## 1. Amounts of Remittances in 2003

## Source details

**Canonical URL:** [1. Amounts of Remittances in 2003](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp06250.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp06250.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp06250.pdf.json)

---

### I. Introduction and context
- Remittances to El Salvador in 2004: US$2.5 billion, representing:
  - 16.6 percent of GDP
  - 76% of exports of goods
  - close to 60 percent of exports of goods and services
  - over 40 percent of imports of goods
- Principal source and scale:
  - 0.8–1.1 million Salvadorans residing in the United States
  - Aggregate personal income of Salvadorans in the U.S.: $13.1 billion (equal to El Salvador’s GDP)
- Household allocation of remittances:
  - Urban: consumption 80 percent; education and health 15 percent; investment 3 percent; savings 2 percent
  - Rural: consumption 90 percent; education and health 7 percent; investment 2 percent; savings 1 percent
- Regional context:
  - Emigration flows reached 10 million persons in the 1990s (Latin America)
  - Remittances to Latin America: US$38 billion in 2003

### II. Remittances by country (2003, in millions of U.S. dollars)
- Brazil: 5,200
- Colombia: 3,067
- Dominican Republic: 2,217
- Ecuador: 1,656
- El Salvador: 1,935
- Guatemala: 2,106
- Jamaica: 1,425
- México: 13,266
- Perú: 1,295

### III. Microeconomic uses and human capital effects
- Education impacts:
  - Rural: a household receiving a US$100 remittance per month → 56 percent lower probability of leaving primary school
  - Urban: corresponding probability is 24 percent lower
- Use for investment and business:
  - Lopez-Calix and Seligson (1990): on average 16 percent of remittances used for investment
  - By 2004: only an average of 3% of remittances used for investment
  - Delgado and Siri (1996): proposed investment funds and micro-enterprise credit lines to channel remittances into businesses
  - Cáceres (2003): explored remittances’ role in rural-sector development
- Household-level magnitude:
  - Average annual remittance received by Salvadoran households in 2002: $1,200, representing 51 percent of income of recipient households
- Distributional and poverty effects:
  - Remittances reduced the Gini coefficient from 0.55 to 0.53 and reduced poverty by 4.2 percent (as reported)

### IV. Macroeconomic effects, vulnerabilities, and “Dutch Disease” concerns
- Potential shock scenario:
  - Rivera Campos (1996): gradual cessation of remittances could push current account deficit to a maximum of 12.5 percent of GDP
- Interest rate and monetary effects:
  - Cáceres (2005a): remittances tend to exert upward pressure on the interest rate with negative effects on growth
  - Cáceres (2005b): remittance uncertainty (GARCH) materially influences money supply
  - Rivera Campos (1998): modeled a “Dutch Disease” phenomenon with contractionary forces
- Trade and income links:
  - Cartagena (2004): 1 percent increase in U.S. GDP → 2.4 percent increase in remittances; 1 percent increase in El Salvador GDP → 0.4 percent reduction in remittances (interpreted as risk-aversion driven migration)

### V. Stylized macro facts (1995–2004): savings, investment, consumption
- Investment and savings ratios:
  - Average investment/GDP: 19 percent in 1970–79 versus 16 percent in 1995–2003
  - National savings declined more than investment; current account deteriorated while consumption grew steadily
- Financing of consumption surge:
  - Consumption financed by remittances, expansion of private credit, and public external borrowing
- Policy concern:
  - Remittances boosted consumption and imports but not investment or domestic production, raising concerns about real exchange rate appreciation, price/wage pressures, and poor export growth

### VI. Key time-series and balance-sheet statistics (selected)
- Trade balance on goods and services vs. remittances (annual, millions of U.S. dollars):
  - 1996: Trade deficit 1,333 / Remittances 1,259
  - 1997: 1,295 / 1,364
  - 1998: 1,455 / 1,534
  - 1999: 1,539 / 1,566
  - 2000: 1,975 / 1,830
  - 2001: 2,182 / 2,022
  - 2002: 2,112 / 2,206
  - 2003: 2,393 / 2,316
  - 2004: 2,739 / 2,548
- Central Bank papers outstanding (millions of U.S. dollars):
  - 1995: 355.9; 1996: 462.5; 1997: 996.6; 1998: 1,028.3; 1999: 1,157.2; 2000: 1,273.0; 2001: 1,363.8; 2002: 1,295.3; 2003: 1,226.1; 2004: 1,317.4
- Commercial banking system balance sheet (as of December 31, 2004; millions of U.S. dollars):
  - Total assets: 9,718.7
  - Reserves: 1,822.3 (BCR reserves: 1,537.5)
  - Credit to private sector: 6,112.3
  - Deposits in checking and savings accounts: 3,050.3
  - Time deposits: 3,036.9
- Banking credit concentration by sector (share of total credit, as of December 31, 2004):
  - Housing: 23.7
  - Commerce: 18.0
  - Consumer loans: 17.3
  - Industry and manufacture: 10.5
  - Other services: 9.2
  - Construction: 8.4
  - Agriculture: 4.2
  - Other sectors: 3.1
  - Financial institutions: 2.7
  - Transportation and communications: 1.6
  - Electricity, gas, water, and other services: 1.2
  - Mines: 0.0
  - Total: 100.0
- Vulnerability note:
  - 59 percent of system loans concentrated in consumer credit, housing, and commerce implies higher financial-sector risk under remittance reductions

