## _wp16189 ## Source details **Canonical URL:** [_wp16189](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2016/_wp16189.pdf) ## Other formats - [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2016/_wp16189.pdf.md) - [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2016/_wp16189.pdf.json) --- ### I. Introduction - Monetary policy effectiveness depends critically on the choice of policy instrument. - In developed countries, indirect instruments—particularly a short-term interest rate—are the consensus tool; empirical evidence supports its effectiveness for controlling inflation and influencing output growth. - In developing countries, central banks typically operate two main instruments: a short-term policy interest rate and the required reserve ratio. - Since the early 2000s many developing-country central banks have de-emphasized reserve requirements in favour of market-based policies focused on short-term interest rates. - Structural impediments in small open developing economies (weak institutions, shallow financial markets, credit market imperfections, oligopolistic banking, and persistent excess liquidity) can weaken the interest rate transmission channel. - Empirical motivation: under structural excess liquidity and credit market imperfections the short-term interest rate may be limited in transmitting monetary policy; direct instruments such as reserve requirements are sometimes used to manage liquidity (examples: Jamaica and Trinidad and Tobago increased required reserve ratios between 2008 and 2010 to absorb excess liquidity). - Paper contribution: estimates a macroeconometric model (restricted VARX) to evaluate relative effectiveness of indirect (short-term interest rate) and direct (required reserve ratio) monetary instruments in Barbados, Jamaica and Trinidad and Tobago. - Main reported results: - Weak transmission of an increase in the short-term interest rate to output growth in Jamaica and Trinidad and Tobago. - In all three countries, a rise in the policy interest rate is not successful in reducing inflation. - A contractionary reserve requirement shock helps alleviate pressure on the exchange rate in Barbados and Trinidad and Tobago. - A contractionary reserve requirement shock reduces credit demand in Jamaica and Trinidad and Tobago. ### II. The Caribbean Context - Policy objectives in Barbados, Jamaica and Trinidad and Tobago: price stability, exchange rate stability, adequate foreign exchange reserves, and financial system stability. - Small, open nature exposes these economies to external shocks (energy price fluctuations, food price spikes, natural disasters, capital account shocks). - Financial structure and exchange rate regimes: - Commercial banking sector dominates financial assets; oligopoly market structure. - Barbados: long-standing fixed exchange rate since the 1980s. - Jamaica: floating exchange rate. - Trinidad and Tobago: managed float. - Central banks often intervene extensively in FX markets to smooth volatility. - Two key impediments to monetary transmission: - Embryonic capital markets; underdeveloped financial markets. - High persistent excess bank liquidity; levels rose to record levels over the last decade. - Causes cited: - Barbados: macro environment, interest rate policy and government money creation (Maynard and Moore (2005)). - Jamaica: increases in government spending facilitated by money creation or borrowing associated with higher bank deposits and excess reserves (Anderson-Reid (2011)). - Trinidad and Tobago: monetization of government deposits at the Central Bank (tax payments from energy companies) leading to "involuntary" excess liquidity (Primus et al. (2014)). - Historical use of instruments (selected facts preserved): - Jamaica: required reserve ratio last changed in 2010 from 14 percent to 12 percent. - Trinidad and Tobago: required reserve ratio last changed in 2008 from 15 percent to 17 percent. - Barbados: reserve requirement on local deposits and government securities last changed from 17 percent to 15 percent in 2007. - Cash reserve ratio and liquid asset ratio in Jamaica reduced from 21 percent and 43 percent, respectively, in 1998, to 12 percent and 26 percent, respectively, in 2014. - Policy implication: Given high liquidity and limited interest-rate transmission, central banks often need to use other measures (including reserve requirements) to achieve objectives. ### III. Econometric Methodology and Data - Model: restricted Vector Autoregressive model with Exogenous Variables (VARX). - Structural VARX form: y_t = A1 y_{t-1} + ... + A_p y_{t-p} + B0 x_t + ... + B_q x_{t-q} + C:D_t + u_t. - Rationale for VARX: - Allows imposing parameter restrictions, accommodates asymmetries, increases degrees of freedom and parameter efficiency, facilitates IRFs and VDs. - VARX specification (three blocks): - Fully interactive endogenous variables (2): output_growth_t (change in log real GDP; quarterly → monthly via Lisman-Sandee with ECOTRIM), infl_t (year-on-year percent change in log CPI or RPI; rebased to 2000). - Partially interactive endogenous variables (9): policy_rate_t (baseline short-term rate; Taylor-type rule), alternative instrument rrr_t (required reserve ratio), plr_t (prime lending rate), spread_t (prime minus deposit rate), exch_rate_t (real exchange rate), credit_t (change in private sector credit), govt_exp_t (growth rate of government expenditure), ertr_t (ratio of excess reserves to total reserves), world_CPI_t and oil_price_t. - Reserve requirement historical changes reiterated: - Jamaica: rrr changed in 2010 from 14 percent to 12 percent. - Trinidad and Tobago: rrr changed in 2008 from 15 percent to 17 percent. - Barbados: statutory ratio on government securities and local currency deposits changed from 17 percent to 15 percent in 2007. - Strictly exogenous variables: constant, time trend, eleven seasonal dummies. - Sample period: monthly data from May 2002 to September 2014. - Data sources: Central Bank of Barbados, Central Bank of Jamaica, Ministry of Finance of Jamaica, Central Bank of Trinidad and Tobago. - Stationarity and seasonality: ADF and PP tests; deseasonalization via X-12-ARIMA in robustness checks. - Estimation details: - Reduced form with two lags on all endogenous variables (optimal endogenous lag length = two based on AIC and FPE). - Inference: Impulse response functions