## _wp1644

## Source details

**Canonical URL:** [_wp1644](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2016/_wp1644.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2016/_wp1644.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2016/_wp1644.pdf.json)

---

### I. INTRODUCTION: scope and summary findings
- International inflows of workers’ remittances rose to approximately US $583 billion in 2014.
- Average workers’ remittances-GDP ratio for all developing countries over 1980-2012: 1.29%.
- Comparable external flow ratios (1980-2012): foreign direct investment 1.95%, other private capital 1.68%, official transfers 0.80.
- Over 1980-2012, remittances exceeded 1% of GDP (on average) for over 74 countries; 7 countries had average remittances-GDP ratios of 15% or higher.
- Country example: in 2000 Jordan remittances accounted for about 20 percent of GDP.
- Relative stability: since the late 1990s remittances have surpassed official transfers and have been more stable year-to-year than private capital or official aid, suffering a milder contraction during the Global Financial Crisis.

- Macro literature summarized:
  - Exchange rates: persistent remittances exert upward pressure on the long-run real exchange rate, producing Dutch Disease effects and declining tradable-sector competitiveness.
  - Fiscal policy: remittances alter the tax base, can increase seignorage and private savings via banking system effects; conventional debt sustainability analysis should be modified for recipient countries.
  - Long-run growth: mixed evidence (null findings and positive effects under different methods and samples).

- Paper’s main question and contribution:
  - Do remittances affect the operability of monetary transmission—especially the bank lending channel—in typical remittance-recipient countries?
  - Two opposing bank effects identified:
    - Positive: remittances expand bank balance sheets, providing a stable, interest-rate insensitive funding source that could strengthen monetary transmission.
    - Negative (dominant empirically): due to asymmetric information and weak institutions, banks do not expand private credit; they hold larger shares of liquid assets and government securities, interbank markets underdevelop, and banks’ marginal cost of funds de-link from the policy rate—weakening monetary transmission.
  - Empirical finding: as remittances increase, transmission of policy rate changes to domestic credit becomes weaker.
  - Policy implication: influences choice of macroeconomic policy framework for remittance-recipient countries and relates to observed preference for fixed exchange rate regimes among remittance recipients.

### II. STYLIZED FACTS: remittances and bank balance sheets (1997–2007 focus)
- Bank balance-sheet size and composition (cross-country comparisons by remittances-to-GDP groups):
  - Banks in higher remittances-to-GDP countries have, on average, higher total deposits to GDP than in lower-remittance countries.
  - Banks in high remittance countries mostly have lower total credit to GDP and lower total assets to GDP.
  - Net: remittance-recipient banks show more liabilities but relatively less credit and assets.

- Liabilities composition:
  - Across all countries, about one-third of deposits are short-term and two-thirds long-term.
  - Countries with average remittances-to-GDP over 5 percent: share of short-term deposits falls to less than 25 percent.
  - The mean difference between “substantial remittance recipients” (remittances-GDP of at least 0.5 percent) and others is statistically significant.
  - Robust to financial development quartiles: banks in higher remittances-to-GDP countries still have higher long-term deposit ratios across levels of financial development.

- Assets composition:
  - Banks in higher remittances-to-GDP countries have, on average:
    - higher ratio of liquid assets to total assets;
    - slightly higher ratio of credit to government to total assets;
    - higher reserves to total assets.
  - Excess reserves (actual minus required using average reserve ratios): higher excess reserves to total assets in higher remittance-to-GDP countries.
  - Most results are statistically significant.

- Volatility and stability:
  - Both total deposits and short-term or long-term deposits are clearly less volatile the higher the remittances-to-GDP ratio in a country.
  - Both total assets and credit to the government reveal lower standard deviations, the higher the remittances-to-GDP ratio.
  - Net implication: bank balance sheets in large-remittance countries appear much more stable.

- Determinants and behavioral implications:
  - Regression evidence: remittances are countercyclical with respect to recipient country income and procyclical with respect to sending country income; remittances are insensitive to interest rate differentials.
  - A substantial portion of remittance inflows end up in the banking system.
  - Many remittance-recipient countries share low quality of institutions, high credit risk and borrower opaqueness, and informational asymmetry problems.
  - Behavioral outcome: despite stable, long-term deposits, banks empirically hold more liquid assets, government securities, and excess reserves—reluctant to extend risky credit.
  - Fiscal correlation: countries with higher remittances-to-GDP tend to have larger government deficits (including and excluding oil exporters).
  - Bank profitability/competition: banks in remittance-recipient countries tend to have higher net interest margins and net interest rate spreads; measures of bank competitiveness (concentration, H-statistic, Lerner index) do not yield clear-cut systematic differences.

### III. THEORETICAL MODEL: banking sector with remittances (overview)
- Model foundations:
  - Based on Mishra et al. (2014) monopolistically competitive banking model suitable for low financial development settings.
  - Remittances (Rem) enter banks' liabilities as part of deposits (D) and are modeled as interest-rate insensitive, reducing deposit-rate sensitivity to deposits.
  - Bank portfolio: loans to private sector (L), government securities (B), reserves (R) with identity B = D − L − R.
  - Two-tier convex lending cost: increasing and convex in L with an inflection at L* (distinguishing loans to large/transparent firms versus opaque borrowers).
  - Fixed required reserve ratio: R = ρD.
  - Banks have market power in loan and deposit markets.

- Mechanism:
  - Interest-insensitive remittances weaken deposit-rate responsiveness to policy-rate changes.
  - Weaker deposit-rate reaction reduces the response of lending rates to policy-rate changes through the bank’s cost-of-funds channel.
  - The model shows formally that remittances impede transmission of the policy rate to the lending rate via deposit-rate insulation.

### IV. EMPIRICAL STRATEGY
- Objective: estimate how remittances magnitudes affect the strength of the bank lending channel (pass-through of central bank discount rate changes to bank lending rates).
- Main specification:
  - Dependent variable: monthly change in lending rate (nominal).
  - Key regressor: monthly change in policy rate (nominal, discount rate).
  - Interaction regressors: remittances-to-GDP (log of five year moving average) interacted with policy-rate change; bank competitiveness and institutional quality dummies (low competitiveness/institutions = 1 if below median).
  - Five year averages used to mitigate endogeneity concerns and monthly remittance-to-GDP unavailability.
  - Robustness: contemporaneous and lagged ratios; alternative measures of competitiveness and institutional quality; deposit-rate regressions; inclusion of high-income countries; alternative empirical approaches (structural panel VAR impulse responses and simple co-movement coefficients).

