## wp17165

## Source details

**Canonical URL:** [wp17165](https://www.imf.org/-/media/files/publications/wp/2017/wp17165.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2017/wp17165.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2017/wp17165.pdf.json)

---

### Key findings
- Migrant remittances help smooth household consumption, with a significantly greater effect during fiscal consolidation episodes.
- Remittances have no impact on delinking changes in consumption from those in income during fiscal expansion episodes, even in high-remittance countries.
- For large fiscal shocks, remittances play a greater role in consumption smoothing during periods of significant fiscal adjustment.
- At the household level in Mexico, remittances contribute to higher consumption after controlling for household characteristics, and remittances and social assistance transfers provided by the government are substitutes.

### Quantitative context and scope
- Migrants account for over 3 percent of the world’s population.
- There are more than 250 million migrants globally.
- Workers’ remittances amount to over US$500 billion a year.
- In 2015, some 84 countries received migrant remittances equivalent to at least 1 percent of GDP, and 19 countries received 10 percent or more.
- Cross-country empirical analysis uses a panel of 149 countries over the period from 1990 to 2014.
- Household-level panel data for Mexico cover the years 2002, 2005-06 and 2009-12.

### Cross-country empirical results and interpretation
- Empirical strategy and main specification:
  - Idiosyncratic household consumption growth: Δc̃_it = Δc_it − Δc̄_t.
  - Idiosyncratic output growth: Δỹ_it = Δy_it − Δȳ_t.
  - Main econometric model:
    - Δc̃_it = β_0 + γ R_it + β_1 Δỹ_it + β_2 (R_it Δỹ_it) + γ X_it + β_3 (X_it Δỹ_it) + λ_t + α_i + ε_it
    - X_it includes controls such as trade and financial openness.
    - Interaction term R_it Δỹ_it (coefficient β_2) measures consumption smoothing facilitated by remittances; a negative β_2 denotes consumption smoothing.
  - Fiscal shocks defined as a change of at least 1.5 percentage points of GDP in the cyclically adjusted primary budget balance (robustness checked for 0.5, 1 and 2 percentage points). Cyclically adjusted primary balance computed using potential GDP estimated with HP filter and λ = 6.25.
  - Robust standard errors clustered at the country level.
- Summary empirical results:
  - Baseline (full sample and high-remittance subsample):
    - The estimated β_2 coefficient is negative and statistically significant for the full sample and for the high-remittance sub-sample.
    - Results robust to inclusion of control variables (trade and financial openness).
    - Interpretation: Workers’ remittances help smooth household consumption beyond standard channels of international risk-sharing.
  - Fiscal stance heterogeneity:
    - Fiscal consolidation episodes:
      - The consumption-smoothing effect of remittances is magnified during fiscal consolidation.
      - Effect is even greater in high-remittance countries.
    - Fiscal expansion episodes:
      - Remittances do not appear to smooth household consumption; β_2 is not statistically significant or robust for expansion periods.
    - Large fiscal shocks:
      - Remittances play a significant role in stabilizing household consumption during large fiscal shocks and large fiscal adjustments.

### Key statistics (cross-country dataset)
- Table 1. Summary Statistics (Full Sample)
  - Number of Observations:
    - Idiosyncratic consumption growth: 2,209
    - Idiosyncratic output growth: 2,209
    - Remittances/GDP: 2,209
    - Financial Openness: 2,084
    - Trade Openness: 2,115
  - Means:
    - Idiosyncratic consumption growth mean: 0.02
    - Idiosyncratic output growth mean: 0.01
    - Remittances/GDP mean: 0.04
    - Financial Openness mean: 0.49
    - Trade Openness mean: 0.81
  - Standard Deviations:
    - Idiosyncratic consumption growth std dev: 0.06
    - Idiosyncratic output growth std dev: 0.04
    - Remittances/GDP std dev: 0.06
    - Financial Openness std dev: 0.35
    - Trade Openness std dev: 0.40
  - Min / Max:
    - Idiosyncratic consumption growth: Min -0.28, Max 0.31
    - Idiosyncratic output growth: Min -0.18, Max 0.17
    - Remittances/GDP: Min 0.00, Max 0.49
    - Financial Openness: Min 0.00, Max 1.00
    - Trade Openness: Min 0.15, Max 3.74
- Table 2. Summary Statistics (High Remittance Countries)
  - Number of Observations:
    - Idiosyncratic consumption growth: 1,169
    - Idiosyncratic output growth: 1,169
    - Remittances/GDP: 1,169
    - Financial Openness: 1,117
    - Trade Openness: 1,111
  - Means:
    - Idiosyncratic consumption growth mean: 0.02
    - Idiosyncratic output growth mean: 0.01
    - Remittances/GDP mean: 0.07
    - Financial Openness mean: 0.44
    - Trade Openness mean: 0.78
  - Standard Deviations:
    - Idiosyncratic consumption growth std dev: 0.06
    - Idiosyncratic output growth std dev: 0.04
    - Remittances/GDP std dev: 0.06
    - Financial Openness std dev: 0.32
    - Trade Openness std dev: 0.31
  - Min / Max:
    - Idiosyncratic consumption growth: Min -0.28, Max 0.30
    - Idiosyncratic output growth: Min -0.18, Max 0.16
    - Remittances/GDP: Min 0.00, Max 0.49
    - Financial Openness: Min 0.00, Max 1.00
    - Trade Openness: Min 0.15, Max 2.11
- Fiscal episode counts:
  - Full sample:
    - Total episodes of fiscal consolidation: 1009 (437 categorized as fiscal shocks)
    - Total episodes of fiscal expansion: 1055 (458 categorized as fiscal shocks)
  - High-remittance countries:
    - Episodes of fiscal consolidation: 532 (232 categorized as fiscal shocks)
    - Episodes of fiscal expansion: 540 (224 categorized as fiscal shocks)

### Quantitative impacts and illustrative calculations (preserved)
- Baseline coefficient β_2 for the regression with the full sample: -2.56.
- Median remittances-to-GDP ratio in high-remittance countries: 4.9 percent.
- Using β_2 (-2.56) multiplied by 4.9 percent implies the country-specific consumption–output correlation is lower by 0.13 (from 0.87 to 0.74).
- Estimated mitigating effects during:
  - Fiscal consolidation: 0.21 (correlation declines from around 1.0 to almost 0.80).
  - Fiscal shocks: 0.22 (correlation declines from around 1.0 to almost 0.80).

### Robustness, dynamics, and heterogeneity
- Dynamic panel methods:
  - System GMM used (Arellano and Bover (1995); Blundell and Bond (1998)) with two sets of equations and instrumentation; three-year averages used to smooth fluctuations; Roodman (2009) strategy followed to limit instruments.
  - Validation via second-order serial correlation test and Hansen J-test.
- GMM / robustness checks summary:
  - Appendix Table 5 (Full Sample, GMM-style) examples:
    - All (column 1): ∆y 0.899*** (4.453); R * ∆y -1.992* (-1.714); Observations 812; Number of Countries 148; AR2 0.915; Hansen 0.254; Instruments 71
    - High_R (column 8): ∆y 1.312*** (3.177); R * ∆y -4.883* (-1.803); Observations 387; Countries 63; AR2 0.146; Hansen 0.114; Instruments 57
  - Appendix Table 6 (Fiscal Consolidation, GMM-style) examples:
    - All (column 1): ∆y 0.853*** (4.130); R * ∆y -2.947* (-1.798); Observations 366; Number of Countries 140; AR2 0.578; Hansen 0.0839; Instruments 89
    - High_R (column 5): ∆y 1.204*** (4.083); R * ∆y -5.650*** (-2.732); Observations 189; Countries 67; AR2 0.148; Hansen 0.640; Instruments 71
  - Appendix Table 7 (Fiscal Expansion, GMM-style) examples:
    - All (column 1): ∆y 0.820*** (4.621); R * ∆y -1.407 (-0.941); Observations 401; Number of Countries 141; AR2 0.738; Hansen 0.189; Instruments 89
  - Appendix Table 8 (Fiscal Shock, GMM-style) examples:
    - All (column 1): ∆y 1.074*** (5.230); R * ∆y -4.205* (-1.686); Observations 174; Number of Countries 91; AR2 0.387; Hansen 0.442; Instruments 71
    - High_R (column 5): ∆y 1.258*** (3.805); R * ∆y -6.769* (-1.697); Observations 75; Countries 41; AR2 0.376; Hansen 0.290; Instruments 29
- Regional and income-group heterogeneity:
  - Main findings hold broadly across regions and income categories, but relationship fails to hold in Africa.
  - Effects more significant for remittance recipients in Latin America and Caribbean, and in Asia.
  - Emerging economies, and especially lower income level countries, tend to benefit more from remittances in stabilizing consumption.

### Household-level empirical results — Mexico
- Rationale and key facts:
  - About 10 percent of Mexico’s population lives outside of Mexico.
  - Almost all (about 98 percent) of Mexican migrants reside in the United States.
  - Migrants send around US$25 billion (or over 2 percent of GDP) annually in remittances.
  - Workers’ remittances declined from a pre-crisis peak of 2.7 percent of GDP to 2.1 percent of GDP in 2015.
- Data and sample:
  - Panel of three survey waves covering 8,440 households over 2002-2012.
  - Balanced sample: 7,332 households responded in all three rounds and form the balanced dataset used.
  - Survey waves: 2002; 2005-2006; 2009-2012.
  - Dataset: Mexican Family Life Survey (MxFLS).
- Baseline household model:
  - Cijt = α + β1 R_iijt + β2 G_iijt + β5 X_iijt + μ_iijt + θ_iijt + ε_iijt
  - Dependent variable Cijt is log per capita consumption.
  - R_iijt is a binary indicator for receiving migrant remittances.
  - G_iijt is a binary indicator for receiving social assistance transfers from the government.
  - X_iijt includes household size, average years of education of adult household members, age of household head, and household net financial worth.
  - Household income excluded due to measurement error concerns; household wealth included instead; results robust to inclusion of household income.
- Classification and prevalence (averages across three survey rounds):
  - About 6 percent of households received remittances.
  - About 9 percent of households indicated receiving government support under one of the social assistance programs.
  - Of households receiving remittances, about 70 percent reported not receiving government support.
  - Of households receiving government support, about 90 percent reported not receiving remittances.
  - Overlap (households receiving both remittances and government support): about 1 percent on average.
- Descriptive expenditure and participation (survey rounds implicit: 2002 | 2005-06 | 2009-12):
  - Total Remittances Receiving Households: 508435394
    - (in percent of total households): 6.95.95.4
  - Remittance Receiving Households (no Government Support): 346407321
    - (in percent of total households): 4.75.64.4
  - Total Government Support Receiving Households: 1002303692
    - (in percent of total households): 13.74.19.4
  - Government Support Receiving Households (no Remittances): 908274619
    - (in percent of total households): 12.43.78.4
  - Remittance and Government Support Receiving Households: 942973
    - (in percent of total households): 1.30.41.0
  - Average Household Expenditure (in thousands of pesos)
    - Remittance Receiving Households: 60.668.062.7
    - Government Support Receiving Households: 34.246.455.9
    - Remittance and Government Support Receiving Households: 44.651.850.3
    - Household does not Receive Remittances or Government Support: 61.760.062.2
  - Mean household size across remittance and government-support receiving households: around 5 members.
- Household regression results (selected coefficients from Table 4 and Table 5):
  - Table 4 (benchmark specifications):
    - Remittances Receiving Household: 0.150*** (0.026) ; 0.143*** (0.027)
    - Government Support Receiving Household: 0.074*** (0.021) ; 0.073*** (0.021)
    - Number of Household Members: 0.071*** (0.006) ; 0.071*** (0.006) ; 0.063*** (0.006) ; 0.063*** (0.006)
    - [Unlabeled variable] 0.001*** (0.000) across columns
    - Age of Household Head: -0.004** (0.002) (in all columns)
    - Household Assets (Ln): 0.066*** (0.004) ; 0.066*** (0.004) ; 0.065*** (0.004) ; 0.065*** (0.004)
    - Constant terms: 8.579*** (0.189) ; 8.565*** (0.189) ; 8.634*** (0.324) ; 8.630*** (0.324)
    - Observations: 19,443 ; 19,274 ; 18,143 ; 18,079
    - Number of households: 7,289 ; 7,273 ; 7,228 ; 7,221
    - Adjusted R-squared: 0.113 ; 0.116 ; 0.108 ; 0.109
  - Table 5 (heterogeneous regressions):
    - Households without Government Support:
      - Remittances Receiving Household: 0.158*** (0.031)
      - Observations: 16,210; Adjusted R-squared: 0.108
    - Households with Government Support:
      - Remittances Receiving Household: 0.055 (0.119)
      - Observations: 1,869; Adjusted R-squared: 0.153
    - Households without Remittances:
      - Government Support Receiving Household: 0.071*** (0.023)
      - Observations: 16,895; Adjusted R-squared: 0.108
    - Households with Remittances:
      - Government Support Receiving Household: 0.137 (0.146)
      - Observations: 1,184; Adjusted R-squared: 0.214
- Household-level interpretation:
  - Main finding: Workers’ remittances and government social assistance programs raise consumption at the household level in Mexico.
  - Substitution result: Remittances and government social assistance transfers are substitutes:
    - For households without government support: remittances have a positive and significant effect on consumption.
    - For households with government support: remittances do not have a significant effect.
    - For households without remittances: government support has a positive and significant effect.
    - For households with remittances: government support effect on consumption is insignificant.

