Fractional cointegration and real exchange rates

This paper uses fractional integration and cointegration to model the DM-US dollar and the yen-US dollar real exchange rates in terms of both monetary and real factors, more specifically real interest rate and labour productivity differentials. We find that whilst the individual series may be integrated of order 1, their long-run relationship might have a fractionally cointegrated structure. This means that mean reversion occurs, consistently with the findings of other studies. However, it also indicates, in contrast to such studies, that the cointegrating relationship possesses long memory. In other words, the error correction term responds slowly to shocks, implying that deviations from equilibrium are long-lived. It appears that only a combination of real and monetary variables can accurately track down the movements of real exchange rates.     

Macroeconomic shocks, structural change and real exchange rates: Evidence from historical data

We present empirical evidence on the forces driving real exchange rates in the long-run. Using data from the US, UK and Italy across different exchange rate regimes, we find support for the hypothesis that productivity and fiscal shocks matter. However, in some cases fiscal shocks cause depreciations, likely triggered by the monetary accommodation of fiscal shocks. We also find that the traditional Harrod–Balassa–Samuelson effect of productivity on real exchange rates is reversed in some cases, which confirms the importance of the distributive sector in driving productivity gains.     

Large versus small foreign exchange interventions

We use non-temporal threshold analysis to investigate the exchange rate effects of large versus small interventions. More than two decades of official daily data on intervention in the JPY/USD market facilitate our analysis. We find no evidence that small interventions exert a discernible influence on the exchange rate while large interventions significantly influence the exchange rate in the theoretically consistent manner. We conclude that small interventions may not be considered a determinant of the exchange rate while large interventions constitute an important element in our understanding, and modeling, of the exchange rate.     

Dynamical clustering of exchange rates

We use techniques from network science to study correlations in the foreign exchange (FX) market during the period 1991–2008. We consider an FX market network in which each node represents an exchange rate and each weighted edge represents a time-dependent correlation between the rates. To provide insights into the clustering of the exchange-rate time series, we investigate dynamic communities in the network. We show that there is a relationship between an exchange rate's functional role within the market and its position within its community and use a node-centric community analysis to track the temporal dynamics of such roles. This reveals which exchange rates dominate the market at particular times and also identifies exchange rates that experienced significant changes in market role. We also use the community dynamics to uncover major structural changes that occurred in the FX market. Our techniques are general and will be similarly useful for investigating correlations in other markets.     

Nonlinear trends in real exchange rates: A panel unit root test approach

We analyze the possibility of nonlinear trend stationarity as the alternative to unit roots in 23 OECD real exchange rates, 1974–1998, by adding nonlinear time terms to the CIPS panel unit root test of Pesaran (2007). We follow a thorough bootstrapping approach and propose a technique to adjust statistical significance for the use of multiple tests over several time trend orders. The unit root null that all real exchange rates have unit roots is rejected at better than the 0.05 level. Bootstrapped results from a procedure of Chortareas and Kapetanios (2009) suggest that the hypothesis that all are stationary is reasonable. We argue that nonlinear trend stationarity is the most likely alternative hypothesis for at least some of the real exchange rates because: (1) the strongest CIPS rejection occurs when quadratic trends are specified; (2) nonlinear time terms are statistically significant at the 0.10 level; (3) the actual CIPS statistics are more consistent with CIPS sampling distributions from bootstrapped nonlinear trend stationary processes than from linear trend or mean stationary processes.     

U.S. real exchange rate fluctuations and relative price fluctuations

Traditional theory attributes fluctuations in real exchange rates to changes in the relative price of nontraded goods. This paper studies the relation between the United States’ bilateral real exchange rate and the associated bilateral relative price of nontraded goods for five of its most important trade relationships. We find that this relation depends crucially on the choice of price series used to measure relative prices and on the choice of trade partner. The relation is stronger when we measure relative prices using producer prices rather than consumer prices. The relation is stronger the more important is the trade relationship between the United States and a trade partner. Even in cases where there is a strong relation between the real exchange rate and the relative price of nontraded goods, however, a large fraction of real exchange rate fluctuations is due to deviations from the law of one price for traded goods.     

The impact of real exchange rate shocks on manufacturing workers: An autopsy from the MORG

We study the impact of large real exchange rate shocks on workers in sectors initially more exposed to international trade using the Current Population Survey’s (CPS) Merged Outgoing Rotation Group (MORG) from 1979 to 2010 combined with new annual measures of imported inputs, a proxy for offshoring. We find that in periods when US relative prices are high, and imports surge relative to exports, workers in sectors with greater initial exposure to international trade were more likely to be unemployed or exit the labor force a year later, but did not experience significant declines in wages conditional on being employed. Contrary to the usual narrative, we find negative wage effects for higher-wage, but not lower-wage workers, particularly for those who are less-educated.     