- Data sources:
  - Remittances: IMF Balances of Payments Statistics (BOPS), workers' remittances category.
  - Interest rates: IMF IFS (lending rate IFS line 60p; policy/discount rate IFS line 60a).
  - Bank competitiveness: GFDD.
  - Institutional quality: Transparency International CPI, CPIA transparency/accountability/corruption rating, Worldwide Governance Indicators regulatory quality.
  - Sample frequency: monthly regressions use monthly data for 1990-2013; interactive variables based on annual figures; stylized facts focus 1997-2007 to minimize Global Financial Crisis effects.

### V. EMPIRICAL RESULTS: panel fixed effects, VARs, and robustness
- Baseline lending-channel strength:
  - Specification (1): a 1 percentage point change in the discount rate is associated with a contemporaneous lending-rate effect with point estimate 0.3 percentage points (reported as 0.284*** (0.088) in Table 8, column (1)).

- Interaction and remittance effects:
  - Baseline interaction result reported: 0.16 with an additional negative effect of remittances-to-GDP of -0.11 in one specification.
  - Across specifications the remittances × policy-rate interaction coefficients are negative and statistically significant in most specifications (examples preserved exactly where reported: -0.719*** (0.017) in one column formatting, and additional reported interaction magnitudes -0.110*** (0.029), -0.114*** (0.028), -0.097*** (0.035), -0.113*** (0.028), -0.107*** (0.030), -0.096** (0.039)).
  - Conclusion: remittance inflows reduce the effectiveness of the lending channel—higher remittances weaken the link between discount and lending rates.

- Thresholds where lending-channel pass-through is zero:
  - Threshold remittances-to-GDP ratios at which the overall reaction of lending rates to the discount rate is zero:
    - Depending on specification: 4-6 percent of GDP.
    - Specification (5), excluding unchanged-lending-rate periods: 6.4 percent.
    - With wider country sample including non-recipients: threshold rises to 5-7 percent of GDP.
  - Examples of converted ln(remittances-to-GDP) thresholds reported: 1.47 → 4.36%, 1.33 → 3.79%, 1.85 → 6.36%, 1.35 → 3.87%, 1.65 → 5.23%, 2.03 → 7.62%.

- Robustness checks and alternative tests:
  - Results hold when extending sample to advanced economies (columns (6)–(8)), and when proxying non-recipients via very low emigration (average ratio less than 0.02).
  - Interacting LIC dummy: LICs generally have weaker transmission; specification (3) shows transmission less than half as strong in LICs relative to non-LICs.
  - Controlling for per capita income: per capita income is positively associated with transmission strength, but remittance interaction remains negative and robust.
  - Deposit-rate regressions (Table 10): remittances similarly weaken transmission from policy to deposit rates (examples preserved: change in policy rate 0.134*** (0.018); interaction coefficients include 0.140*** (0.003) in one column and negative interactions -0.090*** (0.032), -0.098*** (0.028), -0.081*** (0.028), -0.098*** (0.028), -0.085** (0.033), -0.066* (0.039) in others).
  - Structural panel VAR (Mishra et al. (2014)) impulse-response regressions (Table 11): remittances-to-GDP coefficients by quarter largely negative (examples: 1st quarter -0.017 (0.062); 2nd quarter -0.006 (0.048); average -0.006 (0.044); maximum -0.072^^ (0.044)); international financial integration negative and sometimes significant.
  - Supplementing IRs with simple co-movement coefficients expands sample to 71 countries: remittances-to-GDP enters with negative sign and is statistically significant in several specifications (Table 12 examples: -0.006, -0.023** (0.007), -0.017 (0.027), -0.016 (0.030), -0.024** (0.009), -0.028** (0.010)).

- Selected quantitative regression outputs (preserved exactly as reported):
  - Table 8 selected coefficients (dependent variable: monthly changes in lending rate):
    - Change in policy rate: column (1) 0.284*** (0.088); column (2) 0.766*** (0.014); column (3) 0.162*** (0.043).
    - Remittances to GDP × change in policy rate: -0.719*** (0.017) [formatting note in source], and reported interaction magnitudes -0.110*** (0.029), -0.114*** (0.028), -0.097*** (0.035), -0.113*** (0.028), -0.107*** (0.030), -0.096** (0.039).
    - Observations and countries examples: Observations 17,707 (col (1)), 11,294 (col (2)), 5,737 (cols (3) and (4)); Countries 92, 76, 45 respectively. R-squared examples 0.03, 0.04, 0.07.
  - Table 9 (controlling for income): change in policy rate examples 0.138*** (0.043); 0.167** (0.069); 0.189** (0.074). Remittances × change in policy rate examples: -0.129*** (0.034), -0.122*** (0.031), -0.104** (0.040), -0.126*** (0.033), -0.114*** (0.031), -0.108** (0.040). Per capita GDP example coefficient: 0.138*** (0.039). Observations e.g., 6,582; 8,434; 4,065. Countries 55, 56, 56.
  - Table 11 (structural panel VAR IRs): remittances-to-GDP coefficients by quarter examples: 1st quarter -0.017 (0.062); 2nd quarter -0.006 (0.048); 3rd quarter -0.003 (0.044); 4th quarter 0.003 (0.034); average -0.006 (0.044); maximum -0.072^^ (0.044). Observations 46; R-squared 0.10–0.16 ranges.
  - Table 12 (IRs + co-movement): remittances-to-GDP columns examples: -0.006; -0.023** (0.007); -0.017 (0.027); -0.016 (0.030); -0.024** (0.009); -0.028** (0.010). Observations 71; R-squared examples 0.13, 0.20, 0.09.

