## _wp1540

## Source details

**Canonical URL:** [_wp1540](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2015/_wp1540.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2015/_wp1540.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2015/_wp1540.pdf.json)

---

### Introduction — Purpose and research objective
- Develop a set of stylized empirical facts about Ghana concerning the shocks that drive macroeconomic fluctuations.
- Investigate the type of shocks and model features to consider when building a structural dynamic stochastic general equilibrium (DSGE) model for a low-income country.
- Assess the extent to which Ghana is exposed to international versus domestic shocks and motivate inclusion of global counterparts to domestic shocks.
- Compare responses in Ghana with those in South Africa to evaluate whether structural model insights from more advanced economies can be imported.

### Why Ghana and scope
- Ghana chosen because it is the only Low-income Country (LIC) currently operating with an explicit inflation targeting framework in Sub-Saharan Africa and because the Bank of Ghana has had full independence since the bill passed in December2001 and the Monetary Policy Committee was established in September2002.
- Quarterly data on real activity available only from 2006 for Ghana; authors construct their own time series for GDP to create longer series.

### Shocks studied
- Productivity shocks: exogenous shocks to total factor productivity (the standard Solow residual in a DSGE context).
- Credit shocks: exogenous shocks to the supply of credit.
- Commodity price shocks: focus on cocoa and gold; crude oil excluded for most of the sample because exports of crude oil began in 2011.
- Monetary policy shocks not considered in this VAR-based analysis due to changes in monetary policy over the sample period.

### Empirical approach — VAR and identification
- VAR estimated with Bayesian methods (BVAR), using sign restrictions in the spirit of Uhlig (2005) for credit and productivity shocks.
- Commodity price shocks exploit exogeneity for a small economy and use a recursive setup with commodity prices ordered first.
- Global shocks estimated using similar restrictions on data constructed from the first principal component of G7 country-level data.
- Sequential ordering for identification reported: G7-credit shocks, G7-productivity shocks, domestic (SA or GHA)-credit shocks, domestic (SA or GHA)-productivity shocks.
- Penalty function approach based on Mountford and Uhlig (2009) and Uhlig (2005); objective function explicitly modified to allow zero restrictions.
- VAR model setup for credit and productivity shocks: 16 variables in Yt:
  - Six foreign indicators (G7): G7-real GDP; G7-inflation; G7-real credit; G7-short-term interest rates; US-credit spread; US-default rates.
  - Ten home-economy series: six domestic indicators analogous to G7, plus exports, imports, primary commodity prices, and the real exchange rate.
- VAR set so global shocks impact Ghana (or South Africa), but domestic shocks have no impact on global aggregates.
- Sample and transformations:
  - Quarterly data from Ghana, South Africa and G7-countries in 1985 : 1–2010 : 3.
  - Where appropriate series transformed into year to year growth rates.
  - VAR estimated with 3 lags.
- Priors and estimation details:
  - Combination of: i) Normal-inverted Wishart prior; and ii) a Minnesota-type prior that assigns low weights on off-diagonal AR coefficients and zero weights on coefficients related to the home economy’s indicators in the block defined by (primary) commodity prices and G7-factors.
  - Results reported are based on 250 draws; authors state larger number of draws leaves qualitative results unchanged.

### Identification restrictions for shock types
- Credit supply shock identification:
  - Characterized by an increase in the credit spread and a decrease in real credit.
  - Additional restriction: default rates on corporate bonds do not increase.
- Adverse productivity shock identification:
  - Identified as decreases in output that increase inflation (consistent with a New Keynesian DSGE model).
  - Require that real credit does not increase and that default rates do not decrease.
- Zero restrictions used to disentangle South African shocks from global shocks by assuming G7 countries do not respond to shocks originating from South Africa.
- Restrictions imposed over the first four quarters and assume negative shocks for sign restrictions.

### Empirical challenges and data construction
- Two empirical challenges for low-income countries:
  - Data quality is typically poor and time series lengths are short relative to advanced economies.
  - Economic structure is often quite different from advanced economies.
- Authors address data limitations by constructing own GDP series, pooling data of different frequencies, and using common component and now-cast approaches (Giannone et al. (2008)).
- Monetary policy regime context:
  - 1983–1992 Economic Recovery Program (ERP) initiated in April1983; sample starts in 1985 to account for the policy break and start of globalization period.
  - Transition from direct credit controls to monetary targeting from 1983; transition to inflation targeting began in 2001 and was formally adopted in 2007.
  - Exchange rate liberalization culminated in full liberalization in 1990.

