT AND RECENT LEGISLATIVE, AND ADMINISTRATIVE CHANGES TO THE VSS ........................................................................................................................22 Targets for expansion of social insurance coverage.....................................................................25 Modernization of the social insurance system..............................................................................25 III. DELIVERY SYSTEMS ASSESSMENT OF THE VOLUNTARY SOCIAL INSURANCE SCHEME.......29 1. Assess...........................................................................................................................................30 1.1. Outreach................................................................................................................................................................ 30 1.2. Registration .......................................................................................................................................................... 36 2. Enroll.............................................................................................................................................38 2.1. Verification............................................................................................................................................................. 38 2.2. Onboarding........................................................................................................................................................... 39 3. Contribute....................................................................................................................................40 3.1. Contribution.......................................................................................................................................................... 40 3.2. Benefit Eligibility .................................................................................................................................................. 42 4. Provide.........................................................................................................................................43 4.1. Withdrawal (for one-time social insurance allowance).............................................................................. 43 4.2. Benefit payment................................................................................................................................................... 44 5. Manage.........................................................................................................................................45 5.1. Monitoring............................................................................................................................................................. 46 5.2. Exit........................................................................................................................................................................... 47 4 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS IV. SURVEY TO INFORM CHANGES IN DESIGN, DELIVERY, AND INCENTIVES IN VOLUNTARY SCHEME................................................................................................................................................47 a) Participation status of individuals is correlated with some observable socio-economic characteristics..................................................................................................................................52 b) Scheme awareness is most effective through word of mouth but awareness on scheme specifics like government incentives or estimation of benefits is poor........................59 c) Contributors depend on VSS agents for registration and contribution which makes it easier to perform transactions.............................................................................62 d) Lower vesting, additional short-term benefits, more generous state subsidy, and a link to social pension would make the scheme attractive but results point to a heterogeneity in preferences. .......................................................................................................64 RECOMMENDATIONS.........................................................................................................................68 ANNEX 1. LIST OF ADMINISTRATIVE PROCESSES CARRIED OUT BY THE VSS.............................71 ANNEX 2. TARGET OF PROVINCES’ SI PARTICIPANTS FOR THE YEARS 2023-2025 ...........................73 ANNEX 3: POLICY REVIEW.................................................................................................................75 ANNEX 4: BENEFITS AVAILABLE ON CONTRIBUTING TO THE VOLUNTARY SCHEME..................76 ANNEX 5: REGISTRATION FORM FOR VOLUNTARY PENSION SCHEME........................................79 ANNEX 6: SNAPSHOTS OF ONLINE SELF-REGISTRATION PORTALS..............................................81 1. VSS Pulic Service Portal...............................................................................................................81 2. National Public Service Portal....................................................................................................87 ANNEX 7: SAMPLING STRATEGY FOR THE VOLUNTARY SCHEME DESIRABILITY SURVEY (VSDS)...................................................................................................................................92 ANNEX 8: PANEL FIXED-EFFECTS REGRESSION MODEL RESULTS ON FACTORS ASSOCIATED WITH LAST MONTHLY CONTRIBUTION AMOUNT...........................................................................93 ANNEX 9: ROBUSTNESS CHECKS TO VERIFY EXPECTED SIGNIFICANCE OF INCOME LEVEL IN DETERMINING PARTICIPATION STATUS...........................................................................................95 REFERENCES........................................................................................................................................97 5 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS List of Tables Table 1: Revenue, Premium and Support from state budget for voluntary SI..................................................12 Table 2: Additional Support from local government for voluntary SI..................................................................15 Table 3: Premium and State support for voluntary SI (in monthly VND).............................................................16 Table 4: Remuneration for the voluntary SI collection service organizations..................................................19 Table 5: Online public service levels.................................................................................................................................21 Table 6: OLS Regression results with independent variables selected with help of LASSO regression... 48 Table 7: Pension scheme of voluntary SI.........................................................................................................................67 List of Figures Figure 1: Participants of Social Insurance 2011-2023................................................................................................. 17 Figure 2: Age, gender and contribution profile in the voluntary scheme.......................................................... 18 Figure 3: Stylistic benefit-cost ratio calculations based on projected lifetime contribution and benefits for male and female under different contribution scenarios................................................................. 21 Figure 4: Organizational structure of the VSS system................................................................................................ 23 Figure 5: Screenshot of VssID application...................................................................................................................... 28 Figure 6: Voluntary Social Insurance - Business process........................................................................................... 30 Figure 7: Online conference between VSS Thai Nguyen, Provincial Woman’s Union at Phu Luong District People’s Committee, Giang Tien town, Tuc Tranh commune, Yen Do commune...... 35 Figure 8: Voluntary SI leaflet in Thai Nguyen province.............................................................................................. 36 Figure 9: Stylistic diagram of registration via agents vs online.............................................................................. 38 Figure 10: Business process of collecting contributions........................................................................................... 41 Figure 11: Settlement of One-time SI benefit in-person and digitally................................................................. 44 Figure 12: Three provinces for the survey...................................................................................................................... 48 Figure 13: Age and income distribution in the survey and VHLSS........................................................................ 49 Figure 14(a): Type of work; Figure 13(b)Occupation................................................................................................... 50 6 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Figure 15: Saving channels by urban and rural population. ................................................................................... 51 Figure 16: Opinion on income over the next 10 years............................................................................................... 51 Figure 17: Opinion toward retirement age.................................................................................................................... 52 Figure 18a: Participation status by employment status; Figure 17b: by occupation type............................ 53 Figure 19: Count of individuals based on participation, location (urban/rural), and occupation type ....... 53 Figure 20: Average monthly household income by participation status ........................................................... 54 Figure 21 (a) Plan for old-age financial survival and (b) options for meeting emergency spending needs ...................................................................................................................................................................... 57 Figure 22: Frequency of last paid contribution (n=600)............................................................................................ 58 Figure 23. Reasons for not contributing more/stop contributing/not participating in VSIF........................ 59 Figure 24. Share of Participants based on channel that made them aware of scheme................................. 60 Figure 25: Awareness of the VSIF rules (n=900)........................................................................................................... 61 Figure 26: Ability to estimate pension amount by response type, and participation status (n=900)...... 61 Figure 27: Person who initiated registration, by gender (n=600).......................................................................... 62 Figure 28. Registration and contribution channels ................................................................................................... 62 Figure 29: Avenues to resolve queries............................................................................................................................. 63 Figure 30. Ease of VSIF related transactions ................................................................................................................. 63 Figure 31. Respondent preferences on intermediaries to facilitate contribution collections and benefit payments (n=600)................................................................................................................................................... 64 Figure 32: Perception on benefits of the scheme ....................................................................................................... 65 Figure 33: Willingness to participate if they are offered maternity benefits ..................................................... 66 Figure 34: Ranked preference for short term benefits, by gender........................................................................ 66 Figure 35: Count of respondents willing to participate if the Government increased contribution subsidy by 10-15%.................................................................................................................................................................. 67 Figure 36. Willingness to save if non-contributory and contributory design is linked.................................. 68 7 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS LIST OF ABBREVIATIONS DB Defined benefit DC Defined contribution DSS District Social Security GDP Gross Domestic Product GSO General Statistic Office HI Health Insurance MOLISA Ministry of Labour, Invalids, And Social Affairs MPS Ministry of Public Security PSS Provincial Social Security SI Social Insurance NID National Insurance Database VND Vietnam Dong VSIF Voluntary Social Insurance Fund VSDS Voluntary Scheme Desirability Survey VSS Vietnam Social Security Agency VSSID Vietnam Social Insurance Digital Application 8 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS ACKNOWLEDGEMENTS The report and overall study were led by Himanshi Jain from the World Bank and the team comprised of Nga Nguyen Thi, Agastya Yeachuri, Pham Thi Bao Ha, and Dr. Koustav De. The delivery chain assessment was carried out by Pham Thi Bao Ha, Nga Nguyen Thi, and Agastya Yeachuri. The survey was implemented by Vietnam’s General Statistics Office (GSO) and the methodology as well as survey questionnaire was developed by the World Bank team in collaboration with the VSS. Econometric analysis of the survey was carried out by Dr. Koustav De. The team also benefited from the guidance and support of Robert Palacios throughout the study. The report would not have been possible without the excellent support and coordination of the Vietnamese Social Security teams at national and provincial levels who assisted with setting up and conducting of interviews with teams managing the Communication, IT, Call-center team of the Voluntary scheme. The report also includes findings from the field visit to Hai Phong and Thai Nguyen provinces by the task team where we interviewed Mr. Nguyen Van Hanh, Deputy head of Hai Phong VSS; Mr. Nguyen Hong Truong, Deputy head of Thai Nguyen VSS, members of the Craft Cooperative; trade union, women union, farmer union. We express our sincere gratitude to all senior local officials, and partners who have shared remarks in these consultations/interviews. The in-person survey of 900 respondents was carried out in April 2022 by the GSO under the guidance of the management team of the Population and labor department. The team benefitted from excellent comments from formal peer reviewers Ignacio Raul Apella and Oleksiy A. Sluchynskyy. We would also like to thank the VSS management for their support of this task. The report was produced under the overall guidance of Carolyn Turk (Vietnam Country Director at the time of the work); Kathleen A. Whimp (Vietnam Country Manager at the review stage) and Yasser El- Gammal (Practice Manager for Social Protection & Jobs, East Asia & Pacific). 9 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS EXECUTIVE SUMMARY Vietnam has witnessed impressive economic growth in recent decades, but it continues to have a large share of its workers in the informal sector. Economic reforms since 1986, combined with favorable global trends, have helped Vietnam grow rapidly from one of the poorest countries in the world to a low-middle-income country. Between 2002 to 2020, GDP per capita increased 3.6 times, reaching nearly USD 3,7001. Despite this growth, Vietnam (similar to other growing economies in Asia like India, Thailand, Philippines) has high and persistent rates of employment in the informal sector. Around 76 percent of all workers, and 55-60 percent of non-agriculture workers, remain in the informal sector (Cunningham & Pimhidzai, 2018)2. This will likely be exacerbated by the growth of gig economy jobs outside the traditional labor contract model. At the recent rate of change, it is estimated that only around 43 percent of Vietnamese workers will be in contracted-wage jobs by 2040 (Cunningham & Pimhidzai, 2018). The lack of a labor contract for workers in the informal economy excludes them from payroll financed mandatory social insurance coverage. The impacts of informality include lower per capita incomes, greater poverty, less financial development, and weaker growth in output, investment, and productivity (Ohnsorge & Yu, 2022). From the perspective of social insurance, informality presents a coverage challenge as legislative mandates for payroll financed social insurance (the norm in most countries including Vietnam) excludes those without a full-time employer, from protection against idiosyncratic risks like disability, maternity benefits, unemployment, old age, death, and medical expenses. The implementation of the mandate and cost sharing with employers is possible only for workers in medium/large firms (loosely referred to those in formal employment) leaving the rest vulnerable to lifecycle risks. Among lifecycle risks an individual faces, Vietnam has successfully extended health insurance coverage to those in the informal sector, but similar results have not been replicated for pensions and other short-term benefits. Successes in expanding health insurance coverage have been rapid and significant in the East Asia and Pacific (EAP) region (notably for Indonesia, Thailand, and Vietnam), in large part because of contribution subsidies, close to 100 percent for poorest households. However, this success has not been replicated for other risks, including pensions. With regards to pensions for informal sector, countries have created voluntary contributory schemes. Monetary incentives (subsidies or matching contributions), bundled benefits, behavioral nudges, digitally enabled registration/ contributions are some ways countries have tried to incentivize participation rates in voluntary pension schemes (Jain & Palacios, 2022), to varying levels of success. International experience on expanding pension coverage in voluntary schemes suggests that innovation in design, delivery, and incentives is still needed. An open and relevant question is if expansion in health insurance coverage (a salient short-term risk) can be replicated in pensions through fiscal incentives and whether that would provide an adequate pension for older adults, affordably. The rapid pace of aging globally, and concerns around vulnerability of older adults to shocks, non- 1 https://www.worldbank.org/en/country/Viet Nam/overview. 2 According to GSO, the rate of informal employment including agriculture in 2021 was 68.5%. 10 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS communicable diseases, and climate change, is making this question salient for policymakers. The experience is limited, and results are mixed. China and Rwanda are among countries that have achieved coverage upwards of 35 percent of the informal sector, with different pension designs and incentives. China has a defined benefit (DB) scheme, which subsidizes pension contributions of the bottom 40 percent. Rwanda has a defined contribution (DC) scheme that targets its scarce government resources to incentivize people to save for retirement. Both countries deploy their strong multi-tier (central to local) administration network to achieve their coverage targets. Rwanda also offers attractive short- term benefits (withdrawal flexibility, life insurance, loans against savings) and works closely with the civil society, cooperative network, and private sector. Matching contributions in voluntary pension schemes are also offered by countries such as India, Thailand, and Colombia, but have not been as successful. The results achieved in Rwanda and China highlight the importance of focusing on all three aspects - program design, delivery and incentives. Countries aspiring to achieve similar or better outcomes on pension coverage for the informal sector should develop strategies based on their context, capacity, and objectives, borrowing from international experiences where applicable. Vietnam has set a bold target under Resolution 28, of attaining contributor coverage of 60 percent of the labor force and beneficiary coverage of 60 percent of older adults by 2030. The dual pronged strategy of improving coverage of current elderly and future elderly is needed given Vietnam is the fastest aging economy in the region. According to the General Statistics Office, the number of older adults (60 plus) grew from 7.5 million in 2009 to 11.4 million in 2019 (approx. 9 to 12 percent of the total population) and is expected to grow to 17.3 million in 2029 and 28.6 million (approx. 25 percent of the total population) by 20493. Only 3.3 million older adults currently receive pensions from the Vietnam Social Security (VSS), most of whom are civil servants. An additional 1.8 million are beneficiaries of the social pension financed through tax revenues. Guaranteeing income security for current older adults will need government resources, but Vietnam can prepare to sustainably support the growing numbers of older adults in the future by incentivizing current working age adults to contribute in pension schemes. Most contributors in the VSS are from the compulsory scheme, and while coverage in the voluntary scheme has grown it remains at low levels. As of December 2023, total coverage in VSS of compulsory and voluntary scheme stood at 18.2 million individuals (34.8% of labor force), most of whom were participants in the compulsory scheme. The voluntary pension scheme, established in 2006 is a defined benefit design like the compulsory scheme but does not include short term benefits or a cost-sharing arrangement with an employer. It failed to attract a significant number of individuals and covered only about 271,000 people by 2018. To make the scheme attractive, a flat but poverty targeted subsidy was implemented from January 1, 20184, with the total State outlay on matching contributions estimated as VND 1,418 billion during the period of 2018-2024 (VSS report, 2024). The subsidy resulted in a sharp increase in coverage, reaching 1.837 million (~4% of the labor force by 2023). However, coverage in voluntary scheme as a proportion of the informal sector workforce is still low. As per the 2021 Labor Force survey, only 2.1% of all informal workers reported participating in voluntary social insurance. Achieving the Resolution 28 targets will need efforts to increase mandate/compliance in the compulsory scheme and to the extent informality remains persistent, contributors in the voluntary scheme will also need to grow for Vietnam to meet its Resolution 28 targets. 3 https://www.gso.gov.vn/en/data-and-statistics/2021/08/population-ageing-and-older-persons-in-viet-nam/. 4 Based on amendments to Social Insurance Law in 2014. 11 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS The voluntary social insurance scheme (VSIF) introduced in the 2006 Social Insurance Law and implemented from January 1st, 2008, is targeted towards Vietnamese citizens aged 15 years or older who are not eligible to participate in compulsory social insurance. It is similar in design to the compulsory scheme in terms of eligibility and pension benefit calculations albeit with three key differences. First, the voluntary SI scheme offers coverage against two risks (old age and death), while the compulsory scheme offers coverage against five risks.5 Second, the voluntary scheme allows individuals to choose the income level on which contributions are levied with the minimum income level being equal to or greater than the national poverty line. Compulsory scheme participants on the other hand must pay contributions based on the income reported by the employer. Third, the 22 percent of income collected as pension contributions in the compulsory scheme is split into 14 percent paid by employer and 8 percent by employees whereas for the voluntary scheme the individual must pay the 22 percent of their chosen income. Since 2018, the Govt pays a flat amount known as State support to every voluntary scheme participant for a total of ten years. The state support varies based on poverty status of the individual and ranges between VND 33,000 to VND 99,000 for the poorest category of households. Expanding coverage in the Voluntary Scheme will require a multi-stakeholder strategy encompassing the effective use of fiscal incentives, innovation in program design, and strengthening administration as well as service delivery aspects. The report’s conclusions are based on information gathered via a delivery chain assessment using review of operational documents and interviews of VSS staff at central and provincial levels, and the 2023 Voluntary Scheme Desirability Survey (VSDS-2023). The VSDS (2023) is an in-person survey that was designed and implemented for the purposes of informing the Social Insurance Law revision in Vietnam in 2024. It was a thirty minutes in-person survey and included four modules6. The data collection and validation were carried out by Vietnam’s General Statistics Office of Vietnam from April–July 2023. Data quality was monitored using a standard protocol. Section 1 of this report includes an Overview of the Voluntary Pension scheme including institutional arrangements and recent legislative/administrative changes. Section 2 covers the results of the delivery chain assessment while Section 3 includes findings of the VSDS of 900 individuals across three provinces carried out to understand respondents’ awareness and perception of the scheme and their preferences for alternate design options in the VSIF. Section 4 concludes with recommendations for improvement to design, delivery, and incentives in the scheme with the aim of informing the 2024 SI law revision7 and aiding the coverage expansion efforts. 5 The VSS manages five Compulsory Social Insurance Funds (covering sickness, maternity, old age and survivorship, labor accident and occupational disease, and unemployment insurance) and the Voluntary Social Insurance Fund (covering old age and survivorship). 