- Bank balance-sheet summary statistics (unweighted averages, 1997–2007 unless noted):
  - All countries (time frame 1997-2007): Total Deposits to GDP 39.86% (162 countries); Total Credit to GDP 58.84% (122); Total Assets to GDP 81.21% (123).
  - Remittances-GDP >= 5% group: Total Deposits to GDP 39.17% (31); Total Credit to GDP 58.29% (23); Total Assets to GDP 60.67% (24).
  - Liquid assets to total assets: All Countries 20.05 (123); Emerging & Developing 23.07 (105); Remittances-GDP >= 5% 28.67 (23).
  - Excess reserves to total assets: All Countries 4.16% (101); Emerging & Developing 4.53% (94); Remittances-GDP >= 5% 5.94% (23).
  - Volatility: standard deviation of log(total deposit) All Countries 0.293 (169); remittances-GDP >= 5% group 0.221 (31).
  - Bank competitiveness: Lerner index All Countries 0.24 (131); Emerging & Developing 0.26 (103).
  - Government balance to GDP: All Countries -1.12 (154); Remittances-GDP >= 5% -3.20 (27).
  - Net interest margin of banks All Countries 5.20% (163).

### VI. MECHANISM, INTERPRETATION, AND POLICY RECOMMENDATIONS
- Mechanism reiterated:
  - Interest-insensitive remittance inflows expand deposit liabilities; in weak institutional/informational environments banks prefer liquid assets, government securities, and excess reserves rather than risky private lending.
  - This abundant liquidity de-links banks’ marginal cost of loanable funds from policy rate movements and weakens the bank lending channel of monetary policy.

- Policy recommendations:
  - First-best (structural): reduce information asymmetries; improve property rights and contract enforcement to encourage banks to lend remittance-funded deposits.
  - Short- to medium-term instrument choices when policy-rate transmission is weak:
    - Use quantitative targets in place of short-term policy rate if the rate is ineffective.
    - Raise reserve requirements to levels high enough to bind and eliminate excess reserves (weighing initial contraction in lending against greater policy effectiveness).
    - Tax excess reserves to encourage banks to expand lending, with careful monitoring to avoid excessive risk-taking.
    - If independent monetary policy is too challenging, consider retaining a more managed exchange rate regime.
  - Caution: remittances are welfare-enhancing (poverty alleviation, insurance); policies must balance macroeconomic stabilization with the social benefits of remittance flows.
  - Note: policies addressing private capital flows may not directly apply to remittances; remittances should be considered alongside capital flows in policy trilemma assessments.

*Italic source attribution: Content drawn exclusively from the provided IMF PDF excerpt.*

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

### _wp1644 - REFERENCES .............................................................................................................

### I. INTRODUCTION
- International inflows of workers’ remittances have grown from negligible amounts in 1980 to approximately US $583 billion in 2014 (World Bank Migration and Development Brief, 24, 2015).
- Average workers’ remittances-GDP ratio for all developing countries over 1980-2012 is 1.29%; comparable external flow ratios over the same period: foreign direct investment 1.95%, other private capital 1.68%, official transfers 0.80.
- Workers’ remittances exceeded 1% of GDP (on average) for over 74 countries during 1980-2012; 7 of these countries had average workers’ remittances-GDP ratios of 15% or higher.
- Country example: Jordan was the fifth-largest recipient of remittances over 1980-2014; in 2000 remittances accounted for about 20 percent of GDP, more than double FDI inflows, about four times the amount of other private capital flows, and more than three times the official transfers received.
- Since the late 1990s remittances have surpassed official transfers and in some years have been comparable to total non-FDI private capital entering developing countries.
- Compared to private capital or official aid flows, remittances have proved more stable year-to-year and suffered a much milder contraction as a result of the Global Financial Crisis.

- Literature findings on macroeconomic consequences:
  - Exchange rates: persistent remittance inflows exert upward pressure on the long-run real exchange rate, producing Dutch Disease effects and declining competitiveness of tradable sectors (Barajas et al. (2011), Hassan and Holmes (2013), Lartey et al. (2012), Maklouhf and Mughal (2013)).
  - Fiscal policy: remittances alter the tax base, can indirectly increase seignorage and private savings via banking system effects; conventional debt sustainability analysis should be modified for recipient countries (Abdih et al. (2009)); fiscal implications from the 2009 cutback in worldwide remittances estimated for several recipient countries (Abdih et al. (2012a)); remittances can adversely affect institutions via revenue-base expansion and household substitution away from government services (Abdih et al. (2012b)).
  - Long-run growth: mixed evidence
    - No positive effect found using instrumental variables for 67 countries over 1991-2005 (Barajas et al. (2009)).
    - No significant Granger-causality from remittances to growth in 20 Sub-Saharan African countries 1980-2007 (Ahamada and Dramane (2013)).
    - Measurement error, weak power of panel regressions, and offsetting effects of outward migration may explain null findings (Clemens and McKenzie (2014)).
    - Positive effects in low-income countries found using panel GMM for 1990-2006 (Benmamoun and Lehnert (2013)); positive impact in Latin America and Caribbean 1990-2007 with stronger effects in lower-income countries and where institutions/financial development are stronger (Ramirez (2013)); remittances may relax financing constraints in financially underdeveloped countries (Giuliano and Ruiz-Arranz (2009)).

- Main theme: despite welfare benefits to households, remittance inflows pose macro policy challenges—upward pressure on the real exchange rate, weakening institutions around fiscal policy, and doubtful effects on long-term growth.

- Monetary policy literature and gaps:
  - Existing DSGE and general equilibrium studies derive optimal rules or consider welfare implications under assumptions of well-functioning financial systems and operable transmission mechanisms (Chami et al. (2007), Vacaflores (2012), Mandelman (2013)).
  - Many remittance-recipient countries do not satisfy these assumptions; monetary transmission is often substantially weaker in low-income and emerging markets due to underdeveloped financial sectors and markets (Mishra et al. (2012, 2013, 2014)).
  - Monetary transmission channels include the interest rate channel, exchange rate channel, asset price channel, bank lending channel, and balance sheet channel; in low-income countries the bank lending channel is likely the most important.

