Content area
Full text
ABSTRACT
This article examines the relationship between capital reversals and exchange market pressure in Turkey within an autoregressive distributed lag (ARDL) bounds testing and Granger causality framework using monthly data from 1991:12 to 2006:08. The results suggest that capital reversals are in a long-run equilibrium relationship with exchange market pressure. Granger causality tests indicate that there exists short-run and long-run causality running from capital reversals to exchange market pressure, but not vice versa. These findings lend empirical support to the Sudden Stop theory.
JEL Classification: F31, G14
Key words: capital reversals, exchange market pressure, Turkey, ARDL
Kljucne rijeci: pad vrijednosti kapitala, pritisak valutnog trzista, Turska, ARDL (autoregresijski model s distribuiranim vremenskim pomakom)
(ProQuest: ... denotes formulae omitted.)
I. INTRODUCTION
It is widely acknowledged that the IMF-led financial liberalization policies have exposed emerging market economies to short-term speculative capital movements and to currency crises. The mechanisms through which a sudden stop or reversal in capital flows may lead to a currency crisis have been explained through the sudden stop theory suggested by Calvo (1998) and Calvo et al. (2003). Despite the developments on the theoretical front and substantial policy interest, there is surprisingly little empirical literature testing the validity of the sudden stop hypothesis in emerging markets. The related empirical literature has documented evidence of the role played by capital flows in the upsurge of currency crises (See, for example, Kaminsky and Reinhart, 1999; Gazioglu, 2003; Komulainen and Lukkarila, 2003). However, no study to date has focused on the direct causal effects of capital reversals on exchange market pressure.
The present article aims at filling this gap in the literature through investigating the validity of the sudden stop theory in Turkey. In addition, it offers a contribution to the existing literature on the methodological front by applying the autoregressive distributed lag (ARDL) bounds testing procedure (Pesaran, et al. 2001) to investigate the short-run and long-run causal effects of capital reversals on exchange market pressure (hereafter EMP), which has not been done to date. This is a novel approach as the ARDL methodology allows inclusion of a stationary, I(0) EMP index in the analysis while the conventional cointegration techniques can only accommodate I(1) series.
It has been widely argued that Turkish economy became vulnerable to...