and variance decompositions computed using Lütkepohl, Krätzig and Boreiko methods; 95 percent bootstrap confidence bands. - Software: JMulTi econometrics software. ### IV. Econometric Results — Shock to the Policy Rate (one-standard deviation positive shock) - General across countries: - Weak transmission of policy interest rate changes due to high financial system liquidity. - Increase in short-term interest rate often has instantaneous effect on loan rates but does not translate into a significant fall in private sector credit in Barbados and Trinidad and Tobago. - Higher domestic interest rates tend to appreciate the real exchange rate and increase government expenditure in Barbados and Trinidad and Tobago; Jamaica shows real exchange rate depreciation on impact. - Shock has a muted effect on inflation in all three countries. - Timing of output response: immediate contraction in Jamaica; fall realized after 10 months in Trinidad and Tobago. - Barbados (selected magnitudes) - Output growth: contracts insignificantly at first, then fluctuates. - Inflation: initially falls marginally before increasing significantly. - Prime lending rate: increases directly when policy rate rises. - Interest rate spread: reduced by 0:26 percent following an increase in the policy rate. - Real exchange rate: statistically significant appreciation in the second month; effect dissipates after the fifth period. - Credit: falls marginally in the first period but response not significant. - Government expenditure: rises significantly by0:14percent immediately after shock; effect dissipates quickly. - Excess reserves ratio: falls significantly after 5 months. - Variance decompositions (selected shares): - Real GDP: about87percent own innovations; world oil prices account for 6percent. - Inflation: between95percent and40 percent from its own innovation; about20percent from non-fuel world CPI; 15percent from world oil prices. - Prime lending rate: 45percent own shock, 50percent policy rate. - Interest rate spread: 40percent past value; policy rate 30percent; prime lending rate 25percent. - Government expenditure: 85percent own value; policy rate about8percent. - Ratio of excess reserves to total reserves: 95to85 percent own innovation; smaller shares from government expenditure and policy rate. - Jamaica (selected magnitudes) - Policy instrument: 30 day open market rate. - Output: marginal decline by0:03percent. - Inflation: declines but effect not significant. - Interest rate spread: falls because deposit rate rise > prime lending rate rise. - Real exchange rate: depreciates by0:05percent on impact. - Private sector credit: increases on impact before decreasing; fall significant during months 5–7 and after month 11. - Government expenditure: increases significantly after 4 months. - Excess liquidity: quantity of liquidity falls by0:07percent in the second month. - Variance decompositions (selected shares): - Output: 100percent own innovation in first month, declines to75percent over period; world oil prices explain 8percent and policy rate 6percent of variation. - Prime lending rate: 82percent own innovations; also affected by interest rate spread and output growth. - Interest rate spread: prime lending rate responsible for approximately75percent of variation. - Private sector credit: 65percent own innovation; smaller shares from policy rate and excess reserves ratio. - Government expenditure: 86percent own value; policy rate and world oil prices each about4percent. - Excess reserves ratio: about70percent own innovation; policy rate, world oil prices and government expenditure smaller contributors. - Trinidad and Tobago (selected magnitudes) - Output growth: rises before declining significantly during months 10 and 14 (slow transmission). - Inflation: rises, significant during months 5 and 10, then declines (price puzzle/hump-shaped pattern). - Prime lending rate: immediate rise by0:09percent on impact. - Interest rate spread: widens amid high excess liquidity because deposit rates are sticky upward. - Real exchange rate: significant appreciation in second period; dissipates after sixth period. - Private sector credit: falls in first month (insignificantly). - Government expenditure: increases by0:06percent two months after shock. - Excess reserves ratio: declines, but effect not significant. - Variance decompositions (selected shares): - Real output: 100percent own innovations initially, declines to55percent; inflation contribution increases after month 7. - Prime lending rate: about60percent own shocks; ~25percent due to policy interest rate. - Interest rate spread: ~55percent own innovation; 30percent prime lending rate; 12percent policy interest rate. - Private sector credit: ~85percent own shocks; small contributions from prime lending rate, policy rate, and output growth. - Government expenditure: 85percent own innovations; policy rate ~9percent. - Excess reserves ratio: ~80percent past value; 10percent inflation; exchange rate 5percent. - Government fiscal operations account for less than5percent of changes in excess reserves despite being a key contributory factor to excess liquidity. ### V. Econometric Results — Shock to the Required Reserve Ratio (one-standard deviation positive shock) - General across countries: - A higher required reserve ratio is more effective than policy rate shocks at reducing credit demand in Jamaica and Trinidad and Tobago and leads to an instantaneous fall in excess reserves in those countries. - In Barbados, higher reserve requirements reduce inflation significantly (after 9 months) but have limited effect on output. - Required reserve shocks can have stronger effects on bank liquidity metrics compared to policy rate shocks in some cases. - Barbados (selected magnitudes) - Output growth: contracts temporarily after 4 months before rising; effect not significant. - Inflation: significant fall after 9 months. - Interest rate spread: increases by0:17percent on impact because deposit rates fall and loan rates rise. - Real exchange rate: depreciates significantly after 7 months. - Ratio of excess reserves: increases by0:09percent one month after shock before declining; fall not significant. - Variance decompositions (selected shares vs policy-rate experiment): - Inflation: required reserve rate explains about2to4percent after the seventh period; policy rate explained 2to8percent in baseline. - Interest rate spread: required reserve ratio explains about10percent (less than policy rate’s 30to38percent). - Ratio of excess reserves: reserve requirements account for 22percent of movements (larger than policy rate’s contribution). - Jamaica (selected magnitudes) - Output: falls following shock but effect insignificant. - Inflation: falls by0:4percent on impact; significant only in first period. - Interest rate spread: falls by0:09percent as deposit rates fall without significant increase in lending rates. - Exchange rate: depreciates; credit falls from month 2. - Government expenditure: falls in first period then rises almost significantly after 4 months. - Excess reserves ratio: falls marginally more than baseline in month 2 (stronger effect on bank liquidity). - Variance decompositions (selected shares): - Output growth: required reserve ratio explains less than0:5percent of fluctuations (policy rate explained ~6percent in baseline). - Private sector credit: required reserve ratio explains about3percent (policy rate ~10percent). - Prime lending rate, interest rate spread, and government expenditure: larger shares explained by required reserve ratio than by policy rate. - Excess reserves ratio: government expenditure contributes slightly lower share to movements compared to baseline. - Trinidad and Tobago (selected magnitudes) - Output growth: falls significantly by0:05percent in month 2 (contrast: policy rate shock produced +0:07percent on impact). - Private sector credit: contracts by0:05percent in month 2. - Ratio of excess reserves: falls by0:18percent in month 2 (contrasts with insignificant change under policy rate shock). - Prime lending rate and interest rate spread: increase in month 3 (versus month 1 in baseline). - Variance decompositions (selected shares): - World oil prices explain about1percent of real output growth (baseline up to5percent). - Private sector credit and ratio of excess reserves: required reserve ratio explains about7percent and8percent respectively (higher than policy rate’s 3percent and 0:9percent contributions). - Required reserve ratio contributes far less to changes in output and inflation compared to policy rate in some measures. ### VI. Conclusions and Policy Implications - Main conclusions: - Indirect policy instruments (short-term Taylor-type interest rate) have weak transmission in Barbados, Jamaica and Trinidad and Tobago due to high financial system liquidity and credit market imperfections. - Indirect instruments are often insufficient to achieve price stability; shocks to the policy rate have muted effect on inflation across the three countries. - Direct instruments (required reserve ratio) can be more effective at influencing credit demand and bank liquidity, and can produce quicker contractions in output in some cases (e.g., Trinidad and Tobago). - Policy recommendations: - Monetary authorities should use a combination of indirect and direct policy tools to effect monetary policy. - Adjust reserve requirements on local currency deposits more frequently because: - It has a direct effect on credit demand and excess reserves. - It is not associated with an increase in interest costs. - This is particularly important for Barbados, which had not changed the requirement on local currency deposits for over a decade (as noted in the study). - Central Bank of Trinidad and Tobago (and by extension similar authorities) should consider implementing reserve requirements on securities to: - Influence other short-term rates. - Better manage liquidity conditions. - Nuances: - Reserve requirement manipulation can affect spreads, deposit rates and lending behavior (e.g., spreads may increase as reserves function like a tax on banks). - Effectiveness of instruments differs across countries; instrument choice should account for country-specific liquidity conditions and financial sector imperfections. *Italic: Source — _wp16189 (excerpt provided).* ### References .............................................................................................. 23 ### _wp16189 - References .............................................................................................. 23 ### I. Introduction - Monetary policy effectiveness depends critically on the choice of policy instrument. - In developed countries, indirect instruments—particularly a short-term interest rate—are the consensus tool; empirical evidence supports its effectiveness for controlling inflation and influencing output growth. - In developing countries, central banks typically operate two main instruments: a short-term policy interest rate and the required reserve ratio. - Since the early 2000s many developing-country central banks have de-emphasized reserve requirements in favour of market-based policies focused on short-term interest rates. - Structural impediments in small open developing economies (weak institutions, shallow financial markets, credit market imperfections, oligopolistic banking, and persistent excess liquidity) can weaken the interest rate transmission channel. - Empirical motivation: when structural excess liquidity and credit market imperfections prevail, the short-term interest rate may be limited in transmitting monetary policy; therefore direct instruments such as reserve requirements are sometimes used to manage liquidity (example: central banks of Jamaica and Trinidad and Tobago increased required reserve ratios between 2008 and 2010 to absorb excess liquidity). - Paper contribution: estimates a macroeconometric model (restricted VARX) to evaluate the relative effectiveness of indirect (short-term interest rate) and direct (required reserve ratio) monetary instruments in Barbados, Jamaica and Trinidad and Tobago. - Main reported results from the model: - Weak transmission of an increase in the short-term interest rate to output growth in Jamaica and Trinidad and Tobago. - In all three countries, a rise in the policy interest rate is not successful in reducing inflation. - A contractionary reserve requirement shock helps alleviate pressure on the exchange rate in Barbados and Trinidad and Tobago. - A contractionary reserve requirement shock reduces credit demand in Jamaica and Trinidad and Tobago. ### II. The Caribbean Context - Policy objectives in Barbados, Jamaica and Trinidad and Tobago: price stability, exchange rate stability, adequate