6 The modules in the VSDS survey were on (a) Socio-economic and demographic information; (b) Awareness about the voluntary scheme rules, (c) Experience interacting with the scheme, and (d) Design, Delivery or Incentive changes that would make them more likely to save (save more). 7 The findings of this report were used to provide recommendations on the voluntary scheme under Vietnam’s SI law revision process, via national level workshops organized by the World Bank in collaboration with the VSS, National Treasury, and other Govt stakeholders. The SI law was passed in June 29, 2024, prior to the release of this report, and the report therefore references the changes in the law, where appropriate. 12 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Seven key insights emerge from the study: 1. The voluntary scheme’s target group is heterogenous with individuals across multiple occupations with limited disposable incomes but some ability to save, moderate expectations of individual income growth, and an expectation to retire sooner than the mandated retirement age. Although the VSDS-2023 survey was not nationally representative, the findings reveal the importance of understanding characteristics of the target group. 2. Establishing and maintaining trust with scheme participants is critical to coverage expansion efforts. Unmet expectations, ad-hoc changes to scheme parameters, unclear benefits can all impact trust in the scheme. Alternatively, positive testimonials about the scheme are the most effective awareness raising channels according to the survey results. 3. There is a high dependence on the agent network for services throughout the delivery chain, notably outreach, registration, questions on scheme benefits, withdrawals. The in-person presence of agent networks makes them BUT the commissions can be expensive. The agent network’s services should be used at certain points in the delivery chain e.g., initial outreach, and complementing they with low-cost options like digital means for registration & other intermediaries for auto-enrollment or contribution collection. 4. Promoting Technology Use Throughout the Delivery Chain: To date, the number of participants using the VSS website or VSSID for registration, as well as those using bank accounts for contributions, remains low. Encouraging online registration and automatic withdrawals from a bank account for IS contribution could reduce costs and time for both agents and participants. More importantly, these methods can help maintain persistency among active participants. 5. Awareness of state support is extremely low and so is the ability to estimate the pension on retirement. The subsidy on contributions offered by the state and an estimate of benefit are both crucial pieces of information that would go into one’s decision-making process to contribute and continue contributing in the scheme. The survey findings point to a particularly low awareness of both, necessitating a need to focus on them in outreach efforts through clear, easy to understand communication material like leaflets and tools like a digital calculator. 6. Not all incentives are equally valued by all. A state subsidy, while desired by most, does not ensure universal take-up. Similarly, while lump sum maternity benefits is attractive/very attractive to 65 percent of respondents men prefer occupational hazards insurance and women prefer sickness benefit over maternity. While this could reflect the study’s non-representative sample (median age of 37), it shows that an age and gender specific short-term benefit might not lead to wide take-up. 7. It is essential to leverage existing data for analysis to support policy development and evaluate the effectiveness of the program at each stage of implementation. These insights and analysis in the report have guided the report’s recommendations of focusing on three areas to boost coverage expansion efforts in the voluntary scheme. These three areas of focus are summarized in table below: 13 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Focus areas Recommendations Changes in design • Improve predictability in levels of minimum contribution and manage member and incentives to be expectations around increase in minimum. considered for the SI • Increase generosity of state support and focus on raising awareness around the state law revision of 2024 support. • Consider adding short term benefits, ideally as a bundle to make the scheme attractive for different groups. • Offer a flat benefit (or social pension) at an earlier age to those who persistently contribute in voluntary scheme. Updating of VSS’s • Develop a unique branding for the voluntary scheme with a local name/logo to differentiate outreach & business it from other VSS programs. processes along the • Encourage adoption of digital service delivery channels to complement the efforts of delivery chain external service agencies and reduce overall costs. • Set targets for digital take-up and monitor progress across provinces in dashboard. • Eliminate or reduce paper-based forms for registering new VSIF participants. Use available information from the heath Insurance database to auto-populate the online registration forms. This approach will minimize the effort required to recollect information already available in the database.Review the existing data dashboard to ensure that the indicators are effective in decision-making. • Identify appropriate aggregators that can auto-debit member contributions. • Focus on awareness raising through in-person channels (testimonials/word of mouth). • Develop digital pension estimator tool and leaflets to help estimate pension benefits. • Develop a diverse range of outreach material that provides participants clear understanding of rules, how to contribute, benefits. • Simplify the registration form, deleting redundant data fields and include optional fields. • Streamline business processes to make it easier for dormant participants to resume contributions. • Utilize call-center staff in annual spot surveys to understand customer experience. • Use data analytics at the central level to informal outreach strategy which can be finetuned by provinces. Strengthening of • Strengthen monitoring, evaluation and learning across various administrative levels of VSS learning, monitoring to improve scheme design and delivery. and evaluation • Consider introducing a rewards and recognition program for high-performing provinces capacity / districts, including establishing an experience sharing platform, tracking of collection service organization/agents, and develop mentoring models. • Consider collaborating with the General Statistics Office to maximize data collection and utilization; include questions in Labor Force Survey (LFS) and other nationally representative household surveys that allow for better analysis. • Assess cost effectiveness of outreach and communication spending. • Set up impact evaluations to assess increase in coverage and level of contributions because of changes in the 2024 SI law. • Strengthen the analytical capacity of staff to assess trends in the data, design surveys, interpret results and provide recommendations to management. 14 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS I. Overview of the Voluntary Social Insurance scheme The contribution or premium for the voluntary SI scheme is 22 percent of the monthly income level self-reported by the participant. As opposed to the compulsory scheme, contributors of the voluntary SI have the flexibility to choose the income level to which the contribution rate of 22 percent is to be applied. However, the chosen income level cannot be lower than the poverty line of rural areas as prescribed by the Prime Minister (Article 5, 2014 Social Insurance Law). Since January 1, 2022, the poverty line in rural areas is 1,500,000 VND/person/month8. This translates to a minimum monthly contribution in VSIF of VND 330,000 (22 percent of VND 1,500,000). A maximum income level has also been defined viz twenty times of the base salary of civil servants at the time of contribution9. In the absence of an employer, the informal sector worker is liable to pay the full VSIF contribution though in 2018 the government introduced targeted subsidies with the intention of making the scheme more affordable for informal workers. The minimum monthly contribution in the scheme of VND 330,000 (USD 14) is equivalent to ~11% of average household consumption (GSO), 2021. Given that informal sector workers on average have lower and more precarious incomes along with high liquidity needs, paying the minimum contribution each month might not be affordable for all. To incentivize participation, the 2014 Social Insurance Law introduced targeted government subsidies for participants. This subsidy was implemented in 2018, is paid from the state budget10 and provided for a maximum of 10 years to each participant. The subsidy for households in the ‘poor’ category is 30 percent of the minimum contribution level or VND 99,000 (=30% * 22% * VND 1,500,000); the subsidy for near poor is 25 percent of the minimum contribution level or VND 82,500, and for all others it is 10 percent of the minimum contribution level i.e., VND 33,000. Taking the subsidy into account, the minimum monthly contribution to be paid by each contributor in the voluntary scheme can be re-calculated as follows. Recalculated Monthly Premium = 22% x Selected income level - Government Subsidy depending on poverty category While there is a minimum monthly contribution in the scheme there exists flexibility in the frequency of contribution payments. The flexibility in payment of contributions is a unique feature of the VSIF to accommodate the irregular income pattern which informal sector workers face. Voluntary SI participants may choose to pay contributions monthly; every 3 months; every 6 months; every 12 months; or as a one-time contribution for future years but capped to five years; or one-time contribution for the missing years needed to vest pension rights.11 They also enjoy a grace period for paying the contribution once the frequency is selected12. Voluntary SI participants can also change their 8 According to Decree 07/2021/ND-CP stipulating the multi-dimensional poverty line period 2021-2025. 9 From July 1, 2024, the base salary is 2,340,000 VND/month. So, effectively a voluntary SI participant can select a monthly income level between 1,500,000 VND to 46,800,000 VND, escalating in 50,000 VND increments, for the purpose of premium calculation. 10 According to Decree 134/2015/ND-CP. 11 SI participants who have met the age requirements for retirement but not the vesting period (20 years) for pension and are missing less than ten years (120 months) of contribution can pay for their missing years as one-time payment and receive the pension. 12 The grace period is Within the month for the monthly contribution method; Within 3 months for the contribution of every 3 months; In the first 4 months for the contribution of every 6 months; In the first 07 months for the contribution of every 12 months; At the time of registration for the case of one-time contribution. 15 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS chosen contribution frequency or monthly contribution level, over time. They receive partial refunds of their contributions if they become eligible for pension from compulsory social insurance or upon death. If participants fail to contribute the premiums in the stipulated time, they are considered to be suspending the contribution. Participants who have suspended paying contributions may resume after re-registering with the VSS. Participants are also provided the option to compensate for missed contributions albeit with a compound interest rate (equal to the average monthly investment interest of the social insurance fund announced by VSS) applied on such payments. Benefits of participating in the voluntary scheme include pensions, a contribution subsidy for a maximum period of 10 years, lump sum survivor benefits, lump sum funeral allowances, and payment of health premiums during retirement. Individuals who meet the vesting period (currently 20 years for both females and males but will be reduced to 15 years from July 1, 2025, according to the new SI Law) and have reached the retirement age, are eligible to receive old age pensions. Besides the old age pension benefit, they also receive other benefits notably health insurance premiums paid by VSS during their retirement, contribution subsidies by the government for a total of 10 years, lump sum survivor benefit, funeral benefit, and a maternity benefit under the new SI law. The stakeholder consultations reveal that individuals have limited awareness of the other benefits offered by the VSI besides pensions and the leaflets used by the VSS also fail to highlight the other benefits adequately. The pension benefit is calculated using a ‘defined benefit’ formula which considers the years of contribution, and the income on which contributions were paid. The rules for calculation of the pension benefit under the VSI (see Annex 4) are same as those of the compulsory scheme and recent SI reforms to increase retirement ages and equalize vesting period between men and women apply to both the compulsory and voluntary schemes. The provision that allows withdrawal of all SI contributions as a lump sum, before reaching retirement age, is similarly available to both compulsory and voluntary scheme participants. Withdrawals are allowed if social insurance premium payment is suspended after one year of no contributions; total contributions are less than 20 years; individual has settled abroad or is suffering from a life-threatening disease as prescribed by the Ministry of Health. While the rising number of one-time lump sum withdrawals from pension scheme is a cause of concern for policymakers in Vietnam13, the administrative data on withdrawals reveals that voluntary scheme members are less likely to withdraw one-time lump sums (Jain & Palacios, 2022). The number of contributors in the voluntary scheme continues to be low but the counts have more than tripled since 2018. The contributor counts in the VSIF remained low since its inception but since the government subsidy was implemented in 2018, a faster year on year growth is being registered. In 2019 coverage stood at 450,000 in 2019 and grew to 1.45 million by 2021, despite the COVID 19 pandemic which affected household incomes14. By the end of 2023, coverage reached 1.83 million (Figure 1). Besides a growth in counts, there has also been a growth in average contribution amount. The latter is in large part because of the 2021 revision of the rural poverty line which is the basis for the minimum contribution in the VSI. The combined effect of a growing contributor base with higher average contributions is an increase in contribution revenue and consequently in the state budget support to the VSI (Table 1). The 2024 SI law revision once implemented in July 1, 2025 will stop new participants joining the schemes (compulsory or 13  voluntary) from withdrawing lump sums.. According to GSO, in 2020, the number of employed populations at 15 years of age was 53.4 million, a decrease of 1.3 14  million compared to 2019 and the number of formal workers was 15.8 million, a decrease of 21.1 thousand compared to 2019. In 2021, the corresponding figures was 49.0 million, a decrease of 1.0 million compared to 2020; and 15.4 million, a decrease of 469.8 thousand people compared to 2020. 16 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Table 1: Revenue, Premium and Support from state budget for voluntary SI 2016 2017 2018 2019 2020 2021 2022 Voluntary SI Participants (‘000) (a) 203 224 277 558 1,124 1,450 1,462 Total contribution revenue 1,122 1,194 1,288 2,379 4,062 5,496 7,063 (bln VND) (b) Support from State budget 50 83 212 294 541 (bln VND) (c) Compulsory SI Participants (‘000) (a) 12,862 13,596 14,453 15,303 15,064 15,097 16,038 Total revenue (bln VND) (b) 174,490 195,199 221,796 246,250 259,887 268,441 279,678 Source: Report of VSS Figure 1: Participants of Social Insurance 2011-2023 20000 35.00 18000 30.00 16000 14000 25.00 12000 20.00 10000 15.00 8000 6000 10.00 4000 5.00 2000 0 0.00 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 Total compulsory SI participants 1000 person Total voluntary SI participants 1000 person Share of compulsory SI participants in Labour Force % Share of voluntary SI participants in Labour Force % Source: Calculated from VSS data Females currently make up a larger proportion of the voluntary scheme and contribution levels are close to the minimum contribution. The active contributors in the VSI as of 2022, per administrative data, comprised of ~60% females, and the average age of contributors was 40.2 years. The higher proportion of female contributors in the voluntary scheme relative to the compulsory scheme is posited to be a reflection of gender disparities in employment, with more females represented in informal sector jobs. The average monthly income chosen by the VSI participants was VND 1,672,018 for females and VND 1,696,248 for males, viz marginally higher than the required minimum of VND 1,500,000. The monthly selected income on which contributions are paid also does not vary by age (figure 2), suggesting that all participants typically contribute close to the minimum contribution level. 17 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Figure 2: Age, gender and contribution profile in the voluntary scheme 8.00% 2,000,000 Average monthly income in VND Share of contributor base (in %) 7.00% 1,800,000 1,600,000 6.00% 1,400,000 5.00% 1,200,000 4.00% 1,000,000 3.00% 800,000 600,000 2.00% 400,000 1.00% 200,000 0.00% - 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59 61 Contributor age Share of contributors - Female Share of contributors - Male Average Reported Monthly Income - Female Average Reported Monthly Income - Male Source: Calculated from VSS administrative data (2022) The link of minimum contribution to the rural poverty line and the flat government subsidy creates design challenges for the scheme. The minimum contribution to the scheme is linked to the rural national poverty line which is revised once every five years. In the year 2021 the revision was sharp, growing from VND 700,000 to VND 1,500,000. Consultations with VSS agents revealed that members were not expecting an abrupt jump in their minimum contributions and expressed reluctance to continue contributing either because they were unable to pay the higher contributions, or they started distrusting the scheme. At a policy level, the increase in minimum contribution over time is recommended so that the adequacy of pension benefits can be maintained. However, it is important that contributors have clarity on why and how the minimum contribution will grow e.g., see India’s PM- SYM and China’s RURS scheme’s schedule of contributions which clearly shows how contributions are expected to grow with age of contributor. Instead of a large increase in minimum contribution in one year a better approach would be to make the increase gradual and predictable. The subsidy is also a flat amount for the three categories of households (VND 99,000 for poor, VND 82,500 for near-poor and VND 33,000 for all others) irrespective of how much an individual contributes. This flat subsidy, unlike a matching contribution, might not incentivize individuals to contribute more. Despite higher subsidies for poor and near-poor individuals, participation rates remain low indicating a need for additional efforts to understand their needs and cater to them. The share of poor and near-poor in the total population of Vietnam is relatively low to begin with (~10% of the overall population based on the Vietnam Household Living Standards Survey (2022)). Even though this group receives higher subsidies in the VSI scheme they made up less than 5 percent of all active contributors in 2022 (65,481 individuals from poor and near-poor categories out of 1.46 mln contributors). The reasons for the low take up for this group need to be better assessed but consultations reveal that the state support might not be generous enough to motivate these groups to save. Limited digital and financial awareness among these groups could also be a reason behind their low take-up and tailored communication 18 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS strategies should be considered to reach these groups. In 2020, the VSS sent an official letter15 to the Ministry of Finance proposing the Government to increase the support for participants of voluntary social insurance. The letter proposed an increase from 30% to 50% for poor households; from 25% to 30% for near-poor households and from 10% to 20% for the rest. However, the proposal was not approved. With the decentralized arrangements in Vietnam, some provinces have been able to offer additional premium support depending on their budget capacity and policy vision. The Provincial People’s Council can issue a resolution to increase the state support level and/or expand the beneficiary groups. Any increases are to be funded by their local budget, allowing “richer” provinces to make the voluntary SI scheme more attractive for their constituents. By the end of 2022, 13 of 63 provinces provided additional support for participants of voluntary social insurance, focusing on poor and near poor groups (Table 2). Each year provinces are judged against the targets set for them (see Annex 2 for targets) and data needs to be collected and analyzed over time to see if provinces with higher top- ups are more likely to reach their targets. In September 2022, VSS sent an official letter requesting all provinces to balance their budget and support additional contribution for participants of voluntary social insurance, particularly for participants from near-poor households, households of agriculture, forestry, fishery and matches. average standard of living, students and other groups in the locality. Table 2: Additional Support from local government for voluntary SI Additional Top up (over and above the 30% for poor, 25% for near poor and 10% for Province Effective date Calculation basis others offered by central govt Poor hhld Near poor hhld Others 1 Ha Tinh16 Jan 1, 2020 Rural poverty line 10% 2 Vinh Phuc Aug 1, 2020 Rural poverty line 30% 25% 10% 3 Hai Duong Jan 1, 2021 Rural poverty line 10% 10% 4 Bac Ninh Aug 1, 2021 Rural poverty line 30% 30% 30% 5 Bac Giang Jan 1, 2022 Rural poverty line 10% 10% 10% 6 Quang Nam Jan 1, 2022 Rural poverty line 10% 10% 5% 7 Hanoi Aug 1, 2022 Rural poverty line 30% 25% 10% 8 Quang Binh Oct 1, 2022 Rural poverty line 15% 15% 9 Quang Ninh Jan 1,2023 Rural poverty line 20% 20% 20% 10 Hai Phong Jan 1,2023 Rural poverty line 30% 25% 10% 11 Binh Duong Jan 1,2023 Rural poverty line 30% 25% 12 Quang Tri Jan 1,2023 Rural poverty line 20% 15% 10% 13 Son La 17 Jan 1,2023 Rural poverty line 10% 10% 10% 15 Official Letter No. 2620/BHXH-BT dated August 18, 2020. 16 Other groups include households engaged in agricultural, forestry, fishery and salt production. 17 Other groups include households engaged in agriculture, forestry and fishery with average living standards; ethnic minorities. 