Pitfalls in panel tests of purchasing power parity

Pitfalls in Panel Tests of Purchasing Power Parity. —The results of panel unit root tests applied to real exchange rates as a test of long-run purchasing power parity (PPP) diverge much. In particular, due to misspecifications there is little evidence of the convergence of real exchange rates for the German mark. This paper provides evidence of this issue by analyzing large panels of real exchange rates vis-à-vis the German mark and the dollar. In particular, the impact of the base country and various aspects of the dynamic specifications are analyzed. Overall, the results provide strong evidence in favour of PPP as a long-run relationship.     

Real Exchange Rates and Fundamentals: Evidence from 15 OECD Countries

In an extended Balassa–Samuelson model, long-run real exchange rates are determined by relative productivity and terms of trade. We present evidence of systematic long-run relationships between these fundamental variables and real exchange rates in a data set covering 15 OECD countries from 1960 to 1996. High relative productivity is associated with real exchange rate appreciations in most cases. There is less support for the hypothesis that the terms of trade affect equilibrium real exchange rates.     

Maximum likelihood cointegration tests of purchasing power parity: Evidence from seventeen OECD countries

Maximum Likelihood Cointegration Tests of Purchasing Power Parity: Evidence from Seventeen OECD Countries. — This paper examines the relevance of long-run purchasing power parity (PPP) during the recent floating exchange rate period, using Johansen’s maximum likelihood method for estimating and testing steady-state relations in multivariate vector autoregressive models. Thirty-two bilateral intercountry relations are considered and it is found that in many cases there exists a long-run relationship between exchange rates and international price differentials, which, however, significantly deviates from PPP in most instances.     

The trend behavior of real exchange rates: Evidence from OECD countries

The Trend Behavior of Real Exchange Rates: Evidence from OECD Countries. — This paper examines the validity of purchasing power parity (PPP) under the current float using real effective exchange rates of eleven OECD countries. The author employs a test which allows for a one-time change in the intercept and/or in the slope of the trend function. The timing of the structural break is treated as unknown and is endogenously searched from the data. It is found that for a vast majority of countries, the real exchange rate can be characterized as a stationary process with a broken trend. The paper provides support for PPP in the long run.     

Food prices and long-run purchasing power parity in Africa

The purchasing power parity (PPP) theory is a cornerstone of exchange rate models in international economics. PPP is very important for two main reasons: first, it can serve as a prediction model for exchange rates, and second, it can serve as a benchmark in judging the level of exchange rate movements. This article utilised the Johansen cointegration technique in examining whether or not there is empirical support for long-run PPP in Africa. Annual data were used for exchange rates and food price indices in 25 countries covering the 1958-97 period. The empirical evidence showed strong support for long-run PPP in Africa, thereby providing wider acceptance for the applicability of PPP in exchange rate and other macroeconomic adjustment policies.     

Decline in the persistence of real exchange rates,but not sufficient for purchasing power parity

The paper investigates the possibility of decline in the persistence of real exchange rates, or deviations from PPP. To this end, we test the null hypothesis of no decline in the PPP deviation persistence between two subsamples using a fractional integration framework. The test rejects the null at the 10% significance level for nine out of 17 countries, providing solid evidence for a decline in the persistence of real exchange rates. However, the decline is not sufficient for PPP, meaning we fail to reject the unit root hypothesis even in the latter period for all 17 countries. In addition, our rolling-window estimates show that the real exchange rate of many countries have experienced a sharp drop in their persistence once we use samples starting from the mid-1980s. Finally, we examine the relationship between the dynamics of PPP deviation persistence and several economic variables and confirm that the speed of convergence of PPP deviations is highly related to economic/financial integration and world economic stabilization.     

Cointegration Tests of Purchasing Power Parity: the Case of the Thai Baht*

This paper will examine the validity of the purchasing power parity (PPP) hypothesis for the Thai baht vis-à-vis the currencies of Thailand's key trading partners under the new exchange rate regime using the cointegration technique. The major conclusions obtained from this empirical analysis may be broadly summarized as follows. First, the empirical evidence, based on the DF and ADF statistics, seems to suggest that the nominal exchange rates and relative prices are well characterized as non-stationary I(1) processes. Second, the cointegration analysis provides no evidence in support of a long-run equilibrium relationship between bilateral nominal exchange rates for the Thai baht vis-à-vis the currencies of Thailand's major trading partners and the corresponding relative price ratios. This implies rejection of PPP for these countries. If this is the case, considerable care should be taken in assessing the long-run implications for the real exchange rate, and thus competitiveness against Thailand's key trading partners, of shocks to the nominal exchange rate.     

