## wpiea2022092-print-pdf

## Source details

**Canonical URL:** [wpiea2022092-print-pdf](https://www.imf.org/-/media/files/publications/wp/2022/english/wpiea2022092-print-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2022/english/wpiea2022092-print-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2022/english/wpiea2022092-print-pdf.pdf.json)

---

### Major findings on remittances and COVID-19 impact
- Remittances represent a very important source of income for most Central American countries and Mexico; Mexico receives the most remittances (US$ 34.6 billion in 2019) and Guatemala in CAPDR (US$10.5 billion in 2019).
- Scaled over the economy, El Salvador and Honduras remittances represented over 20 percent of GDP in 2019.
- One out of six families in the Dominican Republic, Honduras, and El Salvador rely on remittances as the primary source of income (Keller and Rousse, 2016).
- At the pandemic onset, the World Bank forecasted a 19 percent drop in remittances to Latin American and Caribbean region for 2020; a Rapid Financing Instrument staff report for the Dominican Republic forecasted a 14 percent decline in remittances for 2020.
- Despite a severe pandemic shock, remittances to CAPDR and Mexico were resilient: an initial large contraction in spring 2020 was short-lived, followed by an unprecedented fast rebound such that by end-2020 remittance flows surpassed pre-pandemic levels in many countries; 2021 saw even more elevated levels and pace of growth than pre-pandemic.

### Drivers and mechanisms identified
- Two principal narratives supported by empirical analysis: the “altruism” motive (helping recipients at home) and a change in the mode of transfers (shift from informal to formal/digital channels).
- Kpodar et al. (2021) find remittances responded positively to COVID-19 cases in recipient countries and provide mixed evidence on the role of fiscal support in host countries.
- Lockdowns enforced changes from informal to formal/digital remittance channels (see Dinarte et al., 2021; Frizancho and Parrado, 2021).
- The paper documents an unambiguous effect of U.S. unemployment relief and additional U.S. fiscal support in explaining the increase in the average amount of remittances in the second half of 2020:
  - The average amount remitted increased primarily due to U.S. states unemployment relief, including for COVID-19.
  - Average amounts were higher coming from U.S. states that had additional state relief extended to undocumented migrants and went to Salvadoran departments experiencing the highest increases in poverty, COVID-19 infections, and declines in mobility.
  - Remittance growth peaked in months where U.S. fiscal stimulus checks were received in 2020-21.
  - In the second half of 2021, sustained increases in the average amount remitted are associated with a pick-up in infection cases and additional U.S. fiscal support (tax child credits during July-December 2021) and U.S. labor market developments, especially in migrant-intensive sectors.

### Stylized facts: migration and remittances patterns
- Migration stocks and remittances:
  - 11 million people of Mexican origin live in the U.S.
  - About 1 million from each of the Northern Triangle countries (El Salvador, Guatemala and Honduras) live in the U.S.
  - Over a fifth of Salvadorans, and slightly over a tenth of people from the Dominican Republic live in the U.S.
- Remittances scale and sources:
  - Remittances to CAPDR and Mexico come mostly from the U.S., ranging from 67 percent in Nicaragua to close to 95 percent in Mexico.
  - Aggregate remittances to each country have steadily increased; short-term developments are highly correlated with U.S. labor market developments.
  - At the height of the U.S. lockdown in April 2020, the U.S. Hispanic unemployment rate increased to over 15 percentage points and remittances collapsed by more than a third.
- Regional comparisons:
  - Aggregate remittances to CAPDR and Mexico were as resilient as in other regions in 2020 but recorded some of the highest growth rates in the world in 2021 (e.g., Guatemala remittances grew by 37 percent in 2021).
  - The strong continued growth above pre-pandemic levels is specific to the 2020 recession when compared to 2002 and 2009 U.S. recessions.

### Empirical modeling and traditional determinants
- Data and methods:
  - Panel of five CAPDR countries, Mexico and the U.S., January 2000–December 2020, monthly frequency.
  - Remittances in constant 2000 U.S. dollars; all series in year-over-year growth rates.
  - Explanatory variables included: U.S. Hispanic unemployment; U.S. new housing permits; monthly index of economic activity (IMAE) in manufacturing in recipient countries; interest rate differential; inflation differential; REER.
  - Estimation approaches: fixed effects panel and panel vector autoregression (Abrigo and Love, 2016) with generalized method of moments and Cholesky decomposition (lag order p set at 1).
- Fixed effects results:
  - A decrease in U.S. Hispanic unemployment is significantly associated with an increase in remittances and is the highest coefficient explaining remittances dynamics.
  - Higher real depreciation of the recipient country currency (decline in relative purchasing power) is associated with remittances growth.
  - An improvement in recipient-country IMAE manufacturing and lower inflation were associated with increased remittances in this specification, but authors note these results are likely affected by endogeneity.
- Panel VAR results:
  - Over twenty-one years of monthly data, U.S. Hispanic unemployment explains most of the variation in remittances (7 percent after 3 months, and 95 percent after ten months).
  - Manufacturing activity in the home country explains 4 percent of remittance variation historically.
  - New housing permits explain remittance variations to a much lower extent.
  - The traditional model explained the decline in March–April 2020 and the incipient recovery in May 2020 relatively well.
  - However, traditional variables fail to capture the magnitude of the recovery in remittances in the second half of 2020; they explain only about half of the aggregate remittances recovery observed in CAPDR in that period.