### VII. Monetary policy, dollarization, and central bank operations
- De facto fixed exchange rate regime (1995–2000):
  - CRB purchased foreign exchange to avoid nominal appreciation and sterilized liquidity from remittances via bonds, higher reserve requirements, and mandatory CRB bond holdings by banks
  - Sterilization outcomes:
    - Intensified conditions attracting capital inflow and increasing interest rates
    - High domestic interest rates and tight liquidity constrained benefits of capital inflows for investment and growth
    - Sterilization costly for CRB and could discourage financial intermediation
- Dollarization (2001–2004):
  - Adoption of U.S. dollar as legal tender on January 1, 2001; conversion at 8.75 colones per U.S. dollar
  - CRB continued issuing certificates of liquidity to ensure international reserves exceeded liquidity requirements; such operations are not fully consistent with dollarization and may explain observed interest-rate behavior after remittance shocks

### VIII. Trade links and regional transmission
- Trade composition:
  - Over 80% of El Salvador’s exports are to the U.S. and Central American countries
  - About 65% of imports are from the U.S. and Central American countries
  - Guatemala: El Salvador’s second-biggest trade partner after the U.S., accounting for over 20% of exports and over 10% of imports
  - More than 50% of El Salvador’s regional trade is with Guatemala, followed by Honduras, Costa Rica, and Nicaragua
- Implication:
  - A remittance shock in El Salvador would be transmitted to these economies; trade linkages could be enhanced after full implementation of CAFTA-DR

### IX. Empirical approach: VAR specification and data
- VAR variables (LOGS unless noted): IPCES (consumer price index), IMAE (index of economic activity), REMESASES (remittances), REPOS (rate of interest on repos, percent), M2ES (broad money), and one X variable (REER, other Central American activity, exports/imports, or international reserves)
- Conceptual transmission mechanism:
  - remittances → money supply (via reserves) and personal income → consumption → demand for money → interest rate → economic activity; remittances also affect trade and neighboring countries’ activity
- Exogenous variables:
  - Index of El Salvador’s export prices; dollarization dummy (0: Jan 1995–Dec 2000; 1: Jan 2001 onward)
- Data and sample:
  - Monthly data, January 1995 to August 2004
  - Sources: Central Reserve Bank of El Salvador; Central American Monetary Council
- Stationarity:
  - Most series stationary except international reserves; some series stationary only with constant and linear trend according to Dickey-Fuller tests

### X. VAR impulse-response findings (Appendix I summary)
- Consumer prices:
  - CPI shows steep increases after two and four months and sustains thereafter → remittances impart inflationary effects via non-traded goods prices and wages
- Index of Economic Activity (IMAE):
  - Decreases by 1.2 percent after two months; recovers to 0.005 percent after four months; negative growth thereafter; accumulated response increasingly negative → evidence consistent with Dutch Disease
- Remittances dynamic:
  - Immediate spike, sharp drop after two months, then sustained growth rate
- Interest rate (REPOS):
  - Initial small increase, sharp drop after two months, then increases to a positive value after seven months with rising trend; dollarization dummy coefficient significant and negative (dollarization decreased interest rate)
- Money supply (M2):
  - Consistently negative trend after remittance shock; with M3 re-estimation effect negative first six months then stabilizes positive → suggests substitution toward less liquid assets and central bank reaction reducing excess supply
- International reserves:
  - Initial positive spike; negative responses after three months, stabilizing toward zero → consistent with remittances financing imports
- Real exchange rate:
  - Appreciation in first two months, slight recovery, steady appreciation after three months → helps explain negative economic activity response
- Trade with Central America:
  - Varied short-term responses across partners; accumulated responses show positive accumulated exports to Guatemala, Nicaragua, Costa Rica and positive accumulated imports from Guatemala, Honduras, Costa Rica
- Indices of activity in other Central American countries:
  - Initial drop then recovery to maximum after eight–eleven months; accumulated responses positive for Guatemala, Honduras, Nicaragua (small for Costa Rica) → remittances promote growth in other Central American countries but not in El Salvador

