## wpiea2022185-print-pdf

## Source details

**Canonical URL:** [wpiea2022185-print-pdf](https://www.imf.org/-/media/files/publications/wp/2022/english/wpiea2022185-print-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2022/english/wpiea2022185-print-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2022/english/wpiea2022185-print-pdf.pdf.json)

---

### Digitalization and technological context in Vietnam
- Fixed and mobile-broadbands at 20 and 70 percent, respectively.
- Vietnam is among the first countries to trial 5G.
- Competitive cellular and broadband prices in its region, though internet speed remains relatively low.
- Lags on traditional measures of financial inclusion but performs well on digital measures of financial inclusion (access and usage of internet and mobile accounts).
- E-government initiatives: national public services portal, e-document exchange platform, and a new national financial inclusion strategy.
- Figures cited in source: Fig 1: Mobile and Internet Subscriptions; Fig 2: Mobile and Internet Prices; Fig 3: Financial Inclusion Indices; Fig 4: Coverage, Social Protection and Labor.

### Current social protection coverage and delivery limitations
- Coverage incomplete and delivery fragmented despite ongoing efforts.
- Labor force coverage in 2019: about 35 percent.
- Pensions coverage: 33 percent.
- Unemployment insurance coverage: 27 percent.
- Well-developed social programs for the very poor but relatively little assistance to other vulnerable groups, such as informal urban workers.
- Large informal workforce contributes to low coverage and reliance on self-insurance mechanisms.
- Delivery system relied on manual, time-consuming registration and inefficient payment procedures.
- Limited use of digital tools during the pandemic reduced timeliness and effectiveness of assistance, particularly during lockdowns.
- Aggregate household income lost during the lockdowns in 2021Q3: US$1.9 billion.
- In 2020, only 30 percent of the announced cash transfers in response to Covid-19 were eventually disbursed.
- Millions of informal workers left out of existing safety nets due to strict criteria and lack of verifiable data on the informal sector.

### Macro and micro evidence approach of the paper
- Paper components:
  - Review cross-country experiences on digital tools for identifying and authenticating individuals (identification stage).
  - Review cross-country experiences on digital payments and delivery mechanisms (delivery stage).
  - Use household data (2018 vintage of the Vietnam Household Living Standards Survey (VHLSS)) to simulate distributional implications of the Covid-19 recession and counterfactual social protection policies.
- Simulation methodology:
  - Covid-19 recession modeled as an exogenous income shock primarily affecting service-sector workers (following De Stefani et al. (2022)).
  - Counterfactual income distributions constructed under the current unemployment insurance scheme and under a potential system that provides benefits to informal workers via digital payments.
- Key simulation finding:
  - Implementing a digital payments system that reached informal workers could have reduced the average household income loss by half.
- Data caveat:
  - VHLSS 2018 data collected before the Covid-19 pandemic; more recent datasets covering the worst sections of the pandemic in Vietnam were not yet available.

### Cross-country lessons on identification and data linking
- Trade-offs: exclusion errors (non–take-up) versus inclusion errors (leakage), especially with large informal sectors.
- Cross-linking and cross-checking of databases using unique identifiers facilitates identifying and verifying eligible beneficiaries.
- Automated cross-linking sources can include:
  - beneficiaries of social security sub-systems;
  - beneficiaries of public utilities’ subsidized tariffs;
  - school and health service beneficiary databases;
  - data on the informal sector from local government entities and NGOs;
  - voter/election registration databases.
- Vietnam constraints:
  - Limited data sharing across agencies; data often stored and used only within collecting organizations.
  - Over 1.3 million duplicate health insurance cards issued in 2012-13 due to lack of integration and absence of a nationally standardized user identifier system.
- Digital and biometric IDs:
  - As of end-2018, 175 of 196 surveyed countries had some type of national ID system; most were digitized and around half collected biometric features.
  - Vietnam: partial digital ID but no biometrics collected (Figure 6).
  - Cost estimates for establishing biometric ID systems: 0.6 percent of GDP or about $4-11 per registrant for enrollment and credential issuance, with maintenance costs of 0.1 percent of GDP annually; alternative estimates around $5 per person.
  - Examples: India’s Aadhaar and JAM linking accounts, biometric digital ID, and mobile networks; South Africa eliminated 850,000 ghost beneficiaries in 2014 through biometric registration and halved administrative costs.
- Fiscal gains from digitalization examples:
  - e-KYC can reduce average customer verification costs from $15 to $0.50.
  - Aadhaar-enabled e-KYC reportedly reduced KYC cost from Rs 40 ($0.60) per customer to Rs 5 ($0.07).
  - Indian government estimated fiscal gains of more than $12 bn since 2013 from Aadhaar-enabled direct benefit transfers and related reforms, around nine times the cost of reform implementation.

