## _wp12251 - References

## Source details

**Canonical URL:** [_wp12251 - References](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp12251.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp12251.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp12251.pdf.json)

---

### I. Introduction and research question
- Worldwide remittance flows are estimated to have exceeded $440 billion in 2010, with the share of developing countries around $325 billion.
- Recorded remittances were nearly three times the amount of official development assistance (ODA) and almost on a par with foreign direct investment (FDI) flows to developing countries.
- In Sub-Saharan Africa, remittances range from less than 1 percent of GDP to 25 percent in Lesotho.
- Research question: Do remittance flows affect international business cycle synchronization? Remittances are argued to constitute an additional transmission channel of global shocks.

### II. Baseline specifications of the effect of remittances on business cycle synchronization
- Unilateral (aggregate) specification:
  - Dependent variable: ρi,t = coefficient of correlation between country i business cycle at year t and the weighted sum of business cycles of country i migrant host countries.
  - Key regressor: Remittance inflows as share of GDP Rit.
  - Controls X include trade and financial openness.
  - Fixed effects: country (αi) and year (λt).
- Bilateral specification:
  - Dependent variable: ρi,j,t = yearly correlation coefficient of business cycles of countries i and j.
  - Key regressor: bilateral remittances received by i and sent from j, expressed as percentage of i's GDP (R i j t).
  - Controls X include bilateral trade intensity, summed financial openness, and asymmetry in production (ASP).
  - Country-pair fixed effects ideally included; SEs clustered at country-pair level.

### III. Data and measurement
- Sample periods:
  - Aggregate approach: 1980 to 2010.
  - Bilateral specification: 1980 to 2005.
- Remittance data:
  - Aggregate: World Bank Development Indicators, narrow definition (excludes compensation of employees and migrants’ wealth transfers).
  - Bilateral: Lueth and Ruiz-Arranz (2008) panel for 11 developing countries (Bangladesh, Croatia, Indonesia, Kazakhstan, FYR Macedonia, Moldova, Philippines, Serbia and Montenegro, Slovenia, Tajikistan, Thailand).
  - Bilateral dataset: on average ~16 source countries per recipient and ~9 years of data, yielding almost 1,650 bilateral remittance observations.
- GDP and cycles:
  - GDP measured in PPP terms from WDI.
  - Business cycles measured as log deviation of real GDP from Baxter and King (1999) trend (BK arguments 2 and 8).
  - Robustness: Hodrick-Prescott (HP) filter with smoothness parameter 6.25.
- Key variable definitions:
  - Bilateral trade intensity TIi,j,t = (x i j t + m i j t) / (GDP i t + GDP j t) with exports FOB and imports CIF in U.S. dollars.
  - Financial openness: Chinn and Ito (2008) index rescaled; bilateral measure = sum of individual country openness.
  - Asymmetry in production ASP i j t = sum over k of |s k i t − s k j t| for k = {agriculture, industry, services} using WDI sectoral value added shares.

### IV. Preliminary results (baseline OLS / standardized coefficients)
- Aggregate OLS (country and time dummies included):
  - Using HP cycles:
    - Trade openness: 0.161** (2.345)
    - Financial openness: 0.0978*** (2.883)
    - Remittance ratio: 0.101*** (3.006)
    - Observations: 1,853
    - Number of countries: 70
  - Using BK cycles:
    - Trade openness: 0.184** (2.611)
    - Financial openness: 0.0759** (2.213)
    - Remittance ratio: 0.0879** (2.412)
    - Observations: 1,754
    - Number of countries: 70
- Bilateral OLS (receiving and sending country fixed effects; SEs clustered by country-pair):
  - Using HP cycles:
    - Bilateral trade intensity: 0.105** (2.549)
    - [Financial openness in i] + [Financial openness in j]: 0.267*** (4.136)
    - Asymmetries in production between i and j: -0.196*** (-3.148)
    - Bilateral remittance inflows-to-GDP in i sent from j: 0.0659** (2.571)
    - Observations: 1,541
    - Country pairs: 154
  - Using BK cycles:
    - Bilateral trade intensity: 0.0711* (1.961)
    - [Financial openness in i] + [Financial openness in j]: 0.218*** (3.361)
    - Asymmetries in production between i and j: -0.255*** (-3.377)
    - Bilateral remittance inflows-to-GDP in i sent from j: 0.0627*** (3.817)
    - Observations: 1,116
    - Country pairs: 111
- Interpretation:
  - Remittances significantly increase business cycle synchronization.
  - Magnitude comparable to financial openness and smaller than trade openness in aggregate specifications.