### Decomposition: volume versus average amount
- Aggregate remittances decomposed into number of transactions and average amount remitted (CAPDR central banks data, monthly):
  - Pre-pandemic upward trend in aggregate remittances is explained entirely by continuous increases in the volume of transactions (migration-driven), while average amount remitted was relatively flat.
  - The precipitous drop in spring 2020 was primarily due to a drop in the number of transactions.
  - The rebound since July 2020 was primarily due to increases in the average amount remitted.
- Quantified contributions:
  - By December 2020, peak remittances growth to CAPDR was 28 percent, of which 14 percentage points was due to an increase in the average value of remittances.
  - Prior to the pandemic, increases in average amounts remitted were non-existent until the second half of 2019 when increases were at most 5 percent.
- Drivers of volume changes:
  - The increase in volume of transactions historically is explained by increases in the stock of migrants over time; a pre-pandemic regression of transactions on migrant stock had R² = 0.64 and coefficient significant at the 1 percent level (example: Transactions = -107.5 + 2.2*Migrant_Population).
  - In 2020, pandemic-induced travel restrictions (March–September) reduced inflows of migrants, so the 2020 increase in transactions seems driven by a larger share of existing migrants in the U.S. remitting (including shifts from in-person cash remittances to tracked digital transfers), rather than by new migrants.
- Costs and channel changes:
  - The cost of transferring remittances is about five percent (Bersch and others, 2021).
  - Shift from informal/in-person transfers to formal/digital channels during lockdowns likely increased measurable transactions and raised average remitted values as travel/gift expenditures moved into digital remittances.

### Box 1 — Factors Affecting Change in the Volume of Remittances Transactions during COVID
- Changes in modes of transfer and transaction volumes:
  - The share of registered in-person cash remittance transfers has declined by about 5 percent for the region.
  - Transfers through financial institutions increased, with cross-country variation:
    - Dominican Republic and Honduras: about 10 percent switch away from in-person cash.
    - Guatemala: about 3 percent switch.
    - El Salvador: negligible change in share of in-person cash.
  - El Salvador (monthly index, 2019=100): since May 2020 electronic transfers rebounded, with growth in bank transfer mode remittances more than double growth in other modes.
  - Fintech remittances (example: Xoom) accelerated rapidly during the pandemic, but bank transfers and Fintech remittances together make up less than 1/3 of total remittances, so their overall contribution to change in total remittances was marginal.
- Increase in average remittance value: correlations and explanatory power:
  - Text Table 2 (2011 January - 2019 December) estimates for El Salvador (all variables in natural log form):
    - Constant: 7.01*** (0.30)
    - El Salvador real wage: -0.79*** (0.13)
    - US real wage (construction): 1.42*** (0.37)
    - No. obs.: 108
    - R-squared: 0.39
  - Income factors alone explain only 39 percent of the change in the average remittances over two decades (2011-19), leaving open what explains increases in average amounts remitted, especially in 2020.

### Empirical strategy and estimated determinants (El Salvador corridors)
- Data and variables:
  - Monthly data on average amount remitted for 314 corridors: from 23 U.S. states to 14 El Salvador departments, January 2017 to December 2020.
  - Average amount per corridor = total remittances from a U.S. state to a Salvadoran department divided by number of transactions.
  - Main explanatory variables (natural logarithms, seasonally adjusted) included: ln(RR)S,D,t; ln(W)S,t; ln(UI)S,t; ln(M)S,t; ln(COVID)D,t; ln(Transfers)D,t; uS; vD.
- Estimation:
  - Feasible Generalized Least Squares to address serial autocorrelation; addressed multicollinearity with alternative variable sets; results robust to variable selection and to inclusion/exclusion of unemployment relief, Salvadoran fiscal support.
  - Incorporating the April 2020 U.S. stimulus check was not significant.
- Estimated coefficients (Text Table 3: 2017:1—2020:12):
  - Constant: 0.741*** (0.232)
  - US: Real wage: 0.673*** (0.039)
  - US: Unemployment insurance per unemployed: 0.012*** (0.001)
  - US: State mobility: 0.158*** (0.012)
  - SLV: New COVID cases: 0.037*** (0.003)
  - SLV: Government transfers per household: -0.014*** (0.003)
  - Number of observations: 15072
  - Statistical significance: * p<0.10, ** p<0.05, *** p<0.01