19 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS The government subsidy aims to improve affordability, but the low subsidy level means that a worker in the informal economy earning minimum wage pays more out of pocket than his peer in the formal sector. A formal sector job can be desirable given the secure income and perks like partial social security contributions paid by the employer. For some, working in the informal sector can mean paying fewer taxes and higher disposable incomes because they are not mandated to contribute to social security. Comparing these two groups of workers, who have very different characteristics, is not ideal but it can be illustrative to look at the out-of-pocket costs of social security for those earning close to the minimum wage. Under the current design, a minimum wage worker in the formal sector will pay only VND 260,000 out of pocket in the compulsory scheme while the contributor from the ‘poorest’ household category in the voluntary scheme will pay VND 231,000. (Table 3). This anomaly arises because employers bear a large portion of social security contributions in the compulsory scheme (14 percent of the 22 percent contributions) while the government subsidy in the voluntary scheme is low. Table 3: Premium and State support for voluntary SI (in monthly VND) Minimum Minimum wage worker in Voluntary wage worker scheme in Compulsory scheme Poor HH Near poor HH Other (a) Minimum income level for premium calculation 1,500,000 1,500,000 1,500,000 3,250,000 (b) Minimum premium w/o state support or 330,000 330,000 330,000 715,000 employers [a*22% for voluntary scheme] (c) State or employer contribution (as % of minimum 30% 25% 10% 14% premium) (d) Govt subsidy or employer contribution [b*c] 99,000 82,500 33,000 455,000 (e) Minimum premium worker has to pay with state 231,000 247,500 297,000 260,000 support [b-d] Source: Calculated by author with minimum income levels based on Decree 134/2015/ND-CP Voluntary scheme participants are currently excluded from short term benefits available to those in the compulsory scheme. The compulsory scheme participants pay an additional 3.5% and receive coverage for Maternity, Sickness fund, and the Industrial Accident – Occupational Disease Fund. These short-term benefits are not available to those in the voluntary scheme. The reasoning behind the exclusion being the challenges with verifying eligibility for these benefits for an informal sector worker e.g., what constitutes ‘being sick’ and away from work for a fruit trader? Or what is a clear case of accident at work site for a freelance plumber? Assessing eligibility for short-term benefits can be trickier, costlier and more time consuming for informal sector workers, there are potential solutions to consider. For example, benefits like parental benefits (payable on the birth of a child) or accident insurance (requiring police complaint and vehicle registration details) can be considered as short-term benefits in the voluntary scheme. Experience ratings or limits to how often such benefits can be claimed, could also be used to reduce moral hazard. For e.g., only one accident benefit can be claimed in a period of 3 years of contribution, or the contributions would go up by 0.2% in the year after a claim has been made and go back to 22% after a year of no claim. Such innovation can allow short-term benefits to be offered to 20 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS informal sector workers making the voluntary scheme more attractive18. To further analyze the adequacy of the subsidy and other benefits in the voluntary scheme, illustrative benefit-cost ratio analyses were conducted for projected lifetime contribution and benefits under various scenarios19 (outlined below). The results indicate that the discounted present values of the pension benefits are higher in comparison to the contributions for women over men while the benefit-cost ratio is slightly higher for the shorter contribution of 15 years in comparison to the 20- year period. The design of the program helps explain these findings. Female participants enjoy a higher replacement rate in comparison to male participants and enjoy a longer duration of pension benefits due to higher life expectancy, hence the benefit-cost ratios are correspondingly higher. The differences in the benefit-cost ratios across the contribution periods are attributed to lower accrual rates after the vesting period of 15 years, which lowers the present value of the pension benefits. These findings can help inform future design changes to the scheme to help improve its’ adequacy and subsequently communicate the same to existing and potential beneficiaries to foster better participation. Figure 3: Stylistic benefit-cost ratio calculations based on projected lifetime contribution and benefits for male and female under different contribution scenarios Bene t-cost ratios for Viet Nam voluntary social insurance scheme 1.60 1.40 0.05 0.05 0.05 0.05 1.20 1.00 0.07 0.05 0.04 0.80 0.03 1.17 1.17 0.60 0.78 0.40 0.72 0.20 0.06 0.06 0.07 0.07 - Male - 20 years Female - 20 years Male - 15 years Female - 15 years ofcontribution ofcontribution ofcontribution ofcontribution Subsidy Pensions Life insurance Other bene ts 18 The 2024 SI law revision is considering adding maternity benefit to the voluntary scheme whereby those who contribute in the scheme for 6 months or more in the 12 months before giving birth can receive a maternity benefit of 2,000,000 VND for each child born. 19 The scenarios cover 2 contribution periods (15 years and 20 years) for male and female participants starting at the age of 35. Contributions are projected at the minimum level, factoring in projected increases in the poverty line every five years (averaging an annual growth rate of 6%). Pensions and other lump-sum benefits are based on current statutory provisions. All contributions and benefits are projected in nominal terms and discounted to present values at a nominal rate of 7% (based on a sensitivity analysis conducted with regards to the discount rates, the BCR’s vary between 0.69 (at 8% discount rate) and 2.94 (at 2% discount rate) for the 20-year contribution by a male participant scenario). Life expectancy data is sourced from United Nations, Department of Economic and Social Affairs, Division of Population (2022). World Population Outlook 2022, Online Version. 21 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS II. Institutional arrangement and recent legislative, and administrative changes to the VSS The Ministry of Labor, War Invalids and Social Affairs (MOLISA) is responsible for the management of social insurance and VSS is a State Agency responsible for implementing social insurance and health insurance policies nation-wide. The responsibilities of MOLISA include among others, advising on the formulation, promulgation and implementation of SI policies and laws; guiding the implementation of the information and reporting regime; settling complaints and denunciations, and handling law violations. The Department of Social Insurance is a specialized unit under the MOLISA, responsible for assisting the Minister in performing the function of state management in social insurance. VSS has the functions of organizing the implementation of social insurance policies; managing and using social insurance funds; and inspecting the payment of social insurance in compliance with the laws. VSS is organized and managed according to a vertical system, centralized and unified from central to district levels. At the central level, VSS has thirteen professional and eight non-business units (Figure 3). At the province level, there is a subordinate provincial social security (PSS) office directly reporting to the central level. At the district level, district social security (DSS) offices have been established, directly reporting to the respective provincial social security offices. There is no DSS office in the city or town where the PSS office is located. The administration of social security by VSS is carried out through a system of service organizations and insurance agents. The grassroots work of mobilizing and expanding coverage in voluntary social insurance and household health insurance is carried out almost exclusively through these organizations. In mid-2022, VSS issued Decision 1155/2022/QD-BHXHVN on the regulations of service organizations collecting social insurance (SI) and health insurance (HI) premiums20, replacing an earlier decision21 in 2016. The 2022 regulation paved the way for changes to the existing agent system notably making VNPost, Viettel and PVI Insurance Company as designated collection service organizations by the VSS. Box 1: Transition to a new system of service organizations and insurance agents Hai Phong City Social Security Office has signed the authorization contract with 14 service organizations as well as coordinated with them to absorb the previous collection agency’s staff, handing over the existing participants in preparation for the extension until the end of 2022; and assigning planned target of new participants. Currently, the city has 14 service organizations including 286 collection points and 1,049 collection staff, 100% of communes and wards have at least 01 collection point. Hai Phong City PSS office has conducted 3 training courses for 350 new collection staff on organizing customer meetings in 2022-2024. Source: Report of Hai Phong City Social Security Office. The PSS offices are now allowed to select other service organizations in the respective provinces in accordance with the prescribed standards (see Box 1). Employees of those service organizations also referred as insurance agents receive training by the PSS to improve their knowledge, and communication skills. 20 Besides Decision the service organizations 1155/2022/QD-BHXHVN VSSfrom implemented also coordinates July, with 01, 2022 clearly stakeholders states regulations like Women’s on service Union and organizations and employees collecting SI and HI premium; collection authorization contract; contract signing process; the Farmers’ Union on mobilization activities, by signing the Regulation on Coordination applicable for providing professional training for service organizations; and systems of statistical form, information and reporting regimes. a given period. These stakeholder groups are however not being used to collect contributions on behalf 21 Decision 1599/QĐ-BHXH dated October 28, 2016, regulating the operation of SI and HI collection agents. 22 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS of their members as is done in Rwanda with the cooperatives, allowing the Ejo Heza scheme to count on a steady stream of active contributors. (Guven & Jain, 2023) Figure 4: Organizational structure of the VSS system THE GOVERNMENT THE GOVERNING BODY VIET NAM SOCIAL SECURITY SUPPORTING UNITS NON-BUSINESS AGENCIES Finance The O ce Institute for Social Security Science accounting dept International Media Center Inspection dept cooperation dept Awards & Planning and IT Center Emulation dept investment dept Archive Center Organization & Legal dept personnel dept Center for Medical review & tertiary care payments Fund investment Intemal audit dept dept Customer service Center SI implementation HI implementation Training Institute for social secunty dept dept operation Contribution Collection-Insurance Social secusty magazine Book and Card dept PROVINCIAL SOCIAL SECURITY OFFICE Si implementation division Medical claim review division Information and Communication - Contribution Collection division Coverage Extension division Book and Card issuance division Personnel division Finance planning division Inspection division IT division The O ce DISTRICT SOCIAL SECURITY OFFICE Source: https://vss.gov.vn/english/aboutus/Pages/organization-stucture.aspx The collection service organizations mobilize, and maintain contact with social insurance contributors, receiving commissions in exchange. Agents are paid remuneration according to the 23 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS regulation prescribed by Vietnam Social Security. The latest remuneration level has been issued in 202222 to replace the previous pay scales from 2020 and is paid for each fiscal year and is on average about 14% of the premium paid by voluntary social insurance participants. This level is adjusted based on the region, chosen contributors’ frequency and whether the contribution is for a new or existing participant (Table 4). The commission rates for agents are substantial as a share of collected premiums, making them an expensive marketing strategy for VSS. During the WB-led consultations where agents and VSS district officers were probed about the cost-effectiveness of these commissions, agents mentioned that the rates were reasonable considering it took them ‘on average 3-5 hours to convince an individual to join the voluntary scheme, and multiple follow-ups at different times of the day to collect contributions from the person.’ The agents also noted that they were the point person for any query the individual had on the voluntary scheme. Agents also shared that their work included not only outreach but also receiving potential participants’ dossiers, making the list of overdue contributors each month; urge and guide the participant to pay the premium; collect premium; transfer the participants’ dossier and premium to the social security office; receive SI account books from the social security office and deliver them to the participants. Table 4: Remuneration for the voluntary SI collection service organizations Rate of remuneration Region I23 Region II24 Region III25 Voluntary SI Participants New Existing New Existing New Existing participant participant participant participant participant participant Contribution frequency - Every 12 months 18.0% 7.0% 19.0% 8.0% 20.0% 9.0% - Every 6 months 16.2% 7.0% 17.1% 8.0% 18.0% 9.0% - Every 3 months 13.5% 7.0% 14.3% 8.0% 15.0% 9.0% - Monthly 10.8% 7.0% 11.4% 8.0% 12.0% 9.0% - One-time contribution for coming years, but not exceeding 5 years + The 1st year 18.0% 19.0% 20.0% + Next years 7.0% 8.0% 9.0% 22 Decision 2222/QĐ-BHXH dated August, 16th 2022. 23 Region I (7 provinces): Hà Nội, Hồ Chí Minh, Hải Phòng, Đà Nẵng, Bình Dương, Đồng Nai, Bà Rịa - Vũng Tàu. 24 Region II (27 provinces): Phú Thọ, Thái Nguyên, Bắc Giang, Quảng Ninh, Vĩnh Phúc, Bắc Ninh, Hà Nam, Hưng Yên, Hải Dương, Thái Bình, Nam Định, Ninh Bình, Thanh Hóa, Nghệ An, Hà Tĩnh, Quảng Bình, Quảng Trị, Thừa Thiên - Huế, Khánh Hòa, Lâm Đồng, Tây Ninh, Long An, Cần Thơ, Vĩnh Long, Sóc Trăng, Tiền Giang, Trà Vinh. 25 Region III (29 provinces): Hòa Bình, Sơn La, Điện Biên, Lai Châu, Lào Cai, Yên Bái, Hà Giang, Tuyên Quang, Cao Bằng, Bắc Kạn, Lạng Sơn, Quảng Nam, Quảng Ngãi, Bình Định, Phú Yên, Ninh Thuận, Kon Tum, Gia Lai, Đắk Lắk, Đắk Nông, Bình Phước, Bình Thuận, An Giang, Bến Tre, Bạc Liêu, Cà Mau, Đồng Tháp, Hậu Giang, Kiên Giang. 24 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Rate of remuneration Region I23 Region II24 Region III25 Voluntary SI Participants New Existing New Existing New Existing participant participant participant participant participant participant - One-time contribution for the missing years for SI participants who have fully 7.0% 8.0% 9.0% met the age requirements but in shortage of vesting period not exceeding 10 years Source: Decision 2222/QĐ-BHXH dated August 16th, 2022 Targets for expansion of social insurance coverage Resolution 28-NQ/TW dated May 23, 2018, of the 12th Party Central Committee on reforming social insurance policies is a key milestone towards universal social insurance. The goal of the policy reform is to expand social insurance coverage gradually and steadily. The resolution also includes objectives of improving the capacity, effectiveness and efficiency of state management and developing a lean, professional, modern, reliable, and transparent system for implementing social insurance policies. Resolution 28 has set targets for the share of participants in the voluntary SI participants to go up to 2.5% of labor force in 2025 and 5% by 2030. By 2023, the coverage in the voluntary scheme has exceeded its target with coverage at 1.82 million people or 3.9% of labor force. To increase ownership of the targets in Resolution 28, sub-targets have also been assigned to People’s Committees across all levels of government per Resolution 102/NQ-CP dated August 3, 2018. The People’s Committees are required to ‘develop targets for developing social insurance participants in their localities and submit it to the People’s Council at the same level for inclusion in annual socio- economic development targets, and at the same time develop plans to implement the set targets...’. By the end of 2022, 55 provinces have included the target of social insurance coverage in the Resolution on socio-economic development26. These Resolutions and relevant legal documents have facilitated the mobilization of the political and administrative system at all levels, including political and social organizations, unions, professional associations. Modernization of the social insurance system VSS aims to improve operational efficiency by enacting administrative procedure reforms and digitizing business processes. The number of administrative procedures in the VSS has steadily decreased from 115 procedures in 2012 to 33 procedures in 2015, 32 procedures in 2016, 28 procedures in 2017, 27 procedures in 2019 and 25 procedures in 2021 (Annex 1). VSS has also made significant progress with the digitalization of business processes starting with a pilot in May 2015. Since 2015, digitization of business processes has extended to cover registrations, contribution collections, issuance of SI book and HI cards, as well as other administrative and service delivery functions. This 26 https://hanoimoi.com.vn/tin-tuc/Xa-hoi/1047900/co-13-tinh-thanh-pho-ho-tro-them-muc-dong-cho-nguoi-tham-gia- bao-hiem-xa-hoi-tu-nguyen 25 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS comprehensive digital transformation has led to an increase in digital service maturity by VSS to level 4 in line with guidance from the Central Government (Table 5). Organizations and individuals can use the online Public Service Portal27,28 to enroll in VSS schemes, whereby users are identified as well as authenticated using the Social Insurance Transaction System. VSS has also developed the National Database on Social Security connecting it with the National Database on Population29. They have issued unique Social Security number for individuals and introduced a VssID application to establish a digital social insurance ecosystem. Table 5: Online public service levels Level Concept Level 1 Informational services - The online public service that provides full information about the process and procedures; profile; duration; charges and fees for service provision. Level 2 The level 2 online public service that allows users to download the forms and documents to complete the required registration profile. The completed profile shall be sent directly or by post office to the agencies or organizations providing the services Level 3 The level 3 online public service that allows users to fill in and submit forms online to agencies and organizations providing the services. Transactions in processing profiles and providing services are carried out in the network environment. The payment of fees (if any) and receipt of results shall be made directly to the agencies or organizations providing the services. Level 4 The level 4 online public service that allows users to pay fees (if any) online. Receipt of results can be made online, sent directly or mailed to the user Source: Circular 32/2017/TT-BTTT The Social Security number issued by the VSS serves as a foundational pillar of the VSS digital ecosystem. The Social Security number is a unique identification number issued by the social security office to each participant and is recorded on the social insurance book and health insurance card of each participant30. This unique sequence of 10 digits has the first 2 digits as the province code and can be retrieved from the VSS Portal (at: //www.baohiemxahoi.gov.vn) by providing the individual’s full name, province, date of birth and ID number. The unique number allows for a quick look up of participants‘contribution history, eligibility criteria, or membership status for all social insurance schemes. 27 https://dichvucong.baohiemxahoi.gov.vn. 28 https://baohiemxahoidientu.vn.  he Ministry of Public Security announced the official operation of the National Database on Population from July 1st, 2021 29 T covering 98.7 million Vietnamese citizens, which are digitized, supplemented and regularly updated at grassroot level by local police in nationwide. It is the basis for agencies and organizations to verify and unify information about citizens. 30 Decision 515/QĐ-BHXH dated March 27, 2020. 26 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS The National Database on Social Security31 is another initiative by the VSS under its e-government ecosystem32. The database covers over 98 million individuals and stores information on social insurance, health insurance, unemployment insurance and information on health and social protection. The National Database on Social Security includes information on basic personal data (name, age, gender, ID number, ethnic, nationality, details of immediate household members); information on coverage and claim history for social insurance, health insurance, and unemployment insurance; information on the employer; and basic medical details. VSS’s system has successfully connected with the National database on population, and more than 97.6 million demographic information contained in The National Database on Social Security has been authenticated. Box 2: Health insurance system in Vietnam Vietnam has made significant progress toward achieving universal health coverage with the enactment of the Law on Social Health Insurance in 2008 and its revision in 2014. Following the introduction of compulsory insurance in 2015, the country aims to increase the health insurance coverage rate to 95% by 2025. As of 2022, 93% of the Vietnamese population is covered by health insurance (HI). The Vietnamese HI system is already considered one of the most successful in Asia. The government fully subsidizes premiums for the poor, ethnic minorities, children under 6, the elderly over 80, and other socially vulnerable groups. Near-poor households receive partial subsidies. To further expand coverage, the government has introduced a family package offering a 10% discount for additional family members joining the scheme. In addition to the national subsidy policy, provinces provide top-up support for HI premiums for people in near-poor households, those escaping poverty, the elderly, ethnic minorities, people with HIV/AIDS etc. The National Insurance Database (NID) is one of six major national databases which manages data on social insurance, health, and unemployment for over 98.7 million individuals and households. The HI database is the largest within the NID due to its extensive coverage. As of May 2024, 71 million records in the NID have been authenticated and verified against the national population database. Starting July 2024, with interoperability among different database, services such as birth registration, permanent residence, and health insurance cards for children under 6 can be processed online through the National Public Service Portal or the VNeID application. Due to its extensive coverage, the HI database contains valuable information such as demographic details and health insurance contribution history can be potentially utilized to enhance the VSIF implementation, such as targeting outreach efforts and automatically populating relevant data into VSIF registration forms. In the later stages, it may also be possible to streamline the collection of contributions for both health insurance and VSIF simultaneously, thereby reducing costs for agents. The “VssID - Digital Social Insurance” application on mobile device platform of VSS has also facilitated improved service delivery. The VssID was officially announced in 2020 and as of 2023, 35 million individual electronic transaction accounts on VssID were active across the country. It is one of the 31 Decision 714/QD-TTg dated 22/5/2015. 32 In addition to other key databases such as: (1) National database on population in charged by the Ministry of Public Security; (2) National Database on Land in charged by the Ministry of Natural Resources and Environment; (3) National database on enterprise registration in charged by the Ministry of Planning and Investment; (4) National database on population statistics in charged by the Ministry of Planning and Investment; (5) National Database on Finance in charged by the Ministry of Finance. 27 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS three state agency applications with the largest number of users in Vietnam, along with 2 others: VneID of the Ministry of Public Security and Vietnamese Youth of the Central Youth Union33. The application provides the following functions: • Static Information services: HI card, SI participation process and rules, information on unemployment insurance, industrial accident and occupational disease insurance; information on SI benefits (one-time, sickness, maternity, etc.); history of medical examination and treatment with HI, etc. • Information search services: Social insurance code; Social insurance agency; Medical examination and treatment establishments issuing leave certificates to enjoy social insurance benefits; Medical examination and treatment establishments that sign contracts for medical examination and treatment with health insurance; organization participating in social insurance, etc. • Online public services: Re-issuance of SI books and HI cards; updating location / address to receive pensions and social insurance benefits; changing the form of receiving pensions and social insurance benefits; updating participant personal information, etc. • 24/7 online support: Chatbot support, hotline information (1900.9068), email support, FAQs, etc. • Broadcast information on activities of the social insurance (information obtained from the web portal of VSS), information on social insurance regimes and policies (including updates), application instructions; online payment services, etc. Figure 5: Screenshot of VssID application Source :https://baohiemxahoi.gov.vn/gioithieu/Pages/tai-ung-dung-vssid.aspx 33 MOIT, 2023, Report on the results of national digital transformation implementation in 2023. 28 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS III. Delivery systems assessment of the Voluntary Social Insurance Scheme The objective of the Delivery Chain Assessment is to analyze the different phases of the delivery chain (from Outreach to Exit) in the Voluntary Social Insurance Scheme from the perspective of current and prospective members. Such an assessment can enable policymakers, practitioners, and stakeholders to get a bird’s eye view of the operating environment of the scheme, identify implementation gaps and operational inefficiencies. The assessment uses a framework developed by the Sourcebook on the Foundations of Social Protection Delivery Systems (Lindert et al., 2020) but reimagined for social insurance. In contrast to non-contributory social assistance programs, a social insurance scheme has the distinct function of collecting contributions in addition to the delivery of benefits. This distinct characteristic of social insurance schemes is why analyzing their delivery systems requires looking into both Government-to-person (G2P) payments as well as person-to-Government (P2G) payments. Contributory social insurance schemes must also manage members’ contributions efficiently over long periods. Throughout the contribution collection phase, the social insurance institution must keep accurate records and make them easily available to members. In the payout phase, the benefits need to be calculated based on the scheme’s rules and provided to the right beneficiary when payout is due, in a timely and transparent manner. Increasingly social insurance schemes (health and pensions) also receive subsidies or matching contributions from the Government to boost take-up or ensure coverage of the poorest, such as in the case of the Voluntary Social Insurance Scheme. Due to limited fiscal resources, the incentives are often targeted and provided to individuals who meet some eligibility criteria. Consequently, assessing the eligibility criteria, crediting the incentive to the individual in a timely fashion, and ensuring dynamism and accuracy in the entire process is crucial to the customer experience and building trust among informal sector workers. Gaps in the delivery chain – either through poor operational design, poor communication, or both - can hinder uptake and persistency in voluntary schemes (Guven & Jain, 2023). The Delivery Chain Mapping for Voluntary SI in Vietnam was carried out through a mix of desk research, and discussions with operational staff (at central, provincial and district levels). The reimagined Delivery Chain includes five main stages (Assess, Enroll, Contribute, Provide and Manage) and each stage has two implementation phases as shown in Figure 6. 