• Selected
• Entire List

Abstract:
We re-assess exchange rate prediction using a wider set of models that have been proposed in the last decade: interest rate parity, productivity based models, and a composite specification. The performance of these models is compared against two reference specifications - purchasing power parity and the sticky price monetary model. The models are estimated in first-difference and error correction specifications, and model performance evaluated at forecast horizons of 1, 4 and 20 quarters, using the mean squared error, direction of change metrics, and the "consistency" test of Cheung and Chinn (1998). Overall, model/specification/currency combinations that work well in one period do not necessarily work well in another period.

exchange rates, monetary model, productivity, interest rate parity, behavioral equilibrium exchange rate model, forecasting performance

Abstract:
The rapid mortgage credit growth experienced in recent years in mature and emerging countries has raised some stability concerns. Many European credit institutions in mature markets have reacted by increasing securitization, particularly via mortgage covered bonds. From the issuer's perspective, these instruments have become an attractive funding source and a tool for asset-liability management' from the investor's perspective, covered bonds enjoy a favorable risk-return profile and a very liquid market. In this paper, we examine the two largest jumbo covered bond markets, Germany and Spain. We show how movements in covered bond prices can be used to analyze the credit developments of the underlying issuer and the quality of its mortgage portfolio. Our analysis also suggests that mortgage covered bonds could be of interest to other mature and emerging markets facing similar risks related to mortgage credit.

Mortgage covered bonds, asset swap spreads, market based indicators

Abstract:
This paper generalizes a market-based indicator for financial sector surveillance using a multifactor latent structure in the determination of the default probabilities of an nth-to-default credit default swap (CDS) basket of large complex financial institutions (LCFIs). To estimate the multifactor latent structure, we link the market risk (the covariance of the LCFIs' equity) to credit risk (the default probability of the CDS basket) in a coherent manner. In addition, to analyze the response of the probabilities of default to changing macroeconomic conditions, we run a stress test by generating shocks to the latent multifactor structure. The results unveil a rich set of default probability dynamics and help in identifying the most relevant sources of risk. We anticipate that this approach could be of value to financial supervisors and risk managers alike.

Risk management, market indicators, stress testing, credit default swap (CDS), collateralized debt obligation (CDO), credit risk, large complex financial institutions (LCFIs)

Abstract:
Previous assessments of nominal exchange rate determination have focused upon a narrow set of models typically of the 1970's vintage, including monetary and portfolio balance models. In this paper we re-assess the in-sample fit and out-of-sample prediction of a wider set of models that have been proposed in the last decade, namely interest rate parity, productivitybased models, and "behavioral equilibrium exchange rate" models. These models are compared against a benchmark model, the Dornbusch-Frankel sticky price monetary model. First, the parameter estimates of the models are compared against the theoretically predicted values. Second, we conduct an extensive out-of-sample forecasting exercise, using the last eight years of data to determine whether our in-sample conclusions hold up. We examine model performance at various forecast horizons (1 quarter, 4 quarters, 20 quarters) using differing metrics (mean squared error, direction of change), as well as the "consistency" test of Cheung and Chinn (1998). We find that no model fits the data particularly well, nor does any model consistently out-predict a random walk, even at long horizons. There is little correspondence between how well a model conforms to theoretical priors and how well the model performs in a prediction context. However, we do confirm previous findings that outperformance of a random walk is more likely at long horizons.

Exchange Rates, Monetary Model, Productivity, Interest Rate Parity, Behavioral Equilibrium Exchange Rate Model, Forecasting Performance

Abstract:
Using data from 11 main manufacturing industries in 17 OECD countries, this paper empirically investigates the determinants of cross-country differences in the persistence of productivity differentials Specifically, we focus on the effects of product market structure and technology diffusion. It is found that the manufacturing industries display a wide range of convergence rates. Consistent with theories, the persistence of productivity differentials is found to be positively correlated with the price-cost margin and the intra-industry trade index - the proxies for market monopolistic behavior. The proxies for technology diffusion, however, do not exhibit consistently significant effect. Among the conditioning macro variables, productivity convergence appears to be enhanced by human capital but deterred by government spending.