### XI. Conclusions and economic policy implications
- Observed roles of remittances in El Salvador:
  - Support high private consumption
  - Finance imports of consumption goods and risky bank lending
  - Provide social benefits: reduce poverty rates and increase education rates
- Adverse short-run macro effects identified:
  - Real exchange rate appreciation
  - Increase in consumer prices, interest rates, and imports
  - Decrease in economic activity, money supply, and international reserves
  - Adverse effects intensified by Central Reserve Bank policy reaction
- Contrasts with neighbors:
  - Other Central American countries show increases in indices of economic activity and positive stimulation of exports and imports with El Salvador
- Policy recommendations and priorities:
  - Redesign central bank policy to align with dollarization and preserve remittance benefits
  - Review fiscal policy to promote national saving and investment
  - Strengthen regulation and supervision to limit lending toward consumption and high-risk, high-cost loans
  - Continue structural reforms to improve competitiveness and consolidate public-private initiatives for new investments, higher-productivity activities, and nontraditional products
  - Promote private investment through diverse means, including policy intervention to direct remittance investment via public-private collaborative strategies
  - Increase public investment in infrastructure and education
  - Support innovation in technology and technical education to generate productive capacity to meet remittance-driven demand
- Rationale:
  - Remittances will eventually decrease; without increased productive capacity in a dollarized economy, reductions in remittances could lead to economic and banking collapse and social instability

### XII. Appendix I — Key ADF test outcomes (number of lags = four; significance notation preserved)
- Consumer price index: Intercept = -2.54; Trend and Intercept = -3.521**; None = 5.25***
- Index of economic activity: Intercept = -2.039; Trend and Intercept = -3.43*; None = -0.52
- Remittances: Intercept = -0.053; Trend and Intercept = -8.40***; None = 2.53**
- Money: Intercept = -2.08; Trend and Intercept = -0.69; None = 3.12***
- Interest rate: Intercept = -2.08; Trend and Intercept = -3.72**; None = -1.84*
- Real exchange rate: Intercept = -4.58***; Trend and Intercept = -3.97**; None = -1.86*
- International reserves: Intercept = -2.04; Trend and Intercept = 1.07; None = 1.04
- Imports from the United States: Intercept = -5.17***; Trend and Intercept = -7.06***; None = -1.98**
- Imports from rest of world: Intercept = -4.56***; Trend and Intercept = -6.71***; None = -1.82*
- Index of economic activity, Costa Rica: Intercept = -2.29; Trend and Intercept = -10.01***; None = 0.32
- Index of economic activity, Guatemala: Intercept = -4.60***; Trend and Intercept = -9.77***; None = -0.30
- Index of economic activity, Honduras: Intercept = -1.61; Trend and Intercept = -3.32*; None = 0.69
- Index of economic activity, Nicaragua: Intercept = -3.46; Trend and Intercept = -6.49***; None = 0.14
- Imports from Costa Rica: Intercept = -2.29; Trend and Intercept = -10.01***; None = 0.32
- Imports from Guatemala: Intercept = -1.66; Trend and Intercept = -6.43***; None = 0.60
- Imports from Honduras: Intercept = -1.61; Trend and Intercept = -7.94***; None = 1.05
- Imports from Nicaragua: Intercept = -1.97; Trend and Intercept = -6.01***; None = 0.134
- El Salvador’s exports to Costa Rica: Intercept = -4.99***; Trend and Intercept = -4.97***; None = -0.28
- El Salvador’s exports to Guatemala: Intercept = -0.70; Trend and Intercept = -11.35***; None = 1.24
- El Salvador’s exports to Honduras: Intercept = -1.60; Trend and Intercept = -3.34*; None = 0.60
- El Salvador’s exports to Nicaragua: Intercept = -1.34; Trend and Intercept = -4.31***; None = 0.60

*Source: _wp06250 - 1. Amounts of Remittances in 2003 (excerpt) from the linked IMF PDF.*

### 1. Amounts of Remittances in 2003.......................................................................................

### 1. Amounts of Remittances in 2003

### I. Introduction and context
- Remittances to El Salvador reached US$2.5 billion in 2004, representing:
  - 16.6 percent of GDP
  - 76% of exports of goods
  - close to 60 percent of exports of goods and services
  - over 40 percent of imports of goods
- Principal source: the 0.8–1.1 million Salvadorans residing in the United States, whose aggregate personal income sums to $13.1 billion—an amount equal to El Salvador’s GDP.
- Urban allocation of remittances:
  - consumption: 80 percent
  - education and health: 15 percent
  - investment: 3 percent
  - savings: 2 percent
- Rural allocation of remittances:
  - consumption: 90 percent
  - education and health: 7 percent
  - investment: 2 percent
  - savings: 1 percent
- Broader regional context (Latin America in the 1990s and 2003):
  - emigration flows reached 10 million persons in the 1990s
  - remittances to Latin America: US$38 billion in 2003
  - remittance inflows were larger than foreign investment received by the region (source cited).

### II. Table of remittances by country (2003, in millions of U.S. dollars)
- Brazil: 5,200
- Colombia: 3,067
- Dominican Republic: 2,217
- Ecuador: 1,656
- El Salvador: 1,935
- Guatemala: 2,106
- Jamaica: 1,425
- México: 13,266
- Perú: 1,295
- Source: Inter-American Development Bank, Multilateral Investment Fund, 2004, Sending Money Home: Remittances to Latin America and the Caribbean (Washington).