29 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Figure 6: Voluntary Social Insurance - Business process Source: Created by authors 1. Assess The first stage of the Delivery Chain is the Assess stage which includes two substages – the Outreach and Registration stage. This stage is where the customer has their first interaction with the scheme, and it is advisable for schemes to focus on behavioral nudges that confront the limited attention at the Outreach stage and the complexity of information in the Registration stage. 1.1. Outreach In the outreach stage, raising awareness, building trust, and being present (in-person, digitally, or through partners) to answer questions on the scheme can help informal sector workers make informed decisions about participating in the Voluntary Social Insurance Scheme. At this stage, potential participants typically like to know the design of the scheme, how the scheme meets their needs, why they should trust the scheme, their options for participation, who they can call if they have questions, and so on. Transparently communicating these details helps build trust in the scheme. Outreach is vital for the success of voluntary schemes because unlike mandatory schemes, members must be convinced of the benefits, to decide if they will participate in the scheme. 30 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Outreach efforts in the VSI are supported and guided by a communication strategy set at the national level for a period of 5 years, followed by an annual plan at the provincial and district levels. The communication center at the central level is the focal point for organizing the development and implementation of communication plans; guiding provincial social security offices to develop and implement provincial plans; organizing inspections and providing professional guidance; organizing activities at national level; formulating contracts for coordination with ministries, government agencies, mass organizations; as well as monitoring, and coordinating information in the press. The communication center works closely with other VSS departments which provide periodical data/information that can be used to inform public opinion of the scheme, clarify rules, and create social consensus in the observance of policies and laws. The VSS’s communication plan at the central level outlines initiatives including awareness campaigns and dissemination of legal policies; updates on new regulations; results achieved; promotion of digital payments; instructions on information channels and addresses for receiving requests, filing of complaints, and submitting of feedback. The plans at the province and district level are more detailed, including targets, audiences, measures, tools to implement during the year and the accompanying budget proposal (see Table 6). Table 6: Communication role and responsibilities by level Provincial level Grass root level Central level (Communications District level (Collection service Activities (Communication and customer (Communications organizations, Center is focal point) development officer) mass division) organizations) Developing for the whole country for province communication strategy/plan Training for VSS’s for staff of communication staff at collection service provincial and district organizations, level agents, collaborators Works closely with ministries, central with provincial agencies, mass departments, mass organization, national organizations, media agencies media agencies Carrying out √ √ propaganda, consultation and conferences Propaganda √ √ √ √ publications (Designing, editing) (Distributing (Sharing it with the (Sharing it with the it at district target audience) target audience) level, to service organizations, mass organizations) 31 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Provincial level Grass root level Central level (Communications District level (Collection service Activities (Communication and customer (Communications organizations, Center is focal point) development officer) mass division) organizations) Propaganda on √ social networks and technology platforms Identifying the target √ group for the voluntary scheme Conducting √ propaganda on the grassroots radio system, loudspeakers Customer meetings √ √ Small group √ counselling, home visiting Hot line √ √ √ Source: Compiled based on interviews with VSS staff The provincial level offices of VSS identify target groups and prepare lists of individuals, using data on individuals from partners or other government programs, albeit without a formal process to seek individual consent. The target groups are identified by provincial level teams using their knowledge of the contributors in the scheme include socio-economic situation and observable characteristics like age, gender, employment type, and potential income of workers based on the location of their homes. VSS teams at the provincial level typically target households with stable and higher incomes such as individual business households, small production households and farming households, borrowing from the list of household health insurance participants, members of Women union, Farmer Associations, etc. (Box 2). Data on individuals available with grassroots partners or other government agencies are shared with the VSS to prepare lists. It is unclear if and how consent for sharing individual information is sought. Once the list of potential contributors to the scheme is created, the VSS staff develops outreach plans identifying the appropriate means and methods of mobilization. 32 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Box 3: Interview excerpts from two provinces outlining how targeting of potential customers is carried out “We determined that people over 30 years old are just starting to pay attention to social insurance, because young people often do not think about the distant future… The Women’s Union used to be a collection agent, so we had more access to these groups. In addition, the Farmers’ Union, Labor Confederation, Youth Union depending on the location are also on board… Commune health stations provide the list of people with phone numbers to the district social security office. This is personal information, but we are allowed to be provided with it for the purpose of implementing social protection policies. – Group discussion at Thai Nguyen Social Security “Hai Phong has many industrial zones and factories, so there are many formal employees in the province. We compile the list of employees leaving work, benefiting from unemployment insurance and employees who reserve the process of participating in compulsory social insurance, the list of people who have not participated in voluntary SI. The collection service organizations utilize these lists to develop plans and organize customer meetings, conduct dissemination campaigns and advocacy activities” – Group discussion at Hai Phong Social Security. The identification of potential customers using local level information is useful to target outreach efforts, but VSS should consider a systematic use of data sources that also protects individual privacy. Analysis of representative surveys (household, labor force, business surveys) can reveal characteristics of workers who are more likely to save allowing provincial teams to target efforts based on those observable characteristics. Government data registries e.g. the Rwanda Social Security Board uses the list of cooperatives obtained from the Rwanda Cooperative agency (Box 3) or MoUs with Telecom company whereby consent from individuals is sought before sending them advertisements, can be a way to reach groups that are most likely to save, in a systematic manner while protecting individuals right to privacy. The analytical work and signing of MoUs can be spearheaded by the National VSS office, liaising with the General Statistics Office (GSO) in Vietnam. Sophisticated analysis like using telecom recharge data as a proxy indicator to potentially identify individuals with higher disposable incomes can be considered if data sharing agreements can be put in place. A standardized approach to identifying those with an ability to save would ensure that all provinces, even those with limited capacity, can benefit from some analysis to guide their outreach efforts. 33 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Box 4: The role of cooperatives in the Ejo Heza Voluntary scheme, Rwanda LongTerm Saving Scheme (EJO HEZA) was established by the Government of Rwanda through the Ministry of Finance. It is a defined scheme, established on voluntary basis by opening a savings account with a scheme administrator, the Rwanda Social Security Board (RSSB), and covers both salaried and unsalaried people. Ejo Heza covers 2.1 million individuals (as of April 2022), accounting for 27 percent of the population between ages 15-64. Over 60 percent of current Ejo Heza members were enrolled through cooperatives. Using cooperatives as an aggregator has been key to the success off Ejo Heza to date, as they assume a role similar to that of employers in the formal sector scheme. But unlike a formal sector scheme where contribution is mandated, the cooperative contribution is agreed upon by all its members, and once decided cooperatives auto-debit the contribution from the harvest income of members. Members of cooperatives (for example, in agriculture, fishing, farming, and services) typically have seasonal incomes so the frequency and amount of payment is decided collectively by each cooperative. Cooperative heads send commitment form to the members, members fill them, and the heads collect those forms and keep them as records in their finance office. At the time of harvest, the agreed contribution is auto-deducted based on the amount in the commitment form and deposited in a bank. Ejo Heza then credits the amount into individual accounts of cooperative members and sends a notification (via text) to all members in the cooperative. Members can check their balance any time and they are not charged for doing so. Cooperatives don’t receive any special incentives for enrolling members and have reported that many of their members are part of other savings groups and understand the importance of saving. Source: Guven & Jain (2023) In-person and digital communication channels are used for outreach, but face-to-face communication via agents, and loudspeakers are deemed to be most effective by VSS. Traditional communication channels like pamphlets, loudspeaker/radio, agent network, and word of mouth are used predominantly for outreach of the voluntary scheme. VSS staff and agents engage with the target group via customer meetings, small group counselling and follow up with potential customers via home visits by agents or employees of service organizations. During the consultations, one of the agents remarked that they ‘need to schedule home visits in the evening as those in the informal economy are geographically mobile making it hard to reach them during their work hours.’ The use of agents for face-to-face outreach can be time consuming and costly, they help potential customers understand the design of the scheme, build trust, and can lead to an immediate registration. The loudspeaker system is also deemed effective per the VSS communication staff, especially at the commune and village level where information about the scheme is broadcast at select times of the day. Digital technology can be better leveraged by the VSS as a complement to the agent network to conduct outreach in a cost-efficient manner. As digital awareness and familiarity with digital apps grows in Vietnam, complementing the agent network with digital channels for the purpose of awareness raising, easy registration (like the simple three step registration in schemes of Rwanda, Kenya); and contribution collection can complement the agent network to reduce overall costs. Other initiatives like creating digital groups of agents like the one created on Zalo (an online messaging application) would also allow the network of staff/agents to share their experiences on what works, and exchange answers to commonly asked customer questions. Social distancing measures during Covid-19 also spurred the 34 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS use of virtual technology via conferences between agents across provinces to exchange ideas and share opinions (Figure 6). Figure 7: Online conference between VSS Thai Nguyen, Provincial Woman’s Union at Phu Luong District People’s Committee, Giang Tien town, Tuc Tranh commune, Yen Do commune. Source: https://hoilhpntn.thainguyen.gov.vn A leaflet on the voluntary scheme exists but it is information dense and does not include information that potential customers care most about viz the expected amount of pension benefit. The communication materials (leaflets, videos) for the voluntary scheme are designed and produced at the national level; and distributed to provinces and districts. Local VSS offices can develop their own communication materials after consulting with the national head office. The leaflets (see figure 7) are usually handed over to the audience in customer meetings or group consultations. Panels, posters, and standees can also be found in public places and social insurance offices. The content of communication materials focuses on the rules e.g., required contributions, minimum years of contribution, subsidy by the state but they fail to provide clear information on expected benefits. Agents revealed that potential customers are keen to understand ‘what they will receive if they contribute’ and agents currently lack material, or official tools to answer this question. Some agents currently use back of the envelope calculations to estimate benefits but noted that an official digital tool or a modified leaflet which includes estimates would be impactful. The information in the existing leaflet can be further simplified to highlight the main scheme related information. Additional leaflets can be produced for different objectives like sharing testimonials from existing members or mentioning ways to check last contribution or contribute digitally. A variety of leaflets would also increase its relevance for different target groups and differentiate between new and recurring participants. 35 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Figure 8: Voluntary SI leaflet in Thai Nguyen province Source: Thai Nguyen PSS The funding for outreach activities is allocated to the Communication Center from the VSS’s annual management expenditure34 but there is a need to track spending on voluntary scheme and assess its cost-effectiveness. The VSS does not keep systematic records of amount paid to agents and neither does it track the noteworthy agents across provinces. The total spending on mobilization activities in the voluntary scheme is also not available, as VSS does not separate communication spending by scheme. While it can be economical to use a common pot of resources for outreach, the lack of detailed data on spending by program/activities limits the ability to determine the cost-effectiveness. Since outreach and communications is critical to attract and retain contributors in a voluntary scheme, it is important to be able to use these resources cost-effectively. This will require a separate budgeting mechanisms for the voluntary scheme at minimum as currently it is not feasible to answer questions like how the spending on outreach activities changed over time, which provinces are more effective in using the funding effectively for outreach, what share of the total outreach budget is spent on agent commissions etc. 1.2. Registration The information provided in the outreach phase serves as the foundation for the registration phase. Once individuals have made the decision to participate, they need to know how and where to register, and how to navigate the registration process. Common questions at the registration stage include where to enroll, how to access & complete paperwork, how to use digital means to register, how to submit paperwork or complete a digital transaction, how to set up notifications etc. Individuals can register for the voluntary scheme online or in person, though most registrations currently happen in-person. Interested individuals need to submit a registration form (see Annex 5) accompanied by proof of identification (ID card, citizen ID35 or Passport) for verification purposes. 34 The management expenditure of VSS is calculated based on the percentage of VSS’s revenue and expenditure estimates and assigned by the Standing Committee of the National Assembly. The ratio of social insurance management expenditure by year is: 2019: 2.15%; 2020: 2% and 2021: 1.85%; 2022 maximum 1.59%, 2023 maximum 1.54% and 2024 maximum 1.49%. The citizen ID system (12 digits) is a more recent endeavor and prior to its launch people used ID card (9 digits) for 35  identification ad verification in Vietnam. MPS only issues citizen card for citizen but ID cards are valid until the individual replaces it with a citizen card or the ID card expires. 36 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Registration forms are available at VSS offices or with insurance agents and can also be downloaded electronically. No registration fees are levied, but registrants are required to pay the first premium as confirmation of their participation. The premium contribution can be paid in cash (via agents/staff of VSS) or through mobile banking (for online registration). The self-registration via online portals was introduced after Nov 2022 when VSS issued a decision36 outlining the process of self-registration using VSS’s Public Service Portal (http://dichvucong.baohiemxahoi.gov.vn) or National Public Service Portal (http://dichvucong.gov.vn). (See Annex 6 with snapshots of these portals). The interoperability with the national population database in case of self-registration allows users to complete their registration and payment of first contribution in a matter of minutes, greatly enhancing the user friendliness of the registration process and reducing costs (see Fig 8 for a comparison between registration processes in person and online). Despite the availability of online registration services, its’ adoption is yet to be scaled up. Local VSS staff in Thai Nguyen and Hai Phong noted that currently registration through the VSS insurance agents is common because of the face-to-face interaction which includes support by the agent to fill the registration form, ensure data is entered correctly, submit the information to VSS; and follow-up with the individual to ensure successful registration after the first contribution payment. The relatively recent introduction of self/digital registration, lack of a concerted push to encourage self-registration, and the vested interest of agents who would have to let go of commissions if individuals chose self- registration, are other reasons for the low adoption of digital registration. The registration form (TK1-TS) should be further simplified and include mandatory and optional fields. The form has separate sections for those who have a social insurance numbers and ones who don’t. Individuals who do not have an existing social security number issued by VSS need to provide their personal information (full name; gender; Date of birth; Nationality; Ethnicity; Number of ID card/ citizen ID card/Passport); Phone number; Email (if any); Place of birth registration; and address to obtain the result of registration process. For individuals who already have a social insurance number (either because they were a member of compulsory scheme or are enrolled to receive health insurance) only need to provide the number and their personal data is automatically retrieved from the VSS database. The form currently also asks for information on all household members even though the voluntary scheme only required information on the individual and his/her nominated beneficiary (Annex 5 for a sample registration form). Some of the information like frequency of contribution can be made optional. These simplifications in the registration form will allow for a quicker completion and capturing of information directly relevant for the scheme. On the flip side the form can include some optional data fields as drop downs like where they heard about the scheme, what motivated them to join, or their primary occupation. If members choose to complete these fields, it can give valuable information to the VSS on effectiveness of its outreach and targeting mechanisms. 36 Decision 3511/QĐ-BHXH dated November 21st 2022. 37 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Figure 9: Stylistic diagram of registration via agents vs online Registration via agents Online self-Registration via public service portals Agents of service District Social Registrant Registrant VSS's systems organizations Security O ce 1.1 Fill in 1.1 Fill in e- 1.2. Verify with registration 0. Couseilling registration National DB on form (TK1-TS) form (02-TK) Population 3.1. Receive 1.2. Show ID 1.3. Check the Place of Poverty information No card registration residence status 1.4. Contact local 1.3.Report to authorities for the registrant 2.1. Pay the 14. Make the rst premium list of data update Yes Yes registrants 2.1. Calculate (DOS-TS) the premium 2.2. Issue the 2.2. Pay the 2.4. Notice the invoice rst premium 2.3. Issue time delivering Phase Phase on the Public the invoice the SI book Service Portal cover Source: Compilation from guideline of VSS 2. Enroll The second stage of the Delivery Chain is the Enroll stage which includes two substages – the Verification and Onboarding stage. At this stage the individual has decided to join the scheme but the experience at the verification and onboarding stage will cast an impression on the member. A smooth enrollment process can enhance participant experience and promote improved interactions. 2.1. Verification Once the applicant’s identification is collected at the registration stage it must be verified against the national population database to uniquely identify the prospective participant. Online registration allows for a real-time check of personal information given the interoperability of VSS data systems with the National population database. The National Population Database, managed by the Ministry of Public Security37, collects basic information from all Vietnamese citizens in a standardized digital format, and this data is regularly updated at grassroots level by local police nationwide. The Ministry of Public Security announced the official launch of the National Database on Population on July 1st, 2021, and it already includes digitized data of 98.7 million Vietnamese citizens. If registration was done via an agent or through a walk-in at a district social security office, the detail of the person is added manually to the social insurance database by VSS staff. The registrant is required to show their citizen ID and their personal info and whether they are already enrolled in the compulsory SI is verified 37 Decision 714/QD-TTg dated 22/5/2005. 38 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS in the VSS database using the citizen ID number. The only disadvantage of registration via agents is that it can take an additional day or two as agents need to submit the information to the VSS office and lack laptops/authorization to complete the registration remotely. 2.2. Onboarding Onboarding commences with the notification of the enrolment decision to the individual. The notification is either a confirmation of participation or the participant being wait-listed due to inadequate information or deemed ineligible to participate in the scheme. A positive enrollment decision is near certain upon a satisfactory completion of the registration process. Many of the processes on onboarding are automated, these include: • Creating and managing participant profiles and linking them to the code of the specific management unit at the provincial/district social security office where the participant resides. • Recording the desired monthly premium, recording of any past social insurance contributions to the voluntary scheme, calculation of the amounts net of state support to be paid by the participants (Form C69-HD), determining the remuneration for the service organization collecting the contributions (Form D11-TS) according to regulations. • Determining the amount of remuneration for the service organization/agents (Form D12-TS) according to regulations. • Creating data on the cover sheet of the social insurance book for first-time participants. The social insurance book has both a digital version for those who registered online and a physical book version for those who registered in-person. The automated processes are complemented by manual checks and verifications performed by staff at local levels. Once the software system has completed the steps listed above, the collection and card book staff check the individual profile and data, submit them to the Director of the provincial/ district social security office to sign the cover of the social insurance e-book digitally or print the cover sheet of the SI book in paper. The former is updated in the system and the latter is sent to the receipt and return staff, who return it to the participant to complete the onboarding within five working days. Although the telephone number of the registrant is collected, VSS has not yet used text messages as a channel notify the registrant about the progress of the processing and outcome of their registration. Since the launch of the VSSID application, VSS has encouraged people to install the app and use it as an alternative to the SI book and HI card. The users of VSSID application can check their participation status and history using the digital app. Online public services related to both social insurance and health insurance are also provided through this application. As of July 31, 2022, there were over 26.3 million accounts across the country. However, the issuance of paper-based social insurance books is still applied because almost all voluntary SI participants seem to prefer the paper-based book, based on insights obtained from local VSS in Thai Nguyen and Hai Phong. A proactive digital communication strategy to familiarize individuals with the use of the digital app and outline its benefits, will need to be developed. Some countries like Rwanda have taken the bold decision of not releasing any paper- based book and instead focused on simplifying the website, staffing call centers, and educating 39 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS local governments about the schemes so that participants can get their queries answered and avoid dependence on the paper-based book which is easily lost and is inefficient considering it needs to be updated manually at each contribution. 