Total Factor Productivity, Convergence, Market Structure, Technology Diffusion

Abstract:
This paper reexamines previous favorable results of productivity convergence, accounting for the presence of cross-country correlation. The high levels of correlation among the OECD countries analyzed cause a significant size distortion in the testing procedure. The effect of the correlation structure on the distribution of the panel data unit root (PDUR) test statistic is investigated using Monte Carlo simulation techniques. The change in the distribution of the PDRU statistic switches the inferences drawn by the test: only 2 out of the 9 sectors analyzed show evidence of productivity convergence once the correlation structure is accounted for.

Abstract:
Previous assessments of forecasting performance of exchange rate models have focused upon a narrow set of models typically of the 1970's vintage. The canonical papers in this literature are by Meese and Rogoff (1983, 1988), who examined monetary and portfolio balance models. Succeeding works by Mark (1995) and Chinn and Meese (1995) focused on similar models. In this paper we re-assess exchange rate prediction using a wider set of models that have been proposed in the last decade: interest rate parity, productivity based models, and a composite specification incorporating the real interest differential, portfolio balance and nontradables price channels. The performance of these models is compared against two reference specifications - the purchasing power parity and the Dornbusch-Frankel sticky price monetary model. The models are estimated in error correction and first-difference specifications. Rather than estimating the cointegrating vector over the entire sample and treating it as part of the ex ante information set as is commonly done in the literature, we also update the cointegrating vector, thereby generating true ex ante forecasts. We examine model performance at various forecast horizons (1 quarter, 4 quarters, 20 quarters) using differing metrics (mean squared error, direction of change), as well as the consistency test of Cheung and Chinn (1998). No model consistently outperforms a random walk, by a mean squared error measure; however, along a direction-of-change dimension, certain structural models do outperform a random walk with statistical significance. Moreover, one finds that these forecasts are cointegrated with the actual values of exchange rates, although in a large number of cases, the elasticity of the forecasts with respect to the actual values is different from unity. Overall, model/specification/currency combinations that work well in one period will not necessarily work well in another period.

exchange rates, monetary model, productivity, interest rate parity, purchasing power

Abstract:
We reassess exchange rate prediction using a wider set of models that have been proposed in the last decade. The performance of these models is compared against two reference specificationspurchasing power parity and the sticky-price monetary model. The models are estimated in first-difference and error-correction specifications, and model performance is evaluated at forecast horizons of 1, 4, and 20 quarters, using the mean squared error, direction of change metrics, and the "consistency" test of Cheung and Chinn (1998). Overall, model/specification/currency combinations that work well in one period do not necessarily work well in another period.

exchange rates, monetary model, productivity, interest rate parity, purchasing power parity, forecasting performance

Abstract:
Previous assessments of nominal exchange rate determination have focused upon a narrow set of models typically of the 1970's vintage. The canonical papers in this literature are by Meese and Rogoff (1983, 1988), who examined monetary and portfolio balance models. Succeeding works by Mark (1995) and Chinn and Meese (1995) focused on similar models. In this paper we re-assess exchange rate prediction using a wider set of models that have been proposed in the last decade: interest rate parity, productivity based models, and 'behavioral equilibrium exchange rate' models. The performance of these models is compared against a benchmark model - the Dornbusch-Frankel sticky price monetary model. The models are estimated in error correction and first-difference specifications. Rather than estimating the cointegrating vector over the entire sample and treating it as part of the ex ante information set as is commonly done in the literature, we recursively update the cointegrating vector, thereby generating true ex ante forecasts. We examine model performance at various forecast horizons (1 quarter, 4 quarters, 20 quarters) using differing metrics (mean squared error, direction of change), as well as the 'consistency' test of Cheung and Chinn (1998). No model consistently outperforms a random walk, by a mean squared error measure; however, along a direction-of-change dimension, certain structural models do outperform a random walk with statistical significance. Moreover, one finds that these forecasts are cointegrated with the actual values of exchange rates, although in a large number of cases, the elasticity of the forecasts with respect to the actual values is different from unity. Overall, model/specification/currency combinations that work well in one period will not necessarily work well in another period.