### Interpretation, heterogeneity, and 2021 persistence
- Interpretation of empirical results:
  - The coefficient on U.S. state real wage has the largest magnitude: average remittance values are highly sensitive to migrants’ labor income.
  - Unemployment relief and U.S. state mobility are statistically significant and important in explaining remittance dynamics.
  - Number of new COVID-19 cases in Salvadoran departments and public transfers to Salvadoran households are significant contributors.
  - The model fits observed average remittance values well, especially after June 2020; unexplained component declined since May 2020, reaching almost zero in three of the seven subsequent months.
- Heterogeneity across Salvadoran departments and U.S. states:
  - Home factors: growth in average remittances was highest in the poorest and most mobility-restrained Salvadoran departments with high COVID-19 cases (example: San Vicente had the highest increase in extreme poverty (~8 percentage points), lowest mobility, and received one of the highest increases in average remittances (9 percent)).
  - Host factors: high growth in average remittances observed from U.S. states with robust growth in real wages and unemployment insurance claims; California noted as an example where state-level relief extended to workers regardless of legal status likely increased remittances beyond model predictions.
- 2021 developments and persistence of drivers:
  - Remittances to El Salvador increased 32 percent in 2021 relative to 2019; more than half of the increment attributed to an increase in the average amount remitted.
  - Contributing 2021 factors (many temporary/one-off):
    - Continued U.S. real wage growth, though moderating and starting to decline by year-end.
    - Continued unemployment relief support that largely expired by year-end 2021.
    - COVID-19 infections at home abated with vaccinations, though seasonal rises occurred July–September.
    - Additional fiscal support (CARES Acts stimulus checks and later American Rescue Plan tax child credits) provided further temporary boosts.
    - Employment gains in sectors where CAPDR migrants work: by end-2021 U.S. total non-farm employment gained about five percent, but accommodation (leisure and hospitality) gained 18 percent; accommodation sector average weekly earnings increased with many months above 10 percent in 2021.
  - These host-labor-market developments and one-off fiscal supports explain the extraordinary growth in remittances observed in 2021.

### Policy-relevant implications and outlook
- Short-run:
  - U.S. fiscal support and unemployment relief played a material role in supporting the average value of remittances during 2020–21; timing of stimulus payments correlates with peaks in remittance growth.
  - State-level relief policies that included undocumented migrants were associated with higher average remittances originating from those states, targeting recipients experiencing larger COVID-19 impacts and poverty increases.
- Medium-to-long run:
  - Remittances growth is expected to moderate but remain on an upward trend driven by expected increases in migration and transaction volumes.
  - Further increases in the average amount remitted depend on U.S. labor market developments (real wages and employment gains), especially in sectors where CAPDR and Mexican migrants predominantly work.
  - Over time, reductions in the cost of transferring remittances (financial innovations) could contribute to growth in the average amount remitted.
  - Remittances may spike during recipient-country hardship due to altruism; short-term growth will depend on U.S. sectoral labor developments, especially employment and real wage gains in accommodations.
  - Continued shifts to digital payment modes may increase remittances by lowering transfer costs and capturing migrants without banking services.

*Source: wpiea2022092-print-pdf - REFERENCES _____________________________________________________________ 29*

### REFERENCES _____________________________________________________________ 29

### wpiea2022092-print-pdf - REFERENCES _____________________________________________________________ 29

### Major findings on remittances and COVID-19 impact
- Remittances represent a very important source of income for most Central American countries and Mexico; Mexico receives the most remittances (US$ 34.6 billion in 2019) and Guatemala in CAPDR (US$10.5 billion in 2019).
- Scaled over the economy, El Salvador and Honduras remittances represented over 20 percent of GDP in 2019.
- One out of six families in the Dominican Republic, Honduras, and El Salvador rely on remittances as the primary source of income (Keller and Rousse, 2016).
- At the pandemic onset, the World Bank forecasted a 19 percent drop in remittances to Latin American and Caribbean region for 2020; a Rapid Financing Instrument staff report for the Dominican Republic forecasted a 14 percent decline in remittances for 2020.
- Despite a severe pandemic shock, remittances to CAPDR and Mexico were resilient: an initial large contraction in spring 2020 was short-lived, followed by an unprecedented fast rebound such that by end-2020 remittance flows surpassed pre-pandemic levels in many countries; 2021 saw even more elevated levels and pace of growth than pre-pandemic.

### Drivers and mechanisms identified
- Two principal narratives supported by empirical analysis: the “altruism” motive (helping recipients at home) and a change in the mode of transfers (shift from informal to formal/digital channels).
- Kpodar et al. (2021) find remittances responded positively to COVID-19 cases in recipient countries and provide mixed evidence on the role of fiscal support in host countries.
- Lockdowns enforced changes from informal to formal/digital remittance channels (see Dinarte et al., 2021; Frizancho and Parrado, 2021).
- The paper documents an unambiguous effect of U.S. unemployment relief and additional U.S. fiscal support in explaining the increase in the average amount of remittances in the second half of 2020.
  - The average amount remitted increased primarily due to U.S. states unemployment relief, including for COVID-19.
  - Average amounts were higher coming from U.S. states that had additional state relief extended to undocumented migrants and went to Salvadoran departments experiencing the highest increases in poverty, COVID-19 infections, and declines in mobility.
  - Remittance growth peaked in months where U.S. fiscal stimulus checks were received in 2020-21.
  - In the second half of 2021, sustained increases in the average amount remitted are associated with a pick-up in infection cases and additional U.S. fiscal support (tax child credits during July-December 2021) and U.S. labor market developments, especially in migrant-intensive sectors.

### Stylized facts: migration and remittances patterns
- Migration stocks and remittances:
  - 11 million people of Mexican origin live in the U.S.
  - About 1 million from each of the Northern Triangle countries (El Salvador, Guatemala and Honduras) live in the U.S.
  - Over a fifth of Salvadorans, and slightly over a tenth of people from the Dominican Republic live in the U.S.
- Remittances scale:
  - Remittances to CAPDR and Mexico come mostly from the U.S., ranging from 67 percent in Nicaragua to close to 95 percent in Mexico.
  - Aggregate remittances to each country have steadily increased; short-term developments are highly correlated with U.S. labor market developments.
  - At the height of the U.S. lockdown in April 2020, the U.S. Hispanic unemployment rate increased to over 15 percentage points and remittances collapsed by more than a third.
- Regional comparisons:
  - Aggregate remittances to CAPDR and Mexico were as resilient as in other regions in 2020 but recorded some of the highest growth rates in the world in 2021 (e.g., Guatemala remittances grew by 37 percent in 2021).
  - The strong continued growth above pre-pandemic levels is specific to the 2020 recession when compared to 2002 and 2009 U.S. recessions.

### Empirical modeling and traditional determinants
- Data and methods:
  - Panel of five CAPDR countries, Mexico and the U.S., January 2000–December 2020, monthly frequency.
  - Remittances in constant 2000 U.S. dollars; all series in year-over-year growth rates.
  - Explanatory variables included: U.S. Hispanic unemployment; U.S. new housing permits; monthly index of economic activity (IMAE) in manufacturing in recipient countries; interest rate differential; inflation differential; REER.
  - Estimation approaches: fixed effects panel and panel vector autoregression (Abrigo and Love, 2016) with generalized method of moments and Cholesky decomposition (lag order p set at 1).
- Fixed effects results:
  - A decrease in U.S. Hispanic unemployment is significantly associated with an increase in remittances and is the highest coefficient explaining remittances dynamics.
  - Higher real depreciation of the recipient country currency (decline in relative purchasing power) is associated with remittances growth.
  - An improvement in recipient-country IMAE manufacturing and lower inflation were associated with increased remittances in this specification, but authors note these results are likely affected by endogeneity.
- Panel VAR results:
  - Over twenty-one years of monthly data, U.S. Hispanic unemployment explains most of the variation in remittances (7 percent after 3 months, and 95 percent after ten months).
  - Manufacturing activity in the home country explains 4 percent of remittance variation historically.
  - New housing permits explain remittance variations to a much lower extent.
  - The traditional model explained the decline in March–April 2020 and the incipient recovery in May 2020 relatively well.
  - However, traditional variables fail to capture the magnitude of the recovery in remittances in the second half of 2020; they explain only about half of the aggregate remittances recovery observed in CAPDR in that period.