### Status of digital ID rollout in Vietnam
- As of 2020: partially digitized foundational ID system; half of ASEAN members had fully digitized ID systems (many using smartcards and biometrics).
- National ID near universal but an electronic unique authentication was not yet available at that stage.
- Ministry of Public Security activities:
  - Since July 1st, 2021, chip-based ID cards began to replace paper-based national IDs for more than 50 million citizens.
  - Proposed digital ID scheme intended to enable e-transactions with government agencies and will include biometric identifiers (fingerprints), taxes, and health insurance information.

### Reaching beneficiaries: digital G2P and mobile money lessons
- G2P payments encompass tax refunds, subsidies, social programs, salary, stipends, pensions, scholarships, and emergency assistance.
- Notable digitized G2P programs: Brazil’s Bolsa Familia (established in 2003), Iran’s direct cash electronic transfers in 2010, Sierra Leone’s digital payments during the 2014-15 Ebola crisis.
- Routing G2P through bank or mobile accounts encourages greater financial inclusion.
- Vietnam: government transfer payments predominantly done in cash compared to ASEAN and EMLIDC peers.
- Mobile money advantages:
  - Cost-effective where traditional financial services have limited penetration.
  - Lund et al (2017) estimate digitalizing government payments could save the equivalent of 1 percent of GDP per year.
  - ASEAN infrastructure density per 100,000 adults in 2019: 282 mobile money outlets; 48 ATMs; 24 banks.
  - M-Pesa model progressed from P2P to institutional payments to G2P; agents act as (mobile) ATMs for deposits and withdrawals.
- Mobile G2P typical rollout steps:
  - Government selects MMO(s) with wide coverage.
  - Government identifies eligible recipients and wires money to partnering banks.
  - Banks authenticate beneficiaries via KYC or e-KYC and convert funds into mobile money.
  - Beneficiaries receive funds via mobile wallets (apps or SMS access codes).
  - Ensure branch and ATM efficacy, customer trust, and risk management, including KYC requirements.
  - Cash-out via MMOs access points or partnering local agents.
- Ecosystem effects: P2P, P2B, P2G use cases expand digital payments’ benefits.
- Regulation and enabling environment:
  - Positive relationship between active mobile money accounts and GSMA (2019b) mobile money regulatory index (authorization, consumer protection, transaction limits, KYC, agent networks, investment and infrastructure).
  - Collaboration between government and MMOs essential; regulated non-bank fintechs can accelerate mobile G2P (examples: GCash, Wave Money, GrabPay).

### Vietnam mobile money pilot
- PM 2021 Decision 316-TTg approved a nationwide pilot for mobile money, prioritizing rural and remote areas.
- Pilot features:
  - Duration: runs for 2 years.
  - Allowed services: cash deposits/withdrawals into mobile money accounts linked to customers’ financial institutions or e-wallets; sending/receiving domestic remittances; payments for goods and services.
  - Transaction limits: VND 10 mn (around $430) per month.
  - Eligible providers: telecommunication companies and enterprises with licensed e-wallet services.
  - e-KYC: providers can open mobile money accounts via e-KYC; MoF and SBV tasked with developing e-KYC according to PM 2020 Decision 2289/QD-TTg.
  - Implementation decrees with concrete steps remain to follow.

### Simulating the welfare gains of digital social protection: setup and scenarios
- Motivation: digital payments can increase coverage by reaching workers not registered in traditional social security programs.
- Formal worker support: unemployment insurance (UI) in Vietnam provided 60 percent of lost wages.
- Informality statistics:
  - Approximately 53 percent of households have at least one informal worker.
  - Informal workers account for about 30 percent of workers in the service sector (and higher shares in industry and agriculture).
  - Among households with informal workers: about 55 percent do not have access to financial services.
  - 28 percent are not eligible to receive social security payment because all members are informal workers.
  - Digital access among households with informal workers: 76 percent reported accessing the internet in the month preceding the VHLSS data collection.
- Data: VHLSS 2018 vintage.
- Informal worker definition: without a labor contract and no access to social insurance; if missing, self-employed in household business classified as informal.
- Income under shock specification:
  - y_h(ε) = y_h(0) × [1 − s_hi ε − (1 − θ) s_hf ε]
    - s_hi = share of informal workers in the service sector in household h
    - s_hf = share of formal workers in the service sector in household h
    - θ = share of income received from unemployment benefits (equal to 0.6 in Vietnam’s case)
    - ε = size of the income shock (share of income lost or share of year without income)
- Counterfactual with digital access:
  - y_h^digital(ε) = y_h(0) × [1 − (1 − d_h θ) s_hi ε − (1 − θ) s_hf ε]
    - d_h indicator if household h has internet access
- Shock severities simulated: ε ∈ {0.25, 0.5, 1}
- Four income-distribution cases computed:
  1. Pre-Covid (data)
  2. Covid shock and no transfers
  3. Covid shock and existing transfers (UI only to formal workers)
  4. Covid shock and digital payments (UI to informal workers with digital access)