### Contribution relative to literature
- Documents roles of credit supply and productivity shocks for developing countries.
- Compares domestic versus foreign nature of these shocks and compares Ghana with South Africa.
- Fills gap on macroeconomic impacts of credit supply shocks in Emerging Markets and LICs; prior work identifying credit supply shocks for emerging markets was limited.
- Uses SVAR/BVAR with sign restrictions to distinguish exogenous credit supply shocks from endogenous credit responses.

### Commodity price shocks — Identification and methodology (Section 3.3)
- VAR model with six variables: commodity price; CPI, GDP, Credit; short-term interest rate; real effective exchange rate.
- Recursive identification with commodity prices ordered first.
- Analysis focuses on gold price; nominal commodity price series deflated by the US producer price index (PPI).
- Economic impact of an increase in commodity prices estimated over three sample periods:
  - full sample: 1985 : 1–2010 : 3
  - before the introduction of the inflation targeting framework (IT)
  - after the introduction of the inflation targeting framework (IT)
- Sub-periods:
  - South Africa before and after IT: 1985 : 1–1999 : 4 and 2001–2010 : 3
  - Ghana before and after IT: 1985 : 1–2001 : 4 and 2002–2010 : 3
- Lag length fixed to 3.

### Empirical results — global shocks
- Global credit supply shocks:
  - Significant impact on output in South Africa, but not in Ghana.
  - Modest reduction in credit in both countries.
  - Credit supply shock lowers commodity prices in South Africa, while raising them in Ghana.
  - Interpretation: South Africa is financially integrated such that global credit conditions affect the domestic economy; Ghana is not.
- Global productivity shocks:
  - Similar shapes and magnitudes of response in Ghana and South Africa.
  - Drops in productivity are inflationary.
  - Commodity prices fall in both countries; the impact is not significant in Ghana.
  - World gold price increases following adverse productivity and credit shocks, though the impact is modest.

### Empirical results — domestic shocks
- Domestic credit supply shocks:
  - Generate inflationary pressures in both South Africa and Ghana; quantitatively larger impact in Ghana.
  - Lead to an increase in interest rates in Ghana and South Africa, though only significant for a short period.
  - Inflation effects dominate small output losses caused by financial market disruptions.
  - Similarities suggest monetary policy rules in South Africa can inform modeling for Ghana.
- Domestic productivity shocks:
  - Significant impact on Ghana; more muted on South Africa.
  - By construction these shocks lower output and raise inflation in both countries.
  - Short-term interest rates rise in Ghana and fall in South Africa in response to the shock.
  - Resulting real exchange rate appreciation in Ghana lowers exports and raises imports; in South Africa real exchange rates do not move while exports fall and imports rise.
  - Suggests monetary policy in Ghana focuses more on inflation than on output drops, possibly reflecting stronger policy responses to build credibility.

### Commodity (gold) price shocks — empirical patterns
- Figure 7 (gold price shocks) summarized:
  - Full sample:
    - South Africa: output decreases on impact but later increases significantly and turns positive during several quarters.
    - Ghana: output increases significantly on impact following an increase in gold prices.
  - Before IT vs during IT:
    - During IT the impact response is negative in Ghana as well.
    - For South Africa output response to gold shocks has become less persistent during the IT period.
  - Monetary policy interactions:
    - Before IT in South Africa: inflation increases on impact but decreases significantly following the shocks.
    - During IT in South Africa: inflation shows positive responses; central bank raises the policy rate to respond to inflationary pressure, helping contain the output boom.
    - Bank of Ghana during IT: inflation decreases on impact and increases later; policy rate also decreases first and later increases.
    - Bank of Ghana before IT: inflation and output increase on impact but the policy rate is reduced.

### Variance decomposition (key numerical findings)
- At the 3-year horizon for credit and productivity shocks:
  - Global shocks are more important in both countries, with larger impact in South Africa than in Ghana (Table 3).
  - "For instance, each of these global shock explains about8and11%of output variation in Ghana and South Africa, respectively."
  - Helbling et al. (2011) report global credit and productivity shocks explain each about12%of output fluctuation in G7-countries.
- Short-term dynamics:
  - Domestic credit and productivity shocks dominate their global counterparts for inflation and credit.
  - Domestic productivity shocks are the main drivers of output and inflation fluctuations in the short run.
  - For real credit, domestic credit shocks account for a large portion of the volatility, with productivity shocks contributing as well.
  - For short-term interest rates and real effective exchange rates, global credit and productivity shocks still explain 10 percent of the volatility.
- Commodity price shocks:
  - More important for South Africa than for Ghana (Table 4).
  - The share of variance due to commodity price shocks is in general a bit larger than that of productivity or credit shocks for both countries.
  - There has been an increase of the role of commodity price shocks in macroeconomic fluctuation in recent years.
  - "In the inflation targeting regime commodity price shocks account for 20-25 percent of the volatility of all variables in the VAR."
- Overall accounting:
  - The five shocks (using the IT period of commodity price shocks) account for half the fluctuations in the main macroeconomic variables.
  - A full accounting of macroeconomic volatility in Ghana likely requires both fiscal and monetary policy shocks in addition to the analyzed shocks.

### Conclusions and policy implications
- Global productivity and credit shocks have a greater impact on South Africa than on Ghana.
- Ghana's integration with the world is more through trade channels and less through financial channels.
- Domestic shocks indicate monetary policy in Ghana has responded more strongly to inflation than to output drops, unlike in South Africa.
- Commodity shocks are an important driver of business cycles in both countries; dependence on the commodity sector is central for modeling business cycle stabilization in Ghana.
- Modeling and policy recommendations:
  - Monetary policy rules from South Africa may inform Ghanaian modeling given response similarities.
  - For Ghana, construct models that emphasize the primary goods sector and trade-channel integration.
  - Further study of fiscal shocks would require narrative approaches; regime-switching structural models would be more appropriate to study frequent monetary policy regime changes.

*Source — _wp1540.*

### 1. Introduction. .......................................................................................................

### _wp1540 - 1. Introduction.

### Purpose and research objective
- Develop a set of stylized empirical facts about Ghana concerning the shocks that drive macroeconomic fluctuations.
- Investigate the type of shocks and model features to consider when building a structural dynamic stochastic general equilibrium (DSGE) model for a low-income country.
- Assess the extent to which Ghana is exposed to international versus domestic shocks and motivate inclusion of global counterparts to domestic shocks.
- Compare responses in Ghana with those in South Africa to evaluate whether structural model insights from more advanced economies can be imported.

### Why Ghana and scope
- Ghana chosen because it is the only Low-income Country (LIC) currently operating with an explicit inflation targeting framework in Sub-Saharan Africa and because the Bank of Ghana has had full independence since the bill passed in December2001 and the Monetary Policy Committee was established in September2002.
- Time span and data considerations highlighted: quarterly data on real activity available only from 2006 for Ghana; authors construct their own time series for GDP to create longer series.

### Shocks studied
- Productivity shocks: interpreted as exogenous shocks to total factor productivity (the standard Solow residual in a DSGE context).
- Credit shocks: interpreted as exogenous shocks to the supply of credit.
- Commodity price shocks: focus on prices for cocoa and gold (traditional exports produced for the entire sample); crude oil excluded for most of the sample because exports of crude oil began in 2011.
- Monetary policy shocks not considered in this VAR-based analysis due to changes in monetary policy over the sample period; left for a more structural investigation.

### Empirical approach — VAR and identification
- Use a Vector Autoregression (VAR) estimated with Bayesian methods (BVAR).
- Identification:
  - Sign restrictions in the spirit of Uhlig (2005) for credit and productivity shocks.
  - For commodity price shocks exploit exogeneity for a small economy and use a recursive setup with commodity prices ordered first.
  - Global shocks estimated using similar restrictions on data constructed from the first principal component of G7 country-level data.
- Sequential ordering for identification reported: G7-credit shocks, G7-productivity shocks, domestic (SA or GHA)-credit shocks, domestic (SA or GHA)-productivity shocks (authors note different ordering does not change main results).
- Penalty function approach based on Mountford and Uhlig (2009) and Uhlig (2005) used; objective function explicitly modified to allow zero restrictions.

### Data and model specification
- VAR model setup:
  - For credit and productivity shocks: 16 variables in Yt.
  - Six foreign indicators (G7): G7-real GDP; G7-inflation; G7-real credit; G7-short-term interest rates; US-credit spread; US-default rates.
    - The four G7-factors are estimated by extracting the first principal component from G7 country series.
    - Remaining two indicators (US series) used due to data limitations.
  - Ten home-economy series: six domestic indicators analogous to G7, plus exports, imports, primary commodity prices, and the real exchange rate.
  - VAR is set up so global shocks impact Ghana (or South Africa), but domestic shocks have no impact on global aggregates.
- Sample and transformations:
  - Quarterly data from Ghana, South Africa and G7-countries in 1985 : 1–2010 : 3.
  - Where appropriate series transformed into year to year growth rates.
  - VAR estimated with 3 lags.
- Priors and estimation details:
  - Combination of two types of priors: i) Normal-inverted Wishart prior; and ii) a Minnesota-type prior that assigns low weights on off-diagonal AR coefficients and zero weights on coefficients related to the home economy’s indicators in the block defined by (primary) commodity prices and G7-factors (consistent with the Small Open Economy assumption).
  - Results reported are based on 250 draws; authors state larger number of draws leaves qualitative results unchanged.

### Identification restrictions for shock types
- Credit supply shock identification:
  - Characterized by an increase in the credit spread and a decrease in real credit.
  - Additional restriction: default rates on corporate bonds do not increase (helps isolate exogenous credit supply shocks from endogenous credit responses).
- Adverse productivity shock identification:
  - Identified as decreases in output that increase inflation (consistent with a New Keynesian DSGE model).
  - Require that real credit does not increase and that default rates do not decrease (to discriminate from credit supply shocks and endogenous credit responses).
- Zero restrictions used to disentangle South African shocks from global shocks by assuming G7 countries do not respond to shocks originating from South Africa.
- Restrictions imposed over the first four quarters and assume negative shocks for sign restrictions.

### Empirical challenges and data construction
- Two empirical challenges for low-income countries:
  - Data quality is typically poor and time series lengths are short relative to advanced economies.
  - Economic structure is often quite different from advanced economies.
- Authors address the first challenge by constructing own time series for GDP, pooling data of different frequencies (yearly, quarterly, monthly) and using common component and now-cast approaches (Giannone et al. (2008)) to estimate longer real activity series.
- Monetary policy regime changes described:
  - 1983–1992 Economic Recovery Program (ERP) initiated in April1983; authors start sample in 1985 to account for the policy break and start of globalization period.
  - Monetary policy shifted from direct credit controls to monetary targeting from 1983; transition to inflation targeting began in 2001 and was formally adopted in 2007.
  - Exchange rate liberalization culminated in full liberalization in 1990 when the exchange rate was set in the inter-bank market.

### Contribution relative to literature
- Adds to literature on sources of macroeconomic fluctuations in developing countries by:
  - Documenting roles of credit supply and productivity shocks.
  - Comparing domestic versus foreign nature of these shocks.
  - Comparing economies with similar production structure at different development stages (Ghana vs South Africa).
- Fills a gap regarding macroeconomic impacts of credit supply shocks in Emerging Markets and LICs; to authors' knowledge only Tamasi and Vilagi (2011) explicitly identify credit supply shocks for emerging markets prior to this work.
- Uses SVAR/BVAR with sign restrictions to distinguish exogenous credit supply shocks from endogenous credit responses, complementing regression-based studies that document credit–real activity co-movement but struggle with causality.

*Italic: Source — _wp1540 - 1. Introduction.*

### 3.3  Commodity Price shocks

### 3.3  Commodity Price shocks

### Identification and methodology
- VAR model with six variables: commodity price; CPI, GDP, Credit; short-term interest rate and the real e§ective exchange rate.
- Recursive identiÖcation scheme with commodity prices ordered Örst.
- Analysis focuses on gold price; nominal commodity price series deáated by the US producer price index (PPI).
- Economic impact of an increase in commodity prices estimated over three sample periods:
  - full sample: 1985 : 12010 : 3
  - before the introduction of the ináation targeting framework (IT)
  - after the introduction of the ináation targeting framework (IT)
- Sub-periods:
  - South Africa before and after IT: 1985 : 11999 : 4 and 20012010 : 3
  - Ghana before and after IT: 1985 : 12001 : 4 and 20022010 : 3
- Lag length fixed to 3.