The real exchange rate in the long run: Balassa-Samuelson effects reconsidered

Historical data for over hundred years and 14 countries is used to estimate the long-run effect of productivity on the real exchange rate. We find large variations in the productivity effect across four distinct monetary regimes in the sample period. Although the traditional Balassa-Samuelson model is not consistent with these results, we suggest an explanation of the results in terms of contemporary variants of the model that incorporate the terms of trade mechanism. Specifically we argue that changes in trade costs over time may affect the impact of productivity on the real exchange rate over time. We undertake simulations of the modern versions of the Balassa-Samuelson model to show that plausible parameter shifts consistent with the behavior of trade costs can explain the cross-regime variation of the productivity effect.     

Foreign exchange exposure and multinationality

We examine the relation between firms’ foreign exchange exposure and the extent of their multinationality as a proxy for operational hedging. Using a sample of 953 US firms over the period 1999–2006, we show that there is a nonlinear relation between operational and financial hedging, confirming anecdotal evidence that many highly multinational firms do not hedge with derivatives. We find that operational hedging and financial hedging are significantly inversely related to firms’ foreign exchange exposure, providing evidence that the two hedging techniques are complementary for all but the most highly operationally hedged firms. By comparing our findings for 1999–2006 with 1999–2009, we show that this complementarity breaks down when exchange rate volatility is high – as the effectiveness of financial hedging diminishes. An important message for firms is that operational hedges work, and they potentially provide better protection than financial hedging during times of stress.     

Modelling long-run equilibrium exchange rate in Botswana

The paper estimates the equilibrium real exchange rate for Botswana. It also reviews the country’s exchange rate regimes. Botswana operated a fixed exchange – without adjustable pegs from 1966 to 1976; with adjustable pegs from 1976 to 1980; and with a currency basket from 1980 to date. Using the autoregressive distributed lag bounds testing procedure, the paper found terms of trade and trade openness to determine the equilibrium real exchange rate. The actual real exchange rate has deviated significantly from the equilibrium exchange rate. The estimated speed of adjustment is very slow, which calls for policies that could raise it in order to avoid excess misalignments.     

Global and country-specific movements in real effective exchange rates: Implications for external competitiveness

Using the Bayesian factor model, we decompose real effective exchange rates, which are considered a measure of external competitiveness, into global and country-specific factors. Among several findings, we report a particular global trend in real exchange rates, but a substantial proportion of their variation is found to be country-specific. In line with this finding, we find that structural shifts, when they do exist, are contained in country-specific factors. Furthermore, consistent with economic theory, this global factor is closely related to a trend in the global interest rate, while country-specific factors are closely related to idiosyncratic movements in the countries’ own interest rates. Such a decomposition results in better understanding of the exchange rate-interest rate relationship, and therefore our results can be interpreted as evidence that external competitiveness is heterogeneous among countries and that economic policies can influence countries’ competitiveness.     

The effect of real rates of interest on housing prices

During the late 1970s, U.S. house prices were appreciating rapidly even though mortgage interest rates were climbing. Recently, interest rates have eased but prices have moderated. This study examines the role of appreciation expectations in overcoming the negative effects of nominal mortgage interest rates on house prices. Expectations of future appreciation are important determinants of house sales prices, remaining influential during periods of declining and moderating real prices, not just when prices are rising. The real rate of interest, as viewed by the homebuyer, is the mechanism for affecting change in housing price levels. Because the nominal interest rate is slow to reflect changes in expectations, these real rates vary over time. This ebb and flow of real interest rates appears to explain market price levels. Nominal rates play a role as well, primarily in the formation of appreciation expectations.     

Valuing foreign exchange rate derivatives with a bounded exchange process

Foreign exchange rates have been subjected to periods of tighter or looser controls as various political and economic forces have waxed and waned. When currencies were backed by gold there were fixed exchange rates. In 1973 floating exchange rates were adopted though many countries did try to keep their currency values within certain ranges. More recently the European Economic Community formalized this practice. Free-floating exchange rates might be well characterized by the lognormal distribution which is standard in option pricing. However, this is probably a poor approximation for exchange rates which are kept within some range by the actions of one or both governments or central banks. This paper develops a model which can be used to value options and other derivative contracts when the underlying exchange rate is bounded in a fixed range (a, b). Methods for pricing both European and American style options are developed.The author would like to thank Ken French and Geert Rouwenhorst for their comments and suggestions.     

Accounting for real exchange rates using micro-data

The classical dichotomy predicts that all of the time-series variance in the aggregate real exchange rate is accounted for by non-traded goods in the consumer price index (CPI) basket because traded goods obey the Law of One Price. In stark contrast, Engel (1999) claimed the opposite: that traded goods accounted for all of the variance. Using micro-data and recognizing that final good prices include both the cost of the goods themselves and local, non-traded inputs into retail such as labor and retail space, our work re-establishes the conceptual value of the classical dichotomy. We also carefully show the role of aggregation, consumption expenditure weighting and assignment of covariance terms in the differences between our findings and those of Engel.     