### Simulation results and key statistics
- Table 1: Average income loss (percent, relative to 2018 baseline)
  - ε = 0.25: No transfers −9.3 ; UI transfers −4.3 ; Digital payments −3.8
  - ε = 0.5:  No transfers −18.7 ; UI transfers −8.6 ; Digital payments −7.6
  - ε = 1:    No transfers −37.4 ; UI transfers −17.2 ; Digital payments −15.2
- Main findings:
  - Covid shock shifts income distribution left and compresses around a smaller mean; least affected households are at the bottom (largely agricultural).
  - UI transfers undo a large portion of the Covid effect, but a significant gap relative to initial income remains.
  - Expanding UI via digital payments could have significantly reduced average income loss relative to existing UI.
  - Gains from digital payments increase with duration/intensity of the Covid crisis:
    - For ε = 0.25, average household “gains” 0.5 percent of 2018 annual income.
    - For ε = 1, this number increases to 2 percent.
  - Distributional pattern of gains:
    - Greatest beneficiaries are the urban middle class (middle of the income distribution), where informality and digital access are both pervasive.
    - Top of distribution benefit less from digital expansion because they are more likely formal and already covered by UI.
    - Bottom of distribution benefit less because they are less likely to have digital access.

### Digital access, financial access, and inequality — regression evidence
- Linear probability model: I{digital access_h} = βX_h + δ_m(h) + δ_p(h) + ε_h with explanatory variables including household income, highest education, urban/rural, family size, gender of head, share informal, share industry, share services, plus month and province fixed effects.
- Selected regression coefficients (exactly as in source):
  - log(Income): Digital Access (1) 0.211*** ; (2) 0.176*** ; Financial Access (3) 0.204*** ; (4) 0.165***
  - Education (HH head) Secondary: Digital Access (1) 0.245*** ; (2) 0.232*** ; Financial Access (3) 0.083*** ; (4) 0.068***
  - Education (HH head) College+: Digital Access (1) 0.317*** ; (2) 0.264*** ; Financial Access (3) 0.262*** ; (4) 0.222***
  - Urban: Digital Access (1) 0.030** ; (2) 0.024* ; Financial Access (3) 0.097*** ; (4) 0.075***
  - Family Size: Digital Access (1) 0.029*** ; (2) 0.028*** ; Financial Access (3) −0.029*** ; (4) −0.019***
  - Female HH head: Digital Access (1) 0.035** ; (2) 0.023* ; Financial Access (3) 0.048*** ; (4) 0.048***
  - Share Informal: Digital Access (1) 0.061*** ; (2) −0.064*** ; Financial Access (3) −0.218*** ; (4) −0.358***
  - Share Service: Digital Access (2) 0.205*** 
  - Share Industry: Digital Access (2) 0.181***
  - Digital Access predicts Financial Access: 0.196*** (3) ; 0.167*** (4)
  - Observations: (1)/(3) 4,663 ; (2)/(4) 4,476
  - R-squared: (1) 0.379 ; (2) 0.368 ; (3) 0.408 ; (4) 0.431
- Interpretation:
  - Income and education have the largest positive impact on digital access.
  - Urban location, female-headed households, and higher share of informal workers show mixed associations depending on specification.
  - Low-income, poorly educated, agriculture-dependent households remain largely excluded from digital benefits.
  - Financial exclusion mirrors digital exclusion; digital access positively predicts financial access.

### Caveats, limitations, and risks
- Strong assumptions in shock construction: size of shock on service workers arbitrary; other sectors assumed unaffected.
- Service workers may not have lost entire income; general equilibrium effects and partial impacts in other sectors not modeled.
- Exercise assumes all informal workers were eligible for assistance; in practice, eligibility validation and verification required.
- Risks to digitalization:
  - Digital exclusion, cyberattacks, digital fraud, privacy, and security concerns.
  - Need for appropriate regulation and risk management.