- This paper’s contribution:
  - Focuses on whether remittances affect the operability of monetary transmission, especially the bank lending channel, in typical remittance-recipient countries.
  - Finds remittances have two opposing impacts on banks:
    - Positive: expand bank balance sheets by providing a stable and largely interest-rate insensitive source of funds, potentially strengthening monetary transmission via enhanced financial intermediation.
    - Negative (dominant): due to asymmetric information, weak institutions/regulation, and lack of transparent borrowers, increased liabilities are not matched by private credit expansion; banks hold larger shares of liquid assets and government securities, interbank markets fail to develop, and banks’ marginal cost of funds de-links from the policy rate, weakening monetary transmission.
  - Empirical analysis confirms that as remittances increase, transmission of policy rate changes to domestic credit becomes weaker.
  - Implication: affects choice of macroeconomic policy framework for remittance-recipient countries and helps explain the Singer (2010) finding of a preference for fixed exchange rate regimes among remittance recipients.

- Paper organization:
  - Section II: stylized facts regarding remittances and bank balance sheets.
  - Section III: model of monetary policy in a remittance-recipient country.
  - Section IV: main results from panel fixed effects regressions.
  - Section V: results from an alternative empirical approach based on previously estimated impulse-responses.
  - Section VI: conclusion.

### II. STYLIZED FACTS: BANK BALANCE SHEETS IN REMITTANCE-RECIPIENT COUNTRIES
- Focus and approach:
  - Focus on how well monetary transmission influences financial market targets, concentrating on the bank lending channel as most likely operable in low-income and emerging countries.
  - Bank lending channel operates via: (i) policy rate changes affect bank balance sheets and marginal cost of funds; (ii) banks pass marginal cost changes to borrowers.
  - Strength of the bank lending channel depends on banking competition, institutional environment (e.g., contract protection), and a direct link between policy rates and banks’ marginal cost of funds (interbank markets are often key).
  - Countries where banks hold large amounts of liquid assets tend to lack active interbank markets, weakening efforts by central banks to influence marginal cost of funds via the interbank market.

- Remittances’ multiple effects on bank lending channel:
  - Remittances can influence banking sector development, competitiveness, institutional environment, and bank holdings of liquid assets.
  - Two opposing effects on bank liquid assets:
    - Banks may not expand credit due to asymmetric information and weak institutions, so remittance-driven liability expansion leads banks to accumulate liquid assets.
    - Alternatively, remittances as stable longer-term funding could reduce maturity mismatch and encourage banks to expand credit.
  - Demand-side effect: remittances could substitute for bank loans (households and small firms less financially constrained), de-linking spending from bank credit supply.

- Empirical design notes:
  - Comparison of size and composition of bank balance sheets and competitiveness between remittance recipients and emerging/developing non-recipients, and across remittance inflow levels.
  - To minimize impact of 2008 global financial crisis, analysis focuses on 1997-2007.
  - Country sample split by average remittances-to-GDP ratio during 1990-2013, if at least five years of data reported.

A. Bank Balance Sheets: Size
- Findings:
  - Banks in countries with higher remittances-to-GDP ratios have, on average, a higher ratio of total deposits to GDP than countries with lower ratios.
  - On the asset side, banks in high remittances-to-GDP countries mostly have lower ratios of total credit to GDP than in countries with low or no remittances.
  - The same holds for total assets to GDP.
  - Overall: banks in remittance recipient countries have more liabilities on their balance sheets but relatively less assets compared to non- and particularly low-remittance recipient countries.

B. Bank Balance Sheets: Liabilities
- Findings:
  - Across all countries, about one-third of deposits are short-term and two-thirds are long-term; shares are approximately the same for emerging and developing countries.
  - Countries with higher remittances-to-GDP ratios tend to hold less short-term deposits, particularly where average remittances-to-GDP is over 5 percent, in which case the share of short-term deposits falls to less than 25 percent.
  - The means difference between “substantial remittance recipients” (remittances-GDP of at least 0.5 percent) and the rest of the world is statistically significant.

- Robustness to financial development:
  - Given correlation between remittance inflows and financial depth, sample is also split into quartiles of financial development measured by average ratio of private credit to GDP.
  - While ratio of short-term to long-term deposits decreases with higher financial depth, banks in countries with higher remittances-to-GDP ratios still have, on average, higher long-term deposit ratios across all levels of financial development.
  - Difference between substantial remittance recipients and other countries remains statistically significant.
  - Interpretation: while part of remittances may be consumed or invested immediately, a considerable portion is deposited for longer term or in foreign currency-denominated accounts.

C. Bank Balance Sheets: Composition of Assets
- Findings:
  - Banks in countries with higher remittances-to-GDP ratios have, on average:
    - higher ratio of liquid assets to total assets;
    - slightly higher ratio of credit to government to total assets;
    - higher reserves to total assets.
  - Excess reserves (estimated as actual reserves minus required reserves computed using average required reserve ratios from IMF survey and following Saxegaard (2006) methodology): higher excess reserves to total assets in higher remittance-to-GDP countries.
  - Most results are statistically significant.
  - Summary: banks in higher remittance-to-GDP countries hold more liquid assets, more reserves—excess reserves in particular—and provide slightly more lending to government.

D. Bank Balance Sheets: Volatility
- (Section begins consideration of volatility of different components of bank balance sheets; content truncated in source.)

*Italic source attribution: Content drawn exclusively from the provided IMF PDF excerpt.*

### 5. On the liability  side, both  total  deposits  and short-term or  long-term  deposits  are clearly  less

### _wp1644 - 5. On the liability  side, both  total  deposits  and short-term or  long-term  deposits  are clearly  less

### Stylized facts on remittances and bank balance sheets
- Higher remittances-to-GDP ratios are associated with lower volatility of liabilities:
  - "both total deposits and short-term or long-term deposits are clearly less volatile the higher the remittances-to-GDP ratio in a country."
- Asset-side stability with higher remittances-to-GDP:
  - "both total assets and credit to the government reveal lower standard deviations, the higher the remittances-to-GDP ratio."
- Net implication:
  - "bank balance sheets in countries that receive large remittances-to-GDP seem to be much more stable than in other countries."