### III. Economic and policy implications highlighted
- Benefits:
  - Remittance flows support higher human and capital development.
  - Inflows have helped reduce poverty levels and school desertion rates in El Salvador (cited study).
- Challenges:
  - Channeling external savings into productive projects.
  - Avoiding macroeconomic overheating.
  - Vulnerability to a remittance-flow reversal given the large size of inflows relative to GDP and trade.
- Research objective of the paper:
  - Analyze the transmission mechanism for El Salvador during a period of continuously increasing remittance inflows and poor economic growth (1995-2004).
  - Use monthly data and a model that includes real and financial sector variables of El Salvador and other Central American countries.
  - Answer questions on the macroeconomic effects of remittances on El Salvador and other Central American economies.

### IV. Structure of the study (as presented)
- Section II: brief survey of research on El Salvador’s receipts of remittances.
- Section III: key facts on magnitude of private remittances and relation to macroeconomic variables; nature of economic growth in El Salvador; central bank policy and role of commercial banks in financing consumption.
- Section IV: specification of the model, data issues, and description of the transmission mechanism (how private remittances affect economic activity and the price level).
- Section V: empirical results.
- Section VI: conclusions and policy implications.

*Source: _wp06250 - 1. Amounts of Remittances in 2003 (excerpt) from the linked IMF PDF.*

### 4.2 percent, as well as reduced the Gini coefficient from 0.55 to 0.53.

### _wp06250 - 4.2 percent, as well as reduced the Gini coefficient from 0.55 to 0.53.

### Remittances and human capital, investment, and microeconomic uses
- Education:
  - In rural areas, a child from a household receiving a 100-dollar remittance per month has a 56 percent lower probability of leaving primary school than a child from a household not receiving a remittance.
  - In urban areas, the corresponding probability is 24 percent.
- Use of remittances for investment and business:
  - Lopez-Calix and Seligson (1990) found on average 16 percent of remittances were used for investment purposes.
  - Note: the remittances allocation changed over time; in 2004, only an average of 3% of remittances were used for investment purposes.
  - Delgado and Siri (1996) proposed financial mechanisms (investment funds, micro-enterprise credit lines) to facilitate use of remittances in starting new businesses.
  - Cáceres (2003) explored the potential role of remittances in rural-sector development.
- Household-level magnitude:
  - The average annual amount of remittance received by El Salvador’s households was $1,200 in 2002, which represented 51 percent of the income of households receiving remittances.

### Macroeconomic effects, vulnerabilities, and “Dutch Disease” concerns
- Scenario of remittances ceasing:
  - Rivera Campos (1996) found that a gradual cessation of remittances would lead to increasing deterioration of external accounts, with the deficit in the current account reaching a maximum of 12.5 percent of GDP.
- Remittances and monetary/interest effects:
  - Rivera Campos (1998) formulated a model postulating a “Dutch Disease” phenomenon exerting contractionary forces.
  - Cáceres (2005a) (annual data, 1980–2001, cointegration) found that remittances tend to exert upward pressure on the interest rate in El Salvador, with negative effects on economic growth.
  - Cáceres (2005b) (monthly data) found remittances uncertainty (GARCH methods) was an important determinant of the money supply.
- Remittances and external activity links:
  - Cartagena (2004) (quarterly data 1990:Q1–2003:Q4) estimated a cointegration relationship: a 1 percent increase in the U.S. GDP leads to an increase in remittances of 2.4 percent, whereas an increase of 1 percent in El Salvador’s GDP is associated with a reduction of remittances by 0.4 percent. This was interpreted as evidence of risk-aversion motives in migration flows.

### Stylized macro facts (1995–2004): remittances, savings, investment, and consumption
- Investment and savings:
  - Average rate of investment to GDP: 19 percent in 1970–79 versus 16 percent in 1995–2003.
  - National savings declined more than investment; current account deteriorated while consumption grew steadily.
- Consumption and imports:
  - The surge in consumption was satisfied by increasing imports of consumption goods and was particularly financed by remittances, by an expansion of private sector credit, and by public external borrowing.
- Policy concern:
  - The inflow of remittances boosted consumption and imports but not investment and domestic production — raising concerns about Dutch Disease: upward pressure on prices and wages in non-traded goods and importable goods, real exchange rate appreciation, deterioration of the current account, and poor export growth.