### Policy implications and recommendations
- Roll out digital/biometric IDs and expand their use beyond government agencies.
- Cross-link digital IDs with socioeconomic databases and create a unified beneficiary database securely accessible to relevant ministries/agencies.
- Link digital IDs with bank and/or mobile money accounts to facilitate financial inclusion and e-KYC.
- Translate national digital transformation objectives into actionable plans and targets via implementation decrees.
- Ensure data sharing across government agencies to operationalize plans.
- Use digital payments, including mobile money, to enable quick rollout of government transfers to reach unbanked and informal workers.
- Implement appropriate regulations to manage digital exclusion, cyberattacks, digital fraud, privacy, and security concerns.

*IMF Working Paper — Digitalization and Social Protection: Macro and Micro Lessons for Vietnam (Introduction and selected chapters).*

### 1. Introduction ........................................................................................................

### 1. Introduction

### Digitalization and technological context in Vietnam
- Vietnam has relatively high penetration of mobile phones and internet: "fixed and mobile-broadbands at 20 and 70 percent, respectively".
- Vietnam is among the first countries to trial 5G.
- Vietnam offers competitive cellular and broadband prices in its region, though internet speed remains relatively low.
- Vietnam lags on traditional measures of financial inclusion (access and usage of accounts at financial institutions) but performs well on digital measures of financial inclusion (access and usage of internet and mobile accounts).
- E-government initiatives include a national public services portal, the e-document exchange platform, and a new national financial inclusion strategy.
- Figures cited in source: Fig 1: Mobile and Internet Subscriptions; Fig 2: Mobile and Internet Prices; Fig 3: Financial Inclusion Indices; Fig 4: Coverage, Social Protection and Labor.

### Current social protection coverage and delivery limitations
- Coverage is incomplete and delivery remains fragmented despite ongoing efforts.
- In 2019, only about "35 percent of the labor force was covered by social protection".
- Health coverage is near universal and higher than pensions and unemployment insurance: pensions coverage "33 percent" and unemployment insurance "27 percent".
- Vietnam has well-developed social programs for the very poor, but relatively little assistance to other vulnerable groups, such as informal urban workers.
- Large informal workforce contributes to low coverage and a reliance on self-insurance mechanisms.
- The delivery system relied on manual, time-consuming registration and inefficient payment procedures.
- Limited use of digital tools during the pandemic reduced timeliness and effectiveness of assistance, particularly during lockdowns.
- An estimated "US$1.9 billion" in aggregate household income was lost during the lockdowns in "2021Q3".
- In 2020, only "30 percent" of the announced cash transfers in response to Covid-19 were eventually disbursed.
- Millions of informal workers were left out of existing safety nets due to strict criteria and lack of verifiable data on the informal sector.

### Macro and micro evidence approach of the paper
- The paper provides macro and micro evidence on how digitalization can improve social protection systems in Vietnam by:
  - Reviewing cross-country experiences on digital tools for identifying and authenticating individuals (the identification stage of the social assistance value chain).
  - Reviewing cross-country experiences on digital payments and delivery mechanisms (the delivery stage).
  - Using household data (2018 vintage of the Vietnam Household Living Standards Survey (VHLSS)) to simulate distributional implications of the Covid-19 recession and counterfactual social protection policies.
- Simulation methodology:
  - The Covid-19 recession is modeled as an exogenous income shock primarily affecting service-sector workers (following De Stefani et al. (2022)).
  - Counterfactual income distributions are constructed under the current unemployment insurance scheme and under a potential system that provides benefits to informal workers via digital payments.
- Key simulation finding:
  - Baseline result suggests that implementing a digital payments system that reached informal workers could have reduced the average household income loss "by half".
- Data caveat:
  - The data used (VHLSS 2018) were collected before the Covid-19 pandemic; more recent datasets covering the worst sections of the pandemic in Vietnam were not yet available.