### V. Asymmetric effects of remittances
- Hypothesis: remittances transmit downturns in sending countries more strongly than upswings.
- Aggregate asymmetric OLS:
  - Using HP cycles:
    - Trade openness: 0.164** (2.233)
    - Financial openness: 0.105*** (3.081)
    - Lagged remittance ratio * 1[y* < 0]: 0.0668*** (2.663)
    - Lagged remittance ratio * 1[y* > 0]: 0.0511 (1.140)
    - Observations: 1,773; Number of countries: 70
  - Using BK cycles:
    - Trade openness: 0.195** (2.606)
    - Financial openness: 0.0919** (2.503)
    - Lagged remittance ratio * 1[y* < 0]: 0.0610** (2.291)
    - Lagged remittance ratio * 1[y* > 0]: 0.0100 (0.243)
    - Observations: 1,674; Number of countries: 70
  - Conclusion: aggregate results suggest stronger spillovers during slowdowns; little transmission of upswings in aggregate.
- Bilateral asymmetric OLS:
  - Using HP cycles:
    - Bilateral trade intensity: 0.101** (2.334)
    - [Fin. openness in i] + [Fin. openness in j]: 0.256*** (4.022)
    - Asymmetries in production between i and j: -0.230*** (-2.893)
    - Lagged bilateral remittance ratio * 1[y* < 0]: 0.0429*** (2.940)
    - Lagged bilateral remittance ratio * 1[y* > 0]: 0.0392** (2.170)
    - Observations: 1,452; Country pairs: 154
  - Using BK cycles:
    - Bilateral trade intensity: 0.0740* (1.859)
    - [Fin. openness in i] + [Fin. openness in j]: 0.207*** (3.275)
    - Asymmetries in production between i and j: -0.306*** (-3.530)
    - Lagged bilateral remittance ratio * 1[y* < 0]: 0.0413*** (4.047)
    - Lagged bilateral remittance ratio * 1[y* > 0]: 0.0234** (2.561)
    - Observations: 1,072; Country pairs: 111
  - Conclusion: bilateral results show both upswings and downturns are transmitted via remittances, with stronger channeling of downturns.

### VI. Identification and causal inference
- A. Selection on observables vs. unobservables (Altonji et al. (2005) methodology)
  - Restricted covariates: remittances, trade openness, financial openness, asymmetry in production.
  - Full covariates add: banking crisis dummies, natural disasters, oil rents-to-GDP, inflation rates, differences in government consumption growth rate, reserve money growth rate.
  - Altonji ratios (how much stronger selection on unobservables relative to observables must be to explain away the effect):
    - Cycles: HP Aggregate data — Baseline coefficient 0.101*** (3.006); Additional controls coefficient 0.097* (1.77); Altonji ratio: 28.7
    - Cycles: BK Aggregate data — Baseline 0.0879** (2.412); Additional 0.075* (1.78); Altonji ratio: 6.1
    - Cycles: HP Bilateral data — Baseline 0.0659** (2.571); Additional 0.059* (1.66); Altonji ratio: 8.3
    - Cycles: BK Bilateral data — Baseline 0.0627*** (3.817); Additional 0.049*** (3.80); Altonji ratio: 3.8
  - Interpretation: Altonji ratios range from 3.8 to 28.7; selection on unobservables would need to be at least three times stronger than selection on observables to fully explain away the OLS estimates in several specifications.
- B. Two-stage approach addressing reverse causality (Brückner (2011)-inspired)
  - First stage: remittances regressions include causal responses of remittances to real GDP changes in sending and receiving countries; real GDP growth in recipient instrumented by log(terms of trade).
  - Second stage: use residual variation in remittances (portion not driven by the two cycles) as instrument for remittance ratio to estimate effect on synchronization.
  - Aggregate IV results:
    - First-stage: log(Terms of trade) positively correlated with log(Real GDP) in receiving economies: coefficient 0.0497*** (2.982).
    - Second-stage IV (HP sync): Remittance instrument 0.409*** (9.127); Remittance-to-GDP (instrumented) 0.157** (2.410) appears as determinant of sync in alternative specs.
    - Trade and financial openness remain positive and significant in IV second-stage.
    - Number of countries generally 69; Observations: 1,793 (varies by column).
  - Bilateral IV results:
    - First-stage: log(Terms of trade) strongly positive with log(Real GDP) of recipient: 1.844*** (11.29).
    - Remittance instrument significant: 0.561*** (1,055) and 1.020*** (1,335) in different columns.
    - Second-stage IV: Bilateral remittance-to-GDP (instrumented) 0.139*** (4.641) and 0.0622*** (4.016) appear as determinants of sync in different columns.
    - Bilateral trade intensity and summed financial openness remain positive and significant; asymmetries in production negative and significant.
    - Number of country pairs: 116 (varies by column); Observations: 1,088 and 923 in different columns.
  - Interpretation: IV estimates confirm a positive and significant causal effect of remittance flows on business cycle synchronization; IV estimates generally larger than OLS, consistent with OLS being biased downwards by countercyclical remittances.