### Key time-series and balance-sheet statistics (selected)
- Table 2. Trade Balance on Goods and Services and Remittances (annual data in millions of U.S. dollars):
  - Year / Trade Deficit on Goods and Services / Remittances
  - 1996: 1,333 / 1,259
  - 1997: 1,295 / 1,364
  - 1998: 1,455 / 1,534
  - 1999: 1,539 / 1,566
  - 2000: 1,975 / 1,830
  - 2001: 2,182 / 2,022
  - 2002: 2,112 / 2,206
  - 2003: 2,393 / 2,316
  - 2004: 2,739 / 2,548
  - Note: Table 2 shows the amount of private remittances are similar to the trade deficit on goods and services in millions of U.S. dollars.
- Central Bank papers outstanding stocks (Table 3, in millions of U.S. dollars):
  - 1995: 355.9
  - 1996: 462.5
  - 1997: 996.6
  - 1998: 1028.3
  - 1999: 1157.2
  - 2000: 1273.0
  - 2001: 1363.8
  - 2002: 1295.3
  - 2003: 1226.1
  - 2004: 1317.4
- Commercial banking system balance sheet (as of December 31, 2004; Table 4, in millions of U.S. dollars):
  - Total assets: 9,718.7
  - Total liabilities plus capital: 9,718.7
  - Reserves: 1,822.3 (BCR reserves: 1,537.5)
  - Credit to private sector: 6,112.3
  - Deposits in checking and savings accounts: 3,050.3
  - Time deposits: 3,036.9
- Banking credit concentration by sector (Table 5: share of total credit, as of December 31, 2004):
  - Housing: 23.7
  - Commerce: 18.0
  - Consumer loans: 17.3
  - Industry and manufacture: 10.5
  - Other services: 9.2
  - Construction: 8.4
  - Agriculture: 4.2
  - Other sectors: 3.1
  - Financial institutions: 2.7
  - Transportation and communications: 1.6
  - Electricity, gas, water, and other services: 1.2
  - Mines: 0.0
  - Total: 100.0
- Vulnerability:
  - The concentration of bank lending in consumer credit, housing, and commerce (59 percent of system loans) implies higher risks and costs and increases financial sector vulnerability in a remittance-reduction scenario.

### Monetary policy, dollarization, and central bank operations
- De facto fixed exchange rate regime (1995–2000):
  - CRB intervened by buying foreign exchange to avoid nominal appreciation and attempted to absorb liquidity from remittances via sterilization bonds, increased reserve requirements, and requiring banks to hold CRB bonds.
  - Sterilization outcomes and drawbacks:
    - Sterilization intensified conditions attracting capital inflow, increasing interest rates.
    - High domestic interest rates and tight liquidity deprived the economy of benefits of capital inflows (e.g., increasing investment and growth).
    - Open-market sterilization was costly for the CRB and could discourage financial intermediation.
- Dollarization (2001–2004):
  - On January 1, 2001, El Salvador adopted the U.S. dollar as legal tender; conversion at 8.75 colones per U.S. dollar.
  - After dollarization, the CRB continued issuing certificates of liquidity; these operations aimed to ensure international reserves exceeded liquidity requirements but are not fully consistent with dollarization and could partly explain observed interest-rate behavior after remittance shocks and the apparent negative effect of remittances.

### Trade links and regional transmission
- Trade composition and partners:
  - Over 80% of El Salvador’s exports are to the U.S. and the Central American countries.
  - About 65% of El Salvador’s imports are from the U.S. and the Central American countries during the sample period.
  - Among Central American countries, Guatemala is El Salvador’s second-biggest trade partner after the U.S., accounting for over 20% of El Salvador’s exports and over 10% of its imports.
  - More than 50% of El Salvador’s regional trade is with Guatemala, followed by Honduras, Costa Rica, and Nicaragua.
- Implication:
  - A shock to remittances in El Salvador would be transmitted to these economies in particular; El Salvador is also one of the top trade partners for the rest of Central America.
  - These trade linkages could be enhanced after full implementation of CAFTA-DR.

### Empirical approach: VAR specification and data
- Conceptual VAR structure:
  - VAR variables include: log of consumer price index (IPCES), log of index of economic activity (IMAE), log of remittances (REMESASES), rate of interest on repos (REPOS), log of money supply broadly defined (M2ES), and one X variable (real exchange rate, other Central American countries’ economic activity, exports to/imports from El Salvador, or El Salvador international reserves).
  - Conceptual transmission: remittances → money supply (via reserves) and personal income → consumption → demand for money → interest rate → economic activity; remittances and income also affect trade with Central American countries and their activity.
- Exogenous variables:
  - Index of El Salvador’s export prices.
  - Dummy variable: 0 from January 1995 to December 2000; 1 from January 2001 onward (dollarization period).
  - The export-price index was dropped when REER entered as an X variable.
- Data and sample:
  - Monthly data, January 1995 to August 2004.
  - Data sources: Central Reserve Bank of El Salvador for El Salvador series; Central American Monetary Council for other Central American countries.
- Stationarity and estimation notes:
  - Most series used are stationary, except international reserves.
  - Some series (imports, exports, indices of economic activity) are stationary only when the Dickey-Fuller test includes a constant and a linear trend, indicating deterministic trends.
- Empirical objective:
  - Estimate impulse response functions from the VAR to represent the response of El Salvador’s economic activity, interest rate, consumer prices, and money supply to a remittances shock, and to detect spillover effects on neighboring countries.

*Source: _wp06250 - 4.2 percent, as well as reduced the Gini coefficient from 0.55 to 0.53.*

### Appendix I).

### _wp06250 - Appendix I)

### VAR estimation setup
- Unrestricted VARs estimated using logarithms of variables, except the interest rate which used percentage values.
- Cholesky Impulse Response Functions (IRF) impose ordering reflecting faster reaction of financial variables than real variables.
- Example VAR ordering: IPCES, IMAEES, REMESASES, REPOSES, M2ES, IRES.
- Empirical evidence reported as responses to a one-standard error shock to remittances.