3. Contribute The third stage of the Delivery Chain is the Contribute stage which includes two substages – the Contribution and Benefits Eligibility stage. At the contribute stage, the individual has started making one or more contribution to the scheme. A lack of understanding of the scheme’s benefits can dissuade registered members from contributing. It is important to include incentives in product design to attract workers and ensure they are valued and understood by the target group of the informal sector. 3.1. Contribution Payment of social insurance contributions is facilitated via multiple channels, including insurance agents, internet banking via commercial banks and the National Online public service Portal. Contribution via insurance agents is the most popular channel in rural areas while internet banking is becoming more common in urban areas. Contributors can choose the option to automatically deduct their monthly contributions from their account if they contribute online. While the strong presence of agents has its advantages in terms of ensuring consistent contributions (because agents follow-up personally), digital means of contributing should be encouraged over time given the significant time savings for participants and cost saving for the social security office. VSS depends on agents to send contribution reminders to participants and lacks a systematic process for sending reminders or an auto-enrollment option for in-person contributions. VSS has phone numbers of participants of voluntary social insurance, but it does not use text messages to notify them of the time and premiums of contribution, citing the exorbitant costs of doing so. Participants who are logged in to the VSSID mobile application however receive automatic notifications. For the rest, the district offices export and send the list of the contributors with overdue contributions each month, to the agents of collection service organizations in the respective districts. The agents are then responsible for notifying the participants and collecting the premiums as applicable. The collected premium is recorded in the profile of the participant, commission transferred to the agents, and the total collected premium transferred into the SI Fund for investment. and also processes the remuneration for the service organization (figure 9). 40 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Figure 10: Business process of collecting contributions Contribution (recurring cycle) Participants Agents of service organizations District Social Security O ce 2. Report to 1. Export the list of participants payer from DB 3.1 Pay the periodical premium Cash via agent 5.1. Verify Online and mobile banking 3.3. Check information/ information cash ow National public service portal 3.2. Show ID card/SI 3.4. Issue the 5.2. Record the identi cation invoice contribution 4.1. Make the list 5.3. Update data of payers into DB 6.1 Check information 5.5. Calculate the in VssID remuneration 5.4. Allocate the amount of premium Phase Source: Compilation from interviews with VSS; DB= Database 41 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS During consultations agents revealed that the sharp jumps in minimum contribution level adversely impacts their credibility with participants when they go to collect contributions. The minimum level of contribution in the voluntary scheme is dictated by the poverty line which in turn gets revised every 5 years. The last increase in poverty line was from VND 700,000 to VND 1,500,000 and resulted in a more than double increase in minimum contribution. Agents reported having a hard time explaining to contributors why the minimum they had to pay increased substantially. Moreover, in the absence of a tool to show how the increase in contributions would result in higher benefits, they were at a loss in explaining the long run benefits of having contributions growing in line with living standards. If participants fail to pay premiums within the prescribed time limit, they are considered to have temporarily stopped paying voluntary social insurance premiums (i.e., dormant participants). To support dormant participants in resuming their contributions, the VSS has established the process of re- registration which allows individuals to resume paying contributions even after a gap of some time. Any time there is a change in the contribution premium or contribution method or contribution frequency a similar process like re-registration needs to be followed for participants who contribute using in-person methods. They need to submit a form (TK1-TS) to provide the information relevant to processing such changes. 3.2. Benefit Eligibility Contributing consistently during their working years only to receive a benefit in old age requires individuals to be clear of and value the benefits being offered in the scheme. Governments offer monetary incentives in voluntary schemes via multiple ways - as a flat benefit on registering in the scheme (India’s APY scheme) or a contribution subsidy (Vietnam) or a percentage match to individual’s savings with a cap (Rwanda, Thailand) or as a top-up available at the time of retirement (Colombia). The monetary incentive provided by the government can be poverty targeted if there exists a social registry that can be used to verify household poverty levels. Besides monetary incentives, the scheme can also be made more attractive by including insurance against short term risks like life and funeral insurance, health, accident, or auto insurance. A defined contribution pension scheme which provides returns higher than inflation or at a rate higher than saving accounts in banks can also be desirable for workers in the informal sector, especially women who often keep savings as cash in the home eroding its real value and risking it getting stolen or used in household expenses. Yet, other schemes offer flexibility in the use of accumulated contributions, to attract liquidity constrained informal sector workers e.g., in Rwanda’s Ejo Heza scheme, individuals can take out a housing loan once they have saved at least 4 million RwF (~USD 3,500) into their account or they can choose to withdraw up to 40 percent of the savings in excess of Rwf 4 million Rwf. Vietnam’s voluntary scheme offers contribution subsides and benefits in the long run besides pension but lacks benefits that can be realized in the short run. As of 2024, the benefits offered to those who contribute to the voluntary scheme include a contribution subsidy for ten years that varies based on poverty status, old age pensions, survivorship benefit, health insurance during retirement, and a lump-sum funeral allowance. Besides the contribution subsidy all other benefits are realized in the long run. The contribution subsidy itself is modest relative to subsidies offered by other countries and as found in survey results (Section IV) the knowledge on state subsidies is low. The lack of scheme benefits that can be realized in the short run limits the attractiveness of the scheme for an informal sector worker 42 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS who has a strong present bias (low discount rate) and values present consumption more than the future. Additionally, poor communication, and lack of tools to estimate the benefits a contributor will receive makes it harder to convince informal workers to join the scheme. 4. Provide The fourth stage of the Delivery Chain is the Provide stage which includes two substages – the Withdrawals and Benefit Payout stage. At the Withdrawals substage, the individual interacts with the scheme if he/she is eligible for premature withdrawals and the Benefit Payout stage refers to the substage at which the individual interacts with the scheme to receive the benefit after completing all the eligibility requirements. 4.1. Withdrawal (for one-time social insurance allowance) The voluntary scheme allows withdrawal before retirement in case of death or when an individual chooses to exit the scheme. As of 2024, there is no provision for partial withdrawal which is understandable given the defined benefit nature of the scheme (similar to schemes in India, and China). In defined benefit schemes pension is calculated based on a formula and allowing partial withdrawals would impact the sustainability of the scheme. The only flexibility that can be offered in a defined benefit scheme therefore is to allow participants to withdraw all their contributions as a lump-sum and exit the scheme altogether. The lack of flexibility in withdrawals is a drawback of defined benefit schemes for informal sector workers. Administrative data on lump sum withdrawals in Vietnam’s pension schemes suggests that compulsory scheme participants are more likely to withdraw and exit the scheme than voluntary scheme participants. Most withdrawals are concentrated between the ages of 20 and 3938. The one-time withdrawal and exit feature of the VSS pension schemes hurts the coverage expansion objectives of social insurance and has been the focus of the new law discussed in National Assembly in June 2024. Starting July 1, 2025, those participating in social insurance shall not be able to withdraw social insurance except in select cases39. To apply for one time withdrawal the applicant must prepare a dossier and fill in an application form (14-HSB) that can now be done digitally. The time for verification of the withdrawal is twelve working days from the date that the social security office receives the applicant’s dossier, and time for processing the benefit is five working days after verification. VSS has recently issued the process of settling the one-time withdrawal benefits online; and is piloting digital signature authentication integrated in the VSSID mobile application40 (figure 10). Data and documents that exist in the VSS database are not required to be re-submitted. The applicant can choose to receive the results on paper- based documents or electronically; and receive cash at the social insurance office or through his/her personal bank account. However, the applicant is required to return the paper-based SI book to the social insurance office for storage. 38 http://tapchibaohiemxahoi.gov.vn/bhxh-mot-lan-o-viet-nam-thuc-trang-dang-lo-ngai-91109.html. 39 The selected cases where full withdrawal will be allowed for new joiners include (a) Reach the retirement age but have less than 15 years of contribution; (b) Go abroad to settle; (c) Suffer from one of the following diseases: cancer, polio, decompensated cirrhosis, severe tuberculosis, AIDS; (d) have a working capacity decline of 81% or more; with particularly severe disabilities. 40 Decision No. 3612/QD-BHXH dated December, 9, 2022. 43 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Figure 11: Settlement of One-time SI benefit in-person and digitally Source: Compilation from guideline of VSS 4.2. Benefit payment Processing of Pension payment on retirement still requires submission of paper-based documents. The application form (form 14-HSB) is used to apply for pension benefit and the law prescribes the benefit to be processed within 12 working days of receiving the dossier. Participants of voluntary SI are required to prepare dossiers including: (1) Social insurance book; (2) Original application form No. 14- HSB; (3) Supporting documents for specified cases. The application form requests basic demographic info (name, date of birth, VSS ID, mobile number); the benefit being requested (e.g. pension or funeral allowance or survivorship benefit); address to receive pension; place of registration for medical examination and treatment (for the pensioner’s health insurance card); and preferred channel to receive benefit. Once the dossier is submitted, the social security office verifies the information and calculates the defined benefit based on the participant’s contribution history and benefit calculation rules. Once verification is complete, the result of the settlement viz the monthly pension amount and the health insurance card is sent to the individual at the contact address mentioned in the application form. The payment of pension benefits is primarily through VN Post, but the use of digital channels is increasing, following concerted actions from VSS. The beneficiary selects the preferred payment channel by providing the information in the application form 14-HSB. Currently, the VSS provides monthly pension benefits to nearly 3.3 million participants, but these include very few members 44 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS from the voluntary scheme since it was introduced much later than the compulsory scheme. VN Post is the only service provider to pay pension benefits. To encourage the transition to digital payments, VSS pays the fee to obtain the first ATM card for the pensioner receiving pension via bank account (see Box 4 for experience in Thai Nguyen province). By the end of 2021, about 57% of beneficiaries receiving pensions, and unemployment benefits in urban areas have received their benefits through digital means, an increase of 9% compared to 2020 (and an increase of 28% compared to 2019)41 . In rural areas though, the number of people receiving pensions and monthly allowances at payment points via cash is still very high. During the Covid-19 pandemic, VSS closely coordinated with VN Post to introduce flexibility in the payment of pensions and social insurance benefits, such as making payments every 2 months as opposed to monthly; paying at payment spots or at home in high-risk localities; and encouraging people to change the payment channel to a personal bank account. Box 5: Experience with moving to digital payment of pension benefits in Thai Nguyen provinces Currently. Thai Nguyen province have a total of 76,094 beneficiaries of pensions and social insurance allowances. Of these, 15,720 beneficiaries are receiving the benefit via ATM (bank account); the remaining of 60,374 beneficiaries are receiving cash delivered at payment spots. VSS has also established a target of paying benefits via bank account. For Thai Nguyen province, by the end of 2023, the target on non-cash payment in urban area is: 26% for pensions and monthly social insurance benefits: 76% for lumpsun benefit and 98% for unemployment benefits. The implementation of receiving pensions and social benefits via accounts still faces many challenges. The reason is partly due to the fact that facilities in localities are not met the demand. ATM machines are still not popular. Besides, pensioners are elderly with the habit of using cash. To overcome the above difficulties. Thai Nguyen PSS coordinates with the post office and local authorities to promote propaganda and advocacy on the benefits of payment via ATM cards. The PSS also facilitate beneficiaries to open bank accounts at payment spot, and at the same time agrees with Banks and Post Offices to waive fees for issuing ATM for beneficiaries of pensions, monthly social insurance benefits. Each commercial bank is assigned to deploy in certain areas to avoid overlap and confusion for those registering to open accounts. Source: Interview with That Nguyen PSS 5. Manage Participants and prospective registrants can hear about a scheme from different sources and have multiple interactions with scheme administrators/partners. Monitoring their experiences and learning from those is important to understand what works and what needs improvement. Participants can face challenges at multiple points such as during registration, accessing data, benefit disbursements, withdrawals, claims etc. and the challenges can differ based on the sub-group (e.g. women vs men, urban vs rural, smart phone user’s vs feature phone users). Inclusive monitoring with the intent to learn and correct any loopholes is important to provide a positive experience to participants. 41 Report of VSS. 45 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS 5.1. Monitoring VSS allows participants and beneficiaries of social insurance to update their information in- person or using the VSSID application. Participants can request for a change to demographic data or a change in contribution channel or channel for receiving benefits (from cash to bank account or vice versa). These changes can be requested online or through walk-in to the district social security offices or through the public postal system with the support of insurance agents. Errors or any grievance arising during the different phases of the delivery chain can be brought to VSS’s attention through walk-in, social media, via agents or calling the hotline/call-center. With the digitization of key business processes like contribution payment, benefit disbursement, and status tracking, participants can download and use the VSSID application to stay connected. There is however no known analysis on who is using the VSSID application and if/how it has improved customer satisfaction. VSS has a Call Center with a dedicated hotline number (1900.9068) mentioned on the web, posters, documents, and VSSID applications, but the tellers can be trained to also conduct surveys and convey member concerns to the management. The call center established in 2017 consists of 20 tellers working across two shifts. The management of the call center is outsourced to an agency. The ‘frequently asked question’ bank which tellers use as a reference was compiled by VSS professional units and include nearly 1,000 questions covering areas of social insurance, health insurance policy, issuance of social insurance books and health insurance cards. These questions have been coded into the FAQ Management Software. The tellers are trained in communication skills and are familiar with the scheme rules. As of March 2022, the Call Center has received and processed a total of 1,263,403 calls from customers. The Call Center staff rely on the overall IT system to receive calls, manage customer information, and record incoming calls for quality assurance purposes. They don’t, however, conduct annual or quarterly spot surveys to test awareness, or customer satisfaction. The desk research did not find any regular updates from the call center to the management on frequency/type of calls. A proactive and strategic use of the call center should be considered in the context of the voluntary scheme. Successful voluntary schemes like the Ejo Heza in Rwanda deploy dedicated tellers, trained in answering questions and convincing callers on the benefits of the voluntary scheme. The Ejo Heza scheme also increases their staff capacity in the call centre prior to any change in rules or large-scale awareness campaign. In 2020, VSS set up a separate customer service unit named Customer Care Center (CCC) to perform online public service settlement, customer care and support. Going forward, it will be important to ensure clarity in roles and responsibilities between the call center and CCC staff to avoid duplication of efforts and have clear performance matrices. VSS is using different methods to maintain contact with participants, and over time this should improve the familiarity with VSS schemes. As part of the IT modernization program for the social insurance sector VSS started sending our SMSs to members in April 2019. By the end of December 31st, 2021, the system had provided 5,487,836 messages responding to information requests, 50,089,501 one-time password (OTP) messages and 343,483,796 social insurance brand messages from VSS. Other customer care activities for members include New Year messages; giving gifts to health insurance participants in critical care in hospitals; seeking charitable donations for health insurance cards and contributions to voluntary scheme from enterprises. These efforts will have an overall impact of increasing the familiarity of VSS programs and making it a household name. However, since VSS 46 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS administers multiple schemes, care should be taken that the voluntary scheme’s branding stands out so individuals don’t confuse it with other programs. Having a distinct name to the scheme instead of the generic ‘voluntary scheme’ should be considered. Almost all countries who have launched such schemes in Asia and Africa have given the schemes a name that is intuitive and relevant in the country context for the target population e.g. the Atal Pension Yojana, and the Pradhan Mantri Shram Yogi Mandhan in India; Ejo Heza meaning brighter future in Rwanda, Haba Haba meaning small-small in Kenya. 5.2. Exit Participants exit the system either when their nominated beneficiary dies or when they withdraw all contributions as a one-time social insurance benefit. In the event of a pensioner’s death the nominated dependents is entitled to apply for receiving funeral allowance and survivorship pension (see Annex for benefit details). Up to date beneficiary records and a transparent, dynamic certification process is important for a smooth experience, and to avoid paying pensions to ghost pensioners due to error or fraud. Currently, it is not required for survivors to collect the benefit in person or present proof of life certification. VSS has established a linkage with the database on Civil Registration and Vital Statistics (managed by the Ministry of Justice) for cross-checking death cases to reduce the risk of ‘ghost’ pensioners. IV. Survey to inform changes in Design, Delivery, and Incentives in Voluntary scheme. An in-person survey was conducted in 2023 for the purposes of this study. Data for the VSDS (2023) was collected from 900 individuals in the age group 20-45 years. The records of these individuals were randomly selected from the database of three provinces. The database includes active, dormant, and ‘potential’ individuals in each of the three provinces. Active individuals are those who had made at least one contribution in the last six months, dormant were those who had not contributed in the last six months (but contributed in the past); and potential individuals were the target list kept by the provincial VSS staff. The three provinces chosen for this analysis were Thai Nguyen, Nghe An, and Long An42 (Figure 11) and 300 individuals (100 active, 100 dormant, and 100 potential) were chosen from each province. Gender balance across the provinces was sought at the sampling stage with 52% of the respondents in the sample being female. 42 The three selected provinces were selected to include one each from North, South and Central Vietnam given the differences by region. 47 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Figure 12: Three provinces for the survey Disclaimer: The map shown is for illustration purpose. The boundaries, color, denominations, and other information shown on any map in this work do not imply any judgment on the part of the World Bank Group concerning the legal status of any territory or the endorsement or acceptance of such boundaries Thai Nguyen HOÀNG SA Nghệ An Long An TRƯỜNG SA The primary objective of this survey was to recommend changes to design and incentives in the voluntary scheme, for consideration in the 2024 Social Insurance Law revision process. Additionally, this survey tries to understand the awareness, and customer experience on the key stages of the delivery chain among the three different groups (active, dormant, potential). The survey is motivated by four questions: • What factors or characteristic determine the likelihood of being an active participant? • How aware are individuals of the scheme’s rules and benefits? • How easy or difficult it is for individuals to interact with the scheme through the various stages of delivery chain (registration, contribution, benefit, monitoring etc.)? • What is the design, and incentive changes to be considered for the SI law revision? The survey sample was selected from the administrative data on the voluntary scheme instead of a nationally representative dataset. In Vietnam the household and LFS have a question on whether an individual has contributed to compulsory or voluntary social security, so in theory one could implement a survey to a sub-group of those who said ‘yes’ to contributing to the voluntary scheme. This approach would have two shortcomings- first most provinces had a very low sample size given the overall low share of those who contribute to the scheme (<4% of labor force nationwide); and second our sample would be limited to active contributors and all others. Given the objective of the survey viz to inform the VSS of changes to business processes and to recommend changes in the design and incentives that can boost take-up it was important to also identify dormant and those most likely to contribute to the scheme (potential participants). This objective prompted a discussion with the VSS staff and the GSO on sample selection, and the decision was made to use the VSSs administrative data on voluntary scheme for the survey. Active and dormant individuals were extracted from the database of the VSS. The list of potential contributors was collected from provinces. These were individuals who either attended a customer awareness meeting held by the province or were identified by VSS staff/agents as those likely to contribute, based on local databases the provincial staff has access to. 48 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS The age and income distribution of our sample is not representative of Vietnam’s population. The use of administrative data for our sampling strategy has the limitation that it is not representative of the overall population. Results of the survey should therefore be interpreted with this caveat in mind. Specifically, comparing the age and income distribution between the survey respondents (n=900) and the Vietnam’s nationally representative household survey (VHLSS) shows that the median age of respondents is similar between the two, but the survey sample has a higher share of older individuals relative to the VHLSS. The income distribution of survey respondents also suggests that the survey is more representative of ‘middle-class’ Vietnamese households (Figure 12). Figure 13: Age and income distribution in the survey and VHLSS Age distribution of ages 20-45 in survey sample Household Income distribution between sample and VHLSS and VHLSS 8.00% 35% 7.00% 30% 6.00% 25% 5.00% 20% 15% 4.00% 10% 3.00% 5% 2.00% 0% 1.00% 0 00 0 0 0 0 00 00 00 00 00 00 ,0 ,0 0, 0, 0, 0, 0, 0.00% 00 00 00 00 00 00 00 ,0 ,0 20 22 24 26 28 30 32 34 36 38 40 42 44 5, 5, 0, 5, 0, 10 30 0- -1 -2 -2 -3 0- D 00 00 00 00 D 00 VN Survey sample share ,0 ,0 ,0 ,0 VN 0, 00 00 00 00 n 00 VHLSS share (for ages 20-45) ha ,0 ,0 ,0 ,0 5, 10 15 20 25 rt D te D D D D VN ea VN VN VN VN Gr VSS Sample Survey VHLSS for ages 20-45 Source: VSDS survey (2023); VHLSS (2020) The survey respondents are on average middle-aged and employed, predominantly as own account workers. The median age of survey respondents is 37 and only 2% of the sample is from poor and near-poor category households43. Approximately 58 percent of the survey respondents live in rural areas while 42% were in urban areas. 