Regime-switching models for exchange rates

This study provides evidence of periodically collapsing bubbles in the British pound to US dollar exchange rate in the post-1973 period. We develop two- and three-state regime-switching (RS) models that relate the expected exchange rate return to the bubble size and to an additional explanatory variable. Specifically, we consider six alternative explanatory variables that have been proposed in the literature as early warning indicators of a currency crisis. Our findings suggest that the RS models are, in general, more accurate than the Random Walk model in terms of both statistical and especially economic evaluation criteria for exchange rate forecasts. Our three-state RS model outperforms the two-state models and among the variables considered in our analysis, the short-term interest rate is the optimal variable, closely followed by imports. Results are more promising for one-month predictions and are qualitatively robust over sample spans. However, various robustness checks based on other exchange rates show that the optimal bubble measures and optimal predictors critically depend on the exchange rate.     

The accuracy of reports of foreign exchange intervention by the Bank of Japan: Does Tokyo know more?

This paper examines the accuracy of press reports of foreign exchange intervention by the Bank of Japan. We investigate whether the local news wire (JiJi News) reacted differently from the foreign press (Wall Street Journal) between January 2000 and December 2003. Our results show that the likelihood of intervention being reported given that it actually occurred is higher for the JiJi News than for the Wall Street Journal, but the JiJi News has many more instances of false speculative intervention reports. As such, the underreporting by the Wall Street Journal mitigates its overall errors as compared to the JiJi News. We find that the change of Japan's intervention strategy from the beginning of 2003 has a major impact on the accuracy of press reports. Logit analysis also demonstrates that the likelihood of intervention being firmly reported increased with the size of the intervention and the magnitude of appreciation of the Japanese yen.     

Exchange rates and order flow in the long run

Several recent papers have underlined the importance of microstructure effects in understanding exchange rate behavior by documenting stable long-run relationships between cumulated order flows and spot exchange rates. This stands in contrast to the widely-studied failure of exchange rates to conform to the long-run behavior implied by “conventional” macroeconomic models and is consistent with the prediction of micro-structure models. We re-examine the evidence for stable long-run relationships. We find that such evidence exists only for a small number of the major currencies we examine and that it is statistically fragile. We conclude that this implication of microstructure models does not fit the data as well as previous studies suggest.     

Foreign exchange risk and the cross-section of stock returns

We examine the relation between the cross-section of US stock returns and foreign exchange rates during the period from 1973 to 2002. We find that stocks most sensitive to foreign exchange risk (in absolute value) have lower returns than others. This implies a non-linear, negative premium for foreign exchange risk. Sensitivity to foreign exchange generates a cross-sectional spread in stock returns unexplained by existing asset-pricing models. Consequently, we form a zero-investment factor related to foreign exchange-sensitivity and show that it can reduce mean pricing errors for exchange-sensitive portfolios. One possible explanation for our findings includes Johnson's [2004. Forecast dispersion and the cross-section of expected returns. Journal of Finance, 59, 1957–1978] option-theoretic model in which expected returns are decreasing in idiosyncratic cashflow volatility.     

A stopping time approach to assessing the effectiveness of foreign exchange intervention: An application to Japanese data

I propose a new methodology to assess the effect of foreign exchange (FX) intervention, based on the probability of an FX rate reaching. The variable is the probability of an FX rate reaching a particular threshold before reaching another. Importantly, the probability depends on not only the level, but also the trend and volatility of a current FX rate. When an intervention changes the probability in a desired direction, the intervention is effective. The notable feature of the probability is that it considers both the level and volatility of an FX rate comprehensively, while previous studies have examined these effects of FX intervention separately. Empirical results based on regression and nearest-neighbor analyses applied to Japanese data indicate that publicity and size are significant in the effectiveness of intervention.     

European foreign exchange market efficiency: Evidence based on crisis and noncrisis periods

This study investigates the impact on foreign exchange market efficiency of the 1992 European financial market crisis by studying precrisis, crisis, and postcrisis periods. Long-term relationships among European currency values are identified during the three periods, although the relationships are not stable during the precrisis and the postcrisis periods. These results may be due to one or more of the following: (1) market inefficiency, (2) a risk premium, or (3) common policy guidelines for European monetary system (EMS) members. Evidence of market inefficiency is strong. Forecasting results demonstrate better performance by an error correction model (ECM) than by a random walk model (RWM) for the British pound and German mark, while results for the French franc and Italian lira are mixed. Dominance tests using Granger causality indicate only weak German mark dominance both in the short and long run.