Measuring Hard Currency Index

Lula G. Mengesha, et.al., Int. J. Eco. Res., 2011, 2(2), 61-66 ISSN: 2229-6158

THE DOLLARIZATION OF LESS DEVELOPED COUNTRIES: A NOTE ON THE COMPUTATION OF A HARD CURRENCY INDEX Lula G. Mengesha and Mark J. Holmes Department of Economics, University of Waikato, New Zealand Email: [email protected]

Abstract Numerous studies have investigated the impact of dollarization on the wider economy. In measuring dollarization, many studies use the ratio of foreign currency deposits to broad money. However, such an approach only focuses on one aspect of dollarization. This paper advocates an alternative approach that is holistic in nature. Using this new index, we reflect on the dollarization experienced in the Eritrean economy. Keywords: Hard currency, dollarization, black market, Eritrea, broad money. JEL Codes: E4, E5, F3, O5.

1. Introduction considers onshore dollar deposits and fails Studies such as Girton and Roper (1981), to incorporate offshore dollar deposits as Bahmani-Oskooee and Domac (2003) and well as dollar loans issued by domestic Levy Yeyati, (2004), have argued that the banks either to domestic firms and extent of relying on hard currency or households or to foreign financial dollarization can have significant institutions and firms. Likewise, the consequences for the economy with respect studies by Levy Yeyati (2006), Rennhack to the volatility of the exchange rate, & Nozaki (2006) and Nicolo et al. (2005) inflation and the effectiveness of monetary use the ratio of foreign currency deposits policy. Measuring the degree to which any to total bank deposits rather than broad given economy relies on hard currency is money. The study of Fuentes (2007) crucial before identifying any potential employs the ratio of firms' dollar impact that this might have. An important denominated debt to firms' total asset in issue arises as to whether dollarization analyzing the causes and effects of should be measured in one particular form dollarization on Chilean firms at micro or in a more holistic way. The standard level. In identifying the determinants of measure of dollarization, which is dollarization, the study by Honig (2009) commonly used by IMF and others, is the uses the sum of firm liabilities and bank ratio of foreign currency deposits to broad liabilities to measure dollarization. money in the economy [see, for example, In contrast to these studies, Agenor & Khan (1996), Yinusa (2008), Hausmann et al., (2001) and Reinhart et al., Clements & Schwartz (1992), Viseth, (2003) respectively develop the original (2001), Komarek & Martin (2001), sin index (based on the extent of liabilities Akacay et al., (1997), Melvin & Peiers contracted in foreign currencies.) and (1996) and Balino et al., (2000) among composite indices (based on sum total of others]. the ratios of foreign currency deposits to This method, however, captures broad money, total external debt to GDP only one form of dollarization, namely and domestic government debt financial dollarization. Such a measure denominated in dollars to total domestic neglects other forms of dollarization that government debt). Despite the differences can exist. Although this measure aims to in their methods of measuring dollarization, capture financial dollarization, it only these studies have focused on limited

IJER | Mar - Apr 2011 61 Available [email protected] Lula G. Mengesha, et.al., Int. J. Eco. Res., 2011, 2(2), 61-66 ISSN: 2229-6158

aspects of the process. While the sin index another country's currency as a legal is useful only in capturing the liability tender. De facto dollarization, however, is dollarization, the composite index puts another form of dollarization where the emphasis on dollar loans issued to the domestic currency stays as an official legal government with no inclusion of dollar tender but foreign currency can also be loans issued to the private sector. used for some transactions with or without Moreover, this method neglects the legal approval. Financial dollarization on offshore dollar deposits of domestic the other hand refers to the holdings of investors. It also fails to include foreign currency denominated financial underground dollarization, or the assets and liabilities by the residents. Real dollarization of the black market, which is dollarization is also another form of most common in the economies of third dollarization that refers to the indexing of world countries. prices and wages to the dollar. Liability Bearing in mind the importance of dollarization refers to foreign currency including as many aspects of dollarization denominated debt held by residents as well as possible, this paper constructs the as the government. The dollarization of the holistic approach in measuring Eritrean black market economy has been termed as dollarization. Unlike the above-mentioned underground dollarization. studies, we do not confine measurement to Potentially there are elements of the dollar loans issued to the government. financial, liability, real and underground We also capture dollar loans issued to the dollarization in the Eritrean economy. private sector. In addition to this, our Since real dollarization is in its infant stage, measure also incorporates the dollarization as reflected by the selling and buying of of the black market. This paper is houses in US dollars starting from 2006, organised as follows. The following this paper incorporates the three types of section provides a review of the literature dollarization. In doing so, it is necessary to and discusses the computation of the hard measure the extent of underground currency index. The final section dollarization present. We first consider the concludes. foreign exchange reserves in the Eritrean economy used to finance the import of 2. Constructing a Hard Currency Index goods from abroad and service the external Dollarization has different meanings debt. Let according to different researchers. The E T M t  t FRt (1) studies of Ortiz (1983), Melvin & Peiers where M E is import expenditure, FR T is (1996), Clements & Schwartz (1993) and t t Melvin & de la Parra (1989) denote total foreign exchange reserves, and dollarization as the use of dollars within 0  t 1 is the proportion of total foreign domestic economy. Other studies such as exchange reserves used for financing the Wessels (2004), Viseth (2001) and Zoryan expenditure for imported goods. is (2005) interpret dollarization as the use of used to express import expenditure if there foreign currency in the domestic economy. is no other source of financing import Dollarization has also been defined as expenditure such as dollars from the black official or full (de jure), unofficial or partial (de facto), financial, real and market. The remaining proportion of liability dollarization. According to Gulde can be used to service debt DSt or carry et al., (2004), Minda (2005), Ize & Yeyati foreign exchange reserves forward FR R as (2003), Reinhart et al., (2003) Berkmen & t follows A., (2009) and Wessels (2004) de jure R T dollarization is defined as the adoption of (FRt  DSt )  (1  t )FRt (2) 2

IJER | Mar - Apr 2011 62 Available [email protected] Lula G. Mengesha, et.al., Int. J. Eco. Res., 2011, 2(2), 61-66 ISSN: 2229-6158

Adding equations (1) and (2) provides As discussed above, case one and E T R case three are the most appropriate cases in M t  FRt  FRt  DSt (3) Equation (3) provides a number of determining the amount of hard currency scenarios used to finance imported goods through R E the black market. We now measure the (i) If FR t =0 and DSt > 0 then M t = T overall Eritrean dollarization. FR t - DSt R E Incorporating the three types of (ii) If FR t and DSt > 0 then M t = T R dollarization discussed above- claims on FR t - FR t - DSt R E T foreign commercial banks, external debt (iii) If FR t and DSt = 0 then M t = FR t Of these three cases, (i) and (iii) and hard currency supplied by the black reflect the Eritrean economy more market- we have the following index DL  B  FB realistically than (ii). With the exception of HCI  t t t (5) t M 2  DCC 1996, Eritrea has been servicing its debt t t each year since 1995. It can be noted, where HCIt is the hard currency index, DLt therefore, that DS is greater than zero is dollar loans issued by the Eritrean banks, except during 1996 which is equal to zero. Bt is the amount of hard currency supplied Information on the foreign exchange by the black market, FBt is foreign reserves of Eritrea indicate that there are borrowing, M2t is the money supply and no remaining foreign exchange reserves DCCt is domestic currency in circulation. carried forward to the next period since The data appendix discusses the data there is shortage of hard currency earnings sources. Using quarterly data on all these and high demand for foreign currency to variables, Figure 1 plots the HCI series. finance the importation of goods. While there has been a general The next step is for us to identify upward trend in the HCI over the study the amount of hard currency in the Eritrean period, there was a sharp fall in 1997. This black market which is also used to finance fall was driven by the lower dollar loans the import of goods. Therefore, the actual issued by the banks. There are number of imported goods are not only financed by factors that accompanied this. Namely the the  t share of the total foreign exchange introduction of a new currency in the reserve as determined above, but also by Eritrean economy towards the end of 1997, the hard currency supplied by the black and increased military expenditure driven market. The difference between the actual by the purchase of weapons from abroad in imports of goods M A and the imports of connection with the second war with t Ethiopia. goods financed by total foreign exchange The HCI has continued to rise reserves can be written as A E especially from 2002 onwards. This could Bt  M t  M t (4) also be explained by the issuance of new where Bt denotes the amount of hard currency and the second war with Ethiopia. currency obtained from the black market to Prior to these events, goods were mainly finance import expenditure. Substituting imported from Ethiopia. Transactions for equation (3) into (4) provides the these imports were settled in Birr, which following three scenarios was former local currency unit for both A T R (i) Bt =M t - (FR t - DSt), if FR t =0 countries. After the introduction of new and DSt > 0 currency and the second war, however, A T R R (ii) Bt =M t - (FR t - FR t - DSt), if FR t imports needed to be sourced from other and DSt > 0 countries as Eritrea lost access to the A T R (iii) Bt =M t – (FR t), if FR t and DSt = 0 Ethiopian market. The transactions of imports also needed to be settled in hard currency. In addition to this, the Eritrean 3

