«1. Introduction Even though long-run equilibrium real exchange rates are a function of real variables only, actual real exchange rates respond to ...»
Misaligned? Overvalued? The Untold Story of the Turkish
Deniz Atasoy and Sweta C. Saxena1
Abstract: There is a consensus among scholars that overvalued exchange rates result in
currency crises. This paper estimates the equilibrium real exchange rate for Turkey and
finds that the lira was indeed overvalued before the crises in 1994 and 2001. However,
the actual real exchange rate is at present close to the equilibrium level, exposing the
myth propagated by the Turkish exporters that lira’s overvaluation is responsible for Turkey’s uncompetitive exports. The paper also highlights the role for fiscal adjustment in macroeconomic stability.
Keywords: Turkish lira, overvaluation, equilibrium real exchange rate, misalignment JEL classification: F1, F3, F4 2
1. Introduction Even though long-run equilibrium real exchange rates are a function of real variables only, actual real exchange rates respond to both real and monetary variables (Edwards 1989). The departure of actual real exchange rate from the equilibrium level in the short and medium run due to short run frictions and adjustment costs is common.
However, certain deviations from the equilibrium level could become persistent through time leading to misalignments. The literature has found exchange rate misalignment (namely, overvalued exchange rate) as an important predictor of currency crises (see Kaminsky, Lizondo and Reinhart, 1998 and the literature cited within). The huge costs that these misalignments impose in the form of currency crises make it impossible for policymakers to overlook the problem. Turkey’s crises in the last decade bear a testimony to the devastation caused by such crises.
An overvalued exchange rate causes domestic exports to become uncompetitive in the world markets and puts pressure on governments for protectionism. Turkish exporters have recently raised their concerns alleging that the overvalued lira is making the Turkish exports uncompetitive. Similarly, the U.S. exporters are holding China’s undervalued currency responsible for their huge current account deficits and for prolonging of the recession in the United States. In the wake of the Asian currency crisis in 1997, some emerging countries expressed concern about losing competitiveness to the Asian countries since the Asian currencies experienced huge devaluations. Hence, ascertaining the equilibrium exchange rate, and hence the extent of misalignment, is important in determining the competitiveness of the economy. This is especially relevant in the case of Turkey, given its customs union and accession negotiations with the European Union. An overvalued exchange rate could indeed push Turkey into a severe balance of payments crisis. Turkey has had a Customs Union with the European Union since 1996. This customs union did indeed help Turkey boost its exports by 5 % since 1996. In its recent meeting with the European Council in December 2004, the council was pleased with the progress Turkey is making towards the convergence criteria and has requested to open negotiations in October 2005 (Europa, 2004a). However, European Union has repeatedly stressed that the accession would not take place before 2014. (Europa, 2004b) Therefore, Turkey will be under a great pressure to sustain its good economic record as well as to continue political reforms.
In the event of misaligned exchange rates, policymakers often find themselves entrapped in a dilemma – external competitiveness or internal stability (fiscal and price stability). In the case of Turkey, policymakers have to weigh the need for devaluation to boost exports (if exchange rate is indeed overvalued) versus keeping inflation low because the share of imported raw materials and capital goods in total imports is high (Guncavdi and Orbay, 2001). Kalkan (2002) finds that increasing rates of depreciation cause real depreciations, which he interprets as a dilemma for the government. If the government wants to control inflation by controlling the rate of depreciation, it has to accept the appreciation of real exchange rate and therefore deterioration of trade balance.
While an appreciated exchange rate keeps the level of foreign debt low, devaluation leads to high interest rates (through uncovered interest parity) thus increasing the debt level.