### Decomposition: volume versus average amount
- Aggregate remittances decomposed into number of transactions and average amount remitted (CAPDR central banks data, monthly):
  - Pre-pandemic upward trend in aggregate remittances is explained entirely by continuous increases in the volume of transactions (migration-driven), while average amount remitted was relatively flat.
  - The precipitous drop in spring 2020 was primarily due to a drop in the number of transactions.
  - The rebound since July 2020 was primarily due to increases in the average amount remitted.
- Quantified contributions:
  - By December 2020, peak remittances growth to CAPDR was 28 percent, of which 14 percentage points was due to an increase in the average value of remittances.
  - Prior to the pandemic, increases in average amounts remitted were non-existent until the second half of 2019 when increases were at most 5 percent.
- Drivers of volume changes:
  - The increase in volume of transactions historically is explained by increases in the stock of migrants over time; a pre-pandemic regression of transactions on migrant stock had R² = 0.64 and coefficient significant at the 1 percent level (example: Transactions = -107.5 + 2.2*Migrant_Population).
  - In 2020, pandemic-induced travel restrictions (March–September) reduced inflows of migrants, so the 2020 increase in transactions seems driven by a larger share of existing migrants in the U.S. remitting (including shifts from in-person cash remittances to tracked digital transfers), rather than by new migrants.
- Costs and channel changes:
  - The cost of transferring remittances is about five percent (Bersch and others, 2021).
  - Shift from informal/in-person transfers to formal/digital channels during lockdowns likely increased measurable transactions and raised average remitted values as travel/gift expenditures moved into digital remittances.

### Policy-relevant implications and outlook
- Short-run:
  - U.S. fiscal support and unemployment relief played a material role in supporting the average value of remittances during 2020–21; timing of stimulus payments correlates with peaks in remittance growth.
  - State-level relief policies that included undocumented migrants were associated with higher average remittances originating from those states, targeting recipients experiencing larger COVID-19 impacts and poverty increases.
- Medium-to-long run:
  - Remittances growth is expected to moderate but remain on an upward trend driven by expected increases in migration and transaction volumes.
  - Further increases in the average amount remitted depend on U.S. labor market developments (real wages and employment gains), especially in sectors where CAPDR and Mexican migrants predominantly work.
  - Over time, reductions in the cost of transferring remittances (financial innovations) could contribute to growth in the average amount remitted.

*Source: wpiea2022092-print-pdf - REFERENCES _____________________________________________________________ 29*

### Box 1. Factors Affecting Change in the Volume of Remittances Transactions during COVID

### Box 1. Factors Affecting Change in the Volume of Remittances Transactions during COVID

### Changes in modes of transfer and transaction volumes
- The share of registered in-person cash remittance transfers has declined by about 5 percent for the region.
- Transfers through financial institutions increased, with considerable cross-country variation:
  - Dominican Republic and Honduras: about 10 percent switch away from in-person cash.
  - Guatemala: about 3 percent switch.
  - El Salvador: negligible change in share of in-person cash.
- El Salvador (monthly index, 2019=100): since May 2020 electronic transfers rebounded, with growth in bank transfer mode remittances more than double growth in other modes.
- Fintech remittances (example: Xoom) accelerated rapidly during the pandemic, but:
  - Bank transfers and Fintech remittances together make up less than 1/3 of total remittances, so their overall contribution to change in total remittances was marginal.

### Increase in average remittance value: correlations and explanatory power
- Cross-country correlations and regression evidence:
  - Text Table 2 (2011 January - 2019 December) estimates for El Salvador (all variables in natural log form):
    - Constant: 7.01*** (0.30)
    - El Salvador real wage: -0.79*** (0.13)
    - US real wage (construction): 1.42*** (0.37)
    - No. obs.: 108
    - R-squared: 0.39
  - Income factors alone explain only 39 percent of the change in the average remittances over two decades (2011-19), leaving open what explains increases in average amounts remitted, especially in 2020.
- Pre-pandemic average remitted amounts were relatively constant; pandemic-period increases required additional explanation.

### Empirical strategy: model, data, and variables (El Salvador corridors)
- Data scope:
  - Monthly data on average amount remitted for 314 corridors: from 23 U.S. states to 14 El Salvador departments, January 2017 to December 2020.
  - Average amount per corridor = total remittances from a U.S. state to a Salvadoran department divided by number of transactions.
- Extended “altruism” model predicting higher average remittances when host-country income is higher and home-country hardship is greater.
- Main explanatory variables (natural logarithms, seasonally adjusted):
  - ln(RR)S,D,t: average amount of remittance (adjusted for El Salvador inflation)
  - ln(W)S,t: average weekly real wage in residential construction and landscape sectors in U.S. state S
  - ln(UI)S,t: total unemployment relief benefit (regular + pandemic) divided by stock of unemployed in U.S. state S
  - ln(M)S,t: motor vehicle travel miles in U.S. state S (mobility proxy)
  - ln(COVID)D,t: new COVID-19 cases per 1 million population in Salvadoran department D
  - ln(Transfers)D,t: public transfers to private sector (inflation-adjusted) divided by number of households
  - uS: U.S. state fixed effect; vD: Salvadoran department fixed effect
- Additional controls and data sources:
  - U.S. state real wage and unemployment data from U.S. Bureau of Labor Statistics.
  - Mobility from U.S. federal highway administration (Google Community Mobility correlated closely for Feb–Dec 2020).
  - COVID-19 cases from El Salvador Ministry of Health.
  - Transfers from Ministry of Finance of El Salvador.
- Estimation method and robustness:
  - Feasible Generalized Least Squares to address serial autocorrelation.
  - Addressed multicollinearity by estimating alternative variable sets; results robust to variable selection.
  - Results robust to inclusion/exclusion of unemployment relief, Salvadoran fiscal support, and to alternative wage/unemployment measures. Incorporating the April 2020 U.S. stimulus check was not significant.