### Determinants and characteristics of remittances
- Regression evidence (Table A2, similar to Chami et al. (2008) and Chami et al. (2009)):
  - "remittances are countercyclical (procyclical) with respect to recipient (sending) country income."
  - "remittances ... are insensitive to interest rate differentials."
- Institutional and banking-sector context:
  - Banking sector remains the main recipient of private sector deposits or savings in these countries.
  - "a substantial portion of remittance inflows end up in the banking system."
  - Many remittance-recipient countries share characteristics: "low quality of institutions, high credit risk and opaqueness of borrowers, and informational asymmetry problems."

### Behavioral implications for banks and public finances
- Funding and portfolio choices:
  - Stable, long-term remittance-funded deposits could, in theory, induce banks to expand lending and take more risks.
  - Empirical stylized facts point the other way: banks in remittance-recipient countries hold "more liquid and government securities" and "excess reserves", reflecting reluctance to extend risky credit given institutional weaknesses.
- Fiscal implications:
  - Countries with higher remittances-to-GDP ratios tend to have larger government deficits (both including and excluding oil-exporting countries).
  - Possible explanations:
    - "interest-insensitive remittance flows into the banking system make the financing of these deficits easier by banks that are reluctant to expand risky lending..."
    - Institutional weakening effect of remittances leading governments to "run larger deficits."
- Bank profitability and competition:
  - Banks in remittances-receiving countries tend to have higher net interest margins and net interest rate spreads (results not strictly monotonic), possibly due to:
    - "relatively low bank competition in remittance recipient countries"
    - "interest-insensitive remittances which allow banks to profit from these characteristics."

### Bank competitiveness measures and findings
- Three measures considered (from GFDD):
  - Bank concentration: assets of three largest banks as share of all commercial banks' assets.
  - H-statistic: elasticity of bank revenues relative to input prices.
  - Lerner index: (output prices − marginal costs)/prices.
- Empirical takeaways:
  - "For all three measures, no clear-cut conclusions can be drawn."
  - "There is a tendency for countries with no remittances to exhibit higher competition ... when we consider our bank concentration measure."
  - Opposite tendency found for the H-statistic.
  - "for the Lerner index, we do not find clear differences between our country groups."

### Aggregate interpretation and transmission implications
- Summary of stylized facts:
  - "Remittances provide stable and interest-insensitive funding for banks, as reflected in more long-term and more stable deposits."
  - "Banks hold more liquid assets and excess reserves and, as previous studies showed, they operate in a generally weaker institutional environment."
  - "These empirical regularities suggest that the interbank market is likely to be less developed and active, and therefore the lending channel will be impaired."
  - "Bank competitiveness ... does not seem to be affected in a systematic way by remittances."
- Demand-side caveat:
  - Remittances could reduce credit demand because "a significant share of households find their financing needs satisfied by remittances and no longer need bank loans."
  - Aggregate-level identification is difficult; the analysis focuses on identifying supply-side effects which "might be reinforced by demand-side factors as well."

### Theoretical model: banking sector with remittances (overview)
- Starting point:
  - Based on Mishra et al. (2014) monopolistically competitive banking model in low financial development settings.
- Key modelling features introduced:
  - Remittances (Rem) enter banks' liabilities as part of deposits (D) and are modeled as interest-rate insensitive, reducing deposit-rate sensitivity to deposits.
  - Bank portfolio: loans to private sector (L), government securities (B), reserves (R) with accounting identity B = D − L − R.
  - Two-tiered, convex cost of lending: cost function is increasing and convex in L, with an inflection at L* (loans to large/transparent firms versus more opaque borrowers).
  - Fixed required reserve ratio: R = ρD.
  - Bank has market power in loan and deposit markets (downward loan demand, upward deposit supply).
- Mechanism highlighted:
  - Remittances mitigate the responsiveness of deposit rates to changes in the policy rate because remittance inflows are interest-insensitive.
  - This weaker deposit-rate reaction transmits to a weaker response of lending rates to policy-rate changes.
  - The model shows formally that "the impact of remittances on the transmission of the policy rate to the lending rate runs through its impact on the bank’s cost of funds, that is, the deposit rate."

### Empirical strategy: panel fixed-effects estimates
- Objective:
  - Test empirically "to what extent the magnitude of remittances affects the strength of the bank lending channel."
  - Examine pass-through of changes in central bank discount rates to bank lending rates.
- Specification features:
  - Dependent variable: monthly change in lending rate (nominal).
  - Key regressor: monthly change in policy rate (nominal).
  - Interaction variables: measures of bank competitiveness and institutional quality, and remittances-to-GDP.
  - Remittances-to-GDP enters as the logarithm of a five year moving average (to mitigate endogeneity concerns and because cross-country monthly remittance-to-GDP data are not available).
  - Dummies: low competitiveness and low institutional quality (constant over time), equal to one if below median.
  - Motivation for five year averages: reduces concerns that remittances proxy for other factors such as business cycle swings; robustness tests with contemporaneous and lagged ratios provide similar results.

### Data sources (as used in the analysis)
- Remittances: workers' remittances category from IMF's Balances of Payments Statistics (BOPS).
- Interest rates: IMF IFS.
  - Lending rate: IFS line 60p.
  - Policy rate ("discount rate"): IFS line 60a.
- Bank competitiveness measures: Global Financial Development Database (GFDD).
- Institutional quality measures: Transparency International corruption perception index, CPIA transparency/accountability/corruption rating, Worldwide Governance Indicators regulatory quality.
- Practical choices:
  - Different institutional/competitiveness measures yield similar results; reported results focus on the Lerner index and regulatory quality (based on data availability).

### Key empirical results (panel regressions, Table 8 summary)
- Sample partitions:
  - Specifications (1)–(5): emerging and developing countries (comparable to Mishra et al. (2012)).
  - Specifications (6)–(8): expand sample to include high-income countries.
- Baseline lending-channel strength:
  - Specification (1): a change in the discount rate by 1 percentage point is linked to a contemporaneous and statistically significant effect on the lending rate with a point estimate of 0.3 percentage points.
- Role of remittances:
  - Specification (2): when remittances are included, they initially appear to enhance the lending channel (increasing the lending-rate response to policy-rate changes).
  - However, once the low-banking-competitiveness dummy is included, "the coefficient on the discount rate drops to around" (text breaks off at that point in the supplied content).
- Overall empirical conclusion (as stated earlier in the empirical overview):
  - "As we will show, the results provide evidence that remittance inflows do reduce the effectiveness of the lending channel."