PURCHASING POWER PARITY IN AFRICAN COUNTRIES: EVIDENCE FROM PANEL SURADF TEST

This study reexamines the validity of long-run purchasing power parity (PPP) hypothesis using a battery of panel unit root tests for 11 developing countries in Africa over the period 1980-2007. Based on the conventional panel unit root tests, we found evidence that the monthly real exchange rates in these countries were mean reverting. By contrast, the series-specific unit root test proposed by Breuer et al. (SURADF) reveals that only six of the 11 RERs series were stationary using the US dollar as reference currency. Additionally, our results reveal that there is stronger evidence of the parity condition with the Rand-based rates than in the other currency-based rates like the US dollar or Euro. We conclude that PPP holds in some, but not all, of the African countries according to the SURADF tests.     

Purchasing Power Parity in South East Asian Countries Economies: A Cointegration Approach

This paper presents findings from a study of the long-run purchasing power parity (PPP) conditions in five Asian economies. The cointegration tests using exchange rates and price indices from end-of-quarter observations over the last twenty years reject the PPP proposition for all countries. The absolute version of the PPP hypothesis is tested next by using lower frequency, that is, semi-annual and annual, data. In general these tests also failed to support the long-run PPP hypothesis. Further analysis using the Johansen-Juselius (1990) multivariate approach also failed to support the hypothesis.     

Nonstationarity of Real Exchange Rates in the G7 Countries: Are They Cointegrated with Real Variables?

Using monthly data for the G7 countries in the post-Bretton Woods floating rate period, this paper demonstrates that almost all bilateral real exchange rates have unit roots and, hence, are nonstationary. Consequently, it rejects simple PPP as a long-run relationship. The paper also shows that many of these real exchange rates are cointegrated with other real economic variables such as relative labor productivity, terms-of-trade ratios, real trade balance ratios, and long-term real interest rate differentials. In particular, relative labor productivity is statistically significant with the correct sign for more than half of the country pairs for which cointegration is confirmed. This finding lends support to the Balassa–Samuelson productivity-bias hypothesis. These results imply that nonstationarity of real exchange rates and the consequent rejection of simple PPP can be consistent with the notion that real exchange rates revert to an equilibrium in the long run without deviating arbitrarily far from this equilibrium position.J. Japan. Int. Econ.,December 1997, pp. 523–547. Institute of Social Science, University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113, Japan, and Faculty of Business and Commerce, Meiji University, 1-1 Kanda, Surugadai, Chiyada-ku, Tokyo 101, Japan.     

Nonlinear Purchasing Power Parity under the Gold Standard

Hegwood and Papell (2002) conclude on the basis of analysis in a linear framework that long-run purchasing power parity (PPP) does not hold for 16 real exchange rate series, which were analyzed in Diebold. I lusted, and Rush (1991) for the period 1792-1913 under the Gold Standard. Rather, PPP deviations are mean-reverting to a changing equilibrium—a quasi PPP (QPPP) theory. We analyze the real exchange rate adjustment mechanism for their data set assuming a nonlinear adjustment process allowing for both a constant and a mean shifting equilibrium. Our results confirm that real exchange rates at that time were stationary, symmetric, nonlinear processes that revert to a nonconstant equilibrium rate. Speeds of adjustment were much quicker when breaks were allowed.     

Purchasing Power Parity before and after the Adoption of the Euro

This paper examines the purchasing power parity (PPP) hypothesis for the post–Bretton Woods era including the period after the introduction of the euro. The study applies a new nonlinear unit root test to the bilateral real exchange rates (RERs) of both European and other industrial countries with the French franc and German mark (and the euro after 1998), as well as the US dollar as numeraire currencies. The results of the study provide stronger support for PPP than any earlier studies of bilateral PPP for industrial countries and suggest that (1) PPP tends to hold well within the European Union (EU) even before the adoption of the euro, (2) the evidence for PPP becomes more significant for both EU and non-EU countries when the sample period is extended to the euro era, and (3) convergence toward PPP between the EU countries, especially between the euro-area countries, tends to be nonlinear, while it is likely to be linear for the non-EU industrial countries. JEL no. F31, F33, G15, C22     

Purchasing Power Parity in High-Inflation Countries: A Cointegration Analysis of Integrated Variables with Trend Breaks

This paper examines the long-run validity of purchasing power parity (PPP) for four high-inflation countries. The method of Zivot and Andrews (1992) is employed to detect the time-series behavior of the exchange rates and consumer price indices of these countries. We find that these variables are integrated with some trend breaks. We then utilize these data to test PPP using Johansen's (1988) multivariate cointegration technique. The cointegration tests are conducted with the correction of the finite sample bias and the adjustment for trend breaks. The results are consistent with the argument that, during the recent floating exchange-rate period, PPP holds well, at least in a weak form, in high-inflation countries where the general price level movement overshadows the factors causing deviations from PPP.     

Evidence of long-run purchasing power parity: analysis of real asian exchange rates in terms of the Japanese yen

In this paper we find strong new evidence in favour of the long-run purchasing power parity (PPP) hypothesis in the bilateral real exchange rates between the Japanese yen and the currencies of the most important southeast Asian economies only when the presence of several possible structural breaks of the series is taken into account. Such evidence for PPP is weaker for these southeast Asian exchange rates with the US dollar, the German mark and the Australian dollar.     