### VII. Key findings and statistics
- Remittance significance (selected coefficients and statistics):
  - Aggregate OLS (HP): remittance ratio 0.101*** (3.006)
  - Aggregate OLS (BK): remittance ratio 0.0879** (2.412)
  - Bilateral OLS (HP): bilateral remittance inflows-to-GDP 0.0659** (2.571)
  - Bilateral OLS (BK): bilateral remittance inflows-to-GDP 0.0627*** (3.817)
- Asymmetry (selected coefficients):
  - Aggregate asymmetric OLS (HP): lagged remittance ratio * 1[y* < 0]: 0.0668*** (2.663); * 1[y* > 0]: 0.0511 (1.140)
  - Bilateral asymmetric OLS (HP): lagged bilateral remittance ratio * 1[y* < 0]: 0.0429*** (2.940); * 1[y* > 0]: 0.0392** (2.170)
- Causal evidence and instruments:
  - Altonji ratios: 28.7 (HP aggregate), 6.1 (BK aggregate), 8.3 (HP bilateral), 3.8 (BK bilateral)
  - IV first-stage relevance: terms of trade significant predictor of recipient log(Real GDP) (e.g., 0.0497*** (2.982) aggregate; 1.844*** (11.29) bilateral)
  - IV second-stage: remittance instrument significant (e.g., 0.409*** (9.127) aggregate instrument coefficient)

### VIII. Policy implications and recommendations
- Remittances are a distinct and important channel of integration and vulnerability:
  - Developing countries receiving sizable remittance inflows tend to be more connected and vulnerable to external shocks than suggested by traditional measures (trade and capital account openness) alone.
- Measurement and surveillance:
  - The concept of external openness should be expanded to account for the remittances channel.
  - Measures of openness and integration that include remittance flows would provide a more accurate gauge of exposure to remittance fluctuations and global shocks.
- Policy guidance for practitioners:
  - Policymakers should consider remittances when designing tools to manage external spillovers and to assess vulnerability to global downturns.
  - Awareness that remittance dependency tends to transmit downturns more strongly than upswings should inform macroeconomic and fiscal resilience strategies.

### IX. Appendix — Descriptive statistics and lists
- Aggregate data descriptive statistics (Table A1):
  - Synchronization coefficient (HP): Obs 2100; Mean 0.21; Std. Dev. 1.37; Min -13.99; Max 1
  - Synchronization coefficient (BK): Obs 1958; Mean 0.18; Std. Dev. 1.42; Min -14.28; Max 1
  - Trade-to-GDP ratio: Obs 2058; Mean 0.71; Std. Dev. 0.40; Min 0.06; Max 2.80
  - Financial openness: Obs 2075; Mean 1.62; Std. Dev. 1.31; Min 0.16; Max 4.48
  - Remittance-to-GDP ratio: Obs 1907; Mean 0.02; Std. Dev. 0.04; Min 0.00; Max 0.24
  - Terms of trade (in log): Obs 1999; Mean 4.65; Std. Dev. 0.35; Min 2.36; Max 6.01
- Aggregate sample: List of 70 countries (as presented in source) includes Algeria; Argentina; Bangladesh; Belize; Benin; Bolivia; Botswana; Brazil; Burkina Faso; Burundi; Cameroon; Chile; China; Colombia; Congo, Rep.; Costa Rica; Cote d'Ivoire; Dominican Republic; Ecuador; Egypt; El Salvador; Fiji; Gabon; Gambia; Ghana; Guatemala; Guinea-Bissau; Honduras; India; Indonesia; Iran; Jamaica; Jordan; Kenya; Lesotho; Madagascar; Malawi; Mali; Mauritania; Mauritius; Mexico; Morocco; Nepal; Nicaragua; Niger; Nigeria; Pakistan; Papua New Guinea; Panama; Paraguay; Peru; Philippines; Rwanda; Senegal; Seychelles; Sierra Leone; Sri Lanka; St. Vincent and the Grenadines; Sudan; Suriname; Syrian Arab Rep.; Thailand; Togo; Tunisia; Turkey; Uruguay; Venezuela, RB; Zambia
- Bilateral data descriptive statistics (Table B1):
  - Synchronization coefficient (HP): Obs 1569; Mean 0.29; Std. Dev. 1.36; Min -13.75; Max 1
  - Synchronization coefficient (HP) [second entry]: Obs 1140; Mean 0.20; Std. Dev. 1.45; Min -12.01; Max 1
  - Bilateral trade intensity: Obs 1569; Mean 0.00; Std. Dev. 0.01; Min 0.00; Max 0.05
  - Financial openness in i and j: Obs 1569; Mean 5.44; Std. Dev. 1.52; Min 1.01; Max 8.16
  - Asymmetry of production: Obs 1569; Mean 11.50 6.00; Std. Dev. 0.74; Min 47.83
  - Bilateral remittance ratio: Obs 1569; Mean 0.00; Std. Dev. 0.01; Min 0.00; Max 0.20
  - Terms of trade (in log): Obs 1088; Mean 4.74; Std. Dev. 0.12; Min 4.45; Max 4.96
- Bilateral sample countries (Table B2, 10 countries): Bangladesh; Croatia; Indonesia; Kazakhstan; Macedonia; Moldova; Philippines; Slovenia; Tajikistan; Thailand

*Source: _wp12251 - References (PDF).*

### References .............................................................................................................

### _wp12251 - References

### Tables
- 1. Impact of Remittance Inflows on International Business Cycle Synchronization ................................................................................9
- 2. Bilateral Determinants of Business Cycle Synchronization ................................................10
- 3. Asymmetric Impact of Remittances.....................................................................................12
- 4. Asymmetric Impact of Remittances: Bilateral Data ............................................................12
- 5. Additional Controls and Assessing the Bias due to Selection Based?.................................14
- 6. Instrumental Variable Estimates: Aggregate Data ...............................................................17
- 7. Instrumental Variable Estimates: Bilateral Data ..................................................................18

### Appendix
- I. Descriptive Statistics and List of Countries ..........................................................................23
  - A. Aggregate Data .......................................................................................................23
  - B. Bilateral Data...........................................................................................................24

*Source: _wp12251 - References (PDF).*

### Appendix Tables

### Appendix Tables

### I. Introduction and research question
- Worldwide remittance flows are estimated to have exceeded $440 billion in 2010, with the share of developing countries around $325 billion.
- Recorded remittances were nearly three times the amount of official development assistance (ODA) and almost on a par with foreign direct investment (FDI) flows to developing countries.
- In Sub-Saharan Africa, remittances range from less than 1 percent of GDP to 25 percent in Lesotho.
- The paper asks whether remittance flows affect international business cycle synchronization and argues remittances constitute an additional transmission channel of global shocks.

### II. Baseline specifications of the effect of remittances on business cycle synchronization
- Unilateral (aggregate) specification (equation [1]):
  - Dependent variable: ρi,t = coefficient of correlation between country i business cycle at year t and the weighted sum of business cycles of country i migrant host countries.
  - Key regressor: Remittance inflows as share of GDP Rit.
  - Controls X include trade and financial openness.
  - Fixed effects: country (αi) and year (λt).
- Yearly correlation computed as in equation [2]; weighted host-country cycle as in equation [3].
- Bilateral specification (equation [4]):
  - Dependent variable: ρi,j,t = yearly correlation coefficient of business cycles of countries i and j.
  - Key regressor: bilateral remittances received by i and sent from j, expressed as percentage of i's GDP (R i j t).
  - Controls X include bilateral trade intensity, summed financial openness, and asymmetry in production (ASP).
  - Country-pair fixed effects ideally included but often limited by small within-pair observations; standard errors clustered at country-pair level.

### III. Data and measurement
- Sample periods:
  - Aggregate approach: 1980 to 2010.
  - Bilateral specification: 1980 to 2005.
- Remittance data:
  - Aggregate: World Bank Development Indicators, narrow definition (excludes compensation of employees and migrants’ wealth transfers).
  - Bilateral: Lueth and Ruiz-Arranz (2008) panel for 11 developing countries (Bangladesh, Croatia, Indonesia, Kazakhstan, FYR Macedonia, Moldova, Philippines, Serbia and Montenegro, Slovenia, Tajikistan, Thailand).
  - Bilateral dataset: on average ~16 source countries per recipient and ~9 years of data, yielding almost 1,650 bilateral remittance observations.
- GDP measured in PPP terms from WDI; business cycles measured as log deviation of real GDP from Baxter and King (1999) trend (BK arguments 2 and 8). Robustness: Hodrik-Prescott (HP) filter with smoothness parameter 6.25.
- Bilateral trade intensity (TIi,j,t) defined using DOTS: TIi,j,t = (x i j t + m i j t) / (GDP i t + GDP j t) with exports FOB and imports CIF in U.S. dollars.
- Financial openness: Chinn and Ito (2008) index rescaled to be positive definite; bilateral measure = sum of individual country openness.
- Asymmetry in production (ASP i j t) = sum over k of |s k i t − s k j t| for k = {agriculture, industry, services}, using WDI sectoral value added shares.

### IV. Preliminary results (baseline OLS / standardized coefficients)
- Aggregate results (Table 1; standardized coefficients; country and time dummies included):
  - Using HP cycles:
    - Trade openness: 0.161** (2.345)
    - Financial openness: 0.0978*** (2.883)
    - Remittance ratio: 0.101*** (3.006)
    - Observations: 1,853
    - Number of countries: 70
  - Using BK cycles:
    - Trade openness: 0.184** (2.611)
    - Financial openness: 0.0759** (2.213)
    - Remittance ratio: 0.0879** (2.412)
    - Observations: 1,754
    - Number of countries: 70
- Bilateral results (Table 2; standardized coefficients; receiving and sending country fixed effects included; SEs clustered by country-pair):
  - Using HP cycles:
    - Bilateral trade intensity: 0.105** (2.549)
    - [Financial openness in i] + [Financial openness in j]: 0.267*** (4.136)
    - Asymmetries in production between i and j: -0.196*** (-3.148)
    - Bilateral remittance inflows-to-GDP in i sent from j: 0.0659** (2.571)
    - Observations: 1,541
    - Country pairs: 154
  - Using BK cycles:
    - Bilateral trade intensity: 0.0711* (1.961)
    - [Financial openness in i] + [Financial openness in j]: 0.218*** (3.361)
    - Asymmetries in production between i and j: -0.255*** (-3.377)
    - Bilateral remittance inflows-to-GDP in i sent from j: 0.0627*** (3.817)
    - Observations: 1,116
    - Country pairs: 111
- Interpretation: Remittances significantly increase business cycle synchronization; magnitude similar to financial openness and smaller than trade openness.

### V. Asymmetric effects of remittances
- Hypothesis: remittances transmit downturns in sending countries more strongly than upswings.
- Aggregate asymmetric specification (Table 3):
  - Using HP cycles:
    - Trade openness: 0.164** (2.233)
    - Financial openness: 0.105*** (3.081)
    - Lagged remittance ratio * 1[y* < 0]: 0.0668*** (2.663)
    - Lagged remittance ratio * 1[y* > 0]: 0.0511 (1.140)
    - Observations: 1,773; Number of countries: 70
  - Using BK cycles:
    - Trade openness: 0.195** (2.606)
    - Financial openness: 0.0919** (2.503)
    - Lagged remittance ratio * 1[y* < 0]: 0.0610** (2.291)
    - Lagged remittance ratio * 1[y* > 0]: 0.0100 (0.243)
    - Observations: 1,674; Number of countries: 70
  - Conclusion: aggregate results do not reject that spillovers are stronger during slowdowns; little transmission of upswings in aggregate.
- Bilateral asymmetric specification (Table 4):
  - Using HP cycles:
    - Bilateral trade intensity: 0.101** (2.334)
    - [Fin. openness in i] + [Fin. openness in j]: 0.256*** (4.022)
    - Asymmetries in production between i and j: -0.230*** (-2.893)
    - Lagged bilateral remittance ratio * 1[y* < 0]: 0.0429*** (2.940)
    - Lagged bilateral remittance ratio * 1[y* > 0]: 0.0392** (2.170)
    - Observations: 1,452; Country pairs: 154
  - Using BK cycles:
    - Bilateral trade intensity: 0.0740* (1.859)
    - [Fin. openness in i] + [Fin. openness in j]: 0.207*** (3.275)
    - Asymmetries in production between i and j: -0.306*** (-3.530)
    - Lagged bilateral remittance ratio * 1[y* < 0]: 0.0413*** (4.047)
    - Lagged bilateral remittance ratio * 1[y* > 0]: 0.0234** (2.561)
    - Observations: 1,072; Country pairs: 111
  - Conclusion: bilateral results show both upswings and downturns are transmitted, but dependency on remittances is more effective in channeling downturns.

### VI. Identification and causal inference
- Two approaches to assess causality:
  A. Selection on observables vs. unobservables (Altonji et al. (2005) methodology)
    - Restricted covariates: remittances, trade openness, financial openness, asymmetry in production.
    - Full covariates add: banking crisis dummies, natural disasters, oil rents-to-GDP, inflation rates, differences in government consumption growth rate, reserve money growth rate.
    - Computed Altonji ratio (how much stronger selection on unobservables relative to observables must be to explain away the effect) (Table 5):
      - Cycles: HP Aggregate data — Baseline coefficient 0.101*** (3.006); Additional controls coefficient 0.097* (1.77); Altonji ratio: 28.7
      - Cycles: BK Aggregate data — Baseline 0.0879** (2.412); Additional 0.075* (1.78); Altonji ratio: 6.1
      - Cycles: HP Bilateral data — Baseline 0.0659** (2.571); Additional 0.059* (1.66); Altonji ratio: 8.3
      - Cycles: BK Bilateral data — Baseline 0.0627*** (3.817); Additional 0.049*** (3.80); Altonji ratio: 3.8
    - Interpretation: ratios range from 3.8 to 28.7; selection on unobservables would have to be at least three times stronger than selection on observables to fully explain away the OLS estimates.
  B. Two-stage approach addressing reverse causality (Brückner (2011)-inspired)
    - First stage (equation [7]): remittances regressions include causal responses of remittances to real GDP changes in sending and receiving countries; real GDP growth in recipient instrumented by log(terms of trade).
    - Second stage (equation [8]): use residual variation in remittances (portion not driven by the two cycles) as instrument for remittance ratio to estimate its effect on synchronization.
    - Aggregate IV results (Table 6; standardized coefficients; country and time dummies included):
      - First-stage: log(Terms of trade) positively correlated with log(Real GDP) in receiving economies: coefficient 0.0497*** (2.982).
      - Second-stage IV (HP sync): Remittance instrument 0.409*** (9.127); Remittance-to-GDP (instrumented) 0.157** (2.410) appears as determinant of sync in alternative specs.
      - Trade and financial openness remain positive and significant in IV second-stage.
      - Observations and counts: Number of countries generally 69; Observations: 1,793 (varies by column).
    - Bilateral IV results (Table 7; standardized coefficients; SEs clustered at country-pair level):
      - First-stage: log(Terms of trade) strongly positive with log(Real GDP) of recipient: 1.844*** (11.29).
      - Remittance instrument significant: 0.561*** (1,055) and 1.020*** (1,335) in different columns.
      - Second-stage IV: Bilateral remittance-to-GDP (instrumented) 0.139*** (4.641) and 0.0622*** (4.016) appear as determinants of sync in different columns.
      - Bilateral trade intensity and summed financial openness remain positive and significant; asymmetries in production negative and significant.
      - Number of country pairs: 116 (varies by column); Observations: 1,088 and 923 in different columns.
    - Interpretation: IV estimates confirm a positive and significant causal effect of remittance flows on business cycle synchronization; IV estimates are generally larger than OLS, consistent with OLS being biased downwards by countercyclical remittances.

### VII. Key findings and statistics (enumerated)
- Remittance significance:
  - Aggregate OLS (HP): remittance ratio 0.101*** (3.006); (BK) 0.0879** (2.412).
  - Bilateral OLS (HP): bilateral remittance inflows-to-GDP 0.0659** (2.571); (BK) 0.0627*** (3.817).
- Asymmetry:
  - Aggregate asymmetric OLS (HP): lagged remittance ratio * 1[y* < 0]: 0.0668*** (2.663); * 1[y* > 0]: 0.0511 (1.140).
  - Bilateral asymmetric OLS (HP): lagged bilateral remittance ratio * 1[y* < 0]: 0.0429*** (2.940); * 1[y* > 0]: 0.0392** (2.170).
- Causal evidence:
  - Altonji ratios: 28.7 (HP aggregate), 6.1 (BK aggregate), 8.3 (HP bilateral), 3.8 (BK bilateral).
  - IV first-stage relevance: terms of trade significant predictor of recipient log(Real GDP) (e.g., 0.0497*** (2.982) aggregate; 1.844*** (11.29) bilateral).
  - IV second-stage: remittance instruments significant (e.g., 0.409*** (9.127) aggregate instrument coefficient).

### VIII. Policy implications and recommendations
- Remittance flows are a distinct and important channel of integration and vulnerability:
  - Developing countries receiving sizable remittance inflows tend to be more connected and vulnerable to external shocks than suggested by traditional measures (trade and capital account openness) alone.
- Implications for measurement and surveillance:
  - The concept of external openness should be expanded to account for the remittances channel.
  - Measures of openness and integration that include remittance flows would provide a more accurate gauge of exposure to remittance fluctuations and global shocks.
- Implications for policymakers:
  - Policymakers should consider remittances when designing tools to manage external spillovers and to assess vulnerability to global downturns.
  - Awareness that remittance dependency tends to transmit downturns more strongly than upswings should inform macroeconomic and fiscal resilience strategies.

*Source: _wp12251 - Appendix Tables (IMF Working Paper) — Appendix Tables content unit.*

