Ecuador : quasi fiscal deficits - walking through the maze of subsidies

-- ：계핀 LAO DISCUSSION PAPER SERIES Report No. Tito, Author and Date IDP-3 'An Analysis of the Sources of Earnings Variation Among Brazilian Males' by Marcelo Dabos and George Psacharopoulos, December 1 IDP-4 'The Efficiency and Effectiveness of Export Credit and Export Credit Insurance Programs* by Bruce Fitzgerald and Toffy Monson (Consultant), December 1987 IDP-9 *Export Processing Zones: The Economics of Offshore Manufacturing' by Peter 0. Warr (Consultant), August 1987 IDP-10 *Dumping, Antl-dumplng and Efficiency* by Bruce Yandle and Elizabeth M. Young (Consultants), August 1987 IDP-l i *The Regulation of the Quality of Traded Commodities and Services' by Simon Rotteriberg and Bruce Yandle (Consult"), June 1987 IDP-12 'Argentina: Problems for Achieving Macro Stability* by F. Desmond McCarthy and Alfredo E. Thorne, January 1988 IDP-13 "Argentina: Towards the Year 2000* by F. Desmond McCarthy, June 1987 IDP-14 'Trade Liberalization: The Lessons of Experience", Papers presented In the conference 'Toward a Now Trade Policy for Brazil*, Sao Pa April 11 and 12, 1W8 IDP-16 "Aspects of Privatization: The Case of Argentina 1976-81 * by A. Ludere (consubnt), April 1988 IDP-17 *Aspects of Privatization: The Case of Chile 1974-W", by D. Hachette (oonsuftaM), April 1988 IDP-18 *Privatization In Argentina and Chile: Lessons from a Comparison' D. Hachette and R. Luders (oonsultants), April 19W IDP-19 "Principles of Water Supply Pricing in Developing Countries* by Mohan Munasinghe, June 19M IDP-20 *The Status of Energy Eoonomlcv. Theory and Application* by Mohan Munasinghe, June 1988 IDP-21 *What are the Prospects for Land Rebrm?" by Hans Sinswanger and Miranda Elgin (oonsuftaM), August 19M IOP-24 *Managing Argentina's External Debt The Contribution of Debt Swaps", Carlos Alfredo Rodriguez (consultant), January 1M IDP-29 'Managing Mexico's External Debt. The Contribution of Debt Reduction Schemes*, Allen Sangiln6s (consultant), January IM IDP-30 "Debt Reduction Schemes and the Management of Chilean Debt', Felipe Larra fn (consultam), March 1989 IDP-31 'Managlag Brazil's External Debt The Contribution of Debt Reduction Schemes*, Dionisio D. Camelro and Rogerto LF. Werneck (consulianta), January IM iDP-32 'Leading Economic Indicators for Brazil: At Attempt at Forecasting Turning Points% Antonio Estache, February 1989 IDP-W "Recovering Growth with Equity, World Bank Poverty Alleviation Activides In Latin America% George Peacharopoulos, April 1989 IDP-34 'Brazil - External Debt Development and Prospects'. SlIvina Vatnick, December IM IDP-40 "Cash Debt Buy Backs and the Insurance Value of Reserves", Sweder van Wijnbergen, June 1989 IDP-41 'Growth, External Debt and the Real Exchange Rate In Me)doo*, Sweder van W11ribergen, May iW9 iDP-43 *The Macrosconomics of Social Security In Brazil: Fiscal and Financial Issues', Douglas Puffert (Consuftam) and Emmanuel Y. Jimenez, June 19W IDP49 *Feeding Latin America's Children*, Human Resources Division, Technical Departawnt, October 19M IDP-50 *Marginal Effective Tax Rates on Capital Income In Argentina", Luca Barbone and Michael McKee (Consultant), August 19W IDP-53 *A Bibliography on Poverty and Income Distribution in Latin America% Human Resources Division, Technical Department November low IDP-54 *Ecuador Development Issues and Options for the Amazon Region", Country Department IV, Deoembet 19M IDP-65 "Fiscal Deficits, Inflation and Interest Rates*, Rul Coutinho, June 1989 IDP-W *Regulating Industrial Pollution in Developing Countries: Some Possible Frameworks", Paul R. Portney (Consultant), December IN9 IDP-60 "Strategv for the Promotion of Non-Traditional Agricultural Exports", Martin F. Raine, December 1989 IDP-61 *A Portfolio Model for Brazil', Rul Coutinho, June 1989 IDP-64 'Why Tax Incentives Do Not Promote Investment in Brazil% A. Esteohe and V. Gaspar (consu" , February 19W IDP-W *Public Sector *Debt Distress" In Argentina's Recent Stabilization Efforts, Paul Beckerman, July 1990 IDP-W "Allocating Credit How to Direct, Price or Auction', S. Ramachandran, June 1990 IDP-70 *LAC Economic and Sector Work: A Critical Review of Selected Issues', L Auomheimer (consultant), June 1990 IDP-71 Wricultural Trade In the Caribbean CommunW, John Nash, July 1990 IDP-72 *Poveny Alleviation In Brazil, 1970-87", M. Louise Fox, July 19W IDP-73 *Fmal Deficits, Exchange Rate Crisis and Inflation", Sweder van Wijnbergen, January 1990 IDP-78 *Measuring Price Distortions from Commodity Taxation in Brazil*, A. Estache and A. Fernandes and 1. Roy (consultants), July 1990 IDP-79 *Child Care in Metropolitan Brazil", Deborah Leviison (Consultant), August 1989 ECUADOR: QuS-Piecal Dots Waldng through the Maze of Subsidies Luca Barbone and Mario Vicens Summary This paper discusses the conceptual framework for the analysis of the operating deficits of the Central Bank of Ecuador (identified as Quasi-fiscal deficits), and presents estimates for 1989 and projections for 1990. It also decomposes the estimated deficits identifying the cross-subsidies that are entailed. The magnitudes of the deficits are large, if compared to the consolidated non-financial public sector results; the implicit subsidy system is irrational and anti-economic. ECUADOR: Quasi-Fiscal Deficits Walking through the Haze of Subsidies Luca Barbone and Mario Vicens TABLE OF CONTENTS A. INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Central Bank Net Income and Consolidated Public Sector Deficits . . 