Generalized purchasing power parity and the case of the European Union as a successful currency area

This article uses multicountry purchasing power parity (PPP) tests to study the success of the European monetary system (EMS) in creating a successful currency area for a stable European monetary union. If the EMS has sufficiently integrated the fundamentals within the European Union, then real exchange rates between member countries will share a common stationary trend when denominated by a common outside currency. Previous research using two-country PPP tests have been inadequate in explaining the nonstationary nature of real exchange rates between the EMS countries and nonmember countries. The use of generalized PPP tests can show that even though individual exchange rates within the EMS may appear to be nonstationary with respect to outside countries, some of them will combine to form a currency union with a stable stationary trend.     

Real Exchange Rate Stationarity in Latin America and Relative Purchasing Power Parity: A Regime Switching Approach

This paper tests for long-run purchasing power (PPP) among a sample of six Latin American economies. The key contribution of this paper is in terms of the econometric methodology where non-stationarity of the real exchange rate is tested within a Markov regime-switching framework. In contrast to existing studies, this paper defines two new concepts of PPP where one allows for the possibility that real exchange behaviour either switches between stationary and non-stationary regimes (partial PPP), or switches between stationary regimes characterised by differing degrees of persistence (varied PPP). Whereas standard univariate unit root testing suggests that Latin American real exchange rates are generally non-stationary, employment of the regime-switching methodology indicates that most of the sample is characterised by the existence of two distinct stationary regimes. Further analysis indicates that the high rates of inflation and exchange rate volatility experienced in Latin American have given some impetus towards facilitating long-run PPP.

Do Real Exchange Rates Follow a Nonlinear Mean Reverting Process in Developing Countries?

In an effort to fight relatively high inflation, many developing countries try to manage their nominal exchange rates through official intervention. In addition, developing countries tend to have high transportation costs, tariffs, and nontariff barriers. These factors are among the sources of generating nonlinearity in real exchange rates and hence some nonlinear adjustment toward purchasing power parity (PPP) in developing countries. In this paper, we employ monthly real effective exchange rate (REER) data of 88 developing countries and test the null of nonstationarity versus an alternative of linear stationarity by the means of a conventional unit root test and compare the results with those obtained from a new test in which the null is the same but the alternative hypothesis is nonlinear stationarity. The latter test supports the PPP theory in more developing countries compared with the former test, suggesting that nonlinear adjustment toward PPP in developing countries is an important phenomenon. Reported country characterizations indicate that reversion in REER occurs more often for high-inflation countries and for countries with high flexibility in their exchange rates.     

Fulfilment of purchasing power parity conditions in Africa: the differential role of CFA and non-CFA membership

The paper reports empirical evidence on the fulfilment or otherwiseof purchasing power parity (PPP) conditions in African countries,based on the popular augmented Dickey-Fuller test methodologythat is used to identify whether the nominal exchange rate andprice levels are co-integrated. The empirical analysis is basedon quarterly data for the 1980(I)-1991(IV) period for 35 Africancountries and our finding is that the PPP conditions are metin 17 countries that are almost exclusively non-CFA zone membersbut are not met in the remaining 18 countries comprising ninenon-CFA and nine CFA zone members. Given this lopsidedness inthe distribution of countries where the conditions are met ornot met between the two zones, it is inferred that certain macroeconomicfeatures common to the CFA zone members but which are generallyless pronounced in other countries might have been responsiblefor this finding. A number of such features relating to theextent of monetary growth, inflation rates, rate of depreciationof real and nominal exchange rates, domestic price variability,nominal exchange rate variability and the type of exchange ratesystem being operated were related to the tendency for PPP tobe fulfilled across countries. But, while the direction of theirobserved effects are generally as expected, the effects do notexhibit statistical significance.     

Real exchange rates and unit root tests

Real Exchange Rates and Unit Root Tests. — This paper examines monthly OECD exchange rate data (1979–1997) using univariate and panel data unit root tests. Some of these tests support the hypothesis of a unit root. But tests of cointegration reveal the existence of weak purchasing power parity relationships between bilateral nominal exchange rates and relative prices. We suggest that researchers need not conduct unit root tests on real exchange rate data when a modified version of PPP is used; or if there is a long enough time series. Given the definition of real exchange rates, the indicator should be stationary and should have intrinsic mean reverting behaviour.     

Long-Run Purchasing Power Parity with Asymmetric Adjustment

Tests of purchasing power parity (PPP) that use panel data are more supportive of the theory than are bilateral tests. The article uses threshold cointegration to explore long-run PPP. Using data from the post-Bretton Woods period, we show that cointegration with threshold adjustment holds for a number of European countries on a bilateral basis. Focusing on France and Germany as base countries, we show that the error-correction model has important nonlinear characteristics in that prices and the exchange rate have markedly different adjustment patterns for positive gaps from PPP than negative gaps.