### Cross-country lessons on identification and data linking
- Governments face trade-offs between exclusion errors (non–take-up) and inclusion errors (leakage), especially with large informal sectors.
- Cross-linking and cross-checking of databases using unique identifiers can facilitate identifying and verifying eligible beneficiaries.
- Examples: Pakistan and Thailand cross-checked national IDs against social benefit and revenue databases.
- Automated cross-linking can use multiple sources including:
  - beneficiaries of social security sub-systems;
  - beneficiaries of public utilities’ subsidized tariffs;
  - school and health service beneficiary databases;
  - data on the informal sector from local government entities and NGOs;
  - voter/election registration databases.
- Vietnam’s cross-linking is hampered by limited data sharing across agencies; data often stored and used only within collecting organizations.
  - Example: Over "1.3 million" duplicate health insurance cards issued in 2012-13 due to lack of integration and absence of a nationally standardized user identifier system.
- Digital and biometric IDs:
  - As of end-2018, "175 of 196 surveyed countries had some type of national ID system"; most were digitized and around half collected biometric features.
  - Vietnam currently has a partial digital ID but no biometrics collected (Figure 6).
  - Cost estimates cited for establishing biometric ID systems: "0.6 percent of GDP" or about "$4-11 per registrant" for enrollment and credential issuance, with maintenance costs of "0.1 percent of GDP annually"; alternative estimates around "$5 per person".
  - Examples: India’s Aadhaar and JAM (Jan Dhan-Aadhaar-Mobile) linking accounts, biometric digital ID, and mobile networks; South Africa eliminated "850,000" ghost beneficiaries in 2014 through biometric registration and halved administrative costs.
- Digitalization can yield fiscal savings:
  - e-KYC can reduce average customer verification costs from "$15" to "$0.50".
  - Aadhaar-enabled e-KYC reportedly reduced KYC cost from "Rs 40 ($0.60)" per customer to "Rs 5 ($0.07)".
  - Indian government estimated fiscal gains of more than "$12 bn" since 2013 from Aadhaar-enabled direct benefit transfers and related reforms, around nine times the cost of reform implementation.

### Status of digital ID rollout in Vietnam
- As of 2020, Vietnam had a partially digitized foundational ID system; half of ASEAN members had fully digitized ID systems (many using smartcards and biometrics).
- Vietnam’s national ID is near universal but an electronic unique authentication was not yet available at that stage.
- Ministry of Public Security activities (as described in source):
  - Since "July 1st, 2021", chip-based ID cards began to replace paper-based national IDs for more than "50 million" citizens.
  - The proposed digital ID scheme is intended to enable e-transactions with government agencies and will include biometric identifiers (fingerprints), taxes, and health insurance information.

*IMF Working Paper — Digitalization and Social Protection: Macro and Micro Lessons for Vietnam (Introduction section)*

### 3. Reaching Beneficiaries: Cross-Country

### 3. Reaching Beneficiaries: Cross-Country Experiences

### Overview: digital G2P and G2B payments before and during the pandemic
- G2P payments include payments (or transfers) of tax refunds, subsidies, social programs, salary, stipends, pensions, scholarships, and emergency assistance.
- Notable digitized G2P programs cited:
  - Brazil’s Bolsa Familia (established in 2003) providing monthly transfers via smart cards.
  - Iran’s direct cash electronic transfers to compensate vulnerable households during fuel subsidy reforms in 2010.
  - Sierra Leone’s digital payments to response workers during the 2014-15 Ebola crisis.
- Routing G2P transfers through bank or mobile accounts encourages greater financial inclusion.
- Payments to businesses (G2B) increasingly disbursed digitally; example: in Peru, 59 percent of subnational government G2B procurement payments were made via checks and 41 percent via electronic transfers before the pandemic.
- In Vietnam, government transfer payments are predominately done in cash compared to ASEAN and EMLIDC peers.

### Mobile money as a cost-effective delivery channel
- Mobile money can be cost-effective especially where traditional financial services have limited penetration.
- Mobile money prevalence is particularly high in LIDCs in Sub-Saharan Africa.
- Estimated benefits and cost comparisons:
  - Lund et al (2017) estimate that digitalizing government payments could save the equivalent of 1 percent of GDP per year.
  - The cost of mobile payments is often lower than other disbursement methods due to larger number of access points and wider rural reach.
- Infrastructure density example (ASEAN, per 100,000 adults in 2019):
  - 282 mobile money outlets
  - 48 ATMs
  - 24 banks
- M-Pesa model: an SMS-based system that progressed from P2P to institutional payments to G2P cash transfers (including during Covid-19); agents act as (mobile) ATMs for deposits and withdrawals.

### Mobile G2P program mechanics and typical rollout steps
- Typical sequence for “Mobile G2P” programs:
  - Government selects one or multiple mobile money operators (MMOs) such as MNOs, commercial banks, or fintech firms; MMOs should have high quality of service and wide coverage, especially in rural areas.
  - Government identifies eligible recipients and wires money to the bank(s) partnering with the MMOs.
  - Banks authenticate beneficiaries through KYC or e-KYC processes and convert funds into mobile money.
  - Beneficiaries receive funds via mobile wallets (mobile apps or access codes via SMS).
  - Ensure branch and ATM efficacy, customer trust, and risk management, including in KYC requirements.
  - Mobile wallet owners can cash-out typically through MMOs access points or partnering local (mobile) agents.