*Italic: Content extracted from the provided IMF chapter section.*

### 0.16  with  an  additional  negative  effect  of  remittances-to-GDP  of -0.11.  This  result  is  relatively

### _wp1644 - 0.16  with  an  additional  negative  effect  of  remittances-to-GDP  of -0.11.  This  result  is  relatively

### Major empirical findings
- Baseline interaction result reported: 0.16 with an additional negative effect of remittances-to-GDP of -0.11.  
- The interaction coefficient and main results are stable when adding a dummy for low institutional quality (column (4)).  
- Excluding observation periods with unchanged lending rates for at least one year (specification (5)) yields a slightly higher coefficient on the discount rate change and a slightly lower interaction term with remittances.  
- Across specifications, remittances weaken the pass-through from discount/policy rate to lending rates: higher remittances lead to a weaker link between discount and lending rates, suggesting weaker monetary transmission.

### Thresholds for disappearance of bank lending channel
- Threshold remittances-to-GDP ratios at which the overall reaction of lending rates to the discount rate is zero:
  - Depending on specification: 4-6 percent of GDP.
  - Specification (5), excluding unchanged-lending-rate periods: 6.4 percent.
  - With wider country sample including non-recipients: threshold rises to 5-7 percent of GDP.
- Thresholds reported in Table 8 (converted from log remittances-to-GDP by the natural exponential function) and computed given competitiveness and institutional quality.

### Robustness and alternative specifications
- Results hold when extending the sample to advanced countries (columns (6)–(8)), and when adding ten countries as non-recipients proxied by very low emigration (average ratio less than 0.02 using Panel Data on International Migration, 1975-2000).
- Findings hold when:
  - Controlling for low competitiveness and institutional quality.
  - Interacting a LIC dummy with policy rate changes: LICs generally have weaker transmission; in specification (3) transmission from policy to lending rates is less than half as strong in LICs compared to non-LICs.
  - Controlling for per capita income: per capita income is positively associated with transmission strength, but remittance effects remain negative and robust.
  - Using deposit rate as dependent variable (Table 10): remittances similarly weaken transmission from policy to deposit rates.
- Alternative empirical test using structural panel VAR (Mishra et al. (2014)) over 1978-2013:
  - Impulse responses (IRs) for each country show remittances-to-GDP almost always with a negative coefficient, indicating weaker transmission in higher-remittance countries.
  - Across 36 coefficients estimated (quarterly IRs and related metrics) 32 have expected signs.
- Supplementing IRs with simple country-specific co-movement coefficients (monthly OLS of lending rate change on policy rate change) expands sample to 71 countries; remittances-to-GDP consistently enters with negative sign and is statistically significant in most cases.

### Mechanism and interpretation
- Proposed channel: interest-insensitive remittance inflows expand bank balance sheets in recipient countries; banks facing challenging institutional, informational, and high-risk environments prefer safe, liquid assets and lending to government.
  - This creates ample liquidity, delinking banks’ marginal cost of loanable funds from policy rate movements and weakening the bank lending channel.
- Implication for monetary policy trilemma: remittances reduce effectiveness of independent monetary policy irrespective of private capital mobility, suggesting remittances should be considered alongside capital flows when assessing tradeoffs among macroeconomic policies.

### Policy implications and recommendations
- First-best: ameliorate distortions that prevent banks from lending remittance-funded deposits — reduce information asymmetries; improve property rights and contract enforcement to encourage banks to lend rather than hold safe liquid assets.
- Short- to medium-term instrument choices:
  - Use quantitative targets (in place of short-term policy rate) if the rate is an ineffective instrument.
  - Raise reserve requirements to levels high enough to bind and eliminate excess reserves (weighing the initial contraction of bank lending against greater policy effectiveness).
  - Tax excess reserves to encourage banks to expand lending, with careful monitoring to avoid excessive risk-taking.
  - If effective independent monetary policy proves too challenging, consider retaining a more managed exchange rate regime.
- Note: policies addressing private capital flow impacts may not directly apply to remittances; remittances are welfare-enhancing and important for poverty alleviation and insurance, so policy tradeoffs must balance benefits with macroeconomic challenges.

