## _wp11229

## Source details

**Canonical URL:** [_wp11229](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2011/_wp11229.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2011/_wp11229.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2011/_wp11229.pdf.json)

---

### I. Introduction and context
- Inflation targeting (IT) is a framework that targets inflation directly rather than intermediate variables.
- Twenty-six countries classified as inflation targeters (Roger 2010): nine industrial countries and seventeen emerging market and developing economies.
- In MENA, Turkey adopted IT in 2006; Egypt, Tunisia and Morocco intend to move toward IT in the foreseeable future.
- Four key transmission channels emphasized: interest rate channel, credit (bank lending and balance sheet) channel, exchange rate channel, and asset price channel.
- Data quality limitations and variation in informational content of real rates mean limited empirical studies on the interest rate channel—especially for IT EMEs and non-IT MENA economies.

### Monetary policy transmission mechanism (MTM) — stages and scope
- Key instrument under IT: short-term interest rate.
- Stages of MTM:
  - First stage: policy rate changes → movements in retail interest rates (commercial bank deposit and lending rates).
  - Second stage: retail rate changes → impact private consumption and investment → affect aggregate demand.
  - Third stage: aggregate demand changes → affect domestic activity.
  - Final stage: activity changes → produce desired effect on inflation.
- MTM operates through variables including: real cost of capital, availability of bank credit, exchange rate, household and corporate balance sheets, wealth and monetary aggregates.
- MTMs are complex and involve behavior across all sectors.

### Contribution of the study
- Focus: the traditional interest rate channel.
- Methodological contrast: estimates structural equations for private consumption and private investment (rather than VARs for GDP).
- Comparisons:
  - Interest rate elasticities of private consumption and investment in IT EMEs (prior to IT) versus five potential IT adopters in MENA.
  - Interest rate sensitivities pre-IT and post-IT for IT EME group.
  - Variation of interest rate sensitivities with domestic financial market development.
- Novelty: examination of variation with domestic financial market development not previously done in the literature.

### Main findings (four principal findings)
- First:
  - Real lending rates have a significant impact on private investment in both IT EMEs and MENA countries.
  - Real deposit rates have a significant direct impact on private consumption in IT EMEs but not in MENA countries.
- Second:
  - Adoption of IT did not significantly alter the traditional Keynesian interest rate channel in IT EMEs.
- Third:
  - Interest rate elasticities of private consumption and investment vary with domestic financial sector development.
  - IT EMEs: wealth effect of changes in the real deposit rate increasingly dominates the substitution effect as the financial sector develops.
  - MENA EMEs: substitution effect increasingly dominates as the financial sector develops (opposite to IT EMEs).
  - Financial sector development has no significant impact on private investment in IT EMEs.
  - Financial sector development has a significant impact (direct and indirect) on private investment in MENA EMEs.
- Fourth:
  - Capital account liberalization effects differ across groups:
    - IT EMEs: opening the capital account has a positive impact on private investment but a negative impact on private consumption.
    - MENA EMEs: opening the capital account has a positive impact on private consumption but a negative impact on private investment.

### Channels through which monetary policy affects aggregate demand
- Traditional interest rate channel:
  - Sticky prices → policy nominal rate adjustments → changes in short-term real interest rate and long-term rates.
  - Real interest rate movements affect cost of capital (investment) and consumption via wealth and substitution effects.
- Credit channel:
  - Bank lending channel: contractionary policy reduces usable reserves/deposits → reduces bank lending (especially to small firms) → investment falls.
  - Balance sheet channel: higher rates reduce cash-flow and asset prices → lower net worth/collateral → weaker access to funds → investment falls.
- Exchange rate channel:
  - Under flexible exchange rates and open capital accounts, higher domestic interest rates can appreciate the exchange rate → affects net exports and GDP; impact depends on expectations.
- Asset price channel:
  - Tight monetary policy can depress equity and real estate prices → lower household financial wealth and Tobin’s q → reduce consumption and investment.

### Institutional features of EMEs and implications for transmission
- Common features:
  - Financial systems underdeveloped and bank-dominated; many households and SMEs lack access to services.
  - Weak judicial systems; property rights hard to enforce.
  - Oligopolistic banking environments limit lending rate responsiveness.
  - Thin interbank markets.
  - Bank loans and T-bills are imperfect substitutes.
  - Money supply shocks often occur through external channels (aid, remittances, export revenue volatility).
- Implication: interest rate, bank lending and asset price channels may be weaker in EMEs. The paper tests whether the interest rate channel is weak by estimating structural consumption and investment equations for IT EMEs and MENA EMEs.

### Literature review — empirical findings and gaps (selected)
- Consumption:
  - Mixed findings: Fry (1978) finds elasticity about -0.2 for seven Asian countries (1962–72); Giovannini (1983) does not find significant relationship; Reinhart and Ostry (1995) find no systematic relationship in many developing countries.
  - Funke (2002) finds a small but significant effect of stock market developments on private consumption in 16 EMEs.
  - Noted gap: no comprehensive study on interest rate sensitivity of private consumption in emerging market or MENA economies; no empirical studies on financial development and capital market openness effects on private consumption.
- Investment:
  - Theoretical: higher real interest rates increase borrowing costs → reduce private investment; accelerator links investment to expected future real GDP growth.
  - Greene and Villanueva (1991): one percentage point rise in real interest rate reduces private investment by 0.1 percentage point (23 developing economies, 1975–87).
  - Serven (2003), Aysan et al. (2005), Bader and Malawi (2010), Misati and Nyamongo (2010) find negative relationships in various country samples.
  - Mixed evidence on stock market development and investment efficiency.
- Capital account liberalization:
  - Theory: should lower cost of capital and boost investment.
  - Empirical evidence mixed: Henry (2000a, 2000b) finds positive effects; Achy (2003) and Naceur et al. (2008) find negative or no significant effects in some MENA samples.
- Identified gap: lack of comprehensive panel studies on interest rate sensitivity of private consumption and role of financial development and capital openness in emerging and MENA economies.