### Estimated determinants and magnitudes (Text Table 3: 2017:1—2020:12)
- Estimated coefficients (standard errors in parentheses):
  - Constant: 0.741*** (0.232)
  - US: Real wage: 0.673*** (0.039)
  - US: Unemployment insurance per unemployed: 0.012*** (0.001)
  - US: State mobility: 0.158*** (0.012)
  - SLV: New COVID cases: 0.037*** (0.003)
  - SLV: Government transfers per household: -0.014*** (0.003)
- Number of observations: 15072
- Statistical significance: * p<0.10, ** p<0.05, *** p<0.01

### Interpretation of empirical results and drivers in 2020
- The coefficient on U.S. state real wage has the largest magnitude: average remittance values are highly sensitive to migrants’ labor income.
- Unemployment relief and U.S. state mobility are statistically significant and important in explaining remittance dynamics.
- Number of new COVID-19 cases in Salvadoran departments and public transfers to Salvadoran households are significant contributors.
- The model fits observed average remittance values well, especially after June 2020; unexplained component declined since May 2020, reaching almost zero in three of the seven subsequent months.
- Contributions to the increase in average remittances in 2020 compared to 2019:
  - Predominantly driven by U.S. real wage growth (notably in late 2020).
  - U.S. COVID-19 unemployment relief (state-level regular + federal pandemic relief) also contributed; effect smaller as real wages recovered.
  - Altruism/home hardships: higher infection rates in El Salvador raised average remittances on a relatively constant contribution (higher during July–August 2020 peak).
- Factors that would have reduced average remittances in 2020:
  - Restricted mobility in U.S. states (limited job opportunities or ability to send cash).
  - Salvadoran government one-off cash transfer (US$300 per household) could reduce need for remittances.

### Heterogeneity across Salvadoran departments and U.S. states
- Home factors:
  - Growth in average remittances was highest in the poorest and most mobility-restrained Salvadoran departments with high COVID-19 cases.
  - Example: San Vicente had the highest increase in extreme poverty (~8 percentage points), lowest mobility, and received one of the highest increases in average remittances (9 percent).
  - Other poor departments with mobility constraints and infections (Morazán, Usulután) also saw larger increases.
- Host factors:
  - High growth in average remittances observed from U.S. states with robust growth in real wages and unemployment insurance claims.
  - States above the model-predicted slope (Text Table 3 estimates) exhibited higher-than-predicted increases in average remittances; California noted as an example where state-level relief extended to workers regardless of legal status likely increased remittances beyond model predictions.
- Unexplained components across states attributed partly to additional state-specific relief programs and other policy differences.

### 2021 developments and persistence of drivers
- Remittances to El Salvador increased 32 percent in 2021 relative to 2019; more than half of the increment attributed to an increase in the average amount remitted.
- Contributing 2021 factors (many temporary/one-off):
  - Continued U.S. real wage growth, though moderating and starting to decline by year-end.
  - Continued unemployment relief support that largely expired by year-end 2021.
  - COVID-19 infections at home abated with vaccinations, though seasonal rises occurred July–September.
  - Additional fiscal support (CARES Acts stimulus checks and later American Rescue Plan tax child credits) provided further temporary boosts.
  - Employment gains in sectors where CAPDR migrants work: by end-2021 U.S. total non-farm employment gained about five percent, but accommodation (leisure and hospitality) gained 18 percent; accommodation sector average weekly earnings increased with many months above 10 percent in 2021.
- These host-labor-market developments and one-off fiscal supports explain the extraordinary growth in remittances observed in 2021.