IJER | Mar - Apr 2011 63 Available [email protected] Lula G. Mengesha, et.al., Int. J. Eco. Res., 2011, 2(2), 61-66 ISSN: 2229-6158

economy lost an equivalent amount of Figure 1. The Hard Currency Index for 43.02 million U.S. dollars from export Appendix:Data earnings to Ethiopia. Moreover, the HCI growth of Eritrean GDP declined from an .9 average of 10.8% to an average of 0.2%. .8 The loss in export earnings coupled with .7 the fall in GDP made the economy more .6 reliant on foreign borrowing and imported .5 goods. The rise in the demand for imported .4 goods and the need to settle transaction in .3 hard currency for imported goods therefore .2 increased the reliance on hard currency as .1 reflected in the HCI. .0 96 97 98 99 00 01 02 03 04 05 06 07 08

3. Conclusion To measure the hard currency Measuring the degree to which an index, quarterly data on foreign exchange economy relies on hard currency or reserves are obtained from the IMF dollarization is vital before assessing its International Financial Statistics (IFS). effect in the economy. In finding the M2 was considered in lieu of a extent of dollarization it is also vital to measurement of broad money. However, identify and incorporate the different quarterly data on M2 was also not forms of dollarization that might exist in available for the full study period. the economy. However, this has been a Therefore, broad money was calculated as main shortcoming of the existing literature. the sum total of M1 and quasi money. The majority of the studies have focused Domestic currency in circulation was on some limited aspect of dollarization in deducted from this measure of broad their studies. The most commonly used money in order not to understate the measurement of dollarization, which is the weight of dollar loans issued by the banks. ratio of foreign currency deposit to the Quarterly data on domestic currency in broad money, for example tend to neglect circulation were also obtained from the other forms of dollarization by focusing on IFS database. financial dollarization. It also fails to In computing the dollar loans incorporate offshore dollar deposit by issued by the banks, quarterly data on total including only onshore dollar deposit even foreign assets, foreign exchange reserves though it measures financial dollarization. and gold reserves are required as the data In sharp contrast to this, this paper for dollar loans were not fully available. adopts a holistic approach to measuring the The sum total of foreign exchange reserves dollarization based on the incorporation of and gold reserves then were subtracted all major forms of dollarization that exist from total foreign assets to obtain the in the Eritrean economy. Most importantly dollar loans. Quarterly data on total it has captured the extent of dollarization foreign assets, gold reserves and foreign in the black market economy. The index exchange reserves are obtained from demonstrates that dollarization has Balance of Payments of Eritrea IMF file increased in the Eritrean economy. This number 643 as well as from IFS. new approach has significant potential The amount of foreign borrowing with regard to the study of other less was obtained by adding total multilateral developed economies. loans, cross-border loans from BIS reporting banks and cross-border loans from BIS banks to non-banks. Quarterly 4

IJER | Mar - Apr 2011 64 Available [email protected] Lula G. Mengesha, et.al., Int. J. Eco. Res., 2011, 2(2), 61-66 ISSN: 2229-6158