### Mechanisms and channels
- Remittances reduce dependence of consumption on domestic output fluctuations by providing income usable for ex-post insurance and potentially ex-ante investment in resilience.
- Remittances strengthen financial inclusion (e.g., increased likelihood of opening a bank account), facilitating access to saving and borrowing instruments that aid consumption smoothing.
- Stabilizing impact may vary with inflow size relative to GDP and comovement between migrants’ home and host country business cycles; remittances can be procyclical or countercyclical depending on these factors.

### Policy implications and recommendations
- Fiscal policymakers should consider the cushioning effects of migrant remittances on household consumption, particularly during large fiscal adjustments and in high-remittance countries.
- Fiscal consolidation may not immediately reduce household consumption in high-remittance countries if the pace and composition of adjustment account for household characteristics and include measures to protect vulnerable groups.
- Targeted social assistance programs (such as conditional cash transfers) remain important to raise consumption among the poorest households and incentivize health and education improvements, especially where remittances and government support are substitutes.
- Household-level evidence (Mexico) indicates remittances play a more significant role supporting consumption for households that do not receive government social assistance; policy design should recognize potential interactions between external private transfers and public safety nets.

### Appendix materials (selected)
- Appendix Figures 1–2: Consumption smoothing impact of remittances by region and by income grouping (labels preserved: All Africa Asia Europe LAC MENA; All EME LIC; Fiscal stance categories: Full Time Period, Fiscal Consolidation Periods, Fiscal Shock).
- Appendix Tables 1–4: OLS panel regressions for Full Sample, Fiscal Consolidation, Fiscal Expansion, Fiscal Shock (selected coefficients preserved in main text).
- Appendix Tables 5–8: System GMM robustness checks (selected statistics and tests preserved).
- Appendix Table 9: List of countries with high-remittance countries marked by an asterisk (*).
- Appendix Table 10: List of subsidies or government aid programs in MxFLS and participation by survey wave (2002, 2005-06, 2009-12); program presence entries preserved as Yes/No.

*Source: IMF working paper wp17165.*

### REFERENCES .............................................................................................................

### wp17165 - REFERENCES .............................................................................................................

### Key findings
- Migrant remittances help smooth household consumption, with a significantly greater effect during fiscal consolidation episodes.
- Remittances have no impact on delinking changes in consumption from those in income during fiscal expansion episodes, even in high-remittance countries.
- For large fiscal shocks, remittances play a greater role in consumption smoothing during periods of significant fiscal adjustment.
- At the household level in Mexico, remittances contribute to higher consumption after controlling for household characteristics, and remittances and social assistance transfers provided by the government are substitutes.

### Quantitative context and scope
- Migrants account for over 3 percent of the world’s population.
- There are more than 250 million migrants globally.
- Workers’ remittances amount to over US$500 billion a year.
- In 2015, some 84 countries received migrant remittances equivalent to at least 1 percent of GDP, and 19 countries received 10 percent or more.
- Cross-country empirical analysis uses a panel of 149 countries over the period from 1990 to 2014.
- Household-level panel data for Mexico cover the years 2002, 2005-06 and 2009-12.

### Cross-country empirical results and interpretation
- Using fixed effects and dynamic panel models, workers’ remittances help smooth household consumption even after controlling for standard risk-sharing channels such as trade and financial openness.
- The consumption-smoothing effect of remittances is significantly stronger during fiscal consolidation episodes, where remittances act as a cushion against shocks, especially in high-remittance countries.
- During fiscal expansion episodes, remittances do not contribute to delinking consumption changes from income changes.
- Results are robust when focusing on episodes of large fiscal shocks, showing a greater role for remittances in consumption smoothing during significant fiscal adjustment.

### Household-level (Mexico) empirical results
- Remittance receipt increases household consumption in Mexico, conditional on a wide set of household controls.
- Remittances and government social assistance transfers operate as substitutes: remittances play a more important role in supporting consumption for Mexican households that do not receive government financial support or social assistance.

### Mechanisms and channels
- Remittances reduce dependence of consumption on domestic output fluctuations by providing income that can be used to smooth consumption ex-post (insurance) and potentially ex-ante (investment in resilience).
- Remittances strengthen financial inclusion (e.g., increased likelihood of opening a bank account), which can facilitate consumption smoothing by improving access to saving and borrowing instruments.
- The stabilizing impact of remittances may vary with the size of inflows relative to GDP and with comovement between migrants’ home and host country business cycles; remittances can be procyclical or countercyclical depending on these factors.

### Policy implications
- Fiscal policy makers should consider the cushioning effects of migrant remittances on household consumption, particularly during large fiscal adjustments and in high-remittance countries.
- Fiscal consolidation may not immediately reduce household consumption in high-remittance countries if the pace and composition of adjustment account for household characteristics and include measures to protect vulnerable groups.
- Targeted social assistance programs (such as conditional cash transfers) remain important to raise consumption among the poorest households and incentivize health and education improvements, especially where remittances and government support are substitutes.

### Relation to existing literature
- Builds on literature finding remittances are relatively stable compared to other capital flows and can reduce output growth volatility and consumption volatility in recipient countries.
- Extends prior work by explicitly testing whether the consumption-smoothing role of remittances varies with fiscal policy stance (consolidation vs. expansion).
- Notes mixed prior evidence: some studies find remittances reduce consumption volatility, while others find no role; this paper finds a conditional role that depends on fiscal episodes and remittance intensity.

### Data and empirical strategy overview
- Cross-country analysis: static and dynamic panel models for 149 countries, 1990-2014, controlling for trade and financial openness and other standard risk-sharing channels.
- Country case study: panel household survey data from Mexico for 2002, 2005-06 and 2009-12, assessing remittances’ effect on household consumption and interaction with government support.

*Source: IMF working paper wp17165.*

### 2014. The consumption smoothing impact of remittances is further investigated and contrasted

### wp17165 - 2014. The consumption smoothing impact of remittances is further investigated and contrasted

### Objective and empirical approach
- Purpose:
  - Investigate the impact of migrant remittances (as a share of GDP, R_it) on the co-movements between output and household consumption, and test whether remittances delink household consumption from output growth.
  - Confirm macro results using household survey data from Mexico (reduced-form model) (description of the Mexico analysis is mentioned but details are not included in the supplied excerpt).
- Baseline empirical specification:
  - Idiosyncratic household consumption growth: Δc̃_it = Δc_it − Δc̄_t.  (Equation (1))
  - Idiosyncratic output growth: Δỹ_it = Δy_it − Δȳ_t.  (Equation (2))
  - Main econometric model (Equation (3)):
    - Δc̃_it = β_0 + γ R_it + β_1 Δỹ_it + β_2 (R_it Δỹ_it) + γ X_it + β_3 (X_it Δỹ_it) + λ_t + α_i + ε_it
    - X_it are control variables (including trade and financial openness).
    - The interaction term R_it Δỹ_it (coefficient β_2) measures the extent of consumption smoothing facilitated by remittances.
    - A negative β_2 denotes a consumption smoothing effect of remittances.
  - Robust standard errors clustered at the country level.
- Fiscal episodes and heterogeneity:
  - Estimate Equation (3) separately for periods of fiscal consolidation and fiscal expansion to assess substitutability/complementarity between fiscal policy and remittances.
  - Fiscal shocks defined as a change of at least 1.5 percentage points of GDP in the cyclically adjusted primary budget balance in either direction. (Robustness noted for thresholds of 0.5, 1 and 2 percentage points.)
  - Cyclically adjusted primary balance computed using potential GDP estimated with the Hodrick-Prescott (HP) filter and λ = 6.25.

### Data and sample
- Period covered: 1990-2014.
- Countries:
  - Full sample comprises a broad set of countries (list referenced as Appendix Table 9; not supplied here).
  - High-remittance countries: 69 countries defined as having remittances greater than the median level (1.5 percent of GDP) during 1990-2014.
- Definitions and sources:
  - Workers’ remittances: inflows of in-cash or in-kind current transfers between resident and nonresident households (personal transfers in the secondary income account from the IMF’s Balance of Payments Statistics based on BPM6).
  - Compensation of employees excluded from remittances measure.
  - Data sources: United Nations’ National Accounts Main Aggregates Database (GDP, consumption, population), World Bank’s World Development Indicators (global per capita GDP, trade openness), Chinn-Ito (2006) Index (de jure financial openness), IMF’s Balance of Payments Statistics (remittances), IMF’s Public Finances in Modern History Database and country IMF reports (fiscal data).

### Key dataset summary statistics (preserved exactly as in source)
- Table 1. Summary Statistics (Full Sample)
  - Number of Observations:
    - Idiosyncratic consumption growth: 2,209
    - Idiosyncratic output growth: 2,209
    - Remittances/GDP: 2,209
    - Financial Openness: 2,084
    - Trade Openness: 2,115
  - Means:
    - Idiosyncratic consumption growth mean: 0.02
    - Idiosyncratic output growth mean: 0.01
    - Remittances/GDP mean: 0.04
    - Financial Openness mean: 0.49
    - Trade Openness mean: 0.81
  - Standard Deviations:
    - Idiosyncratic consumption growth std dev: 0.06
    - Idiosyncratic output growth std dev: 0.04
    - Remittances/GDP std dev: 0.06
    - Financial Openness std dev: 0.35
    - Trade Openness std dev: 0.40
  - Min / Max:
    - Idiosyncratic consumption growth: Min -0.28, Max 0.31
    - Idiosyncratic output growth: Min -0.18, Max 0.17
    - Remittances/GDP: Min 0.00, Max 0.49
    - Financial Openness: Min 0.00, Max 1.00
    - Trade Openness: Min 0.15, Max 3.74

- Table 2. Summary Statistics (High Remittance Countries)
  - Number of Observations:
    - Idiosyncratic consumption growth: 1,169
    - Idiosyncratic output growth: 1,169
    - Remittances/GDP: 1,169
    - Financial Openness: 1,117
    - Trade Openness: 1,111
  - Means:
    - Idiosyncratic consumption growth mean: 0.02
    - Idiosyncratic output growth mean: 0.01
    - Remittances/GDP mean: 0.07
    - Financial Openness mean: 0.44
    - Trade Openness mean: 0.78
  - Standard Deviations:
    - Idiosyncratic consumption growth std dev: 0.06
    - Idiosyncratic output growth std dev: 0.04
    - Remittances/GDP std dev: 0.06
    - Financial Openness std dev: 0.32
    - Trade Openness std dev: 0.31
  - Min / Max:
    - Idiosyncratic consumption growth: Min -0.28, Max 0.30
    - Idiosyncratic output growth: Min -0.18, Max 0.16
    - Remittances/GDP: Min 0.00, Max 0.49
    - Financial Openness: Min 0.00, Max 1.00
    - Trade Openness: Min 0.15, Max 2.11

- Fiscal episode counts (based on cyclically adjusted primary budget balance scaled by potential GDP):
  - Full sample:
    - Total episodes of fiscal consolidation: 1009 (437 categorized as fiscal shocks)
    - Total episodes of fiscal expansion: 1055 (458 categorized as fiscal shocks)
  - High-remittance countries:
    - Episodes of fiscal consolidation: 532 (232 categorized as fiscal shocks)
    - Episodes of fiscal expansion: 540 (224 categorized as fiscal shocks)

### Empirical findings (preserved terminology and numerical descriptors)
- Baseline (full sample and high-remittance subsample):
  - The estimated β_2 coefficient (interaction of remittances and idiosyncratic output growth) is negative and statistically significant for the full sample and for the high-remittance sub-sample.
  - These results are robust to inclusion of control variables (trade and financial openness).
  - Interpretation: Workers’ remittances help smooth household consumption beyond standard channels of international risk-sharing (trade and capital account openness). Mechanisms suggested include intertemporal savings, better access to liquidity, and financial services facilitated by remittances, especially during economic difficulty.

- Fiscal stance heterogeneity:
  - Fiscal consolidation episodes:
    - The consumption-smoothing effect of remittances is magnified during fiscal consolidation.
    - Comparison of β_2 across baseline and consolidation regressions indicates remittances are more effective in smoothing household consumption during fiscal austerity.
    - Effect is even greater in high-remittance countries (comparison of β_2 coefficients across columns in Appendix Table 2; specific coefficient values not supplied in the excerpt).
    - Mechanisms: remittances help compensate for lower social transfers and permit intertemporal smoothing via savings or higher consumption of remittances; more developed financial systems in remittance-recipient economies improve access to finance and smoothing.
  - Fiscal expansion episodes:
    - Remittances do not appear to smooth household consumption during periods of fiscal expansion.
    - β_2 is not statistically significant or robust across specifications for expansion periods, including for high-remittance countries (Appendix Table 3).
    - Conclusion: remittances do not help delink consumption from income during fiscal expansion.
  - Large fiscal shocks:
    - Remittances tend to play a significant role in stabilizing household consumption during large fiscal shocks and large fiscal adjustments (Appendix Table 4 and narrative summary; specific coefficient values not supplied in the excerpt).

### Interpretation and implications (as presented in the source)
- Remittances provide a countercyclical, welfare-enhancing buffer:
  - Particularly effective during fiscal consolidation and large fiscal shocks when public transfers and consumption tend to fall.
  - Effect is stronger in countries that rely heavily on remittance inflows (high-remittance countries).
- Remittances do not substitute for consumption smoothing during fiscal expansions.
- The consumption-smoothing role of remittances appears to operate beyond standard international risk-sharing channels (trade and financial openness), suggesting importance of remittance-specific channels (liquidity, savings, access to finance).