### Key statistics and selected results (preserve source figures)
- Interaction baseline: 0.16 (baseline) and additional remittances-to-GDP effect of -0.11.
- Threshold ranges for remittances-to-GDP that nullify policy-to-lending sensitivity: 4-6 percent; specification (5): 6.4 percent; wider sample: 5-7 percent.
- Table 8 (selected coefficients, dependent variable: monthly changes in lending rate):
  - Change in policy rate (column (1)) 0.284*** (0.088); column (2) 0.766*** (0.014); column (3) 0.162*** (0.043).
  - Remittances to GDP x change in policy rate: -0.719*** (0.017) in column (1) [note: formatting in source shows 0.719*** then -0.110***; preserved content indicates strong negative interactions across columns], and -0.110*** (0.029) in another column; other columns show -0.114*** (0.028), -0.097*** (0.035), -0.113*** (0.028), -0.107*** (0.030), -0.096** (0.039).
  - Observations and countries (selected): Observations 17,707 (column (1)), 11,294 (column (2)), 5,737 (columns (3) and (4)); Countries 92, 76, 45 respectively. R-squared values 0.03, 0.04, 0.07.
  - Threshold, ln(remittances-to-GDP) reported e.g., 1.47, 1.33, 1.85, 1.35, 1.65, 2.03 converted to percent: 4.36%, 3.79%, 6.36%, 3.87%, 5.23%, 7.62%.
- Table 9 (controlling for income; dependent variable: monthly changes in lending rate):
  - Change in policy rate examples: 0.138*** (0.043); 0.167** (0.069); 0.189** (0.074); other columns show negative or less precisely estimated coefficients.
  - Remittances to GDP x change in policy rate: -0.129*** (0.034), -0.122*** (0.031), -0.104** (0.040), -0.126*** (0.033), -0.114*** (0.031), -0.108** (0.040).
  - Per capita GDP positive association in some specifications: 0.138*** (0.039); sample observations e.g., 6,582; 8,434; 4,065; Countries 55, 56, 56.
- Table 10 (robustness with deposit rates; dependent variable: monthly changes in deposit rate):
  - Change in policy rate examples: 0.134*** (0.018); 0.227*** (0.002); 0.079 (0.051).
  - Remittances to GDP x Change in policy rate: 0.140*** (0.003) in one column and negative interactions in others: -0.090*** (0.032), -0.098*** (0.028), -0.081*** (0.028), -0.098*** (0.028), -0.085** (0.033), -0.066* (0.039).
  - Observations up to 21,357; Countries up to 112; R-squared examples 0.017, 0.019, 0.036.
- Table 11 (Structural panel VAR; dependent variable: impulse response of log(lending rate) to nominal monetary shock):
  - Remittances to GDP coefficients by quarter: 1st quarter -0.017 (0.062); 2nd quarter -0.006 (0.048); 3rd quarter -0.003 (0.044); 4th quarter 0.003 (0.034); average -0.006 (0.044); maximum -0.072^^ (0.044).
  - International financial integration shows negative and statistically significant coefficients in several specifications: -0.002** (1st quarter) (0.001) and -0.002** (maximum) (0.001).
  - Observations 46 across these regressions; R-squared values from 0.10 to 0.16.
- Table 12 (IRs supplemented with co-movement coefficients):
  - Remittances to GDP coefficients across columns: -0.006 (col 1), -0.023** (col 2) (0.007), -0.017 (col 3) (0.027), -0.016 (col 4) (0.030), -0.024** (col 5) (0.009), -0.028** (col 6) (0.010).
  - Observations increase to 71 in supplemented samples; R-squared examples 0.13, 0.20, 0.09.
- Bank balance sheet stylized facts (selected from Tables 1–7):
  - All countries, unweighted averages (time frame 1997-2007): Total Deposits to GDP 39.86% (162 countries); Total Credit to GDP 58.84% (122); Total Assets to GDP 81.21% (123).
  - Remittances-GDP >= 5% group: Total Deposits to GDP 39.17% (31); Total Credit to GDP 58.29% (23); Total Assets to GDP 60.67% (24).
  - Liquid assets to total assets (Table 4): All Countries 20.05 (123); Emerging & Developing 23.07 (105); Remittances-GDP >= 5%: 28.67 (23).
  - Excess reserves to total assets (Table 4): All Countries 4.16% (101); Emerging & Developing 4.53% (94); Remittances-GDP >= 5%: 5.94% (23).
  - Volatility measures (Table 5): standard deviation of log(total deposit) All Countries 0.293 (169); remittances-GDP >= 5% group 0.221 (31).  
  - Bank competitiveness (Table 6): Lerner index All Countries 0.24 (131); Emerging & Developing 0.26 (103).
  - Government balance to GDP (Table 7): All Countries -1.12 (154); Remittances-GDP >= 5% -3.20 (27). Net interest margin of banks All Countries 5.20% (163).
- Data and sample notes:
  - Frequency and time sample for main monthly regressions: monthly data for 1990-2013, interactive variables (remittances, competitiveness, institutions) based on annual figures.
  - Structural panel VAR IRs estimated over 1978-2013 by Mishra et al. (2014); remittances log average over 1997-2007 if at least five years of data reported.

*Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2016/_wp1644.pdf*

### REFERENCES

### _wp1644 - REFERENCES

### Remittances, economic growth, and financial development
- Abdih,  Y.,  A.  Barajas,  R.  Chami  and  C.  Ebeke  (2012a):  Remittances  Channel and  Fiscal Impact  in  the  Middle  East,  North  Africa,  and  Central  Asia.  IMF Working  Paper  No.  12/104. Washington, DC: International Monetary Fund.
- Abdih,  Y.,  R.  Chami,  J.  Dagher  and  P.  Montiel  (2012b):  Remittances  and  Institutions: Are Remittances a Curse? World Development 40(4), 657-666.
- Aggarwal,  R.,  A.  Demirgürç-Kunt  and  M.  S.  Martínez  Pería  (2011):  Do  Remittances Promote Financial Development? Journal of Development Economics 96, 255-264.
- Barajas, A., R. Chami, C. Fullenkamp and P. Montiel (2009): Do Workers’ Remittances Promote Economic Growth? IMF Working Paper No. 09/153.
- Benmamoun,  M.  and  K.  Lehnert  (2013):  Financing  Growth:  Comparing  the  Effects of  FDI, ODA, and International Remittances. Journal of Economic Development 38(2), 43-65.
- Giuliano, P. and M. Ruiz-Arranz (2009): Remittances, Financial Development, and Growth. Journal of Development Economics 90, 144-152.
- Ramirez,  M.  D.  (2013):  Do  Financial  and  Institutional  Variables  Enhance  the  Impact of Remittances  on  Economic  Growth  in  Latin  America  and  the  Caribbean? A  Panel Cointegration Analysis. International Advances in Economic Research 19, 273-288.
- Brown,  R.  P.  C.,  F.  Carmignani  and  G.  Fayad  (2013):  Migrants'  Remittances  and Financial Development:  Macro- and  Micro-Level  Evidence  of  a  Perverse  Relationship. The  World Economy, 636-660.
- Beck,  T.,  A.  Demirgürç-Kunt  and  R.  Levine  (2000):  A  New  Database  on  Financial Development and Structure. World Bank Economic Review 14, 597-605.
- Djankov, McLiesh and Shleifer (2007): Private Credit in 129 Countries. Journal of Financial Economics  12  (2):  77-99.