### Empirical results — global shocks
- Global credit supply shocks:
  - Significant impact on output in South Africa, but not in Ghana.
  - Modest reduction in credit in both countries.
  - Credit supply shock lowers commodity prices in South Africa, while raising them in Ghana.
  - Suggests South Africa is Önancially integrated such that global credit conditions a§ect the domestic economy; Ghana is not.
- Global productivity shocks:
  - Similar shapes and magnitudes of response in Ghana and South Africa.
  - Drops in productivity are ináationary consistent with New Keynesian predictions.
  - Commodity prices fall in both countries; the impact is not signiÖcant in Ghana.
  - World gold price increases following adverse productivity and credit shocks, though the impact is modest.

### Empirical results — domestic shocks
- Domestic credit supply shocks:
  - Generate ináationary pressures in both South Africa and Ghana; quantitatively larger impact in Ghana.
  - Lead to an increase in interest rates in Ghana and South Africa, though the impact is only signiÖcant for a short period of time.
  - Indicates ináation is more important for both countries than the small output losses caused by Önancial market disruptions.
  - Similarities suggest monetary policy rules in South Africa can be exploited in building a model for Ghana.
- Domestic productivity shocks:
  - SigniÖcant impact on Ghana; more muted impact on South Africa.
  - By construction these shocks lower output and raise ináation in both countries.
  - Short-term interest rates rise in Ghana and fall in South Africa in response to the shock.
  - Resulting real exchange rate appreciation in Ghana lowers exports and raises imports; in South Africa real exchange rates do not move while exports fall and imports rise.
  - Suggests monetary policy in Ghana focuses more on ináation than on output drops, possibly reflecting stronger policy responses to build credibility.

### Commodity (gold) price shocks
- Figure 7 results (gold price shocks):
  - Full sample:
    - South Africa: output decreases on impact but later increases signiÖcantly and turns positive during several quarters.
    - Ghana: output increases signiÖcantly on impact following an increase in gold prices.
  - Before IT vs during IT:
    - During IT the impact response is negative in Ghana as well.
    - For South Africa output response to gold shocks has become less persistent during the IT period.
  - Monetary policy interactions:
    - Before IT in South Africa: ináation increases on impact but decreases signiÖcantly following the shocks.
    - During IT in South Africa: ináation shows positive responses; central bank raises the policy rate to respond to ináationary pressure, helping contain the output boom.
    - Bank of Ghana during IT: ináation decreases on impact and increases later; policy rate also decreases Örst and later increases.
    - Bank of Ghana before IT: ináation and output increase on impact but the policy rate is reduced (harder to explain).

### Variance decomposition
- At the 3-year horizon for credit and productivity shocks:
  - Global shocks are more important in both countries, with larger impact in South Africa than in Ghana (Table 3).
  - "For instance, each of these global shock explains about8and11%of output variation in Ghana and South Africa, respectively."
  - Helbling et al. (2011) report global credit and productivity shocks explain each about12%of output áuctuation in G7-countries.
- Short-term dynamics:
  - Domestic credit and productivity shocks dominate their global counterparts for ináation and credit.
  - Domestic productivity shocks are the main drivers of output and ináation áuctuations in the short run.
  - For real credit, domestic credit shocks account for a large portion of the volatility, with productivity shocks contributing as well.
  - For short-term interest rates and the real e§ective exchange rates, global credit and productivity shocks still explain 10 percent of the volatility.
- Commodity price shocks:
  - More important for South Africa than for Ghana (Table 4).
  - The share of variance due to commodity price shocks is in general a bit larger than that of productivity or credit shocks for both countries.
  - There has been an increase of the role of commodity price shocks in macroeconomic áuctuation in recent years.
  - "In the ináation targeting regime commodity price shocks account for 20-25 percent of the volatility of all variables in the VAR."
- Overall:
  - The five shocks (using the IT period of commodity price shocks) account for half the áuctuations in the main macroeconomic variables.
  - A full accounting of macroeconomic volatility in Ghana likely requires both Öscal and monetary policy shocks in addition to the analyzed shocks.
  - The study focuses on exogenous shocks rather than policy shocks due to data limits on Öscal policy and frequent changes in monetary policy structure.

### Conclusions and policy implications
- Global productivity and credit shocks have a greater impact on South Africa than on Ghana.
- Ghana's integration with the world is more through trade channels and less through Önancial channels.
- Domestic shocks indicate monetary policy in Ghana has responded more strongly to ináation than to output drops, unlike in South Africa.
- Commodity shocks are an important driver of business cycles in both countries; dependence on the commodity sector is central for modeling business cycle stabilization in Ghana.
- Modeling and policy recommendations:
  - Monetary policy rules from South Africa may inform Ghanaian modeling given response similarities.
  - For Ghana, constructing models that emphasize the primary goods sector and trade-channel integration is important.
  - Further study of Öscal shocks would require narrative approaches; regime-switching structural models would be more appropriate to study frequent monetary policy regime changes.

*Source: _wp1540 - 3.3  Commodity Price shocks*

### References

### _wp1540 - References

### References (bibliographic list)
- Abradu-Otoo, P., B. Amoah, and M. Bawumia (2003): ìAn investigation of the transmission mechanisms of monetary policy in ghana: A structural vector error correction analysis,î Working Papers 2003/02, Bank of Ghana.
- Agenor, P.-R., J. McDermott, and E. Prasad (2000): ìMacroeconomic áuctuations in developing countries: Some stylized facts,î World Bank Economic Review, 14, 251ñ285.
- Akinboade, O. A. and D. Makina (2010): ìEconometric analysis of bank lending and business cycles in south africa,î Applied Economics, 42, 3803ñ3811.
- Atta-Mensah, J. and A. Dib (2008): ìBank lending, credit shocks, and the transmission of canadian monetary policy,î International Review of Economics & Finance, 17, 159ñ176.
- Broda, C. (2004): ìTerms of trade and exchange rate regimes in developing countries,î Journal of International Economics, 63, 31ñ58.
- Cashin, P. (2004): ìCaribbean Business Cycles,î IMF working papers 04/136, International Monetary Fund.
- Cetorelli, N. and L. S. Goldberg (2010): ìGlobal banks and international shock transmission: Evidence from the crisis,î Working Paper 15974, National Bureau of Economic Research.
- Chia, W.-M. and J. D. Alba (2006): ìTerms-of-trade shocks and exchange rate regimes in a small open economy,î The Economic Record, 82, S41ñS53.
- Crucini, M., A. Kose, and C. Otrok (2011): ìWhat are the driving forces of international business cycles?î Review of Economic Dynamics, 14, 156ñ175.
- Deaton, A. and R. Miller (1996): ìInternational commodity prices, macroeconomic performance and politics in sub-saharan africa,î Journal of African Economies, 5, 99ñ191.
- Doz, C., D. Giannone, and L. Reichlin (2011): ìA two-step estimator for large approximate dynamic factor models based on Kalman Öltering,î Journal of Econometrics, 164, 188ñ205.
- du Plessis, S. (2006): ìBusiness Cycles in Emerging market Economies: A New View of the Stylised Facts,î Working Papers 02/200, Stellenbosch University, Department of Economics.
- Gerali, A., S. Neri, L. Sessa, and F. M. Signoretti (2010): ìCredit and banking in a dsge model of the euro area,î Journal of Money, Credit and Banking, 42, 107ñ141.
- Giannone, D., L. Reichlin, and D. Small (2008): ìNowcasting: The real-time informational content of macroeconomic data,î Journal of Monetary Economics, 55, 665ñ676.
- Gilchrist, S., V. Yankov, and E. Zakrajsek (2009): ìCredit market shocks and economic áuctuations: Evidence from corporate bond and stock markets,î Journal of Monetary Economics, 56, 471ñ493.
- Harding, D. and A. Pagan (2002): ìDissecting the cycle: a methodological investigation,î Journal of Monetary Economics, 49, 365ñ381.
- Helbling, T., R. Huidrom, M. A. Kose, and C. Otrok (2011): ìDo credit shocks matter? a global perspective,î European Economic Review, 55, 340ñ353.
- Ho§maister, A. W., J. E. Roldos, and P. Wickham (1998): ìMacroeconomic áuctuations in sub-saharan africa,î IMF Sta§ Papers, 45, 132ñ160.
- Ho§maister, A. W. and J. RoldÛs (1997): ìAre business cycles di§erent in asia and latin america,î IMF working papers 97/9, International Monetary Fund.
- Houssa, R. (2008): ìMonetary union in west africa and asymmetric shocks: A dynamic structural factor model approach,î Journal of Development Economics, 85, 319ñ347.
- Houssa, R. (2009): ìAsymmetric shocks in the west african monetary union,î Research paper 187, African Economic Research Consortium.
- Houssa, R., J. Mohimont, and C. Otrok (2013): ìCredit Shocks and Macroeconomic Fluctuations in Emerging Markets,î CESifo Working Paper Series 4281, CESifo Group Munich.
- Houssa, R., C. Otrok, and R. Puslenghea (2010): ìA model for monetary policy analysis for sub-saharan africa,î Open Economies Review, 21, 127ñ145.
- Kose, M. A. (2002): ìExplaining business cycles in small open economies: íhow much do world prices matter?í,î Journal of International Economics, 56, 299ñ327.
- Kose, M. A. and R. Riezman (2001): ìTrade shocks and macroeconomic áuctuations in africa,î Journal of Development Economics, 65, 55ñ80.
- Male, R. (2011): ìDeveloping Country Business Cycles: Characterizing the Cycle,î Emerging Markets Finance and Trade, 47, 20ñ39.
- Meeks, R. (2012): ìDo credit market shocks drive output áuctuations? evidence from corporate spreads and defaults,î Journal of Economic Dynamics and Control, 36, 568ñ584.
- Mendoza, E. G. (1995): ìThe terms of trade, the real exchange rate, and economic áuctuations,î International Economic Review, 36, 101ñ37.
- Mountford, A. and H. Uhlig (2009): ìWhat are the e§ects of Öscal policy shocks?î Journal of Applied Econometrics, 24, 960ñ992.
- Schnabl, P. (2012): ìThe international transmission of bank liquidity shocks: Evidence from an emerging market,î Journal of Finance, 67, 897ñ932.
- Sowa, N. K. and I. K. Acquaye (1999): ìFinancial and foreign exchange markets liberalization in ghana,î Journal of International Development, 11, 385ñ409.
- Tamasi, B. and B. Vilagi (2011): ìIdentiÖcation of credit supply shocks in a bayesian svar model of the hungarian economy,î MNB Working Papers 2011/7, Magyar Nemzeti Bank (The Central Bank of Hungary).
- Uhlig, H. (2005): ìWhat are the e§ects of monetary policy on output? results from an agnostic identiÖcation procedure,î Journal of Monetary Economics, 52, 381ñ419.