### Data and methodology — empirical strategy and variables
- Objective: estimate augmented consumption and investment equations to examine interest rate elasticities.
- Consumption function (dependent variable: private consumption at constant prices scaled by GDP) key regressors:
  - GDP at constant prices (scaled), real deposit rate (rd), real government consumption (scaled), real GDP per capita (y), age dependency ratio (pop aged <=14 and >=65), total population, Chinn-Ito index of capital openness (range: (-1.83) to 2.50), financial market development index, inflation targeting dummy (1 if targeting).
  - Financial development index = stock market capitalization/GDP + private credit by deposit money banks/GDP.
  - Real deposit rate coefficient measures direct interest rate elasticity of private consumption; net effect ambiguous (wealth vs substitution).
- Investment function (dependent variable: private investment at constant prices scaled by GDP) key regressors:
  - GDP (scaled), real lending rate (rl), expected output growth (proxied by one-year ahead WEO GDP growth forecasts), financial development index, Chinn-Ito index, IT dummy.
  - Real lending rate coefficient measures direct interest rate elasticity of private investment.
- Sample:
  - Two groups estimated separately: 14 IT EMEs; five EMEs in MENA region that may adopt IT.
  - Annual data 1990–2009.
- Construction of rates:
  - Real deposit and lending rates = nominal rates minus one-year ahead WEO CPI inflation forecasts (used inside ln(1+ ... ) in Tunisia data definitions).
- Estimation issues:
  - Endogeneity, heteroskedasticity, autocorrelation addressed via IV-GMM using lagged endogenous variables as instruments.
  - Hansen’s J test used for instrument validity.
- Robustness:
  - Excluding economies of transition (Czech Republic, Hungary, Poland, Romania) from IT EMEs did not significantly change results.

### Country groups (primary and additional lists)
- Primary IT EMEs (14): 1. Brazil (1999); 2. Chile (1999); 3. Colombia (1999); 4. Czech Republic (1997); 5. Hungary (2001); 6. Indonesia (2005); 7. Mexico (2001); 8. Peru (2002); 9. Philippines (2002); 10. Poland (1998); 11. Romania (2005); 12. South Africa (2000); 13. Thailand (2000); 14. Turkey (2006).
- Additional MENA EMEs considered: Egypt; Jordan; Lebanon; Morocco; also Tunisia (presented in detail).

### Tunisia — data, stationarity, methodology notes
- Variable definitions (selected):
  - ln(Real private consumption/Real GDP) — WEO.
  - ln(Real private investment/Real GDP) — WEO.
  - ln(Real deposit rate) = ln(1+(Nominal deposit rate – WEO forecast of one period ahead CPI inflation)); nominal rates from IFS.
  - Financial development index = (Stock market capitalization/GDP) + (Private credit by deposit money banks/GDP).
  - Capital openness = Chinn-Ito index.
  - WEO GDP growth forecast = one-year ahead WEO GDP growth forecast.
- Summary statistics (Mean, Std. Dev., Min., Max):
  - ln(Real private consumption/Real GDP): -0.685, 0.624, -7.04, 1.98
  - ln(Real private investment/Real GDP): -1.536, 0.711, -3.08, -1.07
  - ln(Real deposit rate) (as %): 4.512, 1.039, 4.17, 9.15
  - ln(Real lending rate) (as %): 4.229, 1.495, 4.30, 8.44
  - ln(Real government consumption/Real GDP): -1.89, .674, -3.10, 0.585
  - ln(Real GDP per capita): 10.54, 3.06, 6.84, 16.05
  - Financial development index: 0.759, 0.649, 0.216, 3.89
  - (Pop aged <=14, >= 65)/total population (as %): 34.74, 8.961, 28.54, 49.99
  - Dummy for IT: 0.45, 0.49, 0, 1
  - Capital openness: 0.151, 1.41, -1.83, 2.50
  - WEO GDP growth forecast (as %): 5.78, 9.73, -0.66, 91.73
- ADF stationarity test statistics (ADF test statistic, lag by SC):
  - Real private consumption/Real GDP: -3.089** (1)
  - Real private investment/Real GDP: -3.918*** (1)
  - Real deposit rate: -3.143** (1)
  - Real lending rate: -3.821*** (1)
  - Real government consumption/Real GDP: -3.223** (1)
  - Real GDP per capita: -2.673* (5)
  - Financial development index: -4.985*** (1)
  - (Pop aged <=14, >= 65)/total population: -5.415*** (3)
  - Dummy for IT: -5.086*** (1)
  - Capital openness: -4.207*** (1)
  - WEO GDP growth forecast: -5.924*** (6)
  - (*** Significant at 1% level, ** Significant at 5% level, * Significant at 10% level.)
- Conclusion: ADF results reject unit roots for all variables — variables are stationary.
- Estimation approach: IV-GMM to address endogeneity, heteroskedasticity, and autocorrelation; Hansen’s J tests fail to reject instrument exogeneity in reported regressions.