86 percent of the survey respondents reported that they were ‘engaged in work which generated income for them and their family.’ Of those employed, the most common type of employment was an ‘own account worker’ (50.84%) followed by being an ‘employer without any issued labor contracts (24.97%).’ A small share of respondents also reported being ‘employers who hire labor’ and ‘employees’, and while these individuals should by law be contributing to the compulsory scheme, the survey results show that the mandate can be difficult to enforce when labor markets are diverse and fluid (Figure 13a). A third of the respondents reported being “Service and Sales Workers”, followed by elementary occupations, and craft workers, with gender variability in type of occupation (Figure 13b). 43 The share of poor and near poor households in the population is close to 10 percent. 49 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Figure 14(a): Type of work; Figure 13(b)Occupation Type of work (n=777) Respondent Occupation (n=772) 5.15% 300 250 200 205 24.97% 150 85 11.33% 100 50.84% 50 110 87 67 74 0 7.72% ls ls rs rs Tr ork and rs Oc lers , ns b rs na na ke ke ke m to tio es s ad er sio sio or or or en sse era te y W ry pa W sW W t es es s cu p Re she re rt El and e O of of le Fi l, Fo po Pr Pr Sa n r A up ry hi te ra d ta ac cia an lS d tu Employer M ica ul so ice a em ric nd l As er rv Ag Cl ta Own account worker Se d nd an an d ta ille Pl ns Family contracted worker af Sk cia Cr ni ch Employee Te Employer (without labor) Male Female The average monthly disposable income among respondents is VND 3,379,000 (USD 133) viz ten times the minimum monthly contribution in the voluntary scheme. The median monthly household income of survey respondents is VND 10-15 million ($425-640 USD). Household incomes were twice as high for those in urban areas compared to rural areas. Disposable income at the end of the month (i.e., income after paying all expenditures (food, rent, health, education, clothes, transportation, communication, internet) was higher for men (3.64 million VND) than that of women respondents (3.1 million VND)​ . Only five survey respondents reported having no disposable income at the end of the month. The average disposable income was VND 3,379,000 which is more than 10 times the minimum monthly contribution of VND 330,000 in the voluntary scheme. Notably, 88 percent of the respondents had disposable incomes greater than the minimum. The high reported disposable income suggests that a sizeable number of respondents might have the ability to make contributions to the scheme if it met their needs and was perceived to be an attractive scheme. Saving as ‘cash at home’ is the most common channel by respondents. Only 100 out of the 900 survey respondents said their household did not save using any channels while most reported saving in at least 1 or 2 different channels/places. The most common saving channel was ‘cash at home’ for both rural and urban individuals. Females were more likely to save cash at home, comprising 57% of all respondents who save via cash. Saving in a personal bank account was the next common channel, especially for those in urban areas (Figure 14). These results suggest that many in the informal sector, notably women would lose their saving in real terms. They would be better protected against inflation and receive a higher rate of return if they contributed to the voluntary scheme. Use of mobile money for 50 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS savings is particularly low, pointing to the low take-up of digital saving modes among both urban and rural populations. Figure 15: Saving channels by urban and rural population. 80% 70% 60% % of respondents 50% 40% 30% 20% 10% 0% t s ns es e s .) un ng nd tc m rs em tio ,e he ho co vi ie nd itu sa ch /fr Ot ac at ,la st ily ss ey nk sh in ry m ng on ba Ca el al fa vi m w rm al to sa (Je ile on fo ed al ob ts rs n rm an se si Pe M fo lo ng As In sh vi Sa Ca % of urban population % of rural population Respondents expect their incomes to grow over time, albeit modestly. When enquired about their expectation of future income growth almost all respondents expected their individual income to either ‘stay the same’, ‘grow a little’ or ‘grow a lot’ over the next 10 years (Figure 15). The expectation of growth in incomes does not vary by employment type, gender, location (urban/rural) or participation status. The modest expectations around own income growth might translate to respondents being sensitive to any significant jumps in minimum contribution requirements. This concern was underscored by agents who shared that the doubling of poverty line in 2020, and therefore a sharp rise in minimum contributions which discouraged respondents to contribute. Figure 16: Opinion on income over the next 10 years Shrink a lot 0% Shrink a little 2% Stay the same 14% Grow a little 71% Grow a lot 12% 0% 10% 20% 30% 40% 50% 60% 70% 80% Percentage of respondents 51 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Respondents tend to underestimate average life expectancy and prefer to retire between the ages of 50-60. When asked about their expected life expectancy, 34 percent of participants estimated it to be 70 years with no significant variance observed across gender and location. The average life expectancy at birth for Vietnam is estimated at 74 years and life expectancy for those who survive until age 60 is as high as 21.3 (United Nations, 2024). The low estimated life expectancy of respondents relative to what the UN data suggests is consistent with pessimistic bias44 documented in the behavioral science literature. When enquired about the age they would like to retire, a large share (69%) of respondents would like to retire between the ages 50-60 (Figure 16). The retirement age in the voluntary scheme will be 62 for males and 60 for females by 2028, suggesting a potential mismatch between retirement age expectations and the scheme rules. VSS should engage in nationwide communication efforts to explain why an increase in retirement age is justified in the Vietnam context. In the medium term VSS might consider design options which allow those in the voluntary scheme to retire earlier. Figure 17: Opinion toward retirement age I cannot retire because I will need money 9% I never want to retire because I enjoy work 1% Between 60-70 11% Between 50-60 69% Below age 50 10% 0% 10% 20% 30% 40% 50% 60% 70% 80% Percentage of respondents Participation status of individuals is correlated with some observable socio- a)  economic characteristics. Actively contributing individuals in the voluntary scheme are more likely to be employed45 in occupations with higher and predictable incomes. The participation status among respondents is correlated with employment status and the occupation type46. Those in an active participation status were more likely to be employed in occupations with predictable incomes such as professionals or technicians (Figure 17b). Most dormant and potential individuals in our sample are also employed (Figure 17a), but they are more likely to work in skilled agriculture, or in elementary jobs or plant and machine operators, assemblers (Figure 17b) that tend to be low-skilled, or contract-based work. While the share of service and sales workers among active contributors might be lower, the counts of these 44 This type of bias can occur when individuals underestimate positive outcomes or overestimate negative outcomes due to various factors, such as media influence, personal experiences, or lack of knowledge about statistical data. 45 Statistically significant association between employment status and participation status at 99% confidence level; chi- squared value = 182.51; p-value = 0.00. 46 Statistically significant association between occupation type and participation status at 99% confidence level; chi-squared value = 379.61; p-value = 0.00. 52 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS workers can be high if this is a predominant occupation in urban and rural areas (Figure 18). While preparing a targeting strategy, the VSS should therefore consider both the size and relative likelihood of attracting individuals based on occupation status. Figure 18a: Participation status by employment status; Figure 17b: by occupation type Participation status (disaggregated Participation status (disaggregated by occupation by employment status) type - ISCO 08 Major Groups) 100% 100% 90% 80% 80% 70% 60% 60% 50% 40% 40% 30% 20% 20% 10% 0% 0% ls up ion and rs rs Tr ft a ker al, an ine rker ed er , ns active dormant Potential bl ors na ke ke s nd ad nd s s s or ltur tio o lat al em at sio or or ss s pa M es W Re ica ofe cian r ry icu tW W e es As p cu W es he gr d O of or r i Oc Cl te P hn s l Fis A Pr Working Not Working Sa p ia ec d led ry a h d oc T ta Cr ac an lS an il k en ry S ice em er rv s As ta El Se st an re Pl Fo Active Dormant Potential Figure 19: Count of individuals based on participation, location (urban/rural), and occupation type 200 180 160 140 62 120 38 26 100 80 60 52 40 41 56 47 56 20 40 0 Active Dormant Potential Active Dormant Potential Urban Rural Elementary Occupations Service and Sales Workers Plant and Machine Operators, and Assemblers Clerical Support Workers Craft and Related Trades Workers Technicians and Associate Professionals Skilled Agricultural, Forestry and Fishery Workers Professionals Actively contributing individuals have marginally higher household incomes. The self-reported income levels suggests that those in the active status have higher total household income (Figure 19) 53 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS relative to both dormant and potential participants47. The difference in household income levels48 is statistically significant. Interestingly, dormant participants have even lower household incomes than those who are potentially interested to join, suggesting that a drop in income might have been a reason for individuals who joined the scheme to stop participating. Not only do active participants have higher household income, but they also have higher disposable incomes. Specifically, they have VND 1,033,000 higher disposable income in comparison to potential participants. A recent analysis by Trang et a. (2024) using Vietnam’s Labor Force Survey data finds that Individuals with higher income are more likely to participate with a 1% increase in income implies 0.009% and 0.003% increase in the probability of participation in the compulsory and voluntary schemes, respectively. Figure 20: Average monthly household income by participation status 160 140 Active Dormant Count of respondents 120 Potential 100 80 60 40 20 0 VND 0-5,000,000 VND 5,000,000- VND 10,000,000- VND 15,000,000- VND 20,000,000- VND 25,000,000- Greater than VND No income 10,000,000 15,000,000 20,000,000 25,000,000 30,000,000 30,000,000 Monthly household incorne (reported) A LASSO regression was used to select factors that influence participation status. Least Absolute Shrinkage and Selection Operator or LASSO regression is a way to implement linear regression, with a twist. LASSO tries to keep the equation as simple as possible by eliminating variables that do not helping with prediction. The process attaches a penalty parameter to reduce the number of variables used in a regression and also shrinks the size of coefficients. This is a useful methodology to use when one has many variables to consider but suspects that only a few of them are important. In the VSDS (2023) survey, after converting categorical variables into dummies,49 we end up with more than a hundred potential dependent variables. We use LASSO regression to guide our variable selection. We first obtain a reduced list of variables from a LASSO regression complementing it with our knowledge of the economic literature on savings and international experience on voluntary pension schemes as a guide to select variables that we suspect are useful, and dropping those we believe maybe be spurious correlations. After arriving on a list of potential covariates, a linear regression model with 47 Monthly income in 000 VND after paying all expenditures (food, rent, health, education, clothes, transportation, communication, internet). 48 Statistically significant association between monthly household income (in 000 VND) and participation status at 99% confidence level; chi-squared value = 65.42; p-value = 0.00. 49 For instance, the Registration channel variable (C13) is coded 1 through 4 depending on if the response is “Social Insurance Agency,”“VSS Agents,”“Online Self-Registration,” or “Others.” For regression we break this variable down into reg_channel_1, reg_channel_2, reg_channel_3, and reg_channel_1, where each variable takes the value one if the response recorded for the original Registration channel variable was 1, 2, 3 or 4 respectively, or they are set to zero. 54 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS district level fixed effect and robust standard errors is used to determine factors that correlate with participation status. The findings of the LASSO regression suggest that observable characteristics like gender, age50, and disposable income are not significant determinants of participation status but certain awareness channels, and saving methods are correlated with participation status. Awareness spread through word of mouth (from existing SI users and friends) is significant and positively correlated to being a member in VSS (see Table 6, Reg 1 and Reg 2). Awareness via Loudspeaker is also positively associated with likelihood of being active (Reg 1 and Reg 3). Other saving channels used also statistically impact participation status but the differing patterns across groups may require additional investigation. Interestingly those who save in assets (jewelry, land, etc.) are less likely to be active when compared to dormant participants (Reg 3) meanwhile having savings in a personal bank account is associated with higher contribution amounts compared to those who do not have savings in bank account (annex 8).. Finally, participants who had a higher last contribution are more likely to be active vs dormant. Those who count on VSS during old age are more likely to be active while participants who count on their children to support during old age are less likely to do so (forthcoming paper, De & Jain, 2025). Table 6: OLS Regression results with independent variables selected with help of LASSO regression. Reg 1: Active Reg 2: Active + Reg 3: Active =1 | Dependent variable =1|Dormant+ Dormant =1 | Potential Dormant =0 Potential= 0 =0 Observations: 766 766 583 gender_female -0.02 0.01 -0.03 Age 0.00 0.00 0.01 log(1+disposable income) 0.01 0.00 0.01 savings_bank 0.06 -0.09** 0.11 ** savings_mobile -0.02 0.07 -0.07 savings_formal_inst 0.07 0.03 0.05 savings_informal 0.12 * 0.02 0.12 savings_cash 0.05 -0.11 ** 0.11 * savings_cash_loaned_to_family -0.01 -0.30 *** 0.19 * savings_asset -0.12 * 0.21 *** -0.22 *** awareness_tv_channel -0.15 ** -0.06 -0.08 awareness_radio_channel 0.10 0.07 0.12 awareness_loudspeaker_channel 0.11 * -0.07 0.14 * awareness_socialmedia_channel -0.06 -0.08 -0.03 awareness_newspaper_channel 0.02 0.05 0.01  he lack of a correlation to age and gender is at odds with the results from Trang et a. (2024) which uses Labor Force survey 50 T data and find that being female increases the probability to participate in both compulsory and voluntary scheme and that the probability of participating peaks at the age groups 25-39 years old, declines for individuals older than 40 and becomes negative for those older than 55 years old. The difference in findings can be attributed to the sample size for our survey which is restricted to those aged 20-45 and the sampling strategy of our survey. 55 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Reg 1: Active Reg 2: Active + Reg 3: Active =1 | Dependent variable =1|Dormant+ Dormant =1 | Potential Dormant =0 Potential= 0 =0 awareness_friends_channel 0.06 -0.01 0.03 awareness_ SI_users_channel 0.08 * 0.10 ** 0.03 awareness_exisitng_vss_users 0.12 ** 0.11 ** 0.05 log(1+amount_last_contributed) 0.12 *** registration_channel_walk-in_VSS -0.23 registration_channel_agent 0.14 registration_channel_self -0.43 ** plan_old_age_VSS saving 0.14 plan_old_age_ownsaving -0.02 plan_old_age_assets 0.11 plan_old_age_children -0.27 * plan_old_age_work -0.05 RMSE: 0.407 0.353 0.375 Adj. R2 0.268 0.286 0.402 Within R2 0.085 0.116 0.199 Notes: District fixed effects, Robust S.E; Significance: ***0.001; **0.01; *0.05. The lack of significance of income as a factor affecting participation in the regression analysis is an unexpected finding given evidence from similar studies (Rother et al., 2022). However, based on additional checks, this is attributed to either sampling bias in the context of this study, or income potentially not being a determinant of participation for this particular scheme in Viet Nam. Robustness checks were conducted to test the effect of income levels on participation status under varying regression specifications (including variables balance savings after expenses and savings in formal institutions) to arrive at this conclusion (ref Annex 9). Actively contributing participants are more likely to depend on pension from VSS for their financial survival in old age. As seen in Figure 20a, those contributing actively are more likely to report depending on their pension from VSS in old age51. Those in active status also report a lower likelihood of withdrawing contributions to meet emergency spending needs52, noting they would prefer to find other financial sources to meet those needs (figure 20b). Dormant participants are more likely to depend on personal savings, but 17 percent of our sample expects to work in old age or has no plan. Even though personal savings can be an effective source of financial security there is a risk that people outlive their savings. An indexed pension can be a useful complement to one’s own savings. Communication efforts by VSS should focus on this longevity insurance which pensions provide, to motivate individuals to contribute to the scheme. Interestingly almost no one expects the government to support them in old age and few expect to depend on children, reflecting a sense of self-reliance. 51 Statistically significant association between plan for old-age financial survival and participation status at 99% confidence level; chi-squared value = 182.51; p-value = 0.00. 52 30% of dormant participants indicated this option as opposed to 6% of active contributors. 56 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Figure 21 (a) Plan for old-age financial survival and (b) options for meeting emergency spending needs (a) Plan for old age nancial survival, by participation status Others I have no plan I think the Government will support me I will continue working as long as possible 50 My children will support me I have assets I can sell if I need to My savings 138 177 My pension from VSS 187 44 0 50 100 150 200 250 300 350 400 450 Number of respondents active dormant Potential (b) Options to meet emergency spending needs based on participation status I don't know 56 Find other nancial sources and 262 140 do not withdraw the SI saving Withdraw up to 50% of the social insurance saving Withdraw the entire amount of 90 the socialinsurance saving 0 50 100 150 200 250 300 350 400 450 Number of respondents active dormant Among those who have contributed to the scheme the most common frequency of contributions was monthly, followed by once in three months. More than half of participants who have contributed to the scheme at some point made monthly contributions (Figure 21) and 24 percent reported contributing once every three months. The chosen frequency of payments does not vary based on gender, location (urban/rural) or the payment channel used. The auto-debit feature is only available to those who apply online and have linked their contributions to a bank account. 57 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Figure 22: Frequency of last paid contribution (n=600) 1% 0% 9% 9% Every month Every 3 months Every 6 months Every 12 months 57% Other 24% Don't remember The median monthly contribution among respondents is VND 400,000 (~USD 16) and is among the highest contributions in voluntary schemes across developing countries. The voluntary scheme has a minimum monthly contribution of VND 330,000 but depending on the household’s poverty category (poor, near-poor, others) individuals receive a flat state support which reduces their minimum contribution. Those in the ‘others’ category receives the lowest support of VND 33,000 and therefore need to contribute VND 297,000 as minimum. About 25% of survey respondents indicated their last monthly payment was VND 297,000 (~USD 13)53. Males report higher median contribution amounts relative to women (VND 412,000 for males and VND 398,000 for females). The monthly contribution being close to the minimum in the scheme suggests that the minimum acts as a signal for participants on how much to contribute. The monthly contribution in the scheme is significantly higher than that of other voluntary schemes, and this can be credited to the level of the minimum contribution. For example the average contributions in Defined Contribution schemes in Rwanda, Kenya, Thailand is between USD 2-4 per month (Guven et.al 2022), and even in Defined Benefit schemes like India’s PM- SYM the contribution ranges between USD 1-5. The low contribution amounts in the other schemes raises adequacy concerns which are not an immediate challenge for Vietnam’s scheme. However, the high minimum comes at the cost of some individuals not being able to afford it each month. A fixed-effects regression model reveals that living in urban areas, saving in bank accounts, and awareness of vesting period is positively correlated with the amount of last contribution. The model considers various independent variables ranging from the participants’ gender, employment status, geographical location, age, etc. as well as their awareness of the scheme rules, transaction channels used and the reported ease of transactions, controlling for unobserved characteristics at the district level. A caveat however is that dormant in our data include all those who did not contribute to the last six months, including ones who contributed pre 2018 when minimum contribution was much lower. The results indicate that participants from urban areas have higher last contributions (~118,600 VND) in comparison to rural areas, which is noteworthy considering that the average disposable income  he minimum (after considering state support) that poor households need to pay is 231,000 VND/month, near poor is 53 T 247,500 VND and others is 297,000 VND. 58 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS is higher amongst rural respondents54. Participants who save in personal bank accounts made higher contributions (~144,000 VND) than those who do not. Participants with a higher awareness of the vesting period also made higher contributions (~111,900 VND) though awareness of other scheme rules or state support were not found to have a statistically significant impact. These results provide preliminary insights that merit investigation across a more representative sample to understand what determines the level of contribution, contingent on individuals joining the scheme. Limited disposable income was cited as the main reason to not contribute or participate. Under half of the respondents (45%) cited lack of disposable income for not joining the scheme or participating continually (Figure 22). International experience also finds that limited saving capacity is a primary reason for not contributing to the voluntary scheme (Guven & Jain, 2023). Other reasons for not contributing were not seeing the benefits of saving (11%) and preferring the flexibility to have cash in hand (14%). Understanding the reasons why individuals do not join, save, or save more can lead to realistic coverage targets and a discussion on changes that can address the specific barriers to contribute. Figure 23. Reasons for not contributing more/stop contributing/not participating in VSIF. 450 400 350 300 250 200 150 100 50 0 I did not I saved in I don't see Saving in I wanted I did not Others have any too many the bene t the scheme to keep remember more other of saving was not cash in to save in money to places more easy hand the scheme save Male Female Scheme awareness is most effective through word of mouth but awareness on b)  scheme specifics like government incentives or estimation of benefits is poor. General awareness about the scheme is high in our non-representative sample, and primarily driven by word of mouth. Only 15% of respondents reported not being aware of the voluntary scheme, with most of them (87%) being potential participants. It is important to remember though that ours is not a nationally representative sample and there is a miniscule of active and dormant contributors in the overall population. Awareness prominently spreads through word of mouth, facilitated by scheme participants, friends and family (Figure 23). 