### Remittances, exchange rates, Dutch disease, and competitiveness
- Ball,  C. P.,  C.  Lopez  and  J.  Reyes  (2013):  Remittances,  Inflation  and  Exchange Rate Regimes in Small Open Economies. The World Economy, 487-507.
- Barajas,  A.,  R.  Chami,  D.  Hakura  and  P.  Montiel  (2011):  Workers'  Remittances and  the Equilibrium Real Exchange Rate: Theory and Evidence. Economica, 45-94.
- Hassan, G. M. and M. J. Holmes (2013): Remittances and the Real Effective Exchange Rate. Applied Economics 45(35), 4959-4970.
- Lartey,  E.K.,  K.  Federico,  S.  Mandelman  and  P.  A.  Acosta  (2012):  Remittances,  Exchange Rate  Regimes  and  the  Dutch  Disease:  A  Panel  Data  Analysis.  Review of  International Economics 20(2), 377-395.
- Maklouhf,  F.  and  M.  Mughal  (2013):  Remittances,  Dutch  Disease,  and  Competitiveness: A Bayesian Analysis. Journal of Economic Development 38(2), 67-97.
- Singer,  D.  (2010)  Migrant  Remittances  and  Exchange  Rate  Regimes  in  the Developing World. American Political Science Review 104(2), 307–323.
- Kim,  J.  (2013):  Remittances  and  Currency  Crisis:  The  Case  of  Developing  and Emerging Countries. Emerging Markets Finance & Trade 49(6), 88-111.

### Monetary policy, transmission, and remittance fluctuations
- Bernanke, B. and M. Gertler (1995): Inside the Black Box: The Credit Channel of Monetary Policy Transmission. Journal of Economic Perspectives 9, 27-48.
- Mishkin,  F.  (1995):  Symposium  on  the  Monetary  Transmission  Mechanism.  Journal of Economic Perspectives 9, 3-10.
- Mishra, P. and P. Montiel (2013): How Effective is Monetary Transmission in Low-income Countries? A Survey of the Empirical Evidence. Economic Systems 37, 187-216.
- Mishra, P., P. Montiel and A. Spilimbergo (2012): Monetary Transmission in Low- Income Countries: Effectiveness and Policy Implications. IMF Economic Review 60(2), 70-302.
- Mishra,  P.,  P.  Montiel,  P.  Pedroni  and  A.  Spilimbergo  (2014):  Monetary  Policy and  Bank Lending  Rates  in  Low-Income  Countries:  Heterogeneous  Panel  Estimates. Journal  of Development Economics 111, 117-131.
- Mandelman,   F.   S.   (2013):   Monetary   and   Exchange   Rate   Policy   under   Remittance Fluctuations. Journal of Development Economics 102, 128-147.
- Vacaores,  D.  E.  (2012):  Remittances,  Monetary  Policy,  and  Partial  Sterilization. Southern Economic Journal 79(2), 367-387.
- Saxegaard,  M.  (2006):  Excess  Liquidity  and  Effectiveness  of  Monetary  Policy:  Evidence from Sub-Saharan Africa. IMF Working Paper No. 06/115.
- Kumhof,   M. and E. Tanner   (2005):   Government   Debt:   A   Key   Role   in   Financial Intermediation. IMF Working Paper No. 05/57.

### Remittances, volatility, crises, and business cycle transmission
- Barajas,  A.,  R.  Chami,  C.  Ebeke  and  S.  J.  A.  Tapsoba  (2012):  Workers  Remittances: An Overlooked  Channel  of  International  Business  Cycle  Transmission?  IMF Working  Paper 12/251. Washington, DC: International Monetary Fund.
- Chami, R., A. Barajas, T. F. Cosimano, C. Fullenkamp, M. Gapen, and P. Montiel (2008a): Macroeconomic   Consequences   of   Remittances.   IMF   Occasional   Paper No.   259. Washington DC: International Monetary Fund.
- Chami,  R.,  D.  S.  Hakura  and  P.  Montiel  (2012):  Do  Worker  Remittances  Reduce Output Volatility in Developing Countries? Journal of Globalization and Development 3(1).
- Frankel, J. (2011): Are Bilateral Remittances Countercyclical? Open Economies Review 22, 1-16.
- Durdu,  D.  B.  and  S.  Sayan  (2012):  Emerging  Market  Business Cycles with Remittance Fluctuations. IMF Sta_ Papers 57(2).
- Vargas-Silva, C. (2008): Are Remittances Manna from Heaven? A Look at the Business Cycle Properties of Remittances. North American Journal of Economics and Finance 19, 290.
- Combes, J-L. and Ebeke, C., (2011). Remittances and Household Consumption Instability in Developing Countries, World Development, 39(7), 1076-1089.
- Ebeke, C. (2014). Do International Remittances Affect The Level And The Volatility Of Government Tax Revenues? Journal of International Development, 26(7), 1039-1053.
- Ebeke, C. and J.-L. Combes (2012): Do Remittances Dampen the Effect of Natural Disasters on Output Growth Volatility in Developing Countries? Applied Economics 45(16), 2241-2254.
- Ruiz,  I.  and  C.  Vargas-Silva  (2010):  Another  Consequence  of  the  Economic  Crisis:  a Decrease in Migrants' Remittances. Applied Financial Economics 20, 171-182.

### Measurement, methodology, and structural approaches
- Chami,  R.,  C.  Fullenkamp  and  M.  Gapen  (2009):  Measuring  Workers  Remittances: What Should Be Kept In and What Should Be Left Out? mimeo, Washington DC: International Monetary Fund.
- Pedroni, P. (2013): Structural Panel VARs. Econometrics 2, 180-206.
- Deefort,  C.  (2006):  Tendances  De  Long  Terme  des  Migrations  Internationales. Analyse  _a Partir des 6 Principaux Pays Receveurs. mimeo, EQUIPPE, Universitsé de Lille et IRES, Université Catholique de Louvain.
- Narayan,  P.  K.,  S.  Narayan  and  S.  Mishra  (2011):  Do  Remittances  Induce  Inflation? Fresh Evidence from Developing Countries. Southern Economic Journal 77(4), 914-933.
- Termos, A., G. Naufal and I. Genc (2013): Remittance Outflows and Inflation: The Case of the GCC Countries. Economics Letters 120, 45-47.
- Vacaores,  D.  E.  (2012):  Remittances,  Monetary  Policy,  and  Partial  Sterilization. Southern Economic Journal 79(2), 367-387.
- Benmamoun,  M.  and  K.  Lehnert  (2013):  Financing  Growth:  Comparing  the  Effects of  FDI, ODA, and International Remittances. Journal of Economic Development 38(2), 43-65.

*Source: _wp1644 - REFERENCES*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2016/_wp1644.pdf_