### Empirical findings — private consumption (IV-GMM results and interpretation)
- IT EMEs:
  - ln(Real deposit rate) coefficient: -0.258*** (0.044)
  - Financial development index: -1.54*** (0.362)
  - ln(Real deposit rate) * Financial development: 0.077*** (0.168)
  - Capital account openness: -0.022** (0.010)
  - ln(Real government consumption/Real GDP): -0.142*** (0.047)
  - ln(Real GDP per capita): -0.006 (0.005) — not significant
  - (Pop aged <=14, >= 65)/Total population: 0.015*** (0.004)
  - Hansen J χ2 p-value: 0.64
  - Interpretation:
    - Deposit interest rate has a significantly negative elasticity on private consumption.
    - Financial development has a significant negative direct effect on private consumption.
    - Interaction: as financial development increases, the interest rate elasticity becomes less negative (wealth effect starts to dominate substitution effect).
    - Capital account openness negatively impacts private consumption.
- Non-IT MENA EMEs:
  - ln(Real deposit rate): -0.188 (0.203) — not significant
  - Financial development index: 9.33** (3.86)
  - ln(Real deposit rate) * Financial development: -2.06** (0.850)
  - Capital account openness: 0.102*** (0.029)
  - ln(Real government consumption/Real GDP): 0.706*** (0.092)
  - ln(Real GDP per capita): 0.134*** (0.036)
  - (Pop aged <=14, >= 65)/Total population: 0.023* (0.013)
  - Hansen J χ2 p-value: 0.81
  - Interpretation:
    - No significant direct interest-rate effect on consumption at low levels of financial development.
    - Financial development has a significant positive direct effect on private consumption.
    - Interaction: interest rate elasticity becomes more negative as financial development increases (substitution effect dominates).
    - Capital account openness positively impacts private consumption in MENA EMEs.
- Cross-group threshold:
  - For IT EMEs the interest rate elasticity of private consumption turns positive when financial development index > 3.36.
  - Maximum financial development index in MENA EMEs is 3.25 (below 3.36), explaining absence of positive wealth effect there.

### Empirical findings — private investment (IV-GMM results and interpretation)
- IT EMEs:
  - ln(Real lending rate): -0.464*** (0.030)
  - Capital account openness: 0.076*** (0.027)
  - WEO GDP growth forecast: 0.078*** (0.024)
  - Financial development index and its interaction with lending rate: not statistically significant
  - Hansen J χ2 p-value: 0.65
  - Interpretation:
    - Real lending rate has a significant negative effect on private investment.
    - Capital account openness boosts private investment directly.
    - Expected output growth positively affects private investment.
- Non-IT MENA EMEs:
  - ln(Real lending rate): -0.561*** (0.056)
  - Financial development index: 0.693*** (0.157)
  - ln(Real lending rate) * Financial development: -0.166*** (0.029)
  - Capital account openness: -0.181*** (0.035)
  - WEO GDP growth forecast: 0.224*** (0.059)
  - Hansen J χ2 p-value: 0.18
  - Interpretation:
    - Real lending rate has a significant negative effect on private investment.
    - Financial development has a significant positive direct effect on private investment.
    - Interaction: interest rate elasticity of private investment becomes more negative as financial development increases.
    - Capital account openness negatively impacts private investment in MENA EMEs.
- Cross-group comparison:
  - IT EMEs: capital openness supports investment but reduces consumption; financial development does not significantly affect investment.
  - MENA EMEs: capital openness raises consumption but reduces investment; domestic financial development supports both consumption and investment.

### Conclusions and policy-relevant implications
- Interest rate channel is operative:
  - Real lending rate negatively and significantly affects private investment in both IT EMEs and non-IT MENA EMEs.
  - Real deposit rate affects private consumption in IT EMEs; in MENA EMEs the consumption interest-rate effect depends on financial development.
- Adoption of IT:
  - Adoption of IT did not significantly change private consumption or investment behavior, nor their responsiveness to interest rates, in the IT EME sample.
- Role of financial sector development:
  - IT EMEs:
    - Financial development has a significant negative direct effect on private consumption.
    - As financial sector develops, wealth effect increasingly dominates substitution effect (interest-rate elasticity of consumption becomes less negative and can turn positive beyond financial development index = 3.36).
  - MENA EMEs:
    - Financial development has significant positive direct effects on both private consumption and private investment.
    - Interest-rate elasticities become more negative with greater financial development.
- Role of capital account liberalization:
  - IT EMEs: capital openness boosts private investment but reduces private consumption.
  - MENA EMEs: capital openness raises private consumption but reduces private investment.
- Policy implication summary:
  - The traditional Keynesian interest rate channel remains relevant for emerging market economies, including potential IT adopters in the MENA region.
  - Financial sector development and capital account openness have materially different effects across country groups; policy design should account for country-specific financial development and institutional context.

*Source: _wp11229 — IMF staff.*

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

### _wp11229 - References

### I. Introduction and Context
- A growing number of emerging market economies are moving towards inflaton targeting as the guiding framework for monetary policy actions.
- Inflaton targeting (IT) involves targeting inflation directly, in contrast to alternative strategies that target intermediate variables (for example, the growth rate of money aggregates or the level of the exchange rate of an ‘anchor’ currency).
- At present twenty-six countries can be classified as inflation targeters (Roger 2010), including nine industrial countries and seventeen emerging market and developing economies.
- In the MENA region, Turkey adopted IT in 2006, while Egypt, Tunisia and Morocco intend to move toward a fully fledged inflation targeting regime in the foreseeable future.
- The literature emphasizes four key transmission channels: the traditional interest rate channel; the credit or loan supply channel; the exchange rate channel; and the asset price channel.
- Data quality limitations and variation in informational content of real rates of return mean there have not been many empirical studies determining the interest rate channel—especially the impact of the interest rate on private consumption and investment in IT emerging market economies and non-IT MENA economies.

### Monetary Policy Transmission Mechanism (MTM) — Stages and Scope
- Under IT the key monetary policy instrument is the short-term interest rate.
- Stages of the MTM described:
  - First stage: changes in the policy rate lead to movements in retail interest rates (commercial bank deposit and lending rates).
  - Second stage: changes in retail rates impact private consumption and investment, affecting aggregate demand.
  - Third stage: changes in aggregate demand affect domestic activity.
  - Final stage: activity changes produce the desired effect on the inflation rate.
- MTMs operate through a large set of variables, including: the real cost of capital, availability of bank credit, the exchange rate, household and corporate balance sheets, wealth and monetary aggregates.
- MTMs are complex, operating through various channels and involving the behaviour of all sectors of the economy.

### Contribution of This Study
- Focus: the traditional interest rate channel of the monetary policy transmission mechanism.
- Methodological contrast: this study estimates structural equations for private consumption and private investment rather than estimating VARs for overall output (GDP).
- Comparisons made:
  - Interest rate elasticities of private consumption and investment demand in inflation targeting EMEs (prior to their adoption of IT) versus five potential inflation targeters in the MENA region.
  - Interest rate sensitivities of private consumption and private investment pre-IT and post-IT for the IT EME group.
  - How interest rate sensitivities vary with the level of development of the domestic financial market.
- Novelty claim: to the authors’ knowledge, the examination of variation with domestic financial market development has not been done in the existing literature to date.

### Main Findings (four principal findings)
- First:
  - The real lending rates do have a significant impact on private investment in both IT EMEs and MENA countries.
  - Real deposit rates have a significant direct impact on private consumption in IT EMEs but not in MENA countries.
- Second:
  - The adoption of IT did not significantly alter the operation of the traditional Keynesian interest rate channel of the monetary policy transmission process in IT EMEs.
- Third:
  - Interest rate elasticities of private consumption and private investment vary with the level of development of the domestic financial sector.
  - For the IT EMEs: the wealth effect of changes in the real deposit rate increasingly dominates the substitution effect as the financial sector develops.
  - For the MENA EMEs: the exact opposite is the case (the substitution effect increasingly dominates as the financial sector develops).
  - Development of the financial sector has no significant impact on private investment in the IT EMEs, either directly or indirectly through the interest rate channel.
  - The empirical findings suggest a significant impact (both direct and indirect) of financial sector development on private investment in the MENA EMEs.
- Fourth:
  - Liberalization of the capital account has widely different effects on private sector demand in the two groups of countries.
  - In the IT EMEs: opening of the capital account has a positive impact on private investment but a negative impact on private consumption.
  - In the MENA EMEs: opening of the capital account has a positive impact on private consumption, but a negative impact on private investment.

### Structure of the Paper (as stated)
- The paper proceeds with Section II discussing in detail the various channels through which monetary policy can affect aggregate demand and output.

*Source: _wp11229 - References*

### Section III presents

### _wp11229 - Section III presents

### Channels through which monetary policy can affect aggregate demand
- Four channels highlighted:
  - Traditional interest rate channel
    - Prices are sticky and adjust to a monetary shock with a delay.
    - Adjustments to the short-term nominal policy interest rate lead to changes in the short-term real interest rate; long-term interest rates also affected as functions of expected sequences of short-term rates.
    - Movements in real interest rates influence economic agents: cost of capital affects investment; consumption is affected by wealth and substitution effects from changes in deposit interest rates and yields on treasury bonds.
  - Credit channel (split into bank lending channel and balance sheet channel)
    - Bank lending channel: contractionary policy (higher reserve requirements or Central Bank purchases of commercial bank reserve assets) lowers usable reserves and often deposits, reducing supply of bank loans—especially to small firms—adversely affecting investment.
    - Balance sheet channel: higher interest rates reduce cash-flow and asset prices, lowering firms’ net worth and collateral, weakening access to funds and increasing adverse selection/moral hazard, reducing investment.
  - Exchange rate channel
    - Under flexible exchange rates and open capital accounts, an interest rate increase can make domestic deposits more attractive, leading to exchange rate appreciation.
    - Appreciation affects tradable prices via imports and alters net exports, influencing GDP and economic activity; precise impact depends on expectations about domestic/foreign interest rates and inflation.
  - Asset price channel
    - Tighter monetary policy can depress equity, other financial assets, and real estate prices by making bonds relatively more attractive.
    - Falling asset prices lower household financial wealth (reducing consumption) and reduce Tobin’s q (market value vs replacement cost), adversely affecting investment.

### Institutional features of EMEs and implications for transmission
- Common institutional features in EMEs:
  - (i) Financial systems generally underdeveloped and dominated by banks; large share of households and SMEs lack access to financial services.
  - (ii) Judicial systems weak; property rights hard to enforce, reducing investment opportunities for banks.
  - (iii) Banks operate in oligopolistic environments, limiting lending rate responsiveness to monetary shocks.
  - (iv) Interbank markets are usually very thin.
  - (v) Bank loans and T-bills are highly imperfect substitutes due to thin financial markets and lack of competition.
  - (vi) Shocks to money supply often occur through the external channel (volatility of aid, remittances, and export revenues).
- Implications:
  - Given underdeveloped financial systems, the interest rate channel, bank lending channel, and asset price channel may play weaker roles in monetary transmission in EMEs.
  - The paper tests whether the interest rate channel is weak in EMEs by estimating structural equations for private consumption and private investment in IT EMEs and in MENA EMEs that may adopt IT.

### Literature review — empirical findings and gaps
- General:
  - Empirical research on the interest rate channel through cost of capital gives conflicting results, particularly for the MENA region.
  - Most studies are country-based using VAR methodology; panel data-based approaches are scarce.
- Selected empirical findings:
  - Neaime (2008): traditional Keynesian interest rate channel dominates in Morocco and Tunisia; direct linkages between interest rate and inflation are highly significant.
  - Boughrara (2009): credit channel stronger than traditional interest rate channel in Tunisia; exchange rate and asset price channels not operative in Tunisia or Morocco.
  - Jardak and Wrobel (2009): exchange rate channel most important in Tunisia.
  - Mishra et al. (2010): monetary transmission effectiveness in low income countries depends critically on bank lending channel effectiveness given limited financial development; correlation between money market rate and lending rates weaker in low income countries.
- Consumption literature:
  - Fry (1978): negative and statistically significant real interest rate elasticity of domestic consumption; elasticity about -0.2 for seven Asian countries (1962–72).
  - McDonald (1983): similar conclusions for 12 Latin American countries.
  - Giovannini (1983): extending to late 1970s, did not find significant consumption–real interest relationship.
  - Rossi (1988): in low-income developing countries with pervasive liquidity constraints, consumption growth follows income growth more than expected rates of return.
  - Reinhart and Ostry (1995): no systematic relationship in many developing countries between rates of return and private consumption due to poor data and variable informational content of real rates.
  - Funke (2002): panel of 16 EMEs finds a small but statistically significant effect of stock market developments on private consumption.
  - Noted gap: no comprehensive study on interest rate sensitivity of private consumption in emerging market or MENA economies; no empirical studies investigating impact of financial development and capital market openness on private consumption.
- Investment literature:
  - Theoretical expectation: rise in real interest rate raises borrowing costs, reducing private investment; accelerator model links investment positively to expected future real GDP growth.
  - Greene and Villanueva (1991): for 23 developing economies (1975–87), real interest has a significantly negative effect on private investment; a one percentage point rise in the real interest rate reduces private investment by 0.1 percentage point.
  - Serven (2003): private investment in less developed countries responds significantly to interest rate changes (61 less developed countries, 1975–1995).
  - Aysan et al. (2005): real interest rate negatively impacted firms’ investment projects in five MENA countries (Egypt, Iran, Jordan, Morocco and Tunisia) in 1980s–1990s.
  - Bader and Malawi (2010): significant negative impact of real interest rate on private investment in Jordan (1990–2005).
  - Misati and Nyamongo (2010): significant negative relationship between deposit interest rate and private investment for 18 African countries (1991–2004); no impact of stock market turnover on private investment.
  - Misati (2007): stock market development matters for investment efficiency in African countries with relatively developed stock markets; no consistent results for underdeveloped markets.
- Capital account openness and liberalization:
  - Theoretical: liberalization should channel resources to capital-scarce developing countries, lowering cost of capital and boosting investment (Fischer (1998, 2003), Obstfeld (1998), Rogoff (1999), Summers (2000)).
  - Empirical: mixed/contradictory results.
    - Henry (2000a): event study of 11 low income/emerging economies finds large but temporary increase in growth rate of real private investment after stock market liberalization.
    - Henry (2000b): stock market liberalization decreases cost of equity capital and boosts private investment.
    - Achy (2003): significant negative impact of financial liberalization on private investment for five MENA countries (1970–1998), arguing credit allocation distortions favored households over firms.
    - Naceur et al. (2008): fail to find significant impact of stock market liberalization on investment growth for 11 MENA countries (1979–2005).
- Identified empirical gap: lack of comprehensive panel studies on interest rate sensitivity of private consumption and the role of financial development and capital market openness in emerging and MENA economies.

### Data and methodology (empirical strategy and variables)
- Objectives:
  - Estimate augmented versions of standard consumption and investment equations to examine interest rate elasticities of private consumption and private investment.
- Theoretical motivation:
  - Consumption function motivated by life-cycle theory: consumption depends on current real income and expected lifetime income; relationship varies with age.
  - Investment function motivated by accelerator model: current investment is a function of anticipated output growth.
- Empirical specification (variables and interpretation):
  - Consumption function (notation described in source):
    - Dependent variable: private consumption at constant prices (scaled by GDP).
    - Key explanatory variables: GDP at constant prices (scaled), real deposit rate (ݎ஽), government consumption at constant prices (ܩ scaled), real GDP per capita (ݕ), age dependency ratio (݌݋ܲ: population aged 14 and less and 65 and more), total population (݌݋݌_ݐ݋ܶ), Chinn-Ito index of capital openness (݋݇), financial market development index (݂݀), inflation targeting dummy (ܦூ்).
    - Chinn-Ito capital openness index ranges from (-1.83) to 2.50; higher values indicate more financially open economies.
    - Financial development index equals sum of stock market capitalization and credit to the private sector, both measured as shares of GDP.
    - The inflation targeting dummy (ܦூ்) equals one if the country is targeting inflation and zero otherwise.
    - Real deposit rate and its coefficient measure the direct interest rate elasticity of private consumption.
    - The net effect of an increase in real interest rates on consumption is theoretically ambiguous due to substitution and wealth effects.
  - Investment function (notation described in source):
    - Dependent variable: private investment at constant prices (scaled by GDP).
    - Key explanatory variables: GDP at constant prices (scaled), real lending rate (ݎ௅), expected output growth over the coming year (ݕ݃; proxied by one-year ahead WEO GDP growth forecasts), financial market development index (݂݀), Chinn-Ito index (݋݇), inflation targeting dummy (ܦூ்).
    - Real lending rate coefficient represents the direct interest rate elasticity of private investment.
    - Disturbance terms: ߝ (consumption equation) and ߳ (investment equation).
- Sample and data details:
  - Two groups of countries estimated separately:
    - First group: 14 EMEs which have adopted inflation targeting (IT).
    - Second group: five EMEs in the MENA region that may adopt IT in the foreseeable future.
  - Annual data covering the period 1990 to 2009.
  - Dependent variables (private consumption and private investment) are scaled by GDP.
  - Real deposit and real lending rates constructed by subtracting one-year ahead forecasts of consumer price inflation from the IMF’s World Economic Outlook (WEO) nominal deposit and nominal lending rates respectively.
  - Coefficients on real deposit and real lending rates interpreted as direct interest rate elasticities.
  - Expected output growth proxied by one-year ahead WEO GDP growth forecasts in the investment equation.
  - Note: WEO one-year ahead forecasts used as proxies for expected inflation and expected output growth — stated as a novel feature.
  - Remittances were tested as controls but found insignificant for both private consumption and investment for both IT EMEs and potential MENA IT adopters; remittances were dropped from reported estimated equations.
- Robustness note:
  - Excluding the ‘economies of transition’ (Czech Republic, Hungary, Poland and Romania) from the IT EMEs and re-estimating did not have any significant impact on empirical results.

*Source: _wp11229 - Section III presents*

### 1.          Brazil (1999)

### 1.          Brazil (1999)

### Primary country list
- 1.          Brazil (1999)
- 2.          Chile (1999)
- 3.          Colombia (1999)
- 4.          Czech Republic (1997)
- 5.          Hungary (2001)
- 6.          Indonesia (2005)
- 7.          Mexico (2001)
- 8.          Peru (2002)
- 9.          Philippines (2002)
- 10.        Poland (1998)
- 11.        Romania (2005)
- 12.        South Africa (2000)
- 13.        Thailand (2000)
- 14.        Turkey (2006)

### Additional country list
- 1.          Egypt
- 2.          Jordan
- 3.          Lebanon
- 4.          Morocco

*Source: _wp11229 - 1.          Brazil (1999)*

### 5.          Tunisia

### 5.          Tunisia

### Data and variables
- Key variables and sources:
  - ln(Real private consumption/Real GDP) = ln(Private consumption, constant prices/GDP, constant prices) — WEO
  - ln(Real private investment/Real GDP) = ln(Private investment, constant prices/GDP, constant prices) — WEO
  - ln(Real deposit rate) = ln(1+(Nominal deposit rate – WEO forecast of one period ahead CPI inflation)); Nominal deposit rate: IFS
  - ln(Real lending rate) = ln(1+(Nominal lending rate – WEO forecast of one period ahead CPI inflation)); Nominal lending rate: IFS
  - ln(Real government consumption/Real GDP) = ln(Government consumption, constant prices/GDP, constant prices) — WEO
  - ln(Real GDP per capita) — WEO
  - Financial development index = (Stock market capitalization/GDP) + (Private credit by deposit money banks/GDP) — World Bank data on Financial Development and Structure
  - (Pop aged <=14, >= 65) / total population (as %) — WDI
  - Dummy for IT = 1 if the country is targeting inflation, and 0 otherwise (Date of adoption collected from central banks’ website)
  - Capital openness = Chinn-Ito index of capital openness
  - WEO GDP growth forecast = WEO forecast of one year ahead GDP growth — WEO

### Summary statistics and stationarity
- Summary statistics (Mean, Standard Deviation, Min., Max.):
  - ln(Real private consumption/Real GDP): -0.685, 0.624, -7.04, 1.98
  - ln(Real private investment/Real GDP): -1.536, 0.711, -3.08, -1.07
  - ln(Real deposit rate) (as %): 4.512, 1.039, 4.17, 9.15
  - ln(Real lending rate) (as %): 4.229, 1.495, 4.30, 8.44
  - ln(Real government consumption/Real GDP): -1.89, .674, -3.10, 0.585
  - ln(Real GDP per capita): 10.54, 3.06, 6.84, 16.05
  - Financial development index: 0.759, 0.649, 0.216, 3.89
  - (Pop aged <=14, >= 65)/total population (as %): 34.74, 8.961, 28.54, 49.99
  - Dummy for IT: 0.45, 0.49, 0, 1
  - Capital openness: 0.151, 1.41, -1.83, 2.50
  - WEO GDP growth forecast (as %): 5.78, 9.73, -0.66, 91.73
- ADF stationarity test results (ADF test statistic, lag selected by SC):
  - Real private consumption/Real GDP: -3.089** (1)
  - Real private investment/Real GDP: -3.918*** (1)
  - Real deposit rate: -3.143** (1)
  - Real lending rate: -3.821*** (1)
  - Real government consumption/Real GDP: -3.223** (1)
  - Real GDP per capita: -2.673* (5)
  - Financial development index: -4.985*** (1)
  - (Pop aged <=14, >= 65)/total population: -5.415*** (3)
  - Dummy for IT: -5.086*** (1)
  - Capital openness: -4.207*** (1)
  - WEO GDP growth forecast: -5.924*** (6)
- (*** Significant at 1% level, ** Significant at 5% level, * Significant at 10% level.)
- Conclusion: ADF results reject unit roots for all variables — variables are stationary.

### Methodology and estimation challenges
- Endogeneity concerns: Some explanatory variables (e.g., real GDP per capita, expected output growth) may be affected by dependent variables.
- Heteroskedasticity: Breusch-Pagan LM tests indicate heteroskedastic errors in consumption and investment equations (examples provided with Chi-sq P-values in text).
- Estimation approach favored: IV-GMM (General Method of Moments with lagged endogenous variables as instruments) to correct for endogeneity, heteroskedasticity, and autocorrelation.
- Instrument validity: Hansen’s J test p-values fail to reject instrument exogeneity in reported regressions.

### Empirical findings — Private consumption
- For IT EMEs (IV-GMM results unless otherwise noted):
  - ln(Real deposit rate) coefficient: -0.258*** (0.044)
  - Financial development index: -1.54*** (0.362)
  - ln(Real deposit rate) * Financial development: 0.077*** (0.168)
  - Capital account openness: -0.022** (0.010)
  - ln(Real government consumption/Real GDP): -0.142*** (0.047)
  - ln(Real GDP per capita): -0.006 (0.005) — not significant
  - (Pop aged <=14, >= 65)/Total population: 0.015*** (0.004)
  - Hansen J χ2 p-value: 0.64
  - Interpretation:
    - Deposit interest rate has a significantly negative elasticity on private consumption.
    - Financial development has a statistically significant negative direct effect on private consumption.
    - Interaction: as financial development increases the interest rate elasticity becomes less negative (wealth effect starts to dominate substitution effect).
    - Capital account openness negatively impacts private consumption.
- For Non-IT MENA EMEs (IV-GMM results):
  - ln(Real deposit rate): -0.188 (0.203) — not significant
  - Financial development index: 9.33** (3.86)
  - ln(Real deposit rate) * Financial development: -2.06** (0.850)
  - Capital account openness: 0.102*** (0.029)
  - ln(Real government consumption/Real GDP): 0.706*** (0.092)
  - ln(Real GDP per capita): 0.134*** (0.036)
  - (Pop aged <=14, >= 65)/Total population: 0.023* (0.013)
  - Hansen J χ2 p-value: 0.81
  - Interpretation:
    - No significant direct interest-rate effect on consumption at low financial development levels.
    - Financial development has a statistically significant positive direct effect on private consumption.
    - Interaction: interest rate elasticity becomes more negative as financial development increases (substitution effect dominates).
    - Capital account openness positively impacts private consumption in MENA EMEs.
- Cross-group comparison:
  - Threshold: For IT EMEs the interest rate elasticity of private consumption turns positive when financial development index > 3.36.
  - Maximum value of financial development index in MENA EMEs is 3.25, below the 3.36 threshold — explains absence of positive wealth effect in MENA EMEs.

### Empirical findings — Private investment
- For IT EMEs (IV-GMM results):
  - ln(Real lending rate): -0.464*** (0.030)
  - Capital account openness: 0.076*** (0.027)
  - WEO GDP growth forecast: 0.078*** (0.024)
  - Financial development index and its interaction with lending rate: not statistically significant
  - Hansen J χ2 p-value: 0.65
  - Interpretation:
    - Real lending rate has a statistically significant and negative effect on private investment.
    - Capital account openness boosts private investment directly.
    - Expected output growth positively affects private investment.
- For Non-IT MENA EMEs (IV-GMM results):
  - ln(Real lending rate): -0.561*** (0.056)
  - Financial development index: 0.693*** (0.157)
  - ln(Real lending rate) * Financial development: -0.166*** (0.029)
  - Capital account openness: -0.181*** (0.035)
  - WEO GDP growth forecast: 0.224*** (0.059)
  - Hansen J χ2 p-value: 0.18
  - Interpretation:
    - Real lending rate has a statistically significant and negative effect on private investment.
    - Financial development has a statistically significant positive direct effect on private investment.
    - Interaction: interest rate elasticity of private investment becomes more negative as financial development increases.
    - Capital account openness negatively impacts private investment in MENA EMEs.
- Cross-group comparison:
  - IT EMEs: capital openness supports investment but reduces consumption; financial development does not significantly affect investment.
  - MENA EMEs: capital openness raises consumption but reduces investment; domestic financial development supports both consumption and investment.

### Conclusions and policy-relevant implications
- Interest rate channel is operative:
  - Real lending rate has a statistically significant and negative impact on private investment in both IT EMEs and non-IT MENA EMEs.
  - Real deposit rate affects private consumption in IT EMEs; in MENA EMEs the consumption interest-rate effect depends on financial development.
- Adoption of inflation targeting:
  - Adoption of IT did not significantly change private consumption or investment behavior, nor their responsiveness to interest rates, in the sample of IT EMEs.
- Role of financial sector development:
  - IT EMEs: financial development has a significant negative direct effect on private consumption; as the financial sector develops the wealth effect increasingly dominates substitution (interest-rate elasticity of consumption becomes less negative and can turn positive beyond financial development index = 3.36).
  - MENA EMEs: financial development has a significant positive direct effect on both private consumption and investment; interest-rate elasticities become more negative with greater financial development.
- Role of capital account liberalization:
  - IT EMEs: capital account openness boosts private investment but reduces private consumption.
  - MENA EMEs: capital account openness raises private consumption but reduces private investment.
- Policy implication summary:
  - The traditional Keynesian interest rate channel remains relevant for emerging market economies, including potential inflation targeters in the MENA region.
  - Financial sector development and capital account openness have materially different effects across country groups; policy design should account for country-specific financial development and institutional context.

*Source: IMF staff.*

### REFERENCES

### _wp11229 - REFERENCES

### Financial liberalization, financial development, and saving/investment
- Achy, Lahcen, 2004, “Financial Liberalization, Saving, Investment and Growth in MENA Countries”, http://129.3.20.41/eps/fin/papers/0411/0411004.pdf.
- Beck, Thorsten, Asli Demirgüç-Kunt and Ross Levine, (2000), "A New Database on Financial Development and Structure," World Bank Economic Review 14, 597-605.
- Chinn, Menzie D., and Hiro Ito, 2008, “A New Measure of Financial Openness”, Journal of Comparative Policy Analysis: Research and Practice, Vol. 10, pp. 309-322.
- Fischer, Stanley, 1998. “Capital Account Liberalization and the Role of the IMF”, Princeton Essays in International Finance 207, pp. 1-10.
- Fischer, Stanley, 2003. “Globalization and Its Challenges”, American Economic Review, Vol. 93, pp. 1-30.
- Fry, Maxwell J., 1978, “Money and Capital or Financial Deepening in Economic Development?”, Journal of Money, Credit, and Banking, Vol. 10, pp. 464-475.
- Giovannini, Alberto, 1983, The Interest Elasticity of Savings in Developing Countries: The Existing Evidence”, World Development, Vol. 11, pp. 601-607.
- Gruber, Jonathan, 2006, “A Tax-Based Estimate of the Elasticity of Intertemporal Substitution”, NBER Working Paper No. 11945.
- Reinhart, Carmen M. and Jonathan Ostry, 1995, “Saving and Real Interest Rates in Developing Countries”, University Library of Munich, Germany, MPRA Paper Number 13352.
- Reinhart, Carmen M. and Ioannis Tokatlidis, 2003, “Financial Liberalization: The African Experience”, Journal of African Economies, Vol. 12, Supplement 2, pp. 53-88.
- Townsend, Robert M., and Kenichi Ueda, 2010, “Welfare Gains From Financial Liberalization”, International Economic Review, Vol. 51, pp. 553-597.

### Stock market liberalization, equity markets, and investment efficiency
- Henry, Peter Blair, 2000a, “Do Stock Market Liberalization Cause Investment Booms?”, Journal of Financial Economics, Vol. 58, pp. 301-334.
- Henry, Peter Blair, 2000b, “Stock Market Liberalization, Economic Reform, and Emerging Market Equity Prices”, The Journal of Finance, Vol. 55, pp. 529-564.
- Misati, Roseline Nyakerario, 2007, “Liberalization, Stock Market Development and Investment Efficiency in Africa”, International Review of Business Research Papers, Vol. 3, pp. 183-191.
- Naceur, Samy Ben, Samir Ghazouani and Omran Mohammed, 2008, “Does Stock Market Liberalization Spur Financial and Economic Development in the MENA Region?”, Journal of Comparative Economics, Vol. 36, pp. 673-693.

### Monetary policy, transmission mechanisms, and inflation targeting
- Boughrara, Adel, 2009, “Monetary Policy Transmission Mechanisms in Morocco and Tunisia”, Economic Research Forum Working Paper No. 460, April, Tunis.
- Elmendorf, Douglas W., 1996, “The Effect of Interest-Rate Changes on Household Saving and Consumption: A Survey”, Finance and Economics Discussion Series, Board of Governors of the Federal Reserve System (U.S.).
- Funke, Norbert, 2002, “Stock Market Developments and Private Consumer Spending in Emerging Markets”, IMF Working Paper 02/238.
- Jardak, Tarak and Ewa Wróbel, 2009, “The Monetary Transmission Mechanism in Tunisia: Recent Developments”, Banque Centrale de Tunisie, unpublished manuscript.
- Mishra, Prachi, Peter J. Montiel and Antonio Spilimbergo, 2010, “Monetary Transmission in Low Income Countries”, IMF Working Paper WP/10/223, October, International Monetary Fund, Washington, D.C.
- Mohanty, M. S. and Philip Turner, 2008, “Monetary policy transmission in emerging market economies: what is new?”, Bank for International Settlements Papers no. 35, pp. 1-59.
- Neaime, Simon, 2008, “Monetary Policy Transmission and Targeting Mechanisms in the MENA region”, Economic Research Forum Working Paper No. 395, April, Beirut.
- Roger, Scott, 2010, “Inflation Targeting Turns 20”, Finance and Development, Vol. 47, pp. 46-49.
- Taylor, John, 1995, “The Monetary Transmission Mechanism: an Empirical Framework”, Journal of Economic Perspectives, Vol. 9, pp. 11-26.
- Sarno, Lucio and Mark P. Taylor, 1998, “Real Interest Rates, Liquidity Constraints and Financial Deregulation: Private Consumption Behavior in the U.K.”, Journal of Macroeconomics, Vol. 20, pp. 221-242.
- Mishra, Prachi, Peter J. Montiel and Antonio Spilimbergo, 2010, “Monetary Transmission in Low Income Countries”, IMF Working Paper WP/10/223, October, International Monetary Fund, Washington, D.C.

### Determinants of investment, public investment, and crowding-out
- Blejer, Mario I. and Mohsin S. Khan, 1984, “Government Policy and Private Investment in Developing Countries”, IMF Staff Papers, Vol. 31, pp. 379-403.
- Greene, Joshua and Delano Villanueva, 1991, “Investment in Developing Countries: An Empirical Analysis”, Staff Papers – International Monetary Fund, Vol. 38, pp. 33-58.
- Makdisi, Samir, Zeki Fattah and Imed Limam, 2007, “Determinants of Growth in the MENA Countries”, in Jeffrey B. Nugent and Hashem Pesaran (eds.), “Explaining Growth in the Middle East”, Amsterdam; Boston [Mass.]: Elsevier.
- Rossi Nicola, 1988, “Government Spending, the Real Interest Rate, and the Behavior of Liquidity-Constrained Consumers in Developing Countries”, Staff Papers - International Monetary Fund, Vol. 35, pp. 104-140.
- Serven, Luis, 2003, “Real-Exchange-Rate Uncertainty and Private Investment in LDCS”, The Review of Economics and Statistics, Vol. 85, pp. 212-218.
- Sundararajan, V. and Subhash Thakur 1980, “Public investment, crowding-out and growth: A dynamic model applied to India and Korea”, IMF Staff Papers, Vol. 27, pp. 814-855.
- Tun, Wai.U., and Chorng-huey Wong, 1982. “Determinants of Private Investment in Developing Countries.” Journal of Development Studies Vol. 19, pp. 19–36.

### Consumption, liquidity constraints, and household behavior
- Campbell, John Y. and N. Gregory Mankiw, 1989, “Consumption, Income and Interest Rates: Reinterpreting the Time Series Evidence”, NBER Macroeconomics Annual 1989, Volume 4, pp. 185-246.
- Elmendorf, Douglas W., 1996, “The Effect of Interest-Rate Changes on Household Saving and Consumption: A Survey”, Finance and Economics Discussion Series, Board of Governors of the Federal Reserve System (U.S.).
- Giovannini, Alberto, 1983, The Interest Elasticity of Savings in Developing Countries: The Existing Evidence”, World Development, Vol. 11, pp. 601-607.
- McDonald, Donogh, 1983, “The Determinants of Saving Behavior in Latin America”, (unpublished; Washington: International Monetary Fund).
- Rossi Nicola, 1988, “Government Spending, the Real Interest Rate, and the Behavior of Liquidity-Constrained Consumers in Developing Countries”, Staff Papers - International Monetary Fund, Vol. 35, pp. 104-140.
- Sarno, Lucio and Mark P. Taylor, 1998, “Real Interest Rates, Liquidity Constraints and Financial Deregulation: Private Consumption Behavior in the U.K.”, Journal of Macroeconomics, Vol. 20, pp. 221-242.

### Global capital markets, crises, and international institutions
- Obstfeld, Maurice. 1998. “The Global Capital Market: Benefactor or Menace?”, Journal of Economic Perspectives, Vol. 12, pp. 9-30.
- Rogoff, Kenneth. 1999. “International Institutions for Reducing Global Financial Instability”, Journal of Economic Perspectives, Vol. 13(4), pp. 21-42.
- Summers, Lawrence. 2000. “International Financial Crises: Causes, Prevention, and Cures”, American Economic Review, Vol. 90, pp. 1-16.

*Source: _wp11229 - REFERENCES*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2011/_wp11229.pdf_