*Source: Authors’ calculations (wp17165 - 2014. The consumption smoothing impact of remittances is further investigated and contrasted).*

### Appendix Tables 1-4. It visualizes the consumption-smoothing impact of remittances for the full

### Appendix Tables 1-4. It visualizes the consumption-smoothing impact of remittances for the full sample of countries, as well as separately for high-remittance countries, during different phases of the fiscal policy stance categorized as fiscal consolidations, fiscal expansions, and large fiscal shocks.

### Consumption-smoothing findings — aggregate
- Remittances have a significant consumption-smoothing effect, especially in high-remittance countries.
- The smoothing effect is highly significant during periods of fiscal consolidation; remittances do not have a significant role in smoothing household consumption during periods of fiscal expansion.
- The consumption-smoothing impact is asymmetric for periods of fiscal shock: it is amplified during contractionary periods compared with expansionary periods (expanded by more than 1.5 percent of GDP). These results are available upon request.
- Differences in coefficients for the full sample versus fiscal consolidation episodes are statistically significant; this is also true for the high-remittance country subsample. (Chow (1960) test applied.)

### Quantitative impacts and illustrative calculations
- Baseline coefficient 훽2 for the regression with the full sample: -2.56.
- Median remittances-to-GDP ratio in high-remittance countries: 4.9 percent.
- Using the baseline 훽2 (-2.56) multiplied by 4.9 percent implies the country-specific consumption–output correlation is lower by 0.13 (from 0.87 to 0.74).
- Estimated mitigating effects during:
  - Fiscal consolidation: 0.21 (correlation declines from around 1.0 to almost 0.80).
  - Fiscal shocks: 0.22 (correlation declines from around 1.0 to almost 0.80).
- High-remittance country definition: remittance inflows greater than the median 1.5 percent of GDP during the 1990-2014 period.
- Figure 4 note: The bar chart shows the consumption smoothing impact of remittances (훽2) estimated by Equation (3). The symbols *, **, and *** indicate statistical significance at 10, 5, and 1 percent levels, respectively.

### Methodology and robustness checks
- Dynamic panel data estimation used to address endogeneity and reverse causality.
- System GMM technique employed as proposed by Arellano and Bover (1995) and Blundell and Bond (1998).
- System GMM specifics:
  - Two sets of equations: first differences instrumented by lags of levels; levels instrumented by lags of first differences.
  - Use of three-year averages to smooth short-run fluctuations and reduce the number of time periods so that the number of countries exceeds the number of time periods in all regressions (including subsamples).
  - Strategy suggested by Roodman (2009) followed to address weak and excessively numerous instruments.
  - Validation via second-order serial correlation test for residuals and Hansen J-test for overidentifying restrictions.
- System GMM results (Appendix Tables 5-8) confirm that migrant remittances help smooth household consumption, with effects more pronounced during fiscal consolidation and large fiscal shocks, especially in countries more dependent on remittances.
- Caveat: using a large number of instruments weakens the Hansen J-test, while overly restricting the instrument set can lead to imprecise coefficients.

### Regional and income heterogeneity
- Main findings hold broadly across regional and income categories.
- Appendix Figure 1: baseline 훽2 by regional category.
- Appendix Figure 2: baseline 훽2 by income grouping.
- Heterogeneity:
  - Relationship fails to hold in Africa.
  - More significant for remittance recipients in Latin America and Caribbean, and in Asia.
  - Emerging economies, and especially lower income level countries, tend to benefit more from remittances in stabilizing consumption.

### Household-level analysis — Mexico
- Rationale: high emigration and reliance on migrant remittances.
- Key Mexico facts:
  - About 10 percent of Mexico’s population lives outside of Mexico.
  - Almost all (about 98 percent) of Mexican migrants reside in the United States.
  - Migrants send around US$25 billion (or over 2 percent of GDP) annually in remittances.
  - Workers’ remittances declined from a pre-crisis peak of 2.7 percent of GDP to 2.1 percent of GDP in 2015.
- Household panel:
  - Panel of three survey waves covering 8,440 households over 2002-2012.
  - Balanced sample: 7,332 households responded in all three rounds and form the balanced dataset used.
  - Survey waves: 2002; 2005-2006; 2009-2012.
- Baseline household model (Equation (4)): Cijt = α + 훽1 R_iijt + 훽2 G_iijt + 훽5 X_iijt + μ_iijt + θ_iijt + ε_iijt, where:
  - Cijt is log per capita consumption for household i in location j at period t.
  - R_iijt is a binary indicator for receiving migrant remittances.
  - G_iijt is a binary indicator for receiving social assistance transfers from the government.
  - μ_iijt are household-time fixed effects; θ_iijt are location–time fixed effects.
  - X_iijt includes household size, average years of education of adult household members, age of the household head, and household net financial worth.
  - Household income not included directly due to measurement error concerns; household wealth included instead. Results robust to inclusion of household income.
- Classification and prevalence:
  - Households classified as remittance-receiving if at least one household member received monetary transfers from a family member living in the United States during the last year (following Ambrosius (2015)).
  - On average across the three survey rounds, about 6 percent of households received remittances.
  - On average across the three survey rounds, about 9 percent of households indicated receiving government support under one of the social assistance programs.
  - Of households receiving remittances, about 70 percent reported not receiving government support.
  - Of households receiving government support, about 90 percent reported not receiving remittances.
  - Overlap (households receiving both remittances and government support): about 1 percent on average.
  - These patterns suggest remittances and government support may be substitutes.
- Data source: Mexican Family Life Survey (MxFLS), longitudinal panel representative at national, urban, rural, and regional levels; detailed migration and monetary transfer information.

*Source: Authors’ calculations.*

### Appendix Table 10 provides a list of all the programs covered in each round of household survey.

### wp17165 - Appendix Table 10 provides a list of all the programs covered in each round of household survey.

### Data and measurement
- Dataset: MxFLS (Mexican Family Life Survey).
- Age classification: The MxFLS dataset classifies those aged 15 and older as adults.
- Consumption measurement: The length of the period over which consumption patterns are measured varies by goods, ranging from the past 7 days to the past 12 months. Consumption for each good is annualized and summed to obtain annualized consumption in a given survey round.
- Household net worth: Calculated as the total value of all reported household assets including the reported value of housing.
- Note on household categories: Households are categorized by receipt of remittances and/or government support.

### Descriptive statistics (MxFLS household categories and expenditure)
- Survey rounds (column headings implicitly): 2002 | 2005-06 | 2009-12
- Total Remittances Receiving Households: 508435394
  - (in percent of total households): 6.95.95.4
- Remittance Receiving Households (no Government Support): 346407321
  - (in percent of total households): 4.75.64.4
- Total Government Support Receiving Households: 1002303692
  - (in percent of total households): 13.74.19.4
- Government Support Receiving Households (no Remittances): 908274619
  - (in percent of total households): 12.43.78.4
- Remittance and Government Support Receiving Households: 942973
  - (in percent of total households): 1.30.41.0
- Average Household Expenditure (in thousands of pesos)
  - Remittance Receiving Households: 60.668.062.7
  - Government Support Receiving Households: 34.246.455.9
  - Remittance and Government Support Receiving Households: 44.651.850.3
  - Household does not Receive Remittances or Government Support: 61.760.062.2
- Socio-demographic notes:
  - Mean household size across remittance and government-support receiving households: around 5 members (stable across survey rounds).
  - Households that receive remittances have a larger share of adult household members with secondary or higher education.
  - Average age of household head generally consistent across groups; higher for government-support households in the third wave, possibly associated with inclusion of the 70 y más program.

### Empirical results — household-level analysis (Mexico)
- Main finding: Workers’ remittances and government social assistance programs raise consumption at the household level in Mexico; effects are statistically significant when included separately or jointly in baseline regressions.
- Substitution result: Workers’ remittances and social assistance transfers provided by the Mexican government are substitutes.
  - For households without government support:
    - Remittances have a positive and significant effect on household consumption.
  - For households with government support:
    - Remittances do not have a significant effect on household consumption.
  - For households without remittances:
    - Government support has a positive and significant effect on household consumption.
  - For households with remittances:
    - Government support effect on household consumption is insignificant.
- Interpretation: When households receive no government transfers, remittances help support consumption; where government transfers are present, remittances play a diminished role in affecting consumption.

### Key regression estimates (Table 4: Determinants of Household Consumption in Mexico)
- Dependent variable: log per capita household consumption. Robust standard errors in parentheses. Estimates based on equation (4).
- Selected coefficient estimates (benchmark specifications):
  - Remittances Receiving Household: 0.150*** (0.026) ; 0.143*** (0.027)
  - Government Support Receiving Household: 0.074*** (0.021) ; 0.073*** (0.021)
  - Number of Household Members: 0.071*** (0.006) ; 0.071*** (0.006) ; 0.063*** (0.006) ; 0.063*** (0.006)
  - [Unlabeled variable shown as] 0.001*** (0.000) across columns
  - Age of Household Head: -0.004** (0.002) (in all columns)
  - Household Assets (Ln): 0.066*** (0.004) ; 0.066*** (0.004) ; 0.065*** (0.004) ; 0.065*** (0.004)
  - Constant terms: 8.579*** (0.189) ; 8.565*** (0.189) ; 8.634*** (0.324) ; 8.630*** (0.324)
- Sample sizes and fit:
  - Observations: 19,443 ; 19,274 ; 18,143 ; 18,079
  - Number of households: 7,289 ; 7,273 ; 7,228 ; 7,221
  - Adjusted R-squared: 0.113 ; 0.116 ; 0.108 ; 0.109
- Significance: *** p<0.01, ** p<0.05, * p<0.1

### Heterogeneous regressions (Table 5: Role of Government Support and Remittances)
- Note: Dependent variable and estimation framework as in Table 4.
- Selected coefficient estimates across subgroups:
  - Column (1) Households without Government Support:
    - Remittances Receiving Household: 0.158*** (0.031)
    - Government Support Receiving Household: 0.071*** (0.023)
    - Number of Household Members: 0.068*** (0.007)
    - [Unlabeled variable] 0.002*** (0.000)
    - Age of Household Head: -0.004** (0.002)
    - Household Assets (Ln): 0.063*** (0.005)
    - Constant: 8.671*** (0.318)
    - Observations: 16,210
    - Number of folio: 7,110
    - Adjusted R-squared: 0.108
  - Column (2) Households with Government Support:
    - Remittances Receiving Household: 0.055 (0.119)
    - Government Support Receiving Household: 0.137 (0.146)
    - Number of Household Members: 0.069** (0.030)
    - [Unlabeled variable] -0.001 (0.003)
    - Age of Household Head: -0.024 (0.016)
    - Household Assets (Ln): 0.042* (0.023)
    - Constant: 11.070*** (0.846)
    - Observations: 1,869
    - Number of folio: 1,509
    - Adjusted R-squared: 0.153
  - Column (3) Households without Remittances:
    - Remittances Receiving Household: (not applicable / not significant)
    - Government Support Receiving Household: 0.071*** (0.023)
    - Number of Household Members: 0.068*** (0.007)
    - [Unlabeled variable] 0.001*** (0.000)
    - Age of Household Head: -0.004** (0.002)
    - Household Assets (Ln): 0.063*** (0.005)
    - Constant: 8.767*** (0.326)
    - Observations: 16,895
    - Number of folio: 7,139
    - Adjusted R-squared: 0.108
  - Column (4) Households with Remittances:
    - Remittances Receiving Household: (reported as) 0.055 (0.119) in column (2) context; government support coefficient reported as 0.137 (0.146) when relevant
    - Number of Household Members: -0.023 (0.038)
    - [Unlabeled variable] 0.004 (0.004)
    - Age of Household Head: -0.022* (0.011)
    - Household Assets (Ln): 0.058* (0.030)
    - Constant: 11.352*** (0.842)
    - Observations: 1,184
    - Number of folio: 1,013
    - Adjusted R-squared: 0.214
- Significance: *** p<0.01, ** p<0.05, * p<0.1

### Cross-country findings (summary)
- Sample: Panel of 149 countries over 1990 to 2014.
- Main cross-country results:
  - Workers’ remittances help smooth household consumption, controlling for trade and financial openness.
  - Consumption-smoothing effect is significantly greater during fiscal consolidation episodes.
  - Remittances provide a cushion particularly in high-remittance countries and during large fiscal adjustments.
  - During fiscal expansion episodes, remittances show no significant impact on delinking changes in household consumption from income, even in high-remittance countries.

### Policy implications and recommendations
- Policymakers should account for the stabilizing role of migrant remittances when designing fiscal adjustments.
- For high-remittance countries:
  - Fiscal consolidation episodes may not necessarily have an immediate adverse effect on household consumption if the pace and composition of fiscal adjustment consider household characteristics and include measures to protect the most vulnerable.
  - Social assistance programs (such as targeted cash transfers) can be used to raise household consumption among the poorest households and to incentivize outcomes in health and education.
- Household-level evidence (Mexico) indicates:
  - Remittances play a more significant role in supporting consumption for households that do not receive government social assistance.
  - Social assistance and remittances can act as substitutes; policy design should recognize potential interactions between external private transfers and public safety nets.

*Source: MxFLS; Authors' calculations (excerpts from IMF Working Paper wp17165).*

### Appendix Figure 1: Consumption Smoothing Impact of Remittances by Region

### Appendix Figure 1: Consumption Smoothing Impact of Remittances by Region

### Consumption smoothing by region and income group (charts)
- Bar charts show the consumption smoothing impact of remittances (훽훽2) estimated by Equation (3) for each regional category and by fiscal stance.
- Separate charts display:
  - Consumption Smoothing Effect by Region: All, Africa, Asia, Europe, LAC, MENA across "Full Time Period", "Fiscal Consolidation Periods", "Fiscal Shock".
  - Consumption Smoothing Effect by Region (High Remit. Countries): same regional categories for high-remittance countries (those with remittance inflows greater than the median 1.5 percent of GDP during the 1990-2014 period).
- Appendix Figure 2 notes: charts by income grouping (All, EME, LIC) and by fiscal stance; EME refers to emerging markets and LIC stands for low income countries.
- Negative 훽훽2 indicates that remittances help smooth consumption by delinking correlation between country-specific growth in consumption and output.