### REFERENCES

### _wp12251 - REFERENCES

### References
- Abdih, Y., Barajas, A., Chami, R., and Ebeke, C. (2012) Remittance channel and Fiscal Impact in the Middle East, North Africa, and Central Asia. IMF Working Papers, 12/104, International Monetary Fund.
- Abdih, Y., R. Chami, J. Dagher, and P.Montiel. (2011) Remittances and Institutions: Are Remittances a Curse? World Development, 40(4), 657–666.
- Altonji, J., Elder, T., and Taber, C. (2005) Selection on Observed and Unobserved Variables: Assessing the Effectiveness of Catholic Schools. Journal of Political Economy, 113: 151–84.
- Amuedo-Dorantes, C. and S. Pozo. (2004). Workers’ Remittances and the Real Exchange Rate: A Paradox of Gifts. World Development, 32(8), 1407–1417.
- Barajas, A., Chami, R., Fullenkamp, C., Montiel, P. and Gapen, M. T. (2009) Do Workers’ Remittances Promote Economic Growth?, IMF Working Papers 09/153, International Monetary Fund.
- Barajas, A., R. Chami, Hakura, D., and P.Montiel. (2011) Workers’ Remittances and the Equilibrium Real Exchange Rate. Economía, 11(2), 45–94.
- Beine, M., Lodigiani, E., and Vermeulen, R. (2011). Remittances and Financial Openness, DNB Working Papers 317, Netherlands Central Bank, Research Department.
- Bhaskara Rao, B. and Mainul Hassan, G. (2011) A Panel Data Analysis of the Growth Effects of Remittances, Economic Modelling, Volume 28, 1–2, 70–1709.
- Brückner, M. (2011) On the Simultaneity Problem in the Aid and Growth Debate, Journal of Applied Econometrics, forthcoming.
- Calderon, C., Chong, A., Stein, E., (2007) Trade Intensity and Business Cycle Synchronization: Are Developing Countries Any Different?, Journal of International Economics 71, 2–21.
- Catrinescu, N., Leon-Ledesma, M., Piracha, M. and Quillin, B. (2009) Remittances, Institutions, and Economic Growth, World Development, 37, 81–92.
- Cerqueira, P. A. and R., Martins, (2009). Measuring the Determinants of Business Cycle Synchronization Using a Panel Approach, Economics Letters, 102(2), 106–108.
- Chami, R., Fullenkamp, C. and Jahjah, S. (2003) Are Immigrant Remittance Flows a Source of Capital for Development? IMF Working Paper 03/189 (Washington: International Monetary Fund).
- Chami, R., Fullenkamp, C. and Jahjah, S. (2005) Are Immigrant Remittance Flows a Source of Capital for Development, IMF Staff Papers, 52(1), 5–581.
- Chinn, M., and Ito, H. (2008) A New Measure of Financial Openness, Journal of Comparative Policy Analysis, 10(3), 309–322.
- Clark, T.E., and van Wincoop, E., (2001). Borders and Business Cycles. Journal of International Economics 55, 59–85.
- Combes, J-L. and Ebeke, C. (2011) Remittances and Household Consumption Instability in Developing Countries, World Development, 39(7), 1076–1089.
- Ebeke, C. (2011) Remittances, Countercyclicality, Openness, and Government Size, Louvain Economic Review, 77(4), pages 89–114.
- Ebeke, C. (2012a) Do Remittances Lead to a Public Moral Hazard in Developing Countries? Journal of Development Studies, 48(8), 1009–1025.
- Ebeke, C. (2012b) How Do Remittance Flows Behave According to the Business Cycles of the Sending Countries? Unpublished mimeo.
- Ebeke, C. and Combes, J-L. (2012) Do Remittances Dampen the Effects of Natural Disasters in Developing Countries? Applied Economics, forthcoming.
- Flood, R. and Rose, A. K., (2010). Inflation Targeting and Business Cycle Synchronization, Journal of International Money and Finance, 29(4), 704–727.
- Frankel, J. 2011. Are Bilateral Remittances Countercyclical?, Open Economies Review, 22(1), 1–16.
- Frankel, J. and Rose, A., (1998). The Endogeneity of the Optimum Currency Area Criteria. The Economic Journal 108, 1009–1025.
- Frankel, J.A., and Rose, A.K., (1997). Is EMU More Justifiable Ex Post than Ex Ante? European Economic Review 41, 753–760.
- Giuliano, P. and Ruiz-Arranz, M. (2009) Remittances, Financial Development, and Growth, Journal of Development Economics, 90, 144–152.
- Glytsos, N. (2005) The Contribution of Remittances to Growth: A Dynamic Approach and Empirical Analysis, Journal of Economic Studies, 32, 6, 468–496.
- Imbs, Jean (2004) Trade, Finance, Specialization and Synchronization Review of Economics and Statistics, August 86(3), 723–734.
- Inklaar, R., Jong-a-Pin, R., and de Haan, J. (2008) Trade and Business Cycle Synchronization in OECD Countries - a Re-examination European Economic Review, 52(4), 646-666.
- Javorcik, B., Ozden, C., Spatareanu, M., and Neagu, C. (2011) Migrant Networks and foreign Direct Investment, Journal of Development Economics, 94(2): 231–241.
- Kalemli-Ozcan, S., Papaioannou, E., Peydró, J. L.., (2009) Financial Regulation, Financial Globalization and the Synchronization of Economic Activity, NBER Working Papers 14887.
- Kose, M., Prasad, E., Terrones, M., (2003) How Does Globalization Affect the Synchronization of Business Cycles. American Economic Review 93, 57–62.
- Kugler, M. and Rapoport, H. (2011) Migration, FDI, and the Margins of Trade, CID Working Paper No. 222, Center for International Development at Harvard University.
- Laeven, L. and Valencia, F. (2010) Resolution of Banking Crises: The Good, the Bad, and the Ugly, IMF working paper 10/146.
- Lueth, E. and M. Ruiz-Arranz. (2008) Determinants of Bilateral Remittance Flows. The B.E. Journal of Macroeconomics, 8(1) (Topics), Article 26.
- Neagu, I. and M. Schiff. (2009) Remittance Stability, Cyclicality and Stabilizing Impact in Developing Countries World Bank Policy Research Working Paper 5077, Washington, DC.
- Nunn, N. and Wantchekon, L. (2011) The Slave Trade and the Origins of Mistrust in Africa. American Economic Review, 101, 3221–3252.
- Ratha, D. (2009) Remittances: A Lifeline to Poor Countries. Finance and Development, December 2009. International Monetary Fund.
- Ravn, M., and Uhlig, H. (2002) On Adjusting the Hodrick–Prescott Filter for the Frequency of Observations. The Review of Economics and Statistics, 84, 371–376.
- Rose, A., Engel, C., (2002) Currency Unions and International Integration. Journal of Money, Credit, and Banking, 34, 1067–1089.
- Tapsoba, S. J. (2009) Trade Intensity and Business Cycle Synchronicity in Africa, Journal of African Economies, 18(2), 287–318.
- World Bank (2011) Migration and Remittances Factbook 2nd Edition, The World Bank Group.

### Appendix I. Descriptive statistics and list of countries

#### A. Aggregate Data — Table A1. Descriptive Statistics
- Synchronization coefficient (HP): Obs 2100; Mean 0.21; Std. Dev. 1.37; Min -13.99; Max 1
- Synchronization coefficient (BK): Obs 1958; Mean 0.18; Std. Dev. 1.42; Min -14.28; Max 1
- Trade-to-GDP ratio: Obs 2058; Mean 0.71; Std. Dev. 0.40; Min 0.06; Max 2.80
- Financial openness: Obs 2075; Mean 1.62; Std. Dev. 1.31; Min 0.16; Max 4.48
- Remittance-to-GDP ratio: Obs 1907; Mean 0.02; Std. Dev. 0.04; Min 0.00; Max 0.24
- Terms of trade (in log): Obs 1999; Mean 4.65; Std. Dev. 0.35; Min 2.36; Max 6.01

#### A. Aggregate Data — Table A2. List of Countries, 70
- Algeria; Argentina; Bangladesh; Belize; Benin; Bolivia; Botswana; Brazil; Burkina Faso; Burundi; Cameroon; Chile; China; Colombia; Congo, Rep.; Costa Rica; Cote d'Ivoire; Dominican Republic; Ecuador; Egypt; El Salvador; Fiji; Gabon; Gambia; Ghana; Guatemala; Guinea-Bissau; Haiti (not listed in source — omitted); Honduras; India; Indonesia; Iran; Jamaica; Jordan; Kenya; Lesotho; Madagascar; Malawi; Mali; Mauritania; Mauritius; Mexico; Morocco; Nepal; Nicaragua; Niger; Nigeria; Pakistan; Papua New Guinea; Panama; Paraguay; Peru; Philippines; Rwanda; Senegal; Seychelles; Sierra Leone; Sri Lanka; St. Vincent and the Grenadines; Sudan; Suriname; Syrian Arab Rep.; Thailand; Togo; Tunisia; Turkey; Uruguay; Venezuela, RB; Zambia

(Note: the source lists 70 countries in two-column layout; entries above reflect the sequence and names as presented in the source.)

#### B. Bilateral Data — Table B1. Descriptive Statistics
- Synchronization coefficient (HP): Obs 1569; Mean 0.29; Std. Dev. 1.36; Min -13.75; Max 1
- Synchronization coefficient (HP) [second entry]: Obs 1140; Mean 0.20; Std. Dev. 1.45; Min -12.01; Max 1
- Bilateral trade intensity: Obs 1569; Mean 0.00; Std. Dev. 0.01; Min 0.00; Max 0.05
- Financial openness in i and j: Obs 1569; Mean 5.44; Std. Dev. 1.52; Min 1.01; Max 8.16
- Asymmetry of production: Obs 1569; Mean 11.50 6.00; Std. Dev. 0.74; Min 47.83 (as presented in source)
- Bilateral remittance ratio: Obs 1569; Mean 0.00; Std. Dev. 0.01; Min 0.00; Max 0.20
- Terms of trade (in log): Obs 1088; Mean 4.74; Std. Dev. 0.12; Min 4.45; Max 4.96

#### B. Bilateral Data — Table B2. List of Countries, 10
- Bangladesh
- Croatia
- Indonesia
- Kazakhstan
- Macedonia
- Moldova
- Philippines
- Slovenia
- Tajikistan
- Thailand

*Source: _wp12251 - REFERENCES (source PDF content provided).*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp12251.pdf_