### Figures (list of figures and depicted indicators)
- Figure 1: Selected Macroeconomic indicators for Ghana (YoY percent)
  - Common Component of RGDP/RGDP
  - CPI Inflation
  - Base Money growth
  - Monetary Policy Rate
  - Time coverage shown: Q1-80 through Q1-15
  - Axes include numeric scales such as 0 2 4 6 8 10 12 14; 0 20 40 60 80 100 120 140 160 180; -10 0 10 20 30 40 50 60 70; 10 15 20 25 30 35 40 45
- Figure 2: Exchange Rates and International Reserves (YoY percent)
  - International Reserves, Real Effective ER, Nominal Effective Exchange Rate
  - Time coverage shown: Q1-80 through Q1-15 and 1980–2015
  - Axes numeric scales include -100 -50 0 50 100 150 200 250; -250 -200 -150 -100 -50 0 50 100; -100 -80 -60 -40 -20 0 20 40
- Figures 3–7: Impulse response panels for shocks (G7 Credit Shocks, G7 Productivity Shocks, Domestic Credit Shocks, Domestic Productivity Shocks, Commodity Price Shocks)
  - Variables depicted across panels: Output Growth, CPI-Ináation, Real Credit, Short-term Interest Rates, Commodity Prices (Gold, Other Mining for SA, Cocoa for Ghana), REER, Exports, Imports, Real Effective Exchange Rates
  - Horizontal axis labels often show 1 through 12 periods; vertical axis numeric scales vary by panel (examples: -2 to 0.5; -3 to 2; -6.5 to 1.5; -2.5 to 3; -10 to 8; -20 to 20)
  - Figure 7 includes comparisons by sample: Full Sample, Before IT, During IT