### Key chart annotations (preserved labels)
- Region axis labels: All Africa Asia Europe LAC MENA
- Income group labels: All EME LIC
- Fiscal stance categories used in charts: Full Time Period, Fiscal Consolidation Periods, Fiscal Shock
- High remittance threshold: median 1.5 percent of GDP during the 1990-2014 period

---

### Regression results — Appendix Table 1: Full Sample Period (1990 - 2014)
Note: Dependent variable is idiosyncratic growth in consumption; estimates based on equation (3) including country and time fixed effects. Robust standard errors in brackets. High_R denotes high-remittance countries (remittances > median 1.5 percent of GDP).

- All (column 1):
  - ∆y 0.847*** [0.058]
  - R 0.090* [0.048]
  - R * ∆y -2.361*** [0.827]
  - Observations 2,209
  - R-squared 0.205
  - Number of Countries 149
- All (other columns shown for robustness; selected entries):
  - Column 2: ∆y 0.874*** [0.091]; R 0.077 [0.050]; R * ∆y -2.177** [0.869]; Observations 2,084; R-squared 0.205; Countries 139
  - Column 3: ∆y 1.143*** [0.108]; R 0.057 [0.054]; R * ∆y -2.222** [0.926]; Observations 2,115; R-squared 0.213; Countries 144
  - Column 4: ∆y 1.184*** [0.129]; R 0.040 [0.055]; R * ∆y -2.080** [0.930]; Observations 2,003; R-squared 0.217; Countries 135
- High_R (selected columns):
  - Column 5: ∆y 0.869*** [0.101]; R 0.096* [0.050]; R * ∆y -2.558** [0.990]; Observations 1,169; R-squared 0.198; Countries 69
  - Column 8: ∆y 1.300*** [0.173]; R 0.009 [0.063]; R * ∆y -0.990 [1.056]; Observations 1,072; R-squared 0.217; Countries 63

- Additional covariates reported (selected):
  - "Fin. Openness" coefficients and standard errors where included (e.g., 0.009 [0.008], 0.007 [0.008], -0.002 [0.009], 0.001 [0.010])
  - "Trade Openness" where included (e.g., 0.023 [0.014], 0.021 [0.013]; in High_R columns 0.045** [0.019], 0.035** [0.016])

---

### Regression results — Appendix Table 2: Fiscal Consolidation (1990 - 2014)
Note: Sample restricted to periods of fiscal consolidation (narrowing of the cyclically adjusted primary budget balance). Robust standard errors in brackets.

- All (column 1):
  - ∆y 0.860*** [0.089]
  - R 0.190** [0.082]
  - R * ∆y -2.926** [1.234]
  - Observations 1,009
  - R-squared 0.246
  - Number of Countries 145
- All (column 4):
  - ∆y 1.009*** [0.166]
  - R 0.179* [0.096]
  - R * ∆y -3.946*** [1.178]
  - Observations 915
  - R-squared 0.279
  - Countries 131
- High_R (column 5):
  - ∆y 1.001*** [0.164]
  - R 0.192** [0.094]
  - R * ∆y -4.206*** [1.388]
  - Observations 532
  - R-squared 0.232
  - Countries 69
- High_R (column 8):
  - ∆y 1.000*** [0.315]
  - R 0.162 [0.106]
  - R * ∆y -4.181*** [1.147]
  - Observations 486
  - R-squared 0.247
  - Countries 63

- Fin. Openness (where reported): 0.030** [0.012] and 0.031** [0.012] in some columns.

---

### Regression results — Appendix Table 3: Fiscal Expansion (1990 - 2014)
Note: Sample restricted to periods of fiscal expansion (widening of the cyclically adjusted primary budget balance). Robust standard errors in brackets.

- All (column 1):
  - ∆y 0.846*** [0.075]
  - R -0.015 [0.083]
  - R * ∆y -1.917 [1.240]
  - Observations 1,055
  - R-squared 0.218
  - Number of Countries 146
- All (column 4):
  - ∆y 1.071*** [0.196]
  - R -0.014 [0.082]
  - R * ∆y -1.165 [1.138]
  - Observations 969
  - R-squared 0.215
  - Countries 134
- High_R (column 5):
  - ∆y 0.754*** [0.123]
  - R -0.021 [0.097]
  - R * ∆y -1.392 [1.578]
  - Observations 540
  - R-squared 0.232
  - Countries 69
- High_R (column 8):
  - ∆y 1.227*** [0.226]
  - R -0.058 [0.092]
  - R * ∆y 0.480 [1.327]
  - Observations 503
  - R-squared 0.256
  - Countries 63

- Fin. Openness * ∆y significant negative in some columns (e.g., -0.358** [0.175]).

---

### Regression results — Appendix Table 4: Fiscal Shock (1990 - 2014)
Note: Sample restricted to periods of fiscal shocks (change of at least 1.5 percentage points of GDP in the cyclically adjusted primary budget balance). Robust standard errors in brackets.

- All (column 1):
  - ∆y 0.895*** [0.102]
  - R 0.034 [0.080]
  - R * ∆y -3.521** [1.731]
  - Observations 895
  - R-squared 0.229
  - Number of Countries 144
- All (column 4):
  - ∆y 1.290*** [0.241]
  - R 0.027 [0.071]
  - R * ∆y -3.765** [1.636]
  - Observations 791
  - R-squared 0.255
  - Countries 129
- High_R (column 5):
  - ∆y 1.024*** [0.152]
  - R -0.025 [0.094]
  - R * ∆y -4.484** [1.992]
  - Observations 456
  - R-squared 0.287
  - Countries 69
- High_R (column 8):
  - ∆y 1.446*** [0.272]
  - R -0.057 [0.077]
  - R * ∆y -2.656 [1.777]
  - Observations 404
  - R-squared 0.300
  - Countries 62

- Trade Openness often positive and significant in shock subsample (e.g., 0.068*** [0.022] in some High_R columns).

---

### GMM / Robustness checks — Appendix Tables 5–8 (1990 - 2014)
Note: Z-statistics in parentheses; dependent variable is idiosyncratic growth in consumption; estimates based on equation (3) including country and time fixed effects. AR2, Hansen, Instruments reported.