### Broader ecosystem effects and use cases (G, P, B)
- Engagement of public and private actors encourages mobile money adoption across government, people, and businesses:
  - P2P: beneficiaries can digitally send money or receive remittances.
  - P2B: beneficiaries can purchase goods and services at merchants accepting mobile money.
  - P2G: beneficiaries can digitally pay utility or other bills.
- Evidence suggests digital payment methods are gaining traction in Vietnam but Vietnam remains more dependent on cash than ASEAN and EMLIDC peers.
- Wider availability and affordability of digital services would support longer-term benefits from mobile money.

### Regulation, enabling environment, and fintech participation
- Regulation materially affects adoption: positive relationship between number of active mobile money accounts and the GSMA (2019b) mobile money regulatory index.
- The GSMA index covers six broad enabling dimensions: authorization, consumer protection, transaction limits, KYC requirements, agent networks, and investment and infrastructure environment.
- Collaboration between government and MMOs is essential for successful G2P mobile payments.
- Non-bank fintech companies, if adequately regulated, can accelerate mobile G2P programs; examples of tech firms facilitating G2P programs: GCash (Philippines), Wave Money (Myanmar), GrabPay (Malaysia).

### Vietnam pilot for mobile money
- PM 2021 Decision 316-TTg approved a nationwide pilot for mobile money, prioritizing rural and remote areas.
- Pilot features:
  - Duration: runs for 2 years.
  - Allowed services: cash deposits/withdrawals into mobile money accounts linked to customers’ financial institutions or e-wallets; sending/receiving domestic remittances; payments for goods and services.
  - Transaction limits: VND 10 mn (around $430) per month.
  - Mobile money service providers eligible: telecommunication companies and enterprises with licensed e-wallet services.
  - e-KYC: providers can open mobile money accounts via e-KYC, which the MoF and SBV are tasked with developing according to PM 2020 Decision 2289/QD-TTg.
  - Implementation decrees with concrete steps remain to follow.

### Key statistics and figures cited
- Peru subnational G2B procurement payments before pandemic: 59 percent checks, 41 percent electronic transfers.
- Lund et al (2017) estimated savings from digitalizing government payments: equivalent of 1 percent of GDP per year.
- ASEAN infrastructure per 100,000 adults in 2019: 282 mobile money outlets; 48 ATMs; 24 banks.
- Vietnam mobile money pilot: transaction limit VND 10 mn (around $430) per month; pilot duration 2 years.

*Source: IMF Working Paper — "Digitalization and Social Protection: Macro and Micro Lessons for Vietnam", Chapter 3: Reaching Beneficiaries: Cross-Country Experiences.*

### 4. Simulating the Welfare Gains of Digital Social

### 4. Simulating the Welfare Gains of Digital Social Protection for Households During Covid-19

### Context and motivation
- Digital payments can increase coverage of social safety nets by reaching workers not registered in traditional social security programs.
- The Covid-19 pandemic highlighted the need to increase social protection coverage as millions lost jobs or saw incomes reduced.
- For formal workers, unemployment insurance (UI) in Vietnam provided 60 percent of lost wages.
- Informal workers represent approximately 53 percent of households having at least one informal worker and account for about 30 percent of workers in the service sector (and higher shares in industry and agriculture).
- Among households with informal workers:
  - about 55 percent do not have access to financial services (bank accounts, credit cards, among others),
  - 28 percent are not eligible to receive social security payment because all members are informal workers.
- Digital access is pervasive among households with informal workers: 76 percent of households with at least one informal worker reported accessing the internet in the month preceding the VHLSS data collection.

### Data and methodology
- Data source: 2018 vintage of the Vietnam Household Living Standard Survey (VHLSS).
- Definition: Informal workers are those without a labor contract and no access to social insurance; if missing, self-employed in household business are classified as informal.
- Income measurement: household pre-Covid income is y_h(0) (see annex in source).
- Covid shock modeled as an exogenous income shock primarily affecting service-sector workers (retail, restaurants, logistics, hospitality).
- Assumption: non-service workers’ incomes unaffected; service-sector workers lose a portion of income.

### Simulation scenarios and model specification
- Household income under shock:
  - y_h(ε) = y_h(0) × [1 − s_hi ε − (1 − θ) s_hf ε]
    - where s_hi = share of informal workers in the service sector in household h
    - s_hf = share of formal workers in the service sector in household h
    - θ = share of income received from unemployment benefits (equal to 0.6 in Vietnam’s case)
    - ε = size of the income shock (share of income lost or share of year without income)
- Counterfactual with digital access (d_h indicator if household h has internet access):
  - y_h^digital(ε) = y_h(0) × [1 − (1 − d_h θ) s_hi ε − (1 − θ) s_hf ε]
- Three shock severities simulated:
  - ε ∈ {0.25, 0.5, 1}
- Four income-distribution cases computed:
  1. Pre-Covid (data)
  2. Covid shock and no transfers
  3. Covid shock and existing transfers (UI only to formal workers)
  4. Covid shock and digital payments (UI to informal workers with digital access)

### Results and key statistics
- Table 1: Average income loss (percent, relative to 2018 baseline)
  - ε = 0.25: No transfers −9.3 ; UI transfers −4.3 ; Digital payments −3.8
  - ε = 0.5:  No transfers −18.7 ; UI transfers −8.6 ; Digital payments −7.6
  - ε = 1:    No transfers −37.4 ; UI transfers −17.2 ; Digital payments −15.2
- Main findings:
  - The Covid shock shifts the income distribution left and compresses around a smaller mean; least affected households are at the bottom (largely agricultural).
  - UI transfers undo a large portion of the Covid effect, but a significant gap relative to initial income remains.
  - Expanding UI via digital payments could have significantly reduced average income loss relative to existing UI.
  - Gains from digital payments increase with the duration/intensity of the Covid crisis:
    - For ε = 0.25, average household “gains” 0.5 percent of 2018 annual income.
    - For ε = 1, this number increases to 2 percent.
  - Distributional pattern of gains:
    - Greatest beneficiaries are the urban middle class (middle of the income distribution), where informality and digital access are both pervasive.
    - Top of distribution benefit less from digital expansion because they are more likely formal and already covered by UI.
    - Bottom of distribution benefit less because they are less likely to have digital access.

### Digital access, financial access, and inequality
- Linear probability model for digital access (I{digital access_h} = βX_h + δ_m(h) + δ_p(h) + ε_h) uses explanatory variables: household income, highest education, urban/rural, family size, gender of head, share informal, share industry, share services, plus month and province fixed effects.
- Key regression results (Table 2 highlights, coefficients shown exactly as in source):
  - log(Income): Digital Access (1) 0.211*** ; (2) 0.176*** ; Financial Access (3) 0.204*** ; (4) 0.165***
  - Education (HH head) Secondary: Digital Access (1) 0.245*** ; (2) 0.232*** ; Financial Access (3) 0.083*** ; (4) 0.068***
  - Education (HH head) College+: Digital Access (1) 0.317*** ; (2) 0.264*** ; Financial Access (3) 0.262*** ; (4) 0.222***
  - Urban: Digital Access (1) 0.030** ; (2) 0.024* ; Financial Access (3) 0.097*** ; (4) 0.075***
  - Family Size: Digital Access (1) 0.029*** ; (2) 0.028*** ; Financial Access (3) −0.029*** ; (4) −0.019***
  - Female HH head: Digital Access (1) 0.035** ; (2) 0.023* ; Financial Access (3) 0.048*** ; (4) 0.048***
  - Share Informal: Digital Access (1) 0.061*** ; (2) −0.064*** ; Financial Access (3) −0.218*** ; (4) −0.358***
  - Share Service: Digital Access (2) 0.205*** ; (no coefficient in (1))
  - Share Industry: Digital Access (2) 0.181*** ; (no coefficient in (1))
  - Digital Access predicts Financial Access: 0.196*** (3) ; 0.167*** (4)
  - Observations: (1)/(3) 4,663 ; (2)/(4) 4,476
  - R-squared: (1) 0.379 ; (2) 0.368 ; (3) 0.408 ; (4) 0.431
- Interpretation:
  - Income and education have the largest positive impact on digital access.
  - Urban location, female-headed households, and higher share of informal workers appear associated with higher digital access in simple specifications, but these effects change when sector controls are included.
  - Low-income, poorly educated, agriculture-dependent households remain largely excluded from digital benefits.
  - Financial exclusion mirrors digital exclusion; many characteristics predicting lack of digital access also predict lack of financial access.
  - Increasing digital access may alleviate financial exclusion (e.g., via mobile money).

### Caveats and limitations
- Strong assumptions in shock construction: size of shock on service workers is arbitrary; other sectors assumed unaffected.
- Service workers may not have lost entire income; general equilibrium effects and partial impacts in other sectors are not modeled.
- Exercise implicitly assumes all informal workers were eligible for assistance during the pandemic; in practice, eligibility validation and verification are required.
- Results should be interpreted with caution but provide insight into size and distribution of potential gains from digital payments.