Abstract:
This paper proposes a framework to check for consistency between the IMF's standard country surveillance tool, namely medium-term projections of the macroeconomic framework (including the real, fiscal, external, and monetary sectors), and the financial sector. Consistency here entails that the financial sector remain solvent in the medium term under the assumptions of the macroeconomic framework and that the macroeconomic framework is fine-tuned should threats to financial sector solvency arise as a result of assumptions underlying the medium-term macroeconomic framework projections. The proposed framework can also be used to conduct sensitivity analysis of the aggregated financial sector to various types of risks, including foreign exchange, interest rate, and credit risk. For surveillance purposes, this framework can easily be integrated into one of the standard sectoral files so that any update to the macroeconomic framework automatically feeds into the financial sector medium-term projections. We anticipate the proposed framework to be of interest to IMF economists as well as outside analysts.

Financial Programming, Financial Institutions, Stress Testing

Abstract:
This paper investigates output convergence for the G7 countries using multivariate time series techniques. We consider both the null hypotheses of no convergence and convergence. It is shown that inferences on output convergence depend on which one of the two null hypotheses is considered. Further, the no convergence results reported in previous studies using the time series definition may be attributed to the low power of the test procedures being used. Our results also highlight some potential problems on interpreting results from some typical multivariate unit root and stationarity tests.

Output convergence, multivariate test, unit root test, stationarity test

Abstract:
The paper presents a supervisory framework that addresses the vulnerabilities of partially dollarized banking systems. The tendency to underprice systemic liquidity risk and currency-induced credit risk creates vulnerabilities that need supervisory responses. The framework seeks to induce agents to better internalize risks by implementing a risk based approach to supervision, following the risk management guidelines of the Basel Committee, and by establishing buffers to cover higher liquidity and solvency risks. The paper also shows that most dollarized countries have addressed their liquidity vulnerabilities, but few have addressed those arising from currency-induced credit risks.

Bank Supervision, Financial Stability, Risk Management, Dollarization

Abstract:
This paper investigates cross-country productivity convergence at a sectoral level using multivariate unit-root tests. Our empirical analysis counts with three distinctive features. First, it allows all the coefficients in the panel specification to vary across countries. Second, it accounts for the presence of significant cross-country correlations found in the data. Third, when the null hypothesis of non convergence is rejected, a second test determines the number of converging countries. Based on a sample of thirteen OECD countries our results show evidence of convergence in three out of six sectors, namely, agriculture, construction, and transportation and communication services.

Abstract:
This paper investigates output convergence for the G7 countries using panel time-series techniques. We consider both the null hypotheses of no convergence and convergence. It is shown that inferences on output convergence depend on which one of the two null hypotheses is considered. Further, the no convergence results reported in previous studies using the time-series definition may be attributed to the low power of the test procedures being used. Our results also highlight some potential problems on interpreting results from some typical panel unit root and stationarity tests.

Output convergence, panel data test, unit root test, stationarity test

Abstract:
The paper empirically investigates international productivity convergence in the manufacturing sector, which was found not converging in earlier studies. The authors analyze subsectors of aggregate manufacturing in order to compare similar technologies and to avoid the mixing of converging and nonconverging subsectors in the aggregate. Some of the subsectors converge while others, as well as aggregate manufacturing, do not. There is stronger evidence of convergence in subsectors with a smaller number of different industries. The latter serves as a proxy for the variety of technologies. Overall, the results highlight the importance of comparing similar technologies when studying productivity convergence.

Abstract:
This paper investigates productivity convergence in the Manufacturing sector for a set of European countries using a time series framework. We apply a panel data unit root test (PDUR) to total factor productivity (TFP) differentials, where the null hypothesis--the presence of a unit root or a deterministic component--can be interpreted as no convergence. Earlier studies have consistently found the Manufacturing sector non-converging. Our empirical analysis reveals more evidence of convergence as higher levels of disaggregation are explored. We suggest product heterogeneity as a major source for the negative findings of convergence at the aggregate level.