### A. Consumer prices
- Consumer price index shows steep increases after two and four months and reaches a value that is sustained thereafter.
- Interpretation: remittances impart an inflationary effect via increased prices and wages of non-traded goods and slow supply response to rising purchasing power.

### B. Index of Economic Activity (IMAE)
- Index of economic activity decreases by 1.2 percent after two months.
- It recovers to 0.005 percent after four months and experiences negative growth rates thereafter.
- Accumulated response of IMAEES to remittances is increasingly negative.
- Possible mechanism: Dutch Disease effects—higher prices/wages in non-traded goods reduce export growth and raise imports, displacing national production; additional demand met by imports.

### C. Remittances
- Remittances respond by immediately spiking, then dropping sharply after two months, and settling at a sustained growth rate soon thereafter.
- Interpretation: remittances have an inherent growth-driving force.

### D. Interest rate
- Repos interest rate: initial small increase, followed by a sharp drop after two months, then increases to reach a positive value after seven months and shows an increasing trend thereafter.
- Initial decrease attributed to short-lived liquidity from remittances dissipated through imports.
- Dollarization dummy coefficient for interest rate is significant and negative, indicating dollarization decreased the interest rate.
- Net effect: tendency toward an interest rate increase, which may strengthen recessive forces.

### E. Money supply
- Broadly defined money supply (M2) shows a consistently negative trend after a remittances shock.
- Re-estimation using M3: effect negative the first six months and stabilizes to positive values thereafter.
- Interpretation: substitution of money for less liquid assets after remittance shocks; central bank policy reaction aimed at reducing excess supply given lower demand for money during first half of sample.
- Lower demand for money driven by reductions in index of economic activity and increases in price level.
- Since dollarization (second half of sample), central bank aimed to maintain adequate international reserves—consistent with increased interest rate.
- Table 3 documents central bank operations during sample period (1995–2004).

### F. International reserves
- Net international reserves show a positive initial spike, but after three months responses become negative and remain negative, stabilizing toward zero.
- Interpretation: remittances used for imports of consumer goods; central bank policy unnecessarily restricting money supply and increasing interest rate.

### G. Real exchange rate
- Real exchange rate shows an appreciation in first two months, slight recovery, and after three months a steady appreciation.
- Real appreciation helps explain negative response of economic activity to remittances.

### H. El Salvador’s imports by source
- Imports from other Central American countries: sharp drop after two months, then cyclical paths reaching zero after approximately eight months.
- Imports from Honduras: increase sharply after three months, becoming sustained as negative after nine months.
- Imports from Nicaragua: negative values until 10 months, increasing thereafter to reach zero after 15 months.
- Imports from Costa Rica: negative at two months, then cyclical decay to zero after 12 months.
- Accumulated responses (Appendix II) show accumulated responses of exports to Guatemala, Nicaragua, and Costa Rica are positive; accumulated responses of imports from Guatemala, Honduras, and Costa Rica are positive.
- Imports from the United States and rest of the world: sharp initial spikes followed by decreases and cyclical patterns with predominating positive values; accumulated responses indicate large positive responses.

### I. Indices of economic activity of other Central American countries
- Indices show initial drop after two or three months, recovery to a maximum positive value after eight or nine months (11 months for Costa Rica), followed by cyclical decaying trend.
- Amplitude of cycles larger than those for exports/imports with Central America.
- Accumulated responses show positive values at end of period for Guatemala, Honduras, Nicaragua (small for Costa Rica).
- Conclusion: remittances promote economic growth in all Central American countries except El Salvador.

### J. El Salvador’s exports to Central America
- Exports to Guatemala: increase after initial small drop, reach maximum after three months, then decrease to negative values after eight months and remain negative.
- All Central American export responses characterized by initial positive spikes followed by rapid decay to negative values approaching or reaching zero by end of period.
- Net effect on balance of trade with Central American countries is inconclusive.

### VI. Conclusions and economic policy implications
- Remittances in El Salvador:
  - Support high private consumption.
  - Finance imports of consumption goods and risky bank lending.
  - Provide social benefits: reduce poverty rates and increase education rates.
- Adverse short-run macroeconomic effects observed:
  - Appreciation of the real exchange rate.
  - Increase in consumer prices, the interest rate, and imports.
  - Decrease in economic activity, money supply, and international reserves.
  - These adverse effects are intensified by Central Reserve Bank policy reaction.

- Contrasts: Other Central American countries exhibit increases in indices of economic activity; exports to and imports from Central America are positively stimulated (with faster decay in Nicaragua and Costa Rica).

- Policy recommendations and priorities identified:
  - Redesign central bank policy to avoid inconsistency with the dollarization regime and avoid reducing remittance benefits.
  - Review fiscal policy to promote national saving and investment.
  - Strengthen regulation and supervision of the financial system to limit lending toward consumption and higher-risk, higher-cost loans.
  - Continue structural reform agenda to improve competitiveness and consolidate public-private initiatives supporting new investments, higher-productivity activities, and nontraditional products.
  - Promote:
    - Private investment through diverse means, including policy intervention to direct remittance investment (e.g., public-private collaborative strategies).
    - Public investment in infrastructure and education.
    - Support for innovation in technology and technical education to generate productive capacity to satisfy demand created by remittances.
- Rationale: remittances will eventually decrease; without increased productive capacity in a dollarized economy, reductions in remittances could lead to economic and banking collapse and social instability.

### Appendix I — Key Augmented Dickey-Fuller (ADF) test outcomes (data dependent lag; Number of lags = four)
- Consumer price index: Intercept = -2.54; Trend and Intercept = -3.521**; None = 5.25***.
- Index of economic activity: Intercept = -2.039; Trend and Intercept = -3.43*; None = -0.52.
- Remittances: Intercept = -0.053; Trend and Intercept = -8.40***; None = 2.53**.
- Money: Intercept = -2.08; Trend and Intercept = -0.69; None = 3.12***.
- Interest rate: Intercept = -2.08; Trend and Intercept = -3.72**; None = -1.84*.
- Real exchange rate: Intercept = -4.58***; Trend and Intercept = -3.97**; None = -1.86*.
- International reserves: Intercept = -2.04; Trend and Intercept = 1.07; None = 1.04.
- Imports from the United States: Intercept = -5.17***; Trend and Intercept = -7.06***; None = -1.98**.
- Imports from rest of world: Intercept = -4.56***; Trend and Intercept = -6.71***; None = -1.82*.
- Index of economic activity of Costa Rica: Intercept = -2.29; Trend and Intercept = -10.01***; None = 0.32.
- Index of economic activity of Guatemala: Intercept = -4.60***; Trend and Intercept = -9.77***; None = -0.30.
- Index of economic activity of Honduras: Intercept = -1.61; Trend and Intercept = -3.32*; None = 0.69.
- Index of economic activity of Nicaragua: Intercept = -3.46; Trend and Intercept = -6.49***; None = 0.14.
- Imports from Costa Rica: Intercept = -2.29; Trend and Intercept = -10.01***; None = 0.32.
- Imports from Guatemala: Intercept = -1.66; Trend and Intercept = -6.43***; None = 0.60.
- Imports from Honduras: Intercept = -1.61; Trend and Intercept = -7.94***; None = 1.05.
- Imports from Nicaragua: Intercept = -1.97; Trend and Intercept = -6.01***; None = 0.134.
- El Salvador’s exports to Costa Rica: Intercept = -4.99***; Trend and Intercept = -4.97***; None = -0.28.
- El Salvador’s exports to Guatemala: Intercept = -0.70; Trend and Intercept = -11.35***; None = 1.24.
- El Salvador’s exports to Honduras: Intercept = -1.60; Trend and Intercept = -3.34*; None = 0.60.
- El Salvador’s exports to Nicaragua: Intercept = -1.34; Trend and Intercept = -4.31***; None = 0.60.

- Significance notation:
  - * H0 of unit root is rejected at the 10 percent level
  - ** H0 of unit root is rejected at the 5 percent level
  - *** H0 of unit root is rejected at the 1 percent level

### Appendix II — Accumulated responses (summary)
- Accumulated response figures presented for:
  - Consumer prices
  - El Salvador’s index of economic activity
  - El Salvador’s imports from Guatemala
  - Remittances
  - Interest rate
  - Money supply
  - El Salvador’s imports from the U.S.
  - El Salvador’s imports from the rest of the world
  - International reserves
  - Guatemala’s index of economic activity
  - Honduras’s index of economic activity
  - Nicaragua’s index of economic activity
- Note: S.D. :Standard Deviation.

