Measurement of Co-Circulation of Currencies and Dollarization in the Republic of Armenia, The

European Journal of Comparative Economics, The, 2005 by Zoryan, Hakob

The introduction of high denomination 20,000 and 50,000 dram notes (approx. $35 and $85) did not changes the situation. Furthermore, 50,000 dram notes are rare in circulation and 20,000 notes are not popular with the public. This example of high denomination dram notes is a good illustration of the narrow use of drams for larger transactions.

Although it is hard to predict whether the use of drams will expand in the future, the fact is that foreign currency, primarily the U.S. dollar, performs many money functions in Armenia.6 Therefore, estimation of the actual stock of foreign currency in Armenia is essential in evaluating the real money supply

4. The Survey

A. Methodology of the Survey

Foreign currency in circulation (FCC) cannot be obtained through reports as general reporting on import and export of foreign currency at customs has virtually been eliminated. Actual data would be biased even if such reports existed, since firstly, small amounts are not reported and secondly, such reports are usually not accurate. And if all these obstacles were removed, these data would only provide additions to the stock of FCC, not the stock itself.

Surveys provide an alternative for determining the level of FCC. To obtain a more accurate assessment of the level of FCC, we did not ask respondents directly how much foreign curreny they were holding. Rather respondents were asked about their portfolio of cash holding (i.e. the ratio of FCC to DCC or the ratio of FCC to FCD). Since the level of both FCD and DCC is known, it is possible to estimate the level of FCC from this information.

The survey has been conducted among households. While businesses hold substantial amounts of FCC, most non-financial businesses are not legally allowed to hold foreign cash. Therefore, survey results among companies are likely to be biased downward. Instead, we included owners of businesses as households in our sample, where cash holdings of their businesses are considered as holdings of their owners (as is the legal case in Armenia).

Wealth distribution is very unequal in Armenia and, therefore, CSI would vary significantly among different wealth groups. For this reason, the population was split into four wealth groups based on household surveys of 1999 and 2001 and a survey on inequality of consumption and income in 2000, conducted by National Service of Statistics and sponsored by the World Bank. We have also used Lorenz Curve data on distribution of income, constructed by experts of the Central Bank, to estimate holdings of cash by each wealth group. Here we have assumed that the income distribution is the same as distribution of holdings of cash. While wealth and income are normally different, since the market economy is only ten years old in Armenia, the differences between the distribution of income and distribution of holdings of cash should not be high. Furthermore, due to lack of confidence in the banking system most people hold their monetary wealth in the form of cash.

The wealth groups are constructed as follows: