Alternative Measures of Currency and Asset Substitution: The Case of Turkey
By Ozden Birkan
Introduction: The widespread experience in most developing countries of the use of foreign currencies and foreign currency denominated assets both in private transactions and by the governments has been an issue of vast academic interest since 1970s. Despite the lively research on the subject, the literature still suffers from a proper theoretical and empirical distinction between currency substitution, referring to the use of multiple currencies as medium of exchange and asset substitution, referring to the existence of foreign currency denominated financial assets in addition to domestic ones in domestic residents' portfolio as store of value which is associated with international capital mobility.
Different theoretical frameworks are used to analyze the implications of the two phenomena for exchange rate and interest rate determination and to measure their extent. Problems common to most of the research are the difficulty of tackling with the fact that different assets provide different degrees of liquidity services and the difficulty of formally distinguishing between the store of value and medium exchange functions of money in the models. Moreover, when it comes to empirical tests of proposed policy implications, the convention is to use the ratio of foreign currency denominated assets to some monetary aggregate like M2 or M3 as the measure of currency substitution. This is an ad hoc measure, which does not differentiate between currency and asset substitution. Main reason for the common use of this measure without any theoretical justification is the lack of data on foreign currency holdings in economies subject to substitution.
In this study an alternative approach based on the Divisia monetary aggregates literature is considered. The model is based on the Barnett critique, which suggests that money is better measured by a weighted average of the flow of monetary services from a stock of monetaryassets. Making use of a model based in the Divisia literature, it is possible to obtain theoretically substantiated measures for currency and asset substitution and address deliberately the issue of differing degrees of liquidity of different assets.
This paper investigates how such a model can contribute to the analysis of the currency and asset substitution experience in Turkey. First the basic theoretical literature on dollarization and some common characteristics of empirical research in the field are introduced. A brief discussion of the rationale for Divisia Monetary Aggregates is followed by the theoretical model on which distinct measures of currency and asset substitution are founded. After a brief outline of the developments in Turkish financial markets, the relevant measures of substitution are developed and tests of the currency substitution and asset substitution hypotheses are carried out using time series econometrical methods. The question of whether dollarization in Turkey, either in the form of asset substitution or currency substitution, exhibits irreversibility is addressed with an attempt to identify the source of the irreversibility. Finally theoretical and empirical conclusions of the study are presented.
Download: Alternative Measures of Currency and Asset Substitution: The Case of Turkey (PDF, 524.46 K)blog comments powered by Disqus