54 Average disposable household income for active urban participants is 3.711 million VND and 3.839 million VND for active rural participants. For dormant participants, average monthly disposable household income in urban areas is 2.782 million VND while it is 3.97 million VND in rural areas. 59 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Figure 24. Share of Participants based on channel that made them aware of scheme Awareness channels Other 752 Existing members of VSIF 583 Existing social insurance participants 384 Friends/Relatives 430 Newspaper 135 Social Media 240 Loudspeaker 161 Radio 125 Television 218 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Yes No Percentage of respondents The survey finds that half of the respondents are aware about design parameters of the scheme but awareness about state budget support is lower. When enquired about awareness of scheme design parameters like retirement age, vesting period, minimum monthly contribution, close to half of the respondents said they knew about these rules while the rest either didn’t know or weren’t sure (Figure 24). Results from Chi squared tests show that high level of awareness for each design parameter is strongly associated with participation status55. Other factors such as gender and geographical location (urban/rural) are not found to have any statistically significant association with awareness of scheme rules. Relative to awareness on scheme rules, awareness of the state budget support is lower with only 25% of respondents aware of the support. This could be because of the relatively recent introduction of state support (in 2018) or suggest that the way the state support is applied (a reduction in minimum contribution) is not noticeable to members. When those who responded yes to being aware of the state budget support (n=233) were asked to estimate the level of support, 68% of them were able to state the amount of support accurately. These results suggest that more can be done to raise overall awareness of the state budget support and the level of support should be made explicit in communication material, to allow for an easier recall. 55 Results reported at 99% confidence level for the following variables and their association with participation status – 1) Awareness of voluntary social insurance scheme (Chi-squared value = 206.6; p-value = 0.00); 2) knowledge of minimum monthly contribution (Chi-squared value = 217.9; p-value = 0.00); 3) knowledge of retirement age (Chi-squared value = 161.7; p-value = 0.00); 4) knowledge of minimum participation period to enjoy full benefits (Chi-squared value = 194.3; p-value = 0.00); 5) knowledge of level of state budget support (Chi-squared value = 103.71; p-value = 0.00); 6) Ability to estimate pension benefits (Chi-squared value = 148.38, p-value = 0.00). 60 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Figure 25: Awareness of the VSIF rules (n=900) Level of support from the State budget for 233 509 158 participants Minimum period of participation to enjoy 457 329 114 bene ts Retirement age 436 335 129 Minimum monthly contribution 438 338 124 0% 20% 40% 60% 80% 100% Percentage of respondents Yes No Not Sure Only a quarter of the sample said yes to being ability to estimate their pension benefit. Estimating one’s pension at retirement based on contribution history is important for individuals to judge if the scheme is worth contributing to relative to immediate consumption or saving in other financial products. A larger share of active individuals is able to estimate their benefits relative to those in other participation statuses (Figure 25) but it is difficult to comment on causality based on the survey results. For example, are individuals less likely to be active because they cannot estimate pension benefits? or are they unable to estimate benefits because they do not contribute? What is evident, also based on consultations with agents is that the lack of a tool/calculator that can estimate benefits in the defined benefit scheme, makes it difficult to convince people to join the scheme. The survey does not enquire about awareness of benefits besides pensions but there also exist health insurance for pensioners, funeral allowance, and survivorship benefit under the voluntary scheme, awareness, and estimation of which is also needed. Figure 26: Ability to estimate pension amount by response type, and participation status (n=900) 250 166, 18% 250, 28% 200 150 100 50 0 484, 54% active dormant Potential Yes No A little Yes No A little Women are more likely to self-register or register their spouses, and self-registration improves awareness. Females in our sample are more likely to register themselves and also more likely to register their spouses. 40 percent of male respondents reported that their registration was initiated by their 61 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS spouses or their parents (Figure 26). On one hand this could suggest that women are the ones who make saving decisions in the household but as clarified by agents these results might be a function of the fact that agents make home visits during working hours and are more likely to find women or older parents relative to males. Individuals who self-register though have a 10 percent greater awareness of scheme rules (including state premium support) compared to those who are registered by spouses or parents or agents. Figure 27: Person who initiated registration, by gender (n=600) Female 291 Male 165 71 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Percentage of respondents Self Spouse Parent Contributors depend on VSS agents for registration and contribution which makes c)  it easier to perform transactions. The survey finds that VSS agents are commonly used for registration, contribution collection and resolving queries. VSS agents conducted 75% of the registration and 76% of contribution collection, followed by in-person registrations at the Social Insurance Agency (Figure 27). The agent network is what contributors depend on to resolve their queries (Figure 28) suggesting a high overall dependence on agents and minimal use of online channels or visits to VSS offices. Agents are more common in rural areas, where as many as 71% of respondents (n=900) go to agents to resolve queries and only 7% contact the call center. Figure 28. Registration and contribution channels Registration Channels (n=600) Payment channels (n=600) 12, 2% 1, 0% 9, 2% 28, 5% 105, 17% 137, 23% 453, 75% 455, 76% Social Insurance Agency VSS Agents Social Insurance Agency VSS Agent Online Self-Registration Others Online/Bank Transfer Other 62 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Figure 29: Avenues to resolve queries I don't know whom to ask 10 Women's union 12 Trade Union/cooperative 1 My local o cial 98 My agent 639 family/friends 77 VSS Call center 63 0 100 200 300 400 500 600 700 Number of respondents Both active and dormant reported that they find it easy to perform basic transactions. In contrast to international experience where informal sector workers struggle to perform transactions owing to limited digital/financial literacy; too much information or poor customer service; the respondents in the VSDS survey find it ‘easy’ or ‘very easy’ to do basic transactions. On average, a minimum of 80% of respondents indicated that conducting scheme related transactions as “Very Easy” or “Easy” (figure 29). This digression from international experience is plausible given a high dependence on VSS agents which allows for in-person interaction and for transactions to be completed by agents on behalf of individuals. This unique feature of Vietnam’s voluntary scheme and its associated benefits serves a valuable lesson for other countries who implement voluntary scheme. The challenge for VSS is to preserve the in-person agent network and utilize in a cost-effective manner. Figure 30. Ease of VSIF related transactions Track my contributions till date 452 Find out about the state premium subsidy 482 Find out my enrollment information 510 Deposit Contributions 512 Registration 517 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Percentage of respondents Very Easy Easy Neutral Di cult Very Di cult Not Applied Satisfaction with agents is high but respondents are open to using other intermediaries for scheme transactions. When enquired about preferred intermediaries to facilitate transactions, scheme members noted their preference for VSS agents followed by Banks and VN Post (figure 30). These findings reflect positively on the services provided by the agent network and an openness to intermediaries like 63 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Banks and VN Post that can be considered to scale up enrolment and participation. Banks and VN Post could be a more cost-effective channel than agents because of their ground presence, and economies of scale in deploying their staff and systems to collect contributions. To the extent members already have accounts with these institutions an auto-debit option like in case of India’s APY scheme can be considered. A monetary incentive to Banks e.g. a flat amount for every new and returning customer can also be implemented. Interestingly, unions and associations are not considered a desirable intermediary for scheme registration or contribution collection. Figure 31. Respondent preferences on intermediaries to facilitate contribution collections and benefit payments (n=600) Unions and associations 51 VN Post 333 VSS Agent 469 Mobile Network Operators 292 Banks 372 0 100 200 300 400 500 600 Number of respondents Excellent Good Average Not Good Worse Not applicable Lower vesting, additional short-term benefits, more generous state subsidy, and d)  a link to social pension would make the scheme attractive but results point to a heterogeneity in preferences. Respondents indicated a strong preference for a reduction in the vesting period to 15 years and for receiving short term benefits like in the compulsory scheme. As seen in Figure 31, respondents indicated a higher likelihood to participate if the minimum years needed to qualify for the benefit (vesting period) is reduced to 15 years. Over 80% of participants also agreed or strongly agreed with the opinion that the voluntary scheme should offer all the benefits available to compulsory scheme participants, The latter pay a higher contribution but also receive maternity and sickness, unemployment, and work- related hazard insurance from VSS. Majority of participants agree they will need to contribute more if additional benefits are provided. The challenge with providing short term benefits to those in the informal sector is that verification can be difficult, and scheme can be subject to moral hazard issues. For e.g. if the eligibility for an unemployment benefit is only six months of contributions in the scheme, a self-employed individual might start contributing at the start of a difficult year for his/her business. Verifying a work-related injury for a self-employed worker is also difficult. Still innovative alternatives VSS can consider are providing those short-term benefits that are easy to verify like maternity or providing a lump sum benefit whose generosity varies based on years of contribution or introduce an experience rating system i.e. an increase in contributions for a few months after a claim is made (a common practice for auto-insurance). 64 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Figure 32: Perception on benefits of the scheme if minimum participation period is 15 years, I will be more likely to participate If the VSIF scheme o ers more bene ts, I will need to contribute more Is it better to prioritize retirement saving over immediate spending VSIF scheme should o er all bene ts of formal sector pension scheme 0% 20% 40% 60% 80% 100% Strongly agree Agree Neutral Disagree Strongly disagree Lump-sum maternity benefits were considered attractive by most, but sickness benefits were preferred by female and industrial accidents/occupational disease were preferred by males. The maternity benefit was explained to be a benefit where individuals receive a lump sum VND 2,000,000 (USD 79) if they saved for at least six months in the 12 months prior to delivery of baby. This benefit would be available to women beneficiaries of the scheme or can be claimed by a male contributor if his wife was pregnant. 65 percent of respondents said that adding a lump sum maternity benefit to the existing voluntary scheme would make it more attractive or much more attractive (Figure 32). However, when asked to rank three short term benefits (maternity, sickness, occupational hazard) based on their preferences female respondents preferred sickness benefits over maternity, and male respondents preferred industrial accidents/occupational disease coverage (Figure 33). This suggests a difference in risk profile by gender. The lower preference for maternity benefits by both genders might be because of the relatively high median age of the sample is 37. These results suggest that a bundle of short-term benefits might be more attractive for individuals and cost-effective for the scheme, given that the risk profile for short term benefit varies by age and gender of respondents. Such a bundled benefit is offered in Colombia’s BEPS scheme (Guven et.al 2022) via a public-private partnership financing model. 65 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Figure 33: Willingness to participate if they are offered maternity benefits If the voluntary scheme required you to save consistently for six months during 12 months before the birth delivery to be eligible for VND 2 million lump sum maternity bene ts for you or your wife, would it make voluntary more attractive? (n=900) 106, 12% 179, 20% Much more attractive More attractive Not more 139, 15% attractive No opinion 476, 53% Figure 34: Ranked preference for short term benefits, by gender Importance of bene ts - disaggregated by gender 100% 80% 60% 40% 20% 0% Important Level 1 Important Level 2 Important Level 3 Important Level 1 Important Level 2 Important Level 3 Important Level 1 Important Level 2 Important Level 3 Maternity and Industrial accidents and Sickness bene ts paternity bene ts occupational diseases bene ts Male Female More than half of survey respondents would be more motivated to contribute if a 10-15% of contributions was offered as additional subsidy by the government. Vietnam currently offers a flat contribution subsidy which varies based on household poverty status. Since less than 10 percent of households nationally belong to the ‘poor’ or ‘near-poor’ category the effective subsidy for most contributors is VND 33,000 (USD 1.3) viz the subsidy for the others category. This level of subsidy is low relative to the minimum contribution of VND 330,000 and most other countries offer between 50-100% 66 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS of contributions as subsidy in voluntary scheme (Jain & Palacios, 2023). Improving the generosity of matching contributions should be considered given it is a tangible benefit that makes the scheme more affordable for participants. The ability of matching incentives to increase coverage is also supported by analysis of VSS’s administrative data on the voluntary scheme from 2017-2023. The study by Trang et al. (2024) carries out a regression analysis on time series data of contributors in voluntary scheme, taking advantage of variation of government incentives, different characteristics across provinces and different timing of top-up subsidies implemented by local government. The results suggest that increasing the matching contribution to 50 percent of the minimum contribution could triple the number of contributors at a cost of less than 0.1 percent of GDP. It is worth noting though that subsidies alone might not convince all respondents to join, as also noted in the Trang et a. (2024) paper. In the survey for example, 58% expressed willingness to save/continue saving/increase saving if the government offered an additional 10-15% increase in contribution, while 28% remained uncertain (Figure 34). Active contributors in general were more favorable of saving or saving more if this incentive was provided. Figure 35: Count of respondents willing to participate if the Government increased contribution subsidy by 10-15% Would you be save (or save consistently) if the Govt contributes a top-up into your account each month (10-15% of your payment as example)? (n=900) 300 250 200 150 100 50 0 I will most I will I will I am I might No, I will likely save continue participate uncertain not save not save more the current in/come back (save payment more) Male Female The fourth incentive explored in the survey was the possibility of receiving a guaranteed benefit instead of a lump sum as long as the individual continued to contribute to the scheme. As of 2023, Vietnam has in place a monthly flat benefit (social pension) of VND 360,000, given to all older adults above the age of 80 and means tested for ages 60-79. Informal sector workers who are not poor but vulnerable might desire receiving income support before reaching age 80. So, the study explored the option of allowing individuals who participate in the voluntary scheme to receive a flat benefit if they continued to contribute in the scheme and not withdraw all their contributions as a lump sum. China adopts a similar approach to motivate individuals to contribute to the Rural and Urban Resident Scheme (RURS) whereby young individuals who contribute to the voluntary scheme earn the right for their parents in the same household to receive the social pension (Jain & Palacios, 2023). In the VSDS (2023) survey, a potential link between contributions to the scheme and an assured flat benefit was posed to active and dormant participants. Only 17 percent of the sample (n=600) said they would withdraw their 67 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS contributions early if such a link was introduced (figure 36). Future design iterations should explore links between a flat benefit like the social pension and contributions to the voluntary scheme. Figure 36. Willingness to save if non-contributory and contributory design is linked. If you are at retirement age and you do not meet the minimum contribution year for the pension but the government o ered to give you monthly social assistance in return for you not withdrawing the lumpsum, would you like to do so? (n=600) 450 400 350 300 250 200 150 100 50 0 Continue to make full Keep the amount withdraw the I don’t know payment for up to already paid, do not lumpsum 20 years withdraw lumpsum Recommendations The analysis in this report is based on the joint work between the World Bank and VSS teams, with assistance of the GSO on survey implementation. The recommendations, summarized in this section aim to inform policymaking in Vietnam on changes to the design, delivery, and incentives in the voluntary scheme that can be considered for the 2024 Social Insurance Law revision or via changes to VSS’s business processes as it related to the voluntary scheme. The Delivery chain assessment, and the Voluntary Scheme Desirability Survey (VSDS, 2023) analysis has also provided unique insights into best practices in Vietnam’s scheme which should be strengthened and considered by other voluntary schemes. These include (a) strong political will to expand coverage and clear targets as under Resolution 28; (b) use of in-person VSS agents for outreach, and facilitating transactions which translates to a high level of satisfaction among members, (c) a high average monthly contribution by participants in the voluntary scheme (relative to international experience) which will translate to adequate benefits, and (d) introduction of fiscal subsidies since 2018 which has boosted coverage. The recommendations in this section, also informed by international experience, can be grouped into three areas of focus: (a) Changes in design and incentives to be considered for the SI law revision of 2024, (b) Updating of VSS’s business processes along the delivery chain, and (c) Strengthening of learning, monitoring, and evaluation capacity. 68 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Table 7: Summary of key recommendations for voluntary social insurance scheme Focus areas Recommendations Changes in design • Improve predictability in levels of minimum contribution and manage member and incentives to be expectations around increase in minimum. considered for the SI law revision of 2024 • Increase generosity of state support and focus on raising awareness around the state support. • Consider adding short term benefits, ideally as a bundle to make the scheme attractive for different groups. • Offer a flat benefit (or social pension) to those who persistently contribute in voluntary scheme. Updating of VSS’s • Develop a unique branding for the voluntary scheme with a local name/logo to outreach & business differentiate it from other VSS programs. processes along the delivery chain • Encourage adoption of digital service delivery channels to complement the efforts of external service agencies and reduce overall costs. • Eliminate or reduce paper-based forms for registering new VSIF participants. Use available information from the heath Insurance database to auto-populate the online registration forms. This approach will minimize the effort required to recollect information already available in the database. • Review the existing data dashboard to ensure that the indicators are effective in decision-making. • Set targets for digital take-up and monitor progress across provinces in the dash- board. • Identify appropriate aggregators that can auto-debit member contributions. • Focus on awareness raising through in-person channels (testimonials/word of mouth). • Develop digital pension estimator tool and leaflets to help estimate pension benefits. • Develop a diverse range of outreach material that provides participants clear understanding of rules, how to contribute, benefits. • Simplify the registration form, deleting redundant data fields and include optional fields. • Streamline business processes to make it easier for dormant participants to resume contributions. • Utilize call-center staff in annual spot surveys to understand customer experience. • Use data analytics at the central level to informal outreach strategy which can be finetuned by provinces. 69 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Focus areas Recommendations Strengthening of • Strengthen monitoring, evaluation and learning across various administrative levels of learning, monitoring VSS to improve scheme design and delivery. and evaluation capacity • Consider introducing a rewards and recognition program for high-performing provinces / districts, including establishing an experience sharing platform, tracking of collection service organization/agents, and develop mentoring models. • Consider collaborating with the General Statistics Office to maximize data collection and utilization; include questions in Labor Force Survey (LFS) and other nationally representative household surveys that allow for better analysis. • Assess cost effectiveness of outreach and communication spending. • Set up impact evaluations to assess increase in coverage and level of contributions because of changes in the 2024 SI law. • Strengthen the analytical capacity of staff to assess trends in the data, design surveys, interpret results and provide recommendations to management. 