1 B. QUASI-FISCAL DEFICITS: A DEFINITION . . . . . . . . . . . . . . . . . 2 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Balance Sheet and Income Flows . . . . . . . . . . . . . . . . . . 3 Accounting for the Maze of Subsidies . . . . . . . . . . . . . . . 5 Real vs. Nominal Quasi-Fiscal Deficits . . . . . . . . . . . . . . 7 C. ESTIMATES FOR ECUADOR . . . . . . . . . . . . . . . . . . . . . . . . 9 Main Determinants of the Quasi-Fiscal Result . . . . . . . . . . . 9 Composition of the Balance Sheet . . . . . . . . . . . . . . . . . 10 Nominal Quasi-Fiscal Deficit in 1989 and 1990 . . . . . . . . . . . 12 Quasi-fiscal Resul;: Decomposition of Subsidies . . . . . . . . . . 14 Deficit Estimates: Real . . . . . . . . . . . . . . . . . . . . . . 17 D. CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 1/ Senior Economist, The World Bank, Washington D.C., and partner, Macroeconomia, Buenos Aires. The views expressed herein are those of the authors only and are not necessarily shared by the World Bank. A. INTRODUCTION Central Bank Net Income and Consolidated Public Sector Deficits 1. Increasing attention has been paid in recent years to the economic effects of Central Bank deficits, as a result, in part, of the experiences of Argentina and Brazil in this respect.2 The traditional focus of the analysis on the economic effects of non-financial public sector deficits implied that the Central Bank can only contribute to a macroeconomic disequilibrium if its domestic credit policy is not i: line with the behavior of the demand for Central Bank liabilities. This could be labelled the "flow" approach to the role of the Central Bank in macroeconomic equilibrium. The Central Bank, however, is not conceptually different from another public enterprise, with the important exception that it can finance its own deficit, and that it is the first recipient of the proceeds of the inflation tax and of seignorage. In a variety of countries, in addition, the transparency of its accounts is very poor, and over the years a number of expenditure functions have been shifted--perhaps because of the very lack of budgetary control--from the non- financial public sector to the Central Bank. As a result, the balance sheets of the Central Bank have become progressively more burdened by non-performing assets or by the servicing of debts for which no resources are provided. In many cases, the net worth of the Central Bank, at market values, would be negative. This has also led to increasing financing requirements, met by money emission or by increases in other central bank liabilities, which have often compounded the effects of other sources of macroeconomic disequilibrium. Thus, particularly when the financing requirements of the Central Bank are large, a "stock" view of the Central Bank's role becomes the more useful tool for macroeconomic analysis.. The financial needs of the public sector are defined by the consolidation of both the non-financial deficit and the Central Bank net income, the latter being usually referred to as the quasi-fiscal deficit. 21 See Barbone-Beckerman (1989), Piekarz (1987) and World Bank (1989), for a discussion of the quasi-fiscal deficit in Argentina. 2. This paper provides a discussion of issues in the definition and measurement of quasi-fiscal deficits, applied to the Central Bank of Ecuador (BCE). Section B reviews alternative concepts of the deficits, and the associated definitional problems and shortcomings. Section C discusses the structure of the BCE's balance sheet, identifies the sources of gains and losses, and provides an estimate of the quasi-fiscal result for 1989 and 1990, as well as a decomposition of the result by type of subsidy. B. QUASI-FISCAL DEFICITS: A DEFINITION Introduction 3. Despite the variety of existing institutional arrangements (ranging from nominal ownership by financial system entities, to complete independence, to the status of a Treasury agency), Central Banks are public bodies that perform a variety of public functions. It stands to reason, therefore, to include the cost of performing such public functions into the general public sector expenditures. Profits and losses at a Central Bank may result either from activities that are considered "proper" for a monetary authority, or because of the execution of activities that are extraneous to it. The first may include, for instance, gains or losses from the management of the exchange rate, if this is fixed or managed; losses or profits from the exercise of the function of lender of last resort (if this is deemed to be an appropriate function of the central bank); etc. Among the second, we may include the profits and losses accruing from any intermediation activities, be they subsidized or not; the losses derived from the servicing of external debt that was not originally incurred into by the Central Bank; and such non-operational expenditures such as grants and ...n-reimbursable loans to public and private entities for a variety of reascns. 4. In this section, we first define the concept of the nominal quasi-fiscal deficit, linking the structure of the balance sheet to the potential income flow generated by the Central Bank. We then show how the existence of cross -3- subsidies modifies the potential income result; and we finally discuss the issue )f nominal and real measurements of the changes in net worth. Economic and Monetary Effects of the Deficits 5. The quasi-fiscal deficits (as discussed below) have two dimensions, financial and economic. The difference between revenues and expenditures on a cash basis is part of the financial needs of the Central Bank. Usually, monetary programs incorporate the ex-ante financial effects of quasi-fiscal deficits together with the cash flow derived from the expected change of credits, investments and liabilities during a given period. The economic concept of the deficit, i.e. the difference between revenues and expenditures accrued during the period, should reflect the changes in the net worth of the Central Bank. The two concepts are related to each other, although a reconciliation of balance sheet figures is often made difficult by the application of special exchange rates for accounting purposes, or by the recording of interest payments on a cash basis only. Balance Sheet and Income Flows 6. The relationship between Central Bank activities and operational result is best discussed by reference to a stylized balance sheet (Figure 1). A Central Bank will typically have domestic and external assets and liabilities, and monetary and non-monetary liabilities. Figure 1 - Stylized Central Bank Balance Sheet eNFA MB DC CU CR Rfs CR!* CBB CRPr D OA OLf* NW 7. In fig. 1, NFA represents net foreign assets of the Central Bank-- reserves minus short-, medium- and long-term foreign debt--and e is the exchange rate. DC is total domestic credit, composed of credit to the government CRO, credit to the financial sector CR', and direct credit to the private sector CRPr. OA are other assets of the central bank (including revaluation accounts). On the liabilities side, MB is the monetary base, equal to currency in circulation CU plus unremunerated reserves of the banking system, CBB are central-bank bonds, DO are deposits of the public sector, OL S represents other liabilities towards the financial and private sectors (including advanced import deposits), and, finally NW is the net worth.3 8. Associated with each balance-sheet item are income and expenditure flows, as well as valuation changes due to a variety of reasons. In absence of interest rate subsidies and involuntary unremunerated investments by the private, financial and public sectors, the net financial income of the Central Bank would be the sum of the net interest payments on external assets (the external deficit), the net income on domestic operations (equal to the interest rate times domestic credit minus domestic nonmonetary liabilities) and unrequited transfers and operational expenses. We define this as the potential nominal quasi-fiscal deficit (QFDN*): (1) QFDN = - ei NFA + (iCBB + iDQ + OLfs - ICR9 - iCe* - iCRPr) + OE + REG where i is the domestic "market" interest rate, i * is the foreign interest rate, OE are operational expenses, REG are cash grants. This definition (again, in absence of interest rate subsidies) would indicate by how much the net worth of the central bank varies from year to year (net of valuation changes on its net external assets). The presumption is that the domestic component of the overall result would be positive, since domestic non-monetary liabilities are smaller than domestic credit, in view of the existence of the 31 There is a strict connection between the net worth of the Central Bank and the item "Other assets". In many instances, in fact, the balance sheet will show a stable net worth over time, with the residual item OA being used to "activate" losses that are incurred into. - 5 - monetary base. A negative result might emerge if operating expenses or non- refundable grants were to be excessively large, something that may have indeed happened in the past. Whether the external result would be negative or positive, depends on the extent to which the Central Bank may have been burdened with servicing of other entities' debt. Accounting for the Maze of subsidies 9. The definition of nominal quasi-fiscal deficit adopted in (1) is rather straightforward. In practice, however, the calculation is made more difficult by the existence of peculiar practices that characterize--in Ecuador as well as in other countries--central banking operations. Indeed, the assumption of market interest rates charged on all domestic operations as in (1) is a gross simplification of reality. Generally, only private sector assets yield interest, and the income flows between the Central Bank and the rest of the economy can result in transfers from the public sector to private sector or within the public sector. This occurs whenever the interest charged on Central Bank loans or the yield on bond investments are below market rates. The difference can be interpreted as an implicit subsidy or tax. 10. More specifically, in the case of Ecuador, the following "institutional arrangements" can be noted: (i) Deposits of public entities are not remunerated; (ii) non-monetary deposits of private and financial institutions are not remunerated either; (ii) credits to the government are remunerated at below-market-rates; (iii) credits to the private sector were generