Abstract
Building on the celebrated Keynes–Ohlin debate and on Lane and Milesi-Ferretti (Rev Econ Stat 86:841–857, 2004), the paper investigates the transfer problem for the Euro area vis-à-vis the rest of the world. The analysis is developed in a theoretically and statistically consistent way and is intended as a contribution to the empirical literature on EMU. The main result of the paper is that the accumulation of net foreign asset in the Euro area is consistent with real exchange appreciation, largely through the relative price of nontradables rather than through the terms of trade.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
Keynes retorted that “... In so far as Germany can (...) pay reparations without borrowing, these payments will react on the levels of income abroad. (...) But the increased buying power, due to the fact that Germany is paying something with less assistance than before from borrowing, will have been already used up in buying the exports the sale of which has made the reparations payment possible”.
The transfer effect is by no means the only linkage between a country’s real exchange rate and its net foreign assets. As Lane and Milesi-Ferretti (2001) point out, at a shorter horizon the interplay between the exchange rate and the trade balance is complex and less than well understood. In tracking net foreign assets/GDP dynamics, real exchange rates operate through the trade balance, the foreign currency value of domestic GDP (the so-called denominator effect) and the rates of return on foreign assets and liabilities (the so-called valuation effect). On the theoretical and empirical problems related to valuation effect see Gourinchas and Ray (2005).
In this model the real exchange rate is mechanically independent of the terms of trade, since the only tradables which are consumed in the country are produced abroad and their price is held fixed. Movements in the terms of trade may influence the real exchange rate indirectly through a wealth effect on the relative price of nontradables. It may be objected that the terms of trade are endogenous for a large bloc of countries such as the Euro area—but this can be dismissed on the basis of the argument that exogenous oil prices drive the European terms of trade. We thank the anonymous referee for pointing this out.
Available statistical sources make it difficult to obtain an exact counterpart of the variables required by the model. On the basis of this consideration, we have chosen to construct our time-series in the same way as Lane and Milesi-Ferretti (2004) do, maximize the compatibility between the theoretical framework and the empirical analysis.
Troubles on the energy markets in 2004 and in 2005, whose effects might be particularly severe for oil-dependent economies such as the Euro area, suggest to limit our sample to 2003.
While for rer and tra the rest of the world is limited to eight industrial countries outside the Euro area, for tot and nfa the rest of the world includes every country outside the Euro area. Despite its apparent inconsistency, our choice may be justified as follows. First, the eight economies considered as rest of the world for tot and nfa constitute around four-fifths of the external trade of the Euro area. Regional developments in Eastern Europe and Asia are not negligible but of secondary relevance. Second, our measure of rer is highly correlated with other “official” measures of CPI-based real effective exchange rate for the Euro area, as the one used in Fagan et al. (2005). Third, the construction of a world (per capita) output series requires an unavoidable approximation. Using the same aggregating procedure for rer and tra is intended to control that source of arbitrariness.
For the aggregate Euro area, the graph suggests that the terms of trade are driven by fluctuations in the oil price. See also Backus and Crucini (2000) on this point.
Euro area includes: Austria. Belgium, Finland, France, Germany, Ireland, Italy, the Netherlands and Portugal.
Rest of the world includes: Canada, Denmark, Japan, Norway, Sweden, Switzerland, the United Kingdom, the USA.
An increasing rer indicates real exchange rate appreciation.
The terms of trade reasonably approximate the relative price of domestic and foreign tradables only if the following two conditions are met: (1) exports (imports) are a large subset of domestic (foreign) tradables, reflecting the absence of trade barriers and taste differences; (2) producers do not price discriminate across markets.
The optimal lag for each test is determined on the basis of the Final Prediction criterion, with the maximum tested lag set equal to 10. A constant term is included in each regression. Critical values for these tests are provided by Davidson and McKinnon (1993).
The impulse dummies were chosen on the basis of recursive 1-step-ahead Chow test and recursive cointegration tests. The idea of modeling the transition from one policy regime to another by means of a sequence of impulse dummies is proposed by Johansen et al. (2000).
As a robustness check, the PPI-based real exchange rate was tried out en lieu of the CPI-based real exchange rate (rer). Under this alternative specification, however, poor results were obtained. As to this, the CPI nontradable component is larger than its PPI counterpart. This makes PPI unsuitable to measure the kind of transmission from Net foreign assets to the real exchange rate as discussed by Ohlin. The argument whether the CPI or the PPI price level should be the most appropriate price index in the construction of the real exchange rate goes back to Keynes (1932). Nevertheless, it is worth noticing that the CPI–PPI distinction is a subtle one since the two price indices are highly correlated and differences in their movements are, in general, very difficult to explain. On this see Sarno and Taylor (2002).
On this see Lütkepohl and Krätzig (2004).
The fact that tra enters the vector with the opposite sign with respect to theoretical a priori might depend on the labor productivity ratio being a poor proxy for the levels of tradable output. Lane and Milesi-Ferretti (2004) themselves, using the same variable, find a negative or statistically not-significant tra coefficient in many of their regressions. An alternative measure of tradable output we thought of using could have been the sum of the contributions of the primary and secondary sector (net of constructions) to GDP. Cross-country consistent data of this kind, however, are typically only available at a semi-annual or annual frequency. If the time span of the analysis is not extended (something we chose not to do at present), the resulting loss of degrees of freedom might worsen the statistical accuracy of our results.
This means that it should be possible to specify a subset model conditioning on this variable, which acts as a common trend of the system.
In the long-run, productivity differentials changes are offset by opposite movements in the dynamics of real exchange rate; in the short run, productivity differentials growth rates are affected by transitory deviations from the steady-state path.
The size of all the shocks analyzed in this section is set equal to one standard deviation.
The simulation spans 20 quarters. Bootstrapped 95% confidence intervals with 2,500 replications indicate that the responses of the cointegration vector to impulses on the nfa, tra and rer equations are statistically significant up the second year of the simulation. Conversely, the deviations from the baseline path induced by a shock on the tot equation are statistically negligible. These results are robust with respect to both the specification of the model (four and three variable cases) and the estimation horizon (1979–2003 and 1979–1998 samples). On the scant reliability of the confidence bounds for impulse response functions as indicator of the precision of estimates even at short and medium horizons, see Benkwitz et al. (2000).
Half-life is defined as the number of quarters which have to pass before the deviation from the steady-state falls to half the size of the initial shock.
Results (available on request) are substantially unchanged with p = 4.
For an example of cointegration analysis coupled with the exclusion of some observations to take account of a policy-regime change see Juselius (2003).
References
Agresti A-M, Mojon B (2003) Some stylised facts on the Euro area business cycle. In: Angeloni I et al (Eds.) Monetary transmission in the Euro Area, Chapter 1. Cambridge University Press, Cambridge
Backus DK, Crucini MJ (2000) Oil prices and the terms of trade. J Int Econ 50:185–213
Benkwitz, A, Lütkepohl H, Neumann MH (2000) Problems related to confidence intervals for impulse of autoregressive processes. Econom Rev 19:69–103
Bussiere M, Choartareas G, Driver R (2003) Current accounts net foreign assets and the implications of cyclical factors. East Econ J 29:269–86
Davidson R, McKinnon JJ (1993) Estimation and Inference in Econometrics. Oxford University Press, Oxford
Dickey, DA, Fuller WA (1979) Distribution of the estimators for autoregressive time series with a unit root. J Am Stat Assoc 74:427–431
Fagan G, Henry JJ, Mestre R (2005) An area-wide model (AWM) for the Euro area. Econ model 22:39–59
Faruqee H, Laxton D (2000) Life cycle dynasties saving: implications for small open economies, International Monetary Fund, Working paper, 126
Gagnon JE (1996) Net foreign assets and equilibrium exchange rates: panel evidence, Board of Governors of the Federal Reserve System, International Finance discussion paper, 574
Gourinchas P-O, Ray H (2005) International financial adjustment, Centre for Economic Policy Research discussion paper, 4923
Johansen S (1995) Likelihood-based inference in cointegrated vector autoregressive models. Oxford University Press, Oxford
Johansen S, Mosconi R, Nielsen B (2000) Cointegration analysis in the presence of structural breaks in the deterministic trend. Econom J 3:216–249
Juselius K (2003) Wage, price, and unemployment dynamics and the convergence to purchasing power parity in the Euro Area. http://www.econ.ku.dk/okokj/Papers/Eurowideecb.pdf
Keynes J-M (1929) The german transfer problem. Econ J XXXIX:1–7
Keynes, J-M (1932) Essays in persuasion. Harcourt Brace Jovanovich, Orlando, FL
Kraay A, Ventura J (2000) Current accounts in debtor and creditor countries. Q J Econ 115:1137–1166
Kwiatkowski D, Phillips PCB, Schmidt P, Shin Y (1992) Testing the null hypothesis of stationary against the alternative of a unit root. J Econom 54:159–178
Lane P, Milesi-Ferretti G-M (2001) The external wealth of nations: estimates of foreign assets and liabilities for industrial and developing countries. J Int Econ 55:263–294
Lane, P, Milesi-Ferretti G-M (2002) Long-term capital movements. NBER Macroeconomic Annual 16:73–116
Lane P, Milesi-Ferretti G-M (2004) The transfer problem revisited: net foreign assets and real exchange rates. Rev Econ Stat 86:841–857
Lewbel A (1992) Aggregation with log-linear models. Rev Econ Stud 59(3):635–642
Lütkepohl H, Krätzig M (2004) Applied time series econometrics. Cambridge University Press, Cambridge
Mojon B, Peersman G (2003) A VAR description of the effects of monetary policy in the individual countries of the Euro area. In Angeloni I et al (Eds) Monetary transmission in the Euro area, Chapter 3. Cambridge University Press, Cambridge
Ohlin B (1929) The reparation problem: a discussion. Econ J XXXIX:172–182
Pagan A (2003) An examination of some tools for macro-econometric model building, Mimeo, Rep. Natl. Soc Sci Surv
Peersman G, Smets F (2003) The monetary transmission mechanism in the Euro area: evidence from VAR analysis. In: Angeloni I et al (Eds) Monetary transmission in the Euro area, Chapter 2. Cambridge University Press, Cambridge
Pesaran HM, Shin Y (1996) Co-integration and speed of convergence to equilibrium. J Econom 71:117–143
Pesaran HM, Shin Y (1998) Generalized impulse response analysis in linear multivariate models. Econ Lett 58:17–29
Sarno L, Taylor MP (2002) Purchasing power parity and the real exchange rate. IMF Staff Papers 49:65–105
Selaive J, Tuesta V (2003a) Net foreign assets and imperfect pass-through: the consumption-real exchange rate anomaly. Board of Governors of the Federal Reserve System, International Finance discussion paper, 764
Selaive J, Tuesta V (2003b) Net foreign assets and imperfect financial integration: an empirical approach, Working paper 252. Central Bank of Chile, Chile
Acknowledgment
The authors are grateful to the editor (George Tavlas), an anonymous referee, Gustavo Piga and Michele Bagella for helpful comments and suggestions on previous versions of the paper. The usual disclaimers apply. The views expressed do not necessarily reflect those of the Institute for Studies and Economic Analyses (ISAE).
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Girardi, A., Paesani, P. The Transfer Problem in the Euro Area. Open Econ Rev 19, 517–537 (2008). https://doi.org/10.1007/s11079-007-9058-0
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11079-007-9058-0