### Policy implications and recommendations
- Digital IDs and automated cross-linking of databases can improve targeting and coverage of social programs.
- Digital payments, including mobile money, can enable quick rollout of government transfers to reach unbanked and informal workers.
- Vietnam progress noted:
  - Moves toward fully digitized foundational IDs with biometric features.
  - Piloting a nationwide mobile money project and e-KYC facilitation.
- Next steps:
  - Roll out digital/biometric IDs and expand their use beyond government agencies.
  - Cross-link digital IDs with socioeconomic databases and create a unified beneficiary database securely accessible to relevant ministries/agencies.
  - Link digital IDs with bank and/or mobile money accounts to facilitate financial inclusion and e-KYC.
  - Translate national digital transformation objectives into actionable plans and targets via implementation decrees.
  - Ensure data sharing across government agencies to operationalize plans.
- Risk management and regulation:
  - Appropriate regulations are needed to manage digital exclusion, cyberattacks, digital fraud, privacy, and security concerns to fully reap digitalization benefits.

*Source: 4. Simulating the Welfare Gains of Digital Social Protection for Households During Covid-19 (IMF Working Paper).*

### References

### wpiea2022185-print-pdf - References

### References
- Citations focus on digital financial services, digital ID, social protection, and COVID-19 responses in emerging and developing economies.
- Key works include studies and policy papers by: Agur; Bangura; Bazarbash et al.; Cameron et al.; D’Silva et al.; Dabla-Norris et al.; Davidovic et al.; De Stefani et al.; Dercon; Devereux and Vincent; Gelb and coauthors; Gentilini et al.; GSMA; Handayani et al.; ILO; IMF; Lund, White and Lamb; Morgan and Trinh; Muralidharan, Niehaus and Sukhtankar; Prady; Sahay et al.; Townsend; Una et al.; and multiple World Bank publications.
- Document types referenced include: IMF Working Papers (for example, no 20/198 and no 2020/273), BIS Papers no. 106, CGD Working Papers and Policy Papers, World Bank reports, GSMA reports, ADB Sustainable Development Working Paper Series no. 50, and journal articles (e.g., American Economic Review; The Journal of Economic Perspectives).
- Geographical and thematic coverage in the references includes: India, Vietnam, Sierra Leone, Cambodia, Nepal, Mongolia, Sub-Saharan Africa, Southeast Asia, and broader analyses of fintech, mobile money, biometric ID, and government-to-person transfers during the COVID-19 pandemic.

### Annex: Data and Measurements
- Main microdata source: Vietnam Household Living Standards Survey (VHLSS) by Vietnam’s General Statistics Office (GSO).
- VHLSS 2018 vintage used as baseline (latest available year before the Covid-19 crisis).
- Survey scope:
  - Around 46,995 households were surveyed in face-to-face interviews.
  - Only 9,399 households were asked about the full range of questions (income, expenditure, and other issues).
- Final analysis sample after basic data cleaning:
  - 4,713 households retained.
  - 17,478 individuals (members of those households, including children and elderly members).
- Individual data covers: age, education, employment, and other basic demographic questions.
- VHLSS collects information on:
  - Demographic characteristics related to living standards
  - Education
  - Health and healthcare
  - Labor and employment
  - Income
  - Expenditures (consumption and durable goods)
  - Housing, electricity, water, sanitation facilities
  - Participation in poverty alleviation programs
  - Household businesses
  - Commune general characteristics
- Income variable construction: household’s net annual income, aggregating all sources of revenue net of costs; calculated as the sum of:
  - All revenues from employment, including wages, bonuses, subsidies, and other revenues
  - Revenue from education and healthcare aid, as well as from rental of properties
  - Sales of crops and revenues from land, animal husbandry, hunting, agriculture, aquaculture, forestry and other non-farm activities, net of their respective costs of production
- Financial access measurement:
  - A household is considered to have financial access if any household member used any of the following services in the month before the survey: ATM cards, bank accounts, savings accounts, credit cards, life insurance, securities, and other insurance
  - If none of those are used, the household is assumed not to have financial access
- Digital access measurement:
  - A household is considered to have digital access if at least one member claims to have used the internet in the month prior to answering the VHLSS
- Reference for survey details: https://www.gso.gov.vn/en/data-and-statistics/2020/05/result-of-the-vietnam-household-living-standards-survey-2018/

*Source: wpiea2022185-print-pdf - References*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2022/english/wpiea2022185-print-pdf.pdf_