*Source: _wp06250 - Appendix I).*

### References

### References

### Remittances and El Salvador — empirical and policy studies
- Cáceres, Luis René, 2000, “Integración Monetaria en las Regiones Centroamericana y Andina,” Comercio Exterior, Vol. 50, No. 7 (July).
- Cáceres, Luis René, 2003, “Remesas y Desarrollo Rural en Centroamérica,” Estudios Centroamericanos, October
- Cáceres, Luis René, 2005a, “Remesas y Macroeconomía en El Salvador” (forthcoming in Comercio Exterior).
- Cáceres, Luis René, 2005b, “Short Run Macroeconomic Effects of Remittances in El Salvador” (forthcoming in Savings and Development).
- Cartagena, Edgar R., 2004, “Las Remesas Familiares y su Relación con los Ciclos y Shocks Económicos,” Banco Central de Reserva de El Salvador, Documento Ocasional No. 2004/01 (San Salvador).
- Castro, Alvaro, 2004, “El Impacto de las Remesas en la Pobreza e Ingreso de los Hogares Salvadoreños,” Realidad, San Salvador, Universidad Centroamericana José Simeón Canas, Vol. 97 (January/February), pp. 37–66.
- Central Reserve Bank of El Salvador, Revista Trimestral, October–December 2004.
- Cox-Edwards, Alejandra, and Manuelita Ureta, 2003, “Internacional Migration, Remittances and Schooling: Evidence from El Salvador,” Journal of Development Economics, Vol. 72, No. 2 (December), pp. 429–61.
- Delgado, Pedro Abelardo, and Gabriel Siri, 1995, “Uso Productivo de las Remesas Familiares en El Salvador,” Fundación Salvadoreña para el Desarrollo Económico y Social, San Salvador (January).
- Lopez-Calix, Jose Roberto, and Mitchell Seligson, 1990, “Small Business Development in El Salvador: The Impact of Remittances,” Report of the Commission for the Study of International Migration and Cooperative Development (Washington: Government Printing Office).
- Rivera Campos, Roberto, 1996, “La Potencial Contracción de las Remesas y el Financiamiento de la Economía Salvadorena,” Boletín del CEMLA, (November–December).
- Rivera Campos, Roberto, 1998, “Mal Holandés, Esterilización Monetaria y Tasa de Interés Real en El Salvador,” Instituto de Economía Aplicada, Serie Documentos de Investigación, No. 1, San Salvador (July).
- Rivera Campos, Roberto, and Anabella Lardé de Palomo, 2002, “Invirtamos en Educación para Desafiar el Crecimiento Económico y la Pobreza,” Fundación Salvadoreña para el Desarrollo Económico y Social, San Salvador (May).
- Yang, Dean, 2003, “Risk, Migration, and Rural Financial Markets: Evidence from Earthquakes in El Salvador” (unpublished; Ann Arbor, Michigan: Gerald R. Ford School of Public Policy, Department of Economics, University of Michigan).

### Monetary policy, transmission mechanisms, and regional comparisons
- Carlino, Gerald A., 1998, “Monetary Policy and the U.S. States and Regions: Some Implications for the European Monetary Union,” Federal Reserve Bank of Philadelphia, Working Paper No. 98-17 (July).
- Christiano, Lawrence L., Martin Eichenbaum, and Charles L. Evans, 1998, “Monetary Policy Shocks: What Have We Learned and to What End?,” NBER Working Paper No. 6400 (Cambridge, Massachusetts: National Bureau of Economic Research).
- Clements, Benedict, Zenon G. Kontolemis, and Joaquím Levy, 2001, “Monetary Policy Under EMU: Differences in the Transmission Mechanism?” IMF Working Paper 01/102 (Washington: International Monetary Fund).
- Peersman, Pert, 2004, “The Transmission of Monetary Policy in the Euro Area: Are the Effects Different Across Countries?” Oxford Bulletin of Economics and Statistics, Vol. 66, No. 3, pp. 285–308.
- Ramaswamy, Ramana, and Torsten Slök, 1977, “The Real Effects of Monetary Policy in the European Union: What Are the Differences?, IMF Working Paper 97/160 (Washington: International Monetary Fund).
- Jansen, Jos W., 2003, “What Do Capital Inflows Do? Dissecting the Transmission Mechanism for Thailand, 1980–1996,” Journal of Macroeconomics, Vol. 25, No. 4 (December), pp. 457–80.

### Development, self-discovery, and growth strategy
- Hausmann, Ricardo, and Dani Rodrik, 2002, “Economic Development as Self-Discovery,” NBER Working Paper (March).
- Rodrik Dani and Ricardo Hausmann (2003), “Discovering El Salvador’s Production Potential,” part of a report on El Salvador’s economic strategy, Fundación Salvadoreña para el Desarrollo Económico y Social, San Salvador, (September).
- Rodrik Dani and Ricardo Hausmann (2005), “Self Discovering in a Development Strategy for El Salvador,” Harvard University, (September).
- Rajan, Raghuram G., and Arvind Subramanian, 2005, “What Undermines Aid’s Impact on Growth?,” IMF Working Paper 05/126 (Washington: International Monetary Fund).
- Velasco, Andrés, 2004, “Finance in El Salvador: Avoiding Crises, Nurturing Growth,” study prepared for Fundación Salvadoreña para el Desarrollo Económico y Social (FUSADES), San Salvador, El Salvador.

### Regional and institutional studies; finance and exchange rate issues
- Inter-American Development Bank (IADB), Multilateral Investment Fund, 2004, Sending Money Home: Remittances to Latin America and the Caribbean (Washington).
- Saca, Nolvia N., 1990, “Exchange Controls, Transfers and Export Competitiveness: Theory and Some Evidence from El Salvador,” Kiel Institute of World Economy (May).
- Saca, Nolvia N., 1997, Black Market Exchange Rate, Unification of the Foreign Exchange Markets and Monetary Policy: The Case of El Salvador (Bern, Switzerland: Peter Lang, Europaischer Verlag der Wissenschaften).
- Saca, Nolvia N., and Orlando Martinez, 1998, “Fiscal and Monetary Policies and its Impact in El Salvador’s Economy and Financial System,” study prepared for the Banking Association in El Salvador (ABANSA), San Salvador (December).

*Source: _wp06250 - References*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp06250.pdf_