foreign exchange reserves, and financial system stability. - Small, open nature exposes these economies to external shocks (energy price fluctuations, food price spikes, natural disasters, capital account shocks). - Financial structure commonalities: - Commercial banking sector dominates financial assets; operates under oligopoly market structure. - Exchange rate regimes differ: - Barbados: long-standing fixed exchange rate since the 1980s. - Jamaica: floating exchange rate. - Trinidad and Tobago: managed float. - Central banks often intervene extensively in FX markets to smooth volatility. - Two key impediments to monetary transmission: - Embryonic capital markets; financial markets remain underdeveloped. - High persistent excess bank liquidity; levels of excess liquidity rose to record levels over the last decade. - Causes cited: - Barbados: macro environment, interest rate policy and government money creation (Maynard and Moore (2005)). - Jamaica: increases in government spending facilitated by money creation or borrowing associated with higher bank deposits and excess reserves (Anderson-Reid (2011)). - Trinidad and Tobago: monetization of government deposits at the Central Bank (tax payments from energy companies) leading to "involuntary" excess liquidity (Primus et al. (2014)). - Monetary policy instruments historically used: - Barbados: required reserve ratio, discount rate, OMOs, moral suasion, Central Bank policy rate (minimum deposit rate). In 2013 the Treasury bill rate was used to influence banking-sector interest rates; in 2015 the Bank ceased stipulating a minimum interest rate on savings deposits. Reserve requirements are specified on local deposits, foreign deposits and government securities. - Jamaica: OMOs and reserve requirements (open market interest rate on OMOs guides market rates). Reserve ratio includes a cash reserve requirement on local currency liabilities and a liquid assets ratio (government securities up to one year); foreign currency liabilities subject to requirements. - Trinidad and Tobago: short-term interest rate, OMOs and reserve requirements. OMOs introduced in 1996; from 1998 reserve requirements were de-emphasized and recourse to OMOs increased. In 2002 the repo policy rate became the principal instrument. The Central Bank specifies primary and secondary reserve requirements as percent of prescribed liabilities. Commercial banks are not subject to reserve requirements on government securities and foreign currency liabilities. - Observed policy shifts: - Central banks of Barbados and Trinidad and Tobago have shifted toward greater reliance on indirect instruments. - Bank of Jamaica has not explicitly indicated greater reliance, though data show the open market rate is manipulated to indicate stance. - Exception: Jamaica required reserve ratio increased during 2008-2010; otherwise reserve ratios in Jamaica have generally decreased since the late 1990s. - Example numeric adjustments preserved from source: - Jamaica: required reserve ratio last changed in 2010 from 14 percent to 12 percent. - Trinidad and Tobago: required reserve ratio last changed in 2008 from 15 percent to 17 percent. - Barbados: reserve requirement on local deposits and government securities last changed from 17 percent to 15 percent in 2007. - Cash reserve ratio and liquid asset ratio in Jamaica reduced from 21 percent and 43 percent, respectively, in 1998, to 12 percent and 26 percent, respectively, in 2014. - Policy implication: Given high financial system liquidity and limited interest-rate transmission, central banks often need to use other measures (including reserve requirements) to achieve objectives. ### III. Econometric Methodology - Method: restricted Vector Autoregressive model with Exogenous Variables (VARX) to examine transmission of contractionary monetary policy shocks via short-term interest rate and required reserve ratio in Barbados, Jamaica and Trinidad and Tobago. - Standard VARX structural form (as presented): - y_t = A1 y_{t-1} + ... + A_p y_{t-p} + B0 x_t + ... + B_q x_{t-q} + C:D_t + u_t - where y_t = [y_{1t}; ...; y_{Kt}]' endogenous variables, x_t = [x_{1t}; ...; x_{Mt}]' exogenous variables, D_t deterministic variables (constant, linear trend, seasonal dummies), u_t zero mean white noise; A, B, C parameter matrices. - Rationale for VARX: - Allows imposing parameter restrictions based on a priori economic reasoning. - Accommodates asymmetries between variables, reduces parameters to estimate, raises degrees of freedom and parameter efficiency. - Facilitates computation of impulse response functions (IRFs) and variance decompositions (VDs) associated with innovations. - VARX specification for the three countries comprises three blocks: - Fully Interactive Endogenous Variables (2 variables): - output_growth_t: growth rate of real GDP (change in log real GDP). Quarterly GDP converted to monthly via Lisman-Sandee procedure using ECOTRIM. Quarterly data for Trinidad and Tobago in index form; other countries converted to index before interpolation. - infl_t: inflation rate measured as year-on-year percent change in log CPI (or RPI for Trinidad and Tobago). All indices rebased to 2000. - Partially Interactive Endogenous Variables (9 variables): - Monetary policy instrument: - Baseline: policy_rate_t (short-term interest rate). Defined as minimum savings rate for Barbados, OMOs 30-day open market interest rate for Jamaica, repo rate for Trinidad and Tobago. - Policy rate specified using a Taylor-type rule: policy rate depends on inflation, output growth, and its past value. - Alternative experiment: reserve requirement ratio rrr_t as direct instrument; reacts to its past value. Reserve requirement measured using minimum required reserve ratio on domestic currency deposits for all countries except Barbados (Barbados considered combination of local deposits and government securities). - Historical changes preserved: - Jamaica: rrr changed in 2010 from 14 percent to 12 percent. - Trinidad and Tobago: rrr changed in 2008 from 15 percent to 17 percent. - Barbados: statutory ratio on government securities and local currency deposits changed from 17 percent to 15 percent in 2007. - plr_t: prime lending rate (commercial banks' loan rate). Interacts with all variables except non-fuel world CPI and world oil prices. - spread_t: interest rate spread (prime lending rate minus deposit rate). Interacts with all variables except non-fuel world CPI and world oil prices. - exch_rate_t: real exchange rate (nominal rate deflated by U.S. CPI to obtain real effective rate). Interacts with all variables except non-fuel world CPI and world oil prices. - credit_t: change in private sector credit. Interacts with all variables except non-fuel world CPI and world oil prices. - govt_exp_t: growth rate of government expenditure. Reacts with output growth, policy interest rate, exchange rate, world non-fuel CPI, world oil prices and its past value. Total expenditure used to capture interest payment effects on domestic debt. - ertr_t: ratio of excess reserves to total reserves (measure of excess liquidity). Interacts with all variables except non-fuel world CPI and world oil prices. - world_CPI_t: non-fuel world CPI. Depends on its past value and the oil price. - oil_price_t: world price of oil (annual year-on-year rate of change). Depends on its past value only; not influenced by domestic variables. - Strictly Exogenous Variables: - Constant term. - Time trend. - Eleven seasonal dummies to capture seasonality (series with seasonality were deseasonalized using X-12-ARIMA in robustness checks; results were robust). - Interaction structure and economic interpretation: - Policy rate follows a Taylor-type rule reacting to inflation and output growth; changes influence the term structure of banking-sector interest rates (captured via plr_t and spread_t). - Reserve requirement shocks directly affect liquidity conditions and are therefore represented by rrr_t reacting to its own lag. - Private sector credit and banking variables capture credit channel dynamics; excess reserves capture persistent liquidity that may blunt interest-rate effects. - Exchange rate included to capture open-economy transmission (higher domestic interest rates may increase demand for local currency and capital inflows). - Fiscal operations (govt_exp_t) included because fiscal actions can contribute to excess liquidity and affect monetary effectiveness. - Sectors captured by the model: - Real sector: output growth, inflation. - Banking sector: private credit, excess reserves, prime lending rate, interest rate spread. - External sector: real exchange rate. - Monetary authority: short-term interest rate, required reserve ratio. - Fiscal authority: government expenditure. - Sample period: monthly data from May 2002 to September 2014. - Note: the short-term interest rate rule is described as a "Taylor-type" rule in conventional form (Taylor (1993)). ECOTRIM used for temporal disaggregation. *Source: _wp16189 - References .............................................................................................. 23* ### 2002. Also, during the period, the central banks relied more on market-based tools to e§ect ### _wp16189 - 2002. Also, during the period, the central banks relied more on market-based tools to e§ect ### Data and Methodology - Main data sources: The Central Bank of Barbados, The Central Bank of Jamaica, The Ministry of Finance of Jamaica, and The Central Bank of Trinidad and Tobago. - Stationarity and seasonality: - Augmented Dickey Fuller (ADF) and Phillips-Perron (PP) tests used to determine stationarity. - Seasonality tests conducted on each variable; results summarized in Table 1 (source). - Model specification: - Reduced form of the VARX model with two lags on all endogenous variables (optimal endogenous lag length is two based on the Akaike Information Criterion and Final Prediction Error). - Model form: y_t = A1 y_{t-1} + A2 y_{t-2} + C:D_t + u_t. - Baseline instrument: policy_rate_t (Taylor-type rule where it responds to output growth, inflation and its past value). - Alternative instrument: required reserve ratio (rrr_t) included and depends only on its past value; rrr_t placed after policy_rate_t in ordering. - Inference: - Impulse response functions and variance decompositions computed using Lütkepohl, Krätzig and Boreikoís methods. - Confidence bands: 95 percent bootstrap confidence bands. - Estimation software: JMulTi econometrics software. ### Econometric Results — Shock to the Policy Rate (one-standard deviation positive shock) - General summary across countries: - Weak transmission of policy interest rate changes due to high financial system liquidity. - Increase in short-term interest rate often has instantaneous effect on loan rates but does not translate into a significant fall in private sector credit in Barbados and Trinidad and Tobago. - Higher domestic interest rates tend to appreciate the real exchange rate and increase government expenditure in Barbados and Trinidad and Tobago; Jamaica shows a different pattern for the exchange rate. - Shock has a muted effect on inflation in all three countries. - Timing of output response: immediate contraction in Jamaica; fall realized after 10 months in Trinidad and Tobago. - Barbados (section A.1) - Output growth: contracts insignificantly at first, then fluctuates. - Inflation: initially falls marginally before increasing significantly. - Prime lending rate: increases directly when policy rate rises. - Interest rate spread: reduced by 0:26 percent following an increase in the policy rate. - Real exchange rate: statistically significant appreciation in the second month; effect dissipates after the fifth period. - Credit: falls marginally in the first period but response not significant. - Government expenditure: rises significantly by0:14percent immediately after shock; effect dissipates quickly. - Excess reserves ratio: falls significantly after 5 months. - Variance decompositions: - Real GDP: about87percent of its own innovations; world oil prices account for 6percent. - Inflation: between95percent and40 percent from its own innovation; about20percent from non-fuel world CPI; 15percent from world oil prices. - Prime lending rate: 45percent own shock, 50percent policy rate. - Interest rate spread: 40percent past value; policy rate 30percent; prime lending rate 25percent. - Real exchange rate: 50percent own innovation; inflation 40percent. - Government expenditure: 85percent own value; policy rate about8percent. - Ratio of excess reserves to total reserves: 95to85 percent own innovation; smaller shares from government expenditure and policy rate. - Jamaica (section A.2) - Policy instrument: 30 day open market rate. - Output: marginal decline by0:03percent. - Inflation: declines but effect not significant. - Interest rate spread: falls because deposit rate rise > prime lending rate rise; effect on prime lending rate and spread insignificantly different from zero. - Real exchange rate: depreciates by0:05percent on impact. - Private sector credit: increases on impact before decreasing; fall significant during months 5–7 and after month 11. - Government expenditure: increases significantly after 4 months. - Excess liquidity: quantity of liquidity falls by0:07percent in the second month. - Variance decompositions: - Output: 100percent own innovation in first month, declines to75percent over period; world oil prices explain 8percent and policy rate 6percent of variation. - Inflation: initially due solely to its past value; contributions from world oil prices and excess reserves appear over time. - Prime lending rate: 82percent own innovations; also affected by interest rate spread and output growth. - Interest rate spread: prime lending rate responsible for approximately75percent of variation. - Private sector credit: 65percent own innovation; smaller shares from policy rate and excess reserves ratio. - Government expenditure: 86percent own value; policy rate and world oil prices each about4percent. - Excess reserves ratio: about70percent own innovation; policy rate, world oil prices and government expenditure smaller contributors. - Trinidad and Tobago (section A.3) - Output growth: rises before declining significantly during months 10 and 14 (slow transmission). - Inflation: rises, significant during months 5 and 10, then declines (price puzzle/hump-shaped pattern). - Prime lending rate: immediate rise by0:09percent on impact. - Interest rate spread: widens amid high excess liquidity because deposit rates are sticky upward. - Real exchange rate: significant appreciation in second period; dissipates after sixth period. - Private sector credit: falls in first month (insignificantly). - Government expenditure: increases by0:06percent two months after shock. - Excess reserves ratio: declines, but effect not significant. - Variance decompositions: - Real output: 100percent own innovations initially, declines to55percent; inflation contribution increases after month 7. - Inflation: mainly own innovations, policy interest rate and world oil prices also matter. - Prime lending rate: about60percent own shocks; ~25percent due to policy interest rate. - Interest rate spread: ~55percent own innovation; 30percent prime lending rate; 12percent policy interest rate. - Exchange rate: 50percent own innovations; 40percent inflation; policy rate <10percent contribution. - Private sector credit: ~85percent own shocks; small contributions from prime lending rate, policy rate, and output growth. - Government expenditure: 85percent own innovations; policy rate ~9percent. - Excess reserves ratio: ~80percent past value; 10percent inflation; exchange rate 5percent. - Government fiscal operations account for less than5percent of changes in excess reserves despite being a key contributory factor to excess liquidity. ### Econometric Results — Shock to the Required Reserve Ratio (one-standard deviation positive shock) - General summary across countries: - A higher required reserve ratio is more effective than policy rate shocks at reducing credit demand in Jamaica and Trinidad and Tobago and leads to an instantaneous fall in excess reserves in those countries. - In Barbados, higher reserve requirements reduce inflation significantly (after 9 months) but have limited effect on output. - Required reserve shocks can have stronger effects on bank liquidity metrics compared to policy rate shocks in some cases. - Barbados (section B.1) - Output growth: contracts temporarily after 4 months before rising; effect not significant. - Inflation: significant fall after 9 months. - Prime lending rate: increases in first period (effect insignificant). - Interest rate spread: increases by0:17percent on impact because deposit rates fall and loan rates rise. - Real exchange rate: depreciates significantly after 7 months. - Credit and government expenditure: no significant effect. - Ratio of excess reserves: increases by0:09percent one month after shock before declining; fall not significant. - Variance decompositions relative to policy rate shock: - Output: world oil prices account for marginally smaller share (less than5percent). - Inflation: required reserve rate explains about2to4percent after the seventh period; policy rate explained 2to8percent in baseline. - Prime lending rate: own innovation contribution similar under both experiments (45to52percent). - Interest rate spread: required reserve ratio explains about10percent (less than policy rate’s 30to38percent in baseline). - Exchange rate and private sector credit: less attributable to required reserve ratio than to policy rate. - Government expenditure: 3percent determined by required reserve ratio (versus 8percent by policy rate). - Ratio of excess reserves: reserve requirements account for 22percent of movements (larger than policy rate’s contribution). - Jamaica (section B.2) - Output: falls following shock but effect insignificant. - Inflation: falls by0:4percent on impact; significant only in first period. - Interest rate spread: falls by0:09percent as deposit rates fall without significant increase in lending rates. - Exchange rate: depreciates; credit falls from month 2. - Government expenditure: falls in first period then rises almost significantly after 4 months. - Excess reserves ratio: falls marginally more than baseline in month 2 (stronger effect on bank liquidity). - Variance decompositions: - Output growth: required reserve ratio explains less than0:5percent of fluctuations (policy rate explained ~6percent in baseline). - Inflation: world oil price contribution slightly lower under reserve shock; non-fuel world CPI contribution similar. - Prime lending rate, interest rate spread, and government expenditure: larger shares explained by required reserve ratio than by policy rate. - Exchange rate: required reserve ratio explains a slightly lower share than policy rate. - Private sector credit: required reserve ratio explains about3percent (policy rate ~10percent). - Excess reserves ratio: government expenditure contributes slightly lower share to movements compared to baseline. - Trinidad and Tobago (section B.3) - Output growth: falls significantly by0:05percent in month 2 (contrast: policy rate shock produced +0:07percent on impact). - Inflation: rises but effect insignificant. - Prime lending rate and interest rate spread: increase in month 3 (versus month 1 in baseline). - Real exchange rate and government expenditure: no significant effect. - Private sector credit: contracts by0:05percent in month 2. - Ratio of excess reserves: falls by0:18percent in month 2 (contrasts with insignificant change under policy rate shock). - Variance decompositions: - World oil prices explain about1percent of real output growth (baseline up to5percent). - Required reserve ratio contributes far less to changes in output and inflation compared to policy rate (contrary to some literature). - Prime lending rate, interest rate spread, real exchange rate, and government expenditure: reserve requirement explains less of fluctuations than policy interest rate. - Private sector credit and ratio of excess reserves: required reserve ratio explains about7percent and8percent respectively (higher than policy rate’s 3percent and 0:9percent contributions). ### Conclusion and Policy Implications (section V) - Main conclusions: - Indirect policy instruments (short-term Taylor-type interest rate) have weak transmission in Barbados, Jamaica and Trinidad and Tobago due to high financial system liquidity and credit market imperfections. - Indirect instruments are often insufficient to achieve price stability; shocks to the policy rate have muted effect on inflation across the three countries. - Direct instruments (required reserve ratio) can be more effective at influencing credit demand and bank liquidity, and can produce quicker contractions in output in some cases (e.g., Trinidad and Tobago). - Policy recommendations: - Monetary authorities should use a combination of indirect and direct policy tools to effect monetary policy. - Adjust reserve requirements on local currency deposits more frequently because: - It has a direct effect on credit demand and excess reserves. - It is not associated with an increase in interest costs. - This is particularly important for Barbados, which had not changed the requirement on local currency deposits for over a decade (as noted in the study). - Central Bank of Trinidad and Tobago (and by extension similar authorities) should consider implementing reserve requirements on securities to: - Influence other short-term rates. - Better manage liquidity conditions. - Implication nuances: - Reserve requirement manipulation can affect spreads, deposit rates and lending behavior (e.g., spreads may increase as reserves function like a tax on banks). - Effectiveness of instruments differs across countries; instrument choice should account for country-specific liquidity conditions and financial sector imperfections. *Italic: Source — _wp16189 (excerpt provided). * ### References ### _wp16189 - References ### Key cited works and topics - AgÈnor, Pierre-Richard and Karim El Aynaoui (2010), "Excess Liquidity, Bank Pricing Rules, and Monetary Policy," Journal of Banking and Finance, Vol. 34, pp. 923ñ933. - AgÈnor, Pierre-Richard and Peter J. Montiel (2007), "Credit Market Imperfections and the Monetary Transmission Mechanism Part II: Flexible Exchange Rates," Centre for Growth and Business Cycle Research Discussion Paper Series No. 86, University of Manchester. - Anderson-Reid, Karen (2011), "Excess Reserves in Jamaican Commercial Banks: The Implications for Monetary Policy," Working Paper No. 43663, Munich Personal RePEc Archive (MPRA). - Baksh, Sherman and Roland C. Craigwell (1997), "The Monetary Transmission Mechanism in Small Open Economies: A Case Study of Barbados," Savings and Development, Vol. 21, pp. 179-193. - Birchwood, Anthony (2011), ìImplementation of Taylor type Rules in Nascent Money and Capital Markets Under Managed Exchange Ratesî, School of Social Sciences Theses, Brunel University. - Central Bank of Barbados (2013), "The Central Bank of Barbados Issues a Change in its Interest Rate Policy," General Press Release: Central Bank of Barbados. - Central Bank of Trinidad and Tobago (2004), Role and Functions. Central Bank of Trinidad and Tobago. - Central Bank of Trinidad and Tobago (2005), "The Implementation of Monetary Policy in Trinidad and Tobago," Port-of-Spain: Central Bank of Trinidad and Tobago. - Cheong, Daryl and Earl Boodoo (2008), "The Monetary Transmission Mechanism: A Closer Look at the Interest Rate Channel in Trinidad and Tobago," Working Paper, Central Bank of Barbados. - Davoodi, Hamid R., Shiv Dixit, and Gabor Pinter (2013), "Monetary Transmission Mechanism in the East African Community: An Empirical Investigation," Working Paper No. 39, International Monetary Fund. - Glocker, Christian and Pascal Towbin (2012), "Reserve Requirements for Price and Financial Stability: When are they E§ective?," International Journal of Central Banking, Vol. 8, pp. 65-114. - Holden, Paul and Hunt Howell (2009), ìEnhancing Access to Finance in the Caribbeanî, Private Sector Development Discussion Paper No. 4, Inter-American Development Bank. - International Monetary Fund (2015), ìWorld Economic and Financial Surveys, Regional Economic Outlook, Western Hemisphere: Adjusting Under Pressureî, International Monetary Fund, October 2015. - International Monetary Fund (2011), "Trinidad and Tobago: Selected Issues," IMF Country Report No. 74, International Monetary Fund. - Jordan, Alwyn; Sharon Branch; Andrea McQuay; Yvonne Cooper; and Latoya Smith (2012), "An Analysis of Bank Liquidity in The Bahamas (2001-2012)," Working Paper, The Central Bank of The Bahamas. - Khemraj, Tarron (2007a), "Monetary Policy and Excess Liquidity: The Case of Guyana," Social and Economic Studies. Vol. 56, pp. 101-127. - Khemraj, Tarron (2007b), "What Does Excess Bank Liquidity Say About the Loan Market in Less Developed Countries?", DESA Working Paper No. 60, United Nations. - Khemraj, Tarron (2009), "Excess Liquidity and the Foreign Currency Constraint: The Case of Monetary Management in Guyana," Applied Economics, Vol. 41, pp. 2073-2084. - Lattie, Claney (2000), "Monetary Policy Management in Jamaica," Pamphlet No. 1, Bank of Jamaica. - Lebedinski, Lara (2007), "Excess Liquidity and Monetary Policy Under Fixed Exchange Rates: A Theoretical and Empirical Investigation," MSc Dissertation, University of Manchester, United Kingdom. - Lütkepohl, Helmut, Markus Krätzig (2004), "Applied Time Series Econometrics," Cambridge University Press. - Lütkepohl, Helmut, Markus Krätzig and Dmitri Boreiko (2006), "VAR Analysis in JMulTi". Retrieved at http://www.jmulti.de/download/help/var.pdf. - Maynard, Tracy and Winston Moore (2005), "The Demand for Excess Liquid Assets in Barbados," Working Paper, Central Bank of Barbados. - Montoro, Carlos and Ramon Moreno (2011), "The Use of Reserve Requirements as a Policy Instrument in Latin America," Quarterly Review, Bank for International Settlements. - Moore, Winston and Kevon Skete (2010), "The Implications of Monetary Policy Shocks for Government Debt Management in Barbados," Working Paper, Central Bank of Barbados. - Moore, Winston R. and Marlon L. Williams (2008), "Evidence on the Sectoral Monetary Transmission Process under a Fixed Exchange Rate Regime," International Economic Journal, Vol. 22, pp. 387ñ398. - Onyukwu, Onyukwu and Emmanuel Nwosu (2011), "Monetary Policy and Output-ináation Volatility Interaction in Nigeria: Evidence from Bivariate GARCH-M Model," Journal of Economics and Sustainable Development, Vol. 2, pp. 46-64. - Primus, Keyra (forthcoming), "Excess Reserves, Monetary Policy and Financial Volatility," Journal of Banking and Finance. - Primus, Keyra; Anthony Birchwood; and Lester Henry (2014), "The Dynamics of Involuntary Commercial Banksí Reserves in Trinidad and Tobago," The Journal of Developing Areas, Vol. 48, pp. 63-84. - Ramlogan, Carlyn (2007), "Anism of Monetary Policy in Small Developing Countries: An Application to Trinidad and Tobago," The Journal of Developing Areas, Vol. 41, pp. 79-91. - Ramlogan, Carlyn, "The Transmission Mechanism of Monetary Policy: Evidence from the Caribbean," Journal of Economic Studies, Vol. 31 pp. 435 ñ447. - Robinson, Wayne and John W. Robinson (1997), "The Transmission Mechanism of Monetary Policy in the Jamaican Economy," Working Paper, Bank of Jamaica. - Sen Min, Wong (2008), "Monetary Policy Analysis in a Fixed Exchange Rate Economy with Credit Market Imperfections ñA Theoretical and Empirical Exploration for Malaysia," PhD Dissertation, University of Manchester, United Kingdom. - Taylor, John B. (1993), "Discretion versus Policy Rules in Practice," Carnegie-Rochester Conference Series on Public Policy, Vol. 39, pp. 195-214. - Tovar, Camilo E.; Mercedes Garcia-Escribano; and Mercedes V. Martin (2012), "Credit Growth and the E§ectiveness of Reserve Requirements and Other Macroprudential Instruments in Latin America," Working Paper No. 142, International Monetary Fund. - Urquhart Mikhail-Ann F. (2006), "The Reaction of Bank Lending to Monetary Policy: The Jamaican Case," Working Paper, Bank of Jamaica. - Walsh, Carl E. (2010), Monetary Theory and Policy, Third edition, The MIT Press Cambridge, Massachusett s. - Watson, Patrick Kent (1996), "The Monetary Transmission Process in a Small Open Economy: The Case of Trinidad and Tobago," Working Paper, Caribbean Centre for Monetary Studies. - Watson, Patrick Kent (2003), "Evaluating Monetary Policy Measures in a Small Primary-Exporting Economy: The Case of Trinidad and Tobago," Social and Economic Studies, Vol. 52, pp. 143-166. - Worrell, DeLisle (1995) "Monetary Policy in Small Open Economies," Working Paper, Central Bank of Barbados. - Worrell, DeLisle; Harold Codrington; Roland Craigwell; and Kevin Grenidge (2003) "Economic Resilience with an Exchange Rate Peg: The Barbados Experience, 1985-2000," Working Paper No. 168, International Monetary Fund. - Worrell, DeLisle; Michelle Doyle-Lowe; Anton Belgrave; Darrin Downes; and Kester Guy (2012), "Central Bank Intervention and Interest Rate Policy in Barbados," Working Paper, Central Bank of Barbados. ### Table 1. Unit Root and Seasonality Tests — variable orders and seasonal dummies - Countries: Barbados | Jamaica | Trinidad and Tobago - Variables and entries: - output_growth - Barbados: I(1) - Jamaica: I(1) - Trinidad and Tobago: I(0) dummies:  - infl - Barbados: I(0) - Jamaica: I(0) - Trinidad and Tobago: I(0) - policy_rate - Barbados: I(1) - Jamaica: I(1) - Trinidad and Tobago: I(1) - rrr - Barbados: I(0) - Jamaica: I(1) d7 - Trinidad and Tobago: I(1) - plr - Barbados: I(1) d10 - Jamaica: I(1) - Trinidad and Tobago: I(1) - spread - Barbados: I(0) - Jamaica: I(0) dummies:  - Trinidad and Tobago: I(1) d11 - exch_rate - Barbados: I(1) d1, d3, d8 - Jamaica: I(0) dummies:  - Trinidad and Tobago: I(1) d3 - credit - Barbados: I(1) - Jamaica: I(1) - Trinidad and Tobago: I(1) - govt_exp - Barbados: I(0) - Jamaica: I(0) dummies:  - Trinidad and Tobago: I(0) - ertr - Barbados: I(1) - Jamaica: I(0) dummies:  - Trinidad and Tobago: I(0) d9 - world_CPI - Barbados: I(0) - Jamaica: I(0) - Trinidad and Tobago: I(0) - oil_price - Barbados: I(0) - Jamaica: I(0) - Trinidad and Tobago: I(0) ### Figures (captions) - Figure 1. Impulse Response Functions — Barbados — One-Standard Deviation Shock to the Policy Interest Rate. - Figure 2. Variance Decompositions — Barbados — One-Standard Deviation Shock to the Policy Interest Rate. - Figure 3. Impulse Response Functions — Barbados — One-Standard Deviation Shock to the Reserve Requirement Ratio. - Figure 4. Variance Decompositions — Barbados — One-Standard Deviation Shock to the Reserve Requirement Ratio. - Figure 5. Impulse Response Functions — Jamaica — One-Standard Deviation Shock to the Policy Interest Rate. - Figure 6. Variance Decompositions — Jamaica — One-Standard Deviation Shock to the Policy Interest Rate. - Figure 7. Impulse Response Functions — Jamaica — One-Standard Deviation Shock to the Reserve Requirement Ratio. - Figure 8. Variance Decompositions — Jamaica — One-Standard Deviation Shock to the Reserve Requirement Ratio. - Figure 9. Impulse Response Functions — Trinidad and Tobago — One-Standard Deviation Shock to the Policy Interest Rate. - Figure 10. Variance Decompositions — Trinidad and Tobago — One-Standard Deviation Shock to the Policy Interest Rate. - Figure 11. Impulse Response Functions — Trinidad and Tobago — One-Standard Deviation Shock to the Reserve Requirement Ratio. - Figure 12. Variance Decompositions — Trinidad and Tobago — One-Standard Deviation Shock to the Reserve Requirement Ratio. *Source: _wp16189 - References (PDF).* --- _Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2016/_wp16189.pdf_