data on these variables are obtained from Evidence from Chilean the Joint External Hub. The data on the Firms.Unpublished manuscript. import of goods and services as well as Girton, L., & Roper, D. (1981). Theory debt service are available in the form of and Implications of Currency annual frequency. This data were collected Substitution. Journal of Money, from World Development Indicators of the Credit & Banking, 13(1), 12-30. World Bank and then interpolated. Gulde, A.-M., David, H., Alan, I., David, M., & Gaianni, D. N. (2004). References Financial Stability in Dollarized Agenor, P. R., & Khan, M. S. (1996). Economies. International Monetary Foreign Currency Deposits and the Fund Occasional Paper 230. Demand for Money in Developing Hauskrecht, A., & Hai, N. T. (2004). Countries. Journal of Development Dollarization in Economics, 50(1), 101-118. Vietnam.Unpublished manuscript. Akçay, O. C., Alper, C. E., & Karasulu, M. Hausmann, R., & Panizza, U. (2003). On (1997). Currency substitution and the Determinants of Original Sin: An exchange rate instability: The Empirical Investigation. Journal of Turkish case. European Economic International Money and Finance, Review, 41(3-5), 827-835. 22(7), 957-990. Alvarez-Plata Patricia, & Alicia, G.-H. Hausmann, R., Panizza, U., & Stein, E. (2008). To Dollarize or De- (2001). Why Do Countries Float the dollarize:Consequences for Way They Float? Journal of Monetary Policy. Discussion Papers Development Economics, 66(2), 387- of DIW Berlin, German Institute for 414. Economic Research(842). Honig, A. (2009). Dollarization, Exchange Bahmani-Oskooee, M., & Domac, I. Rate Regimes and Government (2003). On the Link between Quality. Journal of International Dollarisation and Inflation: Evidence Money and Finance, 28(2), 198-214. from Turkey. Comparative Ize, A., & Yeyati, E. L. (2003). Financial Economic Studies, 45(3), 306-328 dollarization. Journal of Balino, T., Bennett, A., & Borensztein, E. International Economics, 59(2), 323- (2000). Monetary Policy in 347. Dollarized Economies.Unpublished Komarek, L., & Martin, M. (2001). manuscript. Currency Substitution in the Berkmen, S. P., & A., C. E. (2009). Transition Economy : A Case of the Exchange rate Policy and Liability Czech Republic 1993- Dollarization:What Do the Data 2001.Unpublished manuscript. Reveal About Causality IMF Levy Yeyati, E. (2006). Financial Working Paper. Dollarization: Evaluating the Clements, B., & Schwartz, G. (1993). Consequences. Economic Policy(45), Currency substitution: The recent 61-110. experience of Bolivia. World Melvin, M., & de la Parra, G. A. (1989). Development, 21(11), 1883-1893. Dollar Currency in Latin America: A Clements, B. J., & Schwartz, G. (1992). Bolivian Application. Economics Currency Substitution--The Recent Letters, 31(4), 393-397. Experience of Bolivia.Unpublished Melvin, M., & Peiers, B. (1996). manuscript.Fuentes, M. (2007). Dollarization in Developing Dollarization of Debt Contracts: Countries: Rational Remedy or

IJER | Mar - Apr 2011 65 Available [email protected] Lula G. Mengesha, et.al., Int. J. Eco. Res., 2011, 2(2), 61-66 ISSN: 2229-6158

Domestic Dilemma? Contemporary Economic Policy, 14(3), 30-40. Minda, A. (2005). Offical Dollarization: A last resort solution to financial stability in Latin America. Groupment de Recherches Economiques et Sociales(Working Paper 2, January). Nichols, D. A. (1974). Some Principles of Inflationary Finance. The Journal of Political Economy, 82(2), 423-430. .Nicolَ, G. D., Honohan, P., & Ize, A (2005). Dollarization of bank deposits: Causes and consequences. Journal of Banking & Finance, 29(7), 1697-1727. Ortiz, G. (1983). Currency Substitution in Mexico: The Dollarization Problem. Journal of Money, Credit and Banking, 15(2), 174-185. Reinhart, C. M., Rogoff, K. S., & Savastano, M. A. (2003). Addicted to Dollars.Unpublished manuscript. Rennhack, R., & Nozaki, M. (2006). Financial Dollarization in Latin America.Unpublished manuscript. Viseth, K. R. (2001). Currency Substitution and Financial Sector Developments in Cambodia.Unpublished manuscript. Wessels, G. M. (2004). The Suitability of Dollarization as an Exchange Rate Regime For South Africa. THe South African Journal of Economics, 72 :2 June. Yinusa, D. O. (2008). Between Dollarization and Exchange Rate Volatility: Nigeria's Portfolio Diversification Option. Journal of Policy Modeling, 30(5), 811-826. Zoryan, H. (2005). The Measurement of Co-circulation of Currencies and Dollarization in the Republic of Armenia. European Journal of Comparative Economics, 2(1), 41-65.

IJER | Mar - Apr 2011 66 Available [email protected]