### Tables (data descriptions, identification restrictions, variance decompositions, nowcasting inputs)
- Table 1: Data — Part A (series, transformations, frequency/source by country)
  - Output: year to year (Canada, France, Germany, Italy, Japan, UK, USA, South Africa, Ghana)
  - Real Credit: year to year or level depending on series; sources include IFS, CLAIMS ON PRIVATE SECTOR, CREDIT TO PRIVATE SECTOR, CLAIMS ON OTHER RESIDENT SECTOR, SARB, GSS
  - Short-run interest rates: level (IFS, treasury bill measures, Treasury Bill Tender Rate)
  - CPI: year to year (IFS; specific CPIs listed e.g., IFS, CPI: 108 CITIES; IFS, CPI:ALL JAPAN-485 ITEMS)
  - Notes: G7 indicators represented by First principal component; Ghana backcast data from 1990-2005
- Table 1: Data — Part B (credit spreads, default, REER, Commodity prices, Exports, Imports)
  - USA credit spreads: Macrobond, baa-aaa on US corporate bond yields; Gilchrist et al. (2009) distance to default transformed to inverse distance
  - South Africa: SARB spreads (Eskom bond yield and US-baa bond yield), South Africa Statistics insolvencies (Number), IFS Real Effective Exchange Rate, IFS gold, OECD MEI Exports/Imports of Goods And Services
  - Ghana: SA credit spreads, SA default rates, IFS Real Effective Exchange Rate, IFS cocoa, IMF DOTS Exports/Imports of Goods And Services
- Table 2: Identification Restrictions (sign restrictions imposed on the first quarters)
  - Indicators listed 1 to 16, including G7-Real GDP, G7-Ináation, G7-Real Credit, G7-Tbil, US-Credit Spread, US-Default, (SA or GHA)-Real GDP, (SA or GHA)-Ináation, (SA or GHA)-Real Credit, (SA or GHA)-Tbil, SA-Credit Spread, SA-Default, (SA or GHA)-Commodity Price, REER of the rand or the GHS, (SA or GHA)-Export, (SA or GHA)-Import
  - Sign markers in matrix include , +, 0 indicating imposed signs or zeros
  - Note: The sign restrictions were imposed on the Örst quarters.
- Table 3: Variance Decomposition for Credit and Productivity Shocks
  - Columns: G7-Credit, G7-Productivity, Dom.-Credit, Dom.-Productivity
  - Rows organized by Horizons and variables (Output Growth, CPI-Ináation, Short-term Interest Rates, Real Credit, Real Effective Exchange Rates) with numeric entries as presented (examples extracted verbatim):
    - Output Growth: 13.082.250.688.35; 46.656.461.325.73
    - South Africa rows: 89.3810.401.473.97; 1210.1011.241.263.07
    - Ghana rows: 12.862.781.244.41; 45.436.161.133.51
    - CPI-Ináation rows: 13.023.2011.5018.37; 44.064.029.3614.46
    - Short-term Interest Rates rows: 13.824.031.211.46; 45.545.741.542.03
    - Real Credit rows: 12.392.2716.386.71; 45.095.6512.725.53
    - Real Effective Exchange Rates rows: 15.115.972.252.25; 46.386.322.342.32
  - Table spans multiple country-specific entries with corresponding numeric decompositions
- Table 4: Variance Decomposition for Commodity Price Shocks (Full Period / Before IT / During IT)
  - Variables and horizons similar to Table 3 with numeric entries presented verbatim (examples):
    - Output Growth: 13.682.182.58; 44.116.138.43
    - South Africa rows: 810.6212.6113.18; 1213.8814.6616.32
    - Ghana rows: 12.0711.016.15; 44.6213.1919.44
    - CPI-Ináation rows: 11.4111.362.35; 41.707.266.36
    - Short-term Interest Rates rows: 15.914.013.12; 49.369.156.60
    - Real Credit rows: 10.579.585.84; 41.977.689.14
    - Real Effective Exchange Rates rows: 13.669.062.60; 45.2610.377.31
- Table 5: Data used in the Nowcasting
  - Series and frequency/sources:
    - Goods Exports (NSA, Mil.Rand) — Monthly — IMF, DOTS
    - Goods Imports (NSA, Mil.Rand) — Monthly — IMF, DOTS
    - Commodity exports — Monthly — Bank of Ghana
    - Commodity imports — Monthly — Bank of Ghana
    - Sales — Monthly — Bank of Ghana
    - Port Harbour Operations — Monthly — Bank of Ghana
    - Domestic VAT — Monthly — Bank of Ghana
    - Gross Domestic Product (SA, Mil.2006.Cedis) — Quarterly — Ghana Stat. Service
    - Real GDP growth — Yearly — IMF, WEO

*Italic: Source document content as provided in _wp1540 - References*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2015/_wp1540.pdf_