### Conclusions and outlook
- Remittances to CAPDR and Mexico are well explained by an “altruism” model: remittances rise with improved host-country labor income and with deteriorating recipient-country conditions.
- Over time the number of transactions trended up (migration increase), while average amount remitted historically remained relatively constant but is highly elastic to U.S. real wage changes.
- During the pandemic total remittances until June 2020 were largely explained by number of transactions; from the second half of 2020 onward the increase was equally due to more transactions and higher average amounts remitted.
- The increase in average amounts remitted in the second half of 2020 is primarily explained by U.S. real wage growth, U.S. unemployment relief, and pandemic developments in Salvadoran departments; additional U.S. fiscal stimulus and state-level supports also contributed.
- Looking forward:
  - Remittances growth to CAPDR and Mexico is expected to moderate.
  - Longer-term growth will be driven primarily by an increase in transaction volumes as migration continues.
  - Average remitted amounts will grow with U.S. real wages.
  - Remittances may spike during recipient-country hardship due to altruism.
  - Short-term remittances growth will depend on U.S. sectoral labor developments, especially employment and real wage gains in accommodations.
  - Continued shifts to digital payment modes may increase remittances by lowering transfer costs and capturing migrants without banking services.

*Sources: National authorities, IPUMS USA, Haver Analytics, Google Community Mobility, U.S. Bureau of Labor Statistics, U.S. federal highway administration, El Salvador Ministry of Health, Ministry of Finance of El Salvador, and authors’ calculations.*

### Annex I.  Alternative Measure of Economic

### Annex I. Alternative Measure of Economic Conditions in the Recipient Country: the IMAE Agriculture

### Annex II. CARES Act’s Unemployment Insurance Support
- The CARES Act, passed in March 2020, introduced over US$3.4 trillion in federal spending to bolster state-administered unemployment programs, support to businesses through low interest, and Paycheck Protection Loans (PPP Loans).
- The CARES Act and its followed iterations (Consolidated Appropriations Acts, CARES Act II), along with the American Rescue Plan introduced in 2021 sustained many households with stimulus checks, and businesses.
- The combined federal spending allowed for US$1.52 trillion unemployment benefits and employee retention (US$590 billion and US$930 billion allocations, respectively). They expired in November 2021.
- Expanded eligibility:
  - Track I: eligible recipients of traditional unemployment insurance (UI) benefits (used existing State Administered UI programs, with usually coverage of 26 weeks on average depending on the state). After exhausting regular UI benefit, eligible for additional (non-overlapping) benefits for a total duration of additional 53-60 weeks (depending on the state), via:
    - Pandemic Emergency Unemployment Compensation (PEUC)
    - Extended Benefit
    - Pandemic Unemployment Insurance (PUA)
  - Track II: non-eligible recipients of traditional UI benefits (self-employed, part-time, insufficient work history) — able to participate in the PUA program only for a duration up to 39 weeks.
- Supplementary programs included Federal Pandemic Unemployment Compensation (FPUC), Mixed Earners Unemployment Compensation (MEUC), Trade Readjustment Allowances (TRA), and additional state benefits.
- Aggregate outcome: 5 out of 6 workers received benefits in excess of their previous earnings — for the median worker benefits amount to 134 percent of earnings.

### Annex III. U.S. State-Run Support Programs (Undocumented and Excluded Workers)
- Federal COVID-19 relief packages apply to documented workers only; some state and local jurisdictions extended benefits by occupation or special allocations to reach undocumented workers. At least 12 states extended coverage.
- California:
  - Program estimated to have reached 150,000 undocumented workers via direct one-time cash payments.
  - Disaster Relief Assistance for Immigrants Fund (DRAI): one-time cash bonus of US$500 capped at US$1,000 per household.
  - California legislature designated US$75 million; private donations added US$50 million.
- New York City:
  - Allocation of US$2.1 billion for excluded workers.
  - One-time payments of up to US$15,600 per household.
  - Eligibility required: lived in the state prior to March of 2020, resided in the state at time of application, excluded from other unemployment or COVID-19 income related benefits, gross annual income below US$26,208, and loss of 50 percent or more of this income.
- Washington, DC:
  - US$5 million fund allocated to support undocumented workers via local non-profit organizations, in addition to the US$15 million program for documented workers.
  - One-time benefits of US$1,000 per family distributed through US$500 pre-loaded debit cards.
- Maryland (Montgomery County):
  - Single individuals could receive as much as US$500.
  - Families: up to US$1,000 for a family with one child, and US$150 for each additional child, with a maximum benefit of US$1,450.
  - Program ran between April 2020 and June 30, 2021.
- Other non-status-restricted programs:
  - California: US$100 million rental assistance subsidy and US$50 million for small business relief (rental assistance estimated to help 50,000 households).
  - Chicago, Illinois: US$2 million fund for housing assistance and US$100 million fund for forgivable small business loans; accessible regardless of citizenship status.