Selected results (All / High_R focus on R * ∆y):

- Appendix Table 5: Full Sample (GMM-style)
  - All (column 1): ∆y 0.899*** (4.453); R * ∆y -1.992* (-1.714); Observations 812; Number of Countries 148; AR2 0.915; Hansen 0.254; Instruments 71
  - High_R (column 8): ∆y 1.312*** (3.177); R * ∆y -4.883* (-1.803); Observations 387; Countries 63; AR2 0.146; Hansen 0.114; Instruments 57

- Appendix Table 6: Fiscal Consolidation (GMM-style)
  - All (column 1): ∆y 0.853*** (4.130); R * ∆y -2.947* (-1.798); Observations 366; Number of Countries 140; AR2 0.578; Hansen 0.0839; Instruments 89
  - High_R (column 5): ∆y 1.204*** (4.083); R * ∆y -5.650*** (-2.732); Observations 189; Countries 67; AR2 0.148; Hansen 0.640; Instruments 71

- Appendix Table 7: Fiscal Expansion (GMM-style)
  - All (column 1): ∆y 0.820*** (4.621); R * ∆y -1.407 (-0.941); Observations 401; Number of Countries 141; AR2 0.738; Hansen 0.189; Instruments 89
  - High_R (column 8): ∆y 1.165*** (3.674); R * ∆y -4.075 (-1.510); Observations 193; Countries 61; AR2 0.210; Hansen 0.239; Instruments 57

- Appendix Table 8: Fiscal Shock (GMM-style)
  - All (column 1): ∆y 1.074*** (5.230); R * ∆y -4.205* (-1.686); Observations 174; Number of Countries 91; AR2 0.387; Hansen 0.442; Instruments 71
  - High_R (column 5): ∆y 1.258*** (3.805); R * ∆y -6.769* (-1.697); Observations 75; Countries 41; AR2 0.376; Hansen 0.290; Instruments 29

---

### Interpretation from regression summaries (preserved statements)
- A negative 훽훽2 (R * ∆y) indicates that remittances help smooth consumption by delinking correlation between country-specific growth in consumption and output.
- High_R refers to countries with remittance inflows greater than the median 1.5 percent of GDP during the 1990-2014 period.
- R denotes remittances as percent of GDP; ∆y is the country-specific output growth.
- "Fin. Openness" is the normalized index of financial openness obtained from Chinn-Ito (2006).
- "Trade Openness" is defined as the sum of exports and imports in GDP derived from the World Bank’s World Development Indicator Database.
- *, **, and *** denote statistical significance at 10, 5, and 1 percent levels, respectively.

---

### Appendix Table 9: List of Countries (high remittance countries distinguished by *)
- Full country list provided; high remittance countries are marked with an asterisk (*).
- Examples of entries (preserving exact formatting and stars):
  - Afghanistan
  - Albania*
  - Algeria
  - Antigua and Barbuda
  - Argentina
  - Armenia*
  - Austria
  - Azerbaijan*
  - Bangladesh*
  - Barbados
  - Belarus
  - Belgium
  - Belize*
  - Benin*
  - Bhutan
  - Bolivia*
  - Bosnia and Herzegovina*
  - Botswana
  - Brazil
  - Bulgaria*
  - Burkina Faso*
  - Burundi*
  - Cambodia*
  - Cameroon
  - Cape Verde*
  - Chad
  - China
  - Colombia*
  - Comoros*
  - Congo, Republic of
  - Costa Rica
  - Croatia
  - Cyprus
  - Czech Republic
  - Côte d'Ivoire
  - Djibouti
  - Dominica*
  - Dominican Republic*
  - Ecuador*
  - Egypt*
  - El Salvador*
  - Estonia
  - Ethiopia
  - FYR Macedonia*
  - Fiji*
  - France
  - Gabon
  - Gambia, The*
  - Georgia*
  - Germany
  - Ghana
  - Greece
  - Grenada*
  - Guatemala*
  - Guinea
  - Guinea-Bissau*
  - Guyana*
  - Haiti*
  - Honduras*
  - Hungary
  - Iceland
  - India*
  - Indonesia
  - Iraq
  - Ireland
  - Italy
  - Jamaica*
  - Japan
  - Kazakhstan
  - Kenya*
  - Kiribati*
  - Korea
  - Kosovo*
  - Kyrgyz Republic*
  - Lao P.D.R.
  - Latvia
  - Lebanon*
  - Liberia*
  - Libya
  - Lithuania
  - Luxembourg
  - Macao SAR, China
  - Madagascar*
  - Malawi
  - Mali*
  - Malta
  - Marshall Islands*
  - Mauritania
  - Mexico*
  - Micronesia, Fed. States of *
  - Moldova*
  - Mongolia*
  - Montenegro, Rep. of*
  - Morocco*
  - Mozambique
  - Myanmar
  - Namibia
  - Nepal*
  - Netherlands
  - New Zealand
  - Nicaragua*
  - Niger
  - Norway
  - Oman
  - Pakistan*
  - Palau
  - Panama
  - Papua New Guinea
  - Paraguay*
  - Peru
  - Philippines*
  - Poland
  - Portugal
  - Qatar
  - Russia
  - Rwanda
  - Samoa*
  - Senegal*
  - Serbia*
  - Seychelles
  - Sierra Leone*
  - Slovak Republic
  - Slovenia
  - Solomon Islands*
  - Spain
  - Sri Lanka*
  - St. Kitts and Nevis
  - St. Vincent and the Grenadines*
  - Suriname
  - Sweden
  - Switzerland
  - Syria*
  - Tajikistan*
  - Tanzania
  - Thailand
  - Togo*
  - Tonga*
  - Trinidad and Tobago
  - Tunisia*
  - Turkey
  - Tuvalu*
  - Uganda*
  - Ukraine*
  - United Kingdom
  - Uruguay
  - Vanuatu*
  - Venezuela
  - Yemen*
  - Zambia

---

### Appendix Table 10: Subsidies or Government Aid Programs in MxFLS (program participation by survey wave)
- Columns correspond to survey waves: 2002, 2005-06, 2009-12 (preserved labels).
- Programs and participation (Yes/No) entries preserved exactly as shown:
  - Oportunidades (earlier Progresa) 1/ : Yes Yes Yes
    - Note: 1/ Included in the survey, but results not published.
  - PROCAMPO: Yes Yes Yes
  - VivAh (programa Ahorro y Subsidio a la Vivienda Progresiva): Yes Yes Yes
  - Crédito a la Palabra: Yes Yes No
  - Programa de Coinversión Social: Yes Yes Yes
  - Programa de Empleo Temporal: Yes Yes Yes
  - Programa Alianza para el Campo: Yes Yes Yes
  - Fondo de Apoyo para la Micro, Pequeña y Mediana Empresa: Yes Yes No
  - Fondo Nacional de Apoyo para las Empresas de Solidaridad: Yes Yes Yes
  - 70 y más: No No Yes
  - Apoyo alimentario: No No Yes
  - Opciones productivas: No No Yes
  - Other: Yes Yes Yes

*Source: "wp17165 - Appendix Figure 1: Consumption Smoothing Impact of Remittances by Region" (PDF chapter).

---


_Source: https://www.imf.org/-/media/files/publications/wp/2017/wp17165.pdf_