70 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Annex 1. List of administrative processes carried out by the VSS I. Collection of Social insurance, health insurance, unemployment insurance • Register and adjust premium of compulsory social insurance, health insurance, unemployment insurance and occupational accident and occupational disease insurance; issue social insurance books, health insurance cards • Register, re-register and adjust voluntary social insurance contributions; issue social insurance books • Register and issue health insurance cards for people only participating in health insurance. II. Issuance of social insurance books and health insurance cards -Re-issue, change, adjust information on social insurance books, health insurance cards III. Implementation of social insurance policies • Settlement of monthly allowances according to Decision No. 613/QD-TTg • Settlement of one-time social insurance benefits • Settlement of sickness benefits • Settlement of maternity benefits • Settlement of survivorship benefits • Settlement of benefits after sickness, maternity, occupational accidents, occupational diseases • Settlement of one-time allowance for people who are enjoying monthly pension or social insurance allowance going abroad to settle down and foreign citizens who are enjoying monthly pension or social insurance allowance no longer residing in Vietnam male • Settlement for continued enjoyment of monthly pensions and social insurance allowances for those who have completed their imprisonment sentences, those who illegally exit the country and return to their country of lawful settlement, or who are declared missing by the court. • Settlement of relocation for people moving to another location for monthly pensions and social insurance allowances, one-time occupational accident and occupational disease benefits, one-time survivorship benefits and people waiting for monthly pension or allowance • Settlement of adjustment, cancellation of decision, termination of social insurance benefits • Settlement of benefits for retired teachers who have not yet received seniority allowances in their pensions according to Decree No. 14/2020/ND-CP of the Government • Settlement of monthly pension and allowance for commune officials • Settlement of entitlement to occupational accident and occupational disease regimes for those who have suffered from occupational accidents and diseases and now continue to suffer from occupational accidents or diseases 71 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS • Settlement of entitlement to occupational accident and occupational disease benefits due to recurrent injury or illness • Settlement of entitlement to the occupational accident and occupational disease regime for the first case of occupational accident and disease IV. Payment of social insurance schemes • Beneficiaries receiving social insurance benefits in cash transfer to receive by personal accounts and vice versa, or change personal information • Authorization to receive pension, social insurance benefits, unemployment benefits • Retrieve pensions and social insurance benefits of beneficiaries who have died but have not yet received their benefits V. Implementation of health insurance policy • Sign the contract for medical examination and treatment with health insurance • Direct payment of medical examination and treatment expenses covered by health insurance • Payment of medical examination and treatment expenses covered by health insurance with health care facilities covered by insurance Source: https://baohiemxahoi.gov.vn/dich-vu-cong/Pages/default.aspx 72 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Annex 2. Target of provinces’ SI participants for the years 2023-2025 Unit: person 2023 2024 2025 Compulsory Voluntary Compulsory Voluntary Compulsory Voluntary SI SI SI SI SI SI Whole country 17,935,139 2,960,948 19,086,262 3,849,232 20,330,481 5,004,002 1 Hà Nội 2,222,276 141,592 2,378,968 184,069 2,546,481 239,290 2 TP Hồ Chí Minh 2,873,781 162,919 3,100,859 211,794 3,344,084 275,332 3 An Giang 128,476 53,684 135,754 69,789 143,750 90,725 4 Bà Rịa- Vũng Tàu 239,264 27,992 254,587 36,389 271,211 47,306 5 Bạc liêu 42,212 28,522 44,514 37,079 46,964 48,203 6 Bắc Giang 392,511 52,139 420,653 67,781 450,284 88.115 7 Bắc Kạn 24,787 21,542 25,673 28,005 26,602 36,407 8 Bắc Ninh 522,485 27,561 560,130 35,829 599,873 46,578 9 Bến Tre 110,170 25,114 116,194 32,649 122,744 42,443 10 Bình Dương 1,213,326 37,331 1,305,845 48,531 1,407,466 63,090 11 Bình Định 135,833 41,118 143,520 53,453 152,034 69,489 12 Bình Phước 155,306 27,806 164,544 36,148 174,288 46,993 13 Bình Thuận 105,460 34,152 111,161 44,398 117,435 57,717 14 Cà Mau 67,449 54,336 71,172 70,637 75,361 91,829 15 Cao Bằng 38,322 31,144 39,758 40,487 41,298 52,633 16 Cần Thơ 151,949 45,956 162,015 59,742 173,072 77,665 17 Đà Nẵng 253,299 27,955 269,935 36,342 287,779 47,244 18 Đăk Lăk 115,212 38,709 120,984 50,321 127,209 65,418 19 Đăk Nông 35,775 28,329 37,179 36,828 38,696 47,876 20 Điện Biên 41,059 28,563 42,685 37,132 44,369 48,272 21 Đồng Nai 968,108 52,394 1,032,649 68,112 1,102,093 88,545 22 Đồng Tháp 110,591 42,446 116,561 55,180 123,132 71,734 23 Gia Lai 86,660 37,398 90,922 48,617 95,521 63,202 24 Hà Giang 49,498 24,684 51,420 32,089 53,511 41,716 25 Hà Nam 182,195 27,639 195,909 35,930 210,870 46,709 26 Hà Tĩnh 104,069 69,795 109,977 90,733 116,194 117,953 27 Hải Dương 439,624 72,956 469,207 94,842 501,199 123,295 28 Hải Phòng 536,895 57,285 576,610 74,470 620,138 96,811 29 Hậu Giang 67,332 34,047 71,179 44,260 75,393 57,539 30 Hoà Bình 88,035 29,355 92,582 38,162 97,592 49,610 31 Hưng Yên 276,971 31,702 296,398 41,213 318,024 53,577 32 Khánh Hoà 154,296 36,793 163,795 47,831 174,201 62,180 33 Kiên Giang 110,078 54,620 116,097 71,006 122,839 92,308 34 Kon Tum 43,734 23,316 45,441 30,310 47,243 39,404 73 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS 2023 2024 2025 Compulsory Voluntary Compulsory Voluntary Compulsory Voluntary SI SI SI SI SI SI 35 Lai Châu 32,043 21,154 33,244 27,500 34,539 35,750 36 Lạng Sơn 59,469 32,496 62,101 42,245 64,901 54,918 37 Lào Cai 65,870 21,455 68,800 27,892 71,920 36,259 38 Lâm Đồng 96,681 36,613 102,244 47,597 108,204 61,876 39 Long An 412,002 41,117 443,878 53,452 478,606 69,487 40 Nam Định 249,724 48,258 266,902 62,735 285,763 81,555 41 Nghệ An 293,843 175,370 312,418 227,980 332,824 296,375 42 Ninh Bình 163,070 36,850 174,349 47,905 186,711 62,277 43 Ninh Thuận 43,215 21,210 45,449 27,573 47,902 35,844 44 Phú Thọ 207,372 74,270 219,600 96,552 232,877 125,517 45 Phú Yên 64,014 37,259 67,451 48,437 71,266 62,968 46 Quảng Bình 79,275 58,097 83,664 75,526 88,422 98,184 47 Quảng Nam 217,106 46,837 232,265 60,888 248,847 79,154 48 Quảng Ngãi 141,640 37,904 150,765 49,275 160,888 64,057 49 Quảng Ninh 278,308 46,253 293,892 60,129 310,535 78,168 50 Quảng Trị 60,599 36,659 63,782 47,657 67,357 61,954 51 Sóc Trăng 74,692 37,008 78,280 48,111 82,230 62,544 52 Sơn La 64,717 59,167 67,307 76,917 70,144 99,992 53 Tây Ninh 254,266 29,767 271,763 38,697 290,705 50,307 54 Thái Bình 254,612 70,293 271,549 91,380 289,674 118,794 55 Thái Nguyên 255,414 46,495 272,615 60,444 291,287 78,577 56 Thanh Hoá 464,280 136,778 495,573 177,811 529,121 231,154 57 Thừa Thiên- Huế 142,062 40,672 149,985 52,874 158,607 68,736 58 Tiền Giang 228,427 49,881 244,302 64,845 261,593 84,299 59 Trà Vinh 77,399 36,631 81,775 47,621 86,465 61,907 60 Tuyên Quang 64,412 34,190 67,459 44,447 70,743 57,781 61 Vĩnh Long 113,038 37,041 119,480 48,153 126,475 62,599 62 Vĩnh Phúc 264,457 38,860 281,817 50,517 301,403 65,673 63 Yên Bái 59,095 41,472 61,676 53,914 64,521 70,088 Total of all provinces 16,868,139 2,960,948 18,019,262 3,849,232 19,263,481 5,004,002 Source: Plan 1813/KH-BHXH dated July 6th, 2022 of VSS 74 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Annex 3: Policy review No. Year Name of Documents Implementation Status 1 2014 Law 58/2014/QH13, Law on Social Insurance passed Implemented by the National Assembly on November 20, 2014 2 2015 Decree 134/2015/ND-CP detailing articles of the Law Implemented and dictated voluntary on Social Insurance on voluntary social insurance. social insurance schemes and the premium subsidy for voluntary social insurance 3 2018 Resolution 28-NQ/TW dated May 23, 2018, of the Implemented and Proposed solutions to Communist Party of Vietnam on reforming social increase the coverage of social insurance insurance policies; participants, including voluntary social insurance 4 2018 Resolution 125/NQ-CP dated October 8, 2018, on the Implemented action plan to implement Resolution 28-NQ/TW 5 2018 Resolution 102/NQ-CP dated August 3, 2018, on Implemented assigning targets for developing social insurance participants; 6 2019 Decision 357/QD-BHXH dated April 02, 2019, on Implemented and assigned targets assigning targets to develop the number of people for the development of the number of participating in social insurance and unemployment people participating in social insurance, insurance in the period of 2019 - 2021 unemployment insurance for provincial social security offices in the period of 2019 - 2021 7 2021 Decree 07/2021/ND-CP stipulating multidimensional Implemented and led to the increase in poverty line for the period of 2021 - 2025 premium contribution and state support for contribution 8 2022 Resolution 69/NQ-CP dated May 19, 2022, on Implemented assigning targets for developing social insurance participants 9 2022 Decision 1155/QD-BHXH dated May 12, 2022, on Implemented promulgating regulations on management of activities of service organizations collecting social insurance and health insurance. 10 2022 Decision 19/2022/QD-TTg dated September 22, Implemented 2022, on management costs of social insurance, unemployment insurance and health insurance for the period 2022-2024 75 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Annex 4: Benefits available on contributing to the voluntary scheme. Participants of the voluntary social insurance scheme are eligible to receive various benefits subject to meeting the following eligibility criteria: • Vesting period: 20 years (decrease to 15 years according to the new Law from July, 1st 2025) • Pension age: 60 years old for men or 55 years old for women, from 2021 gradually rising (3 months per year for man and 4 months per year for women) to 62 years old for men by 2028 or 60 years old for women by 2035. Pension Benefits: The pension benefits are calculated based on the premium and number of years of contribution by the participant, subject to annual statutory revisions from 2018 onwards (refer table 7 below). From 2022 onwards, it is equal to 45% of the insured’s average monthly covered income level (i.e., replacement rate) for 20 years (for men) or 15 years (for women); for each exceeding year of contributions, the pension benefit increases by additional 2% of the insured’s average monthly covered income level, to a maximum of 75%. From July 1st, 2025, the replacement rate for men contributing 15 years will be 40%, plus 1% for each additional year to 45% for 20 years of contribution. To receive the maximum pension at retirement age (i.e., 75% of the average monthly covered income level), male participants must contribute to the VSI for the entire 35 year period and female participants must contribute for 30 years. If the contribution period exceeds this, resulting in a higher replacement rate than the maximum statutory rate of 75%, the participants shall enjoy an additional allowance upon retirement, amounting to 0.5 times the average monthly covered income level for each exceeding year. Table 7: Pension scheme of voluntary SI Unit: Year Year of retirement 2018 2019 2020 2021 2022 2023 2024 July, 2025 Vesting period (year) Men 20 20 20 20 20 20 20 15 Women Pension age (years old) Men 60 60 60 60.25 60.5 60.75 61 61.25 Women 55 55 55 55.33 55.66 56 56.33 56.66 Minimum replacement rate (%) Men 45 45 45 45 45 45 45 40 Women 55 55 55 55 55 55 55 45 Contribution years to receive the Men 16 17 18 19 20 20 20 20 rate of 45% of average monthly Women 15 15 15 15 15 15 15 15 covered income level (years) Contribution years to receive Men 31 32 33 34 35 35 35 35 maximum rate (75% of average monthly covered income level) Women 30 30 30 30 30 30 30 30 (years) Source: According to 2014 Social Insurance Law and 2024 Social Insurance Law 76 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS One-time allowance (withdrawal): In addition to the pension benefits, participants are also eligible for one-time allowances prior to reaching the specified vesting period, to afford participants liquidity to cover any contingent expenses. This one-time benefit is calculated as follows: Under 1 year of participation: Allowance = 22% × Total of the insured’s average monthly covered income – Government subsidy Beyond 1 year of participation: Allowance = 1.5 × The insured’s average monthly covered income × participation years before 2014 + 2 × the insured’s average monthly covered income × participation years after 2014 – Government subsidy Survivorship and Funeral allowance: Funeral allowance: When voluntary SI participants are deceased or declared dead by the Court, a funeral allowance is awarded to the individual managing the same. Participants who have contributed voluntary SI premiums for full 60 months or more; and pensioners are eligible for this allowance. The allowance is a lump-sum benefit equal to 10 times the base salary as specified by the Government, which may be subject to statutory changes (can be revised / removed or replaced in the future). last monthly covered income reported during the contribution period. Survivorship Allowance: If a person who is participating in social insurance, on reservation or enjoying the pension, dies, his/her relatives will be entitled to the survivorship allowance in the form of a lump- sum. Benefit: • People participating in social insurance or in reservation: Under 1 year of participation: Allowance = 2 × Total of the insured’s average monthly covered income From 1 year of participation: Allowance = 1.5 × The insured’s average monthly covered income × participation years before 2014 + 2 × the insured’s average monthly covered income × participation years after 2014 77 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS • Pensioners: Death within the first 2 months of receiving pension: • Allowance = 48 × monthly pension. Death after 2 months of receiving pension: • Allowance = (48 – 0.5 × months of receiving pension) × monthly pension. The minimum amount of survivor allowance is equal to 03 months of last drawn pension amount. Health Insurance (HI) premiums: Pensioners are also issued free Health Insurance cards by the Social Insurance Fund, and monthly health insurance premiums are paid by the VSS, amounting to 4.5% of the monthly pension amount, which covers up to 95% of medical expenses56. Maternity allowance According to the new Law on Social Insurance 2024, voluntary SI participants (both male and female) who has paid social insurance for 6 months or more in the 12 months before giving birth, shall receive an allowance of 2 million VND for each child born or a 22 weeks or older fetus that dies in utero or during giving birth. 56 https://vss.gov.vn/english/news/Pages/vietnam-social-security.aspx?CateID=0&ItemID=11026#:~:text=In%20 addition%20to%20monthly%20retirement,by%20the%20Health%20Insurance%20Fund. 78 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Annex 5: Registration Form for Voluntary Pension Scheme 79 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS 80 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Annex 6: Snapshots of Online Self-registration portals 1. VSS Pulic Service Portal Step 1: Login VSS Pulic Service Portal at https://dichvucong.baohiemxahoi.gov.vn/ Select “Individual”, entry infomation: Account (Social Security Number), password, check code, then click “Login” 81 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Step 2: Select menu “Pay e-SI” Step 3. Select “Register to voluntary SI” Step 4: Declare information to register to voluntary social insurance - Personal information taken from the individual’s electronic transaction registration information including Social Security number 82 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS - Permanent residence address information is verified from the National database on Population, including house number, street/village; commune; district; province/city. - Select the category of voluntary social insurance participants whose contributions are supported by the state: poor households, near-poor households, other (not from poor or near-poor households). 83 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS - Input the payment amount and select the payment method: - Select where and how to receive the social insurance book cover: 84 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS - Select the service organization (if any): Step 5: Select “Confirm” to confirm the declared information 85 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS The system will check registrant’s information, calculate the amount to pay according to regulations and display the following screen: Step 6: Select “Pay” to start making payment 86 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS - Select the bank of the Social Security agency (where to receive payment). Then select “Payment” to continue. The system will navigate to the Payment Platform: - Select a bank or payment intermediary (where the participant has an account) to make payment. 2. National Public Service Portal Step 1: Log in to the National Public Service Portal at: https://dichvucong.gov.vn, select Login: 87 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS - Select the account type (for example: an account issued by the National Public Service Portal), enter Username, Password, Authentication Code and select Login. - Enter the OTP code sent to the registered phone number, then press Confirm 88 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Step 2: Select Online Public Services - Enter search keywords (for example: voluntary social insurance), press Search 89 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Step 3. Select online public service “Register, re-register, adjust voluntary social insurance contributions; issue social insurance book” Step 4. Select List of public services 90 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Step 5. Select Submit online at Public Service “Register to participate in voluntary social insurance contributions - For individuals”, the system shall navigate to the VSS Public Service Portal. Step 6. To continue registering to participate in voluntary social insurance, registrants continue from Step 3, Part I. 91 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Annex 7: Sampling Strategy for the Voluntary Scheme Desirability Survey (VSDS) The sample size will include 900 people. From the list of VSS, (Thai Nguyen, Nghe An, Long An), each province will be selected 300 people, including a mix of about active (100), dormant (100) and unregistered individuals (100). This section guides the provincial state office (PSO) of the GSO to review the enumeration areas and update the list of respondents as follows: Step 1: Review the enumeration area - Based on the list sent from the General Statistics Office, the PSO will select a list of respondents aged from 20 to 45 years old (ie, born between 1978 and 2003) including urban and rural areas and try to achieve 50% male and 50% female, each province will select 300 people, including a mix of about active (100), dormant (100) and unregistered individuals (100). An extra backup lists and mark separately the two types of lists as official and backup lists. Step 2: Update the list of respondents- Based on the list reviewed in step 1, the PSO prioritizes to review and update the current situation of the respondents to check whether they are in living in the province and can be accessed to do survey for the official listing? Contact the local social insurance officer (information provided by the social insurance agency) or local authorities (commune and village/hamlet level) to conduct a check and review with the phone numbers that do not have a specific address, or the phone number does not have or the phone number has but is incorrect, to verify and add information update the listing. In case, after updating the list of respondents, there are many cases where the address cannot be found or the address is moved to another province or hard-to-reach area, the PSO will continue to check the backup list until it is enough sampling as required in step 1. PSO try to achieve 50% male and 50% female in each type of participants (3 types) Step 3: The PSO completes the update of the list in the Excel file sent by GSO (specified in the column note: contacted to interview, changed address and updated, supplement specific addresses, wrong addresses and corrected, wrong phone number and corrected, unable to contact, have been moved and unable to access, no phone numbers and added) and sent to the GSO. The GSO will check the list and inform to PSO to prepare for the field work. 92 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Annex 8: Panel fixed-effects regression model results on factors associated with last monthly contribution amount The following table provides the results of the panel fixed-effects regression model used to understand the factors associated with the amounts of participants’ last reported monthly contribution amount towards the VSIF, utilizing various independent variables related to the participants’ socio-economic characteristics and awareness and interactions with the scheme, controlling for unobservable characteristics at the district level. Amount of last monthly Reference Assessed categories (for payment in 000 VND Independent Variables category categorical variables) Coefficient Standard Error Urban / Rural Urban Rural -118.6** (14.13) Near poor -35.83 (146.0) Type of household Poor Other 27.79 (88.38) Gender Male Female -17.90 (20.82) Employment Status Working Not Working -37.76 (18.46) Age 0.550 (2.427) Balance income after expenditures in 000 VND 0.00720 (0.00947) Participation status Active Dormant -130.9** (23.11) Between 5 million to 10 18.60 (58.95) million Between 10 million to 15 26.30 (35.96) million Between 15 million to 20 Household average total income per 102.0 (67.01) million month in VND Under 5 million Between 20 million to 25 192.0 (241.7) million Between 25 million to 30 40.38 (159.0) million Above 30 million 77.63 (242.7) No income 46.77 (197.7) Savings in personal Yes No -143.5** (22.09) bank account Mobile money savings Yes No 177.2** (28.77) Saving schemes in Yes No 32.21 (99.48) formal institutions Informal savings Savings Yes No 58.05 (44.72) schemes Channels Cash at home Yes No -41.41 (32.08) Used Cash loaned to family/ Yes No -57.35 (32.18) friends Assets - Jewelry, land, housing, vehicles, Yes No -69.34 (25.77) livestock, etc. Others Yes No 25.38 (165.0) 93 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Amount of last monthly Reference Assessed categories (for payment in 000 VND Independent Variables category categorical variables) Coefficient Standard Error Spouse -16.70 (30.87) Person initiating registration Self Parent 108.1 (108.2) Others 185.0 (181.6) VSS Agents -93.75 (63.49) Social Insurance Registration channel Online Self-Registration -112.9 (67.43) Agency Others -220.7 (113.2) Every 3 months -37.86 (41.19) Chosen frequency of last payment Monthly Every 6 months -74.71 (67.51) Every 12 months 20.51 (122.7) VSS Agent -30.04 (47.32) Social Insurance Payment channel Online/Bank Transfer -26.76 (39.90) Agency Other 197.8 (108.0) Satisfied 36.10 (21.60) Satisfaction with agent support for Highly Satisfied Neutral 223.0 (96.81) registration Unsatisfied -167.0 (228.9) Do you know about the minimum No -29.20 (63.20) Yes monthly contribution? Not Sure -27.55 (33.69) Do you know the retirement age for No -15.57 (31.56) Yes the VSS? Not Sure 38.71 (104.3) Do you know the minimum period No -111.9* (30.51) of participation in VSS to enjoy the Yes Not Sure -93.41 (65.58) benefits? Do you know the level of support No -196.6 (128.2) from the State budget for VSS Yes Not Sure -145.4 (91.23) participants? My savings -15.16 (17.64) I have assets I can sell if I -30.28 (11.23) need to My pension from Plan for old age financial survival My children will support me -103.9 (98.38) VSS I will continue working as -18.23 (100.6) long as possible I have no plan 63.33 (37.03) Grow a little 113.8 (75.90) Opinion on income growth over next Grow a lot Stay the same 90.51 (72.93) 10 years Shrink a little -75.53 (39.35) Easy 103.2* (32.58) Ease of task – Registration Very easy Neutral -51.21 (63.03) Difficult 74.09 (44.39) Easy 80.48 (81.92) Ease of task - Deposit Contributions Very easy Neutral 115.5 (111.1) 94 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS Amount of last monthly Reference Assessed categories (for payment in 000 VND Independent Variables category categorical variables) Coefficient Standard Error Easy -111.2** (20.83) Ease of task - Find out my enrollment Neutral -155.1 (55.63) Very easy information Difficult -288.2 (162.8) Very difficult -720.7* (213.6) Ease of task - Find out about the state Easy -260.1 (258.0) premium subsidy Very easy Neutral -208.2 (255.1) Difficult -238.4 (262.0) Easy 240.4 (263.0) Ease of task - Track my contributions Very easy Neutral 179.8 (281.2) till date Difficult 164.7 (252.4) Can you estimate the pension you No 44.09 (63.25) could get if you continued to save in Yes A little 57.22 (19.63) VSIF? Ability to accurately estimate VSIF Yes No 248.5 (164.0) state subsidy Constant Constant 456.1 (350.9) Observations 600 R-squared 0.216 Number of Provinces Fixed effects 3 Robust standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1 Annex 9: Robustness checks to verify expected significance of income level in determining participation status The LASSO regression conducted to identify factors correlated with participation status (ref table 6) indicated that income levels did not have a statistically significant correlation with participation status – a finding expected based on a review of existing literature (Brodersohn & Guven, 2019; Rother et al., 2022). Consequently, additional checks were required to be conducted under varying specifications to ensure the robustness of this finding and to identify potential causes for the same. Based on a simple regression specification where participation status is assumed to be correlated to gender, age and monthly household income, using a fixed effects model based on the district the respondent is located in, income is found to have a positive coefficient with participation status but is statistically insignificant as shown below - Model specification - participation status ~ gender + age + log (1+ monthly household income) | district fixed effects 95 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS As an additional test, balance income after expenses and savings in formal institutions are included as additional covariates in the above model. However, income remains statistically insignificant, but the coefficient becomes negative, indicating that the positive effect identified in the earlier model is explained by other variables (balance income and savings). Model specification – participation status ~ gender + age + log (1+ monthly household income) + log (1+ balance income after expenses) + savings in formal institutions | district fixed effects This indicates that income is not a statistically significant predictor of participation status for the available sample. This can be attributed to two possibilities viz. i) In the overall population in Vietnam, income is not a determinant of participation for the voluntary social insurance scheme and/or 2) there is sampling bias in the survey sample. 96 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS REFERENCES GSO 2021, The Population and Housing Census 2019: Population Ageing And Older Persons in Viet Nam. Lindert et al 2020, Sourcebook on the Foundations of Social Protection Delivery Systems. (GSO), G. S. O. (2021). Result of the Viet Nam household living standards survey 2020. S. P. House. Cunningham, W., & Pimhidzai, O. (2018). Vietnam’s Future Jobs. World Bank: Washington, DC, USA. Guven, M., Jain, H., & Joubert, C. (2021). Social Protection for the Informal Economy. Guven, M. U., & Jain, H. (2023). The Promise of Ejo Heza: A Brighter Future for All Rwandans-Executive Summary. Jain, H., & Palacios, R. J. (2022). Assessment of VSS’s Voluntary Pension Scheme and Policy Recommendations. Lindert, K., Karippacheril, T. G., Caillava, I. R., & Chávez, K. N. (2020). Sourcebook on the foundations of social protection delivery systems. World Bank Publications. O’Higgins, N., & Viegelahn, C. (2021). Informal employment in Viet Nam: trends and determinants. Ohnsorge, F., & Yu, S. (2022). The long shadow of informality: Challenges and policies. World Bank Publications. VSS IT Center, 2023, National Database on Social Security Trang L, & Nguyen T.N, & Palacios. R. (2024). Expanding pension coverage to the informal sector: the case of Viet Nam 97 Expanding Coverage in Vietnam’s Voluntary Social Insurance scheme: INSIGHTS AND RECOMMENDATIONS 98