### Annex IV. Remittances Data per Corridor
- Data sample representative for aggregate remittances; used data from two major remittances operators which account for 17 percent of transactions and 14 percent of value of remittances in 2019.
- For El Salvador:
  - Most growth in remittances was due to increase in the average amount of remittances.
  - Growth in the average value of remittances per transaction in the two-operator sample tracks well the growth rates for the country (all operators).
- Graphical indicators (described):
  - El Salvador: Growth in Average Remittances — Percent change (Year-over-year) shows monthly variation (JAN...DEC) with series for Two operators and Aggregate (all operators).
  - El Salvador: Growth in Remittances and Contributions — Annual percent change and percentage points (Relative to 2019) decomposed into Change in transactions, Change in the average amount, Total change in remittances.

### Annex V. Correlation Matrix for Model Explaining of Average Amount Remitted (Selected numeric correlations)
- Sample period 2017:1--2020:12 (correlations shown):
  - US: Average remittances — correlation with US: Real wage = -0.01221
  - US: Average remittances — correlation with US: Unemployment insurance per unemployed = 0.09610.30491 (matrix layout preserved as in source)
  - US: Average remittances — correlation with US: State mobility = 0.090.05-0.16911
  - US: Average remittances — correlation with SLV: New COVID cases = 0.18230.19930.6413-0.04011
  - US: Average remittances — correlation with SLV: Government transfers per household = -0.00660.12010.4105-0.08810.2081
- Subperiods:
  - 2017:1--2019:12 (selected):
    - US: Average remittances — correlation with US: Real wage = -0.03941
    - US: Average remittances — correlation with US: Unemployment insurance per unemployed = -0.01320.22451
    - US: Average remittances — correlation with US: State mobility = 0.09970.0738-0.19481
    - US: Average remittances — correlation with SLV: Government transfers per household = 0.08960.0894-0.04720.0061.1
  - 2020:1--2020:12 (selected):
    - US: Average remittances — correlation with US: Real wage = -0.041
    - US: Average remittances — correlation with US: Unemployment insurance per unemployed = 0.08810.21951
    - US: Average remittances — correlation with SLV: New COVID cases = 0.24690.13190.51290.0241
    - US: Average remittances — correlation with SLV: Government transfers per household = -0.2404-0.08570.1619-0.1396-0.34491

### Annex VI. Robustness of Results for the Model Explaining Average Amount Remitted (Key regression results and model details)
- Table 4: Determinants of evolution of average value of remittances (2017:1--2020:12) — selected coefficient estimates (standard errors in parentheses) and significance:
  - Constant:
    - Examples across specifications: -0.187 (0.193); 0.396* (0.230); -0.670*** (0.217); -0.385* (0.207); -0.979*** (0.201); -1.874*** (0.203); -1.983*** (0.185); -1.018*** (0.239); 2.158*** (0.232); 0.749*** (0.216); -0.385* (0.232); 0.741*** (0.232).
  - US: Real wage:
    - Coefficients: 0.937*** (0.032); 0.825*** (0.040); 0.856*** (0.037); 0.861*** (0.037); 1.073*** (0.035); 1.093*** (0.033); 1.137*** (0.033); 1.099*** (0.031); 0.548*** (0.041); 0.641*** (0.039); 0.861*** (0.037); 0.673*** (0.039).
  - US: Unemployment insurance per unemployed:
    - Coefficients shown where included: 0.006*** (0.001); 0.019*** (0.001); 0.019*** (0.001); -0.003* (0.001); 0.011*** (0.001); 0.019*** (0.001); 0.012*** (0.001).
  - US: State mobility:
    - Coefficients: 0.216*** (0.008); 0.144*** (0.012); 0.219*** (0.010); 0.178*** (0.007); 0.202*** (0.008); 0.144*** (0.012); 0.158*** (0.012).
  - US: Unemployment rate:
    - Selected coefficients: -0.427*** (0.039); 0.308*** (0.050).
  - SLV: New COVID cases:
    - Coefficients: -0.023*** (0.003); -0.044*** (0.002); -0.023*** (0.003); -0.014*** (0.003).
  - SLV: Government transfers per household:
    - Coefficients: 0.053*** (0.003); 0.041*** (0.003); 0.037*** (0.003).
- Sample and estimation details:
  - Number of US states: 23 (across specifications)
  - Number of SLV departments: 14
  - Number of panels: 322
  - Number of observations: 15072 (repeated across specifications)
  - All variables are in natural logarithm form.
  - Estimated with Feasible Generalized Least Squares with panel-specific autocorrelation AR(1) specification and heteroscedastic error structure.
  - Significance: * p<0.10, ** p<0.05, *** p<0.01

*Evolution of Remittances to CAPDR Countries and Mexico During the COVID-19 Pandemic — Working Paper No. WP/2022/092*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2022/english/wpiea2022092-print-pdf.pdf_