In order to solve the policy dilemma, we feel the need to estimate the equilibrium real exchange rate for Turkey to assess if the recent crises in 1994 and 2001 were caused 3 by overvalued exchange rate and to test the claim of the Turkish exporters that the overvalued exchange rate is making the Turkish economy uncompetitive. The rest of the paper is organized as follows. Section 2 gives a brief account of Turkey’s economic history. Section 3 reviews the literature on Turkish exchange rate. Section 4 describes the theoretical underpinnings for our empirical model. Section 5 presents the estimation and the results from the error correction models. Section 6 estimates the equilibrium real exchange rate for Turkey. Section 7 concludes with policy implications in light of the recent monetary policy.
2. Turkey’s economic history
Prior to the stabilization program in 1980, Turkey followed a highly protective and inward-looking policy, plagued with inflationary pressures. Under this strategy, domestic producers enjoyed protection while multiple exchange rate practices were in place to favor imports of inputs for domestic industry. The economy performed well until the late 1970s, when worsening of inflation and current account was aggravated by the increasing public deficits and the two oil crises, thereby creating pressure on the exchange rate. Foreign borrowing was used extensively to finance current account deficits. High inflation along with a fixed exchange rate led to losses in competitiveness and a balance of payments crisis in 1977-78. However, the post-1980 period witnessed a policy of abolition of most price controls, continual real depreciation (at least until the end of 1989) to boost exports and liberalization of exchange and payments system (see Asikoglu and Uctum, 1992, and Erol and van Wijnbergen, 1997 for description of the exchange rate policy). By late 1980s, both current account and capital account were liberalized.
In the early years of the program, impressive growth rates, burgeoning exports and falling inflation rate together with improved fiscal position created a buoyant economic environment. However, after 1987, Turkey experienced the boom-bust cycles, resembling the Southern Cone experience – using exchange rate for stabilizing inflation, without correcting the underlying budget deficits – which would result in speculative attacks (see Calvo and Vegh, 1999, for a survey). Trade liberalization led to unwanted consequences on the trade balance because it was not supported by appropriate fiscal corrections (Kale, 2001). High public deficits kept interest rates high, which led to appreciation of the lira and burgeoning current account deficits, which culminated in the financial collapse in 1994.
The 1994 financial collapse led to the adoption of a stabilization and structural adjustment program, consisting of fiscal retrenchment to reduce inflation and to improve the external balance (Kale, 2001). However, political uncertainties, combined with loose fiscal and monetary policies, undermined the credibility of the disinflation program.
In 1999, the newly established government embarked on a new disinflation program. The program in essence was an exchange rate based stabilization program (with a fixed exit date). The program also aimed at fiscal discipline and structural reforms.
Exchange rate was predetermined in line with the targeted inflation rate. Liquidity creation by the central bank was tied to the foreign exchange purchases.2 The initial effects of the program were typical in that inflation slowed down, interest rates declined, consumption boomed and current account deficit surmounted. The 4 program relied on the sustainability of capital inflows. Like the 1994 crisis, banks borrowed recklessly from abroad with short maturities. After capital began to flow out, the fragile banking sector was pushed into a banking crisis in November of 2000. The loss of confidence precipitated the currency crisis of February 2001. As a result, Turkish authorities decided to let the lira float. From then on, the program has been implemented under a floating exchange rate regime.3 After one and half decade of economic instability in the Turkish economy, the recent stable performance has raised hopes. Inflation has declined from 45% in 2002 to 25% in 2003 – mainly due to the appreciation of the lira. However, increasing current account deficits and short term capital inflows are causing concerns. In addition, “overvalued lira” has become the center stage of discussions in the economic circle. Our aim in this paper is to estimate the equilibrium real exchange rate to test if overvalued lira was responsible for the crises in 1994 and 2001 and also to test the claim about present overvaluation of the lira.
3. Literature Review
We admit that ours is not the first study examining the equilibrium real exchange rate in Turkey, but it is the only paper that estimates the equilibrium using the correct methodology. We now review some of the empirical studies on Turkish real exchange rate and point to their shortcomings. Alper and Saglam (2000) estimate the equilibrium using cointegration method – the study is flawed in terms of omitting the main determinant of equilibrium real exchange rate for Turkey, namely, government spending.
They use a bilateral exchange rate (vis-à-vis the U.S. dollar), instead of the effective exchange rate, when Turkey’s main trading partner is Germany. In addition, they use the actual values of the fundamentals to construct the equilibrium exchange rate. While Dordoodian, Jung and Yucel (2002) use a theoretical model similar to ours, they use a dated econometric technique, moving-average method as used by Edwards (1989) to estimate the long run equilibrium. Using PPP in a non-linear model, Sarno (2000) finds that the real Turkish lira adjusts non-linearly towards its equilibrium level for the period 1980-97. Using a time varying parameter model, Ozlale and Yeldan (2002) find that lira remained structurally undervalued for most of 2000. Their estimation suffers from misspecification of the model, which ignores the impact of terms of trade and productivity increases on the real exchange rate. In addition, they estimate the equilibrium real exchange rate by multiplying the coefficient vector with the actual values of the regressors. This includes both the temporary and the permanent components in the regressors, which by definition, is not the equilibrium. Using a structural VAR model, Erlat and Erlat (1998) find that real shocks explain the fluctuations in the real exchange rate, and once displaced from the equilibrium, it takes about 3 – 4 years for the real and nominal exchange rates to return to the equilibrium value. However, this model does not explicitly estimate the equilibrium real exchange rate. Due to the shortcomings in the previous studies, our paper attempts to improve on the existing literature by accounting for all the long run and the short run variables that affect real exchange rates and by estimating the equilibrium real exchange rate, where only the permanent components of the long run fundamentals are included.
4. Theoretical Underpinnings
We use the models developed by Montiel (1997), Edwards (1989, 1994) and Elbadawi (1994) to determine the real fundamentals affecting the long-run real exchange rate. The equilibrium real exchange rate is one that is consistent with simultaneous
internal and external balance. The predictions from these models are summarized below:4
• Composition of government spending: If government spending is directed mainly toward traded goods and services, the trade balance deteriorates. To bring the external balance in equilibrium, the REER must depreciate (expected sign is negative).
Conversely, spending directed mainly toward non-traded goods and services generates excess demand in the non-traded sector. To restore the sectoral balance, there must be an appreciation of the REER (expected sign is positive).
• Terms of trade: If terms of trade deterioration shifts the demand away from importables and into the nontradables, this would put an upward pressure on the real exchange rate, hence we would expect a negative sign. On the other hand, if the income effect from the terms of trade deterioration dominates the substitution effect, we would expect a depreciation of the real exchange rate and hence a positive sign.
• As exchange and trade controls in the economy decrease, the demand for imports leads to external and internal imbalances, which require real depreciation to correct them. Using the ratio of import tariff revenue to imports as the proxy for exchange and trade controls, as trade barriers are reduced (a reduction in the value of this proxy), the total amount of trade will increase. Accordingly, a reduction in controls should be associated with real depreciation, and the expected sign is positive. If, however, the share of exports plus imports in the GDP is used as a proxy for openness, then a reduction in trade barriers will be associated with higher trade, requiring a real depreciation (negative sign).5
• The long-run effect of a reduction in capital controls is ambiguous.6 The reduction in capital controls is equivalent to a decrease in the tax on foreign borrowing that generates a positive wealth effect, which increases consumption in all periods. Hence, an appreciation is required (positive sign) for equilibrium to hold. On the other hand, by the intertemporal substitution effect, future consumption is lower than present consumption, which exerts a downward pressure on the future (long-run) price of non-tradables, and hence a depreciation of the REER is required (negative sign). The overall sign of the equilibrium depends on which effect dominates.
• Technological progress (Balassa-Samuelson): Higher differential productivity growth in the traded goods sector leads to increased demand and higher real wages for labor in that sector. The traded goods sector expands, causing an incipient trade surplus. To restore both internal and external balance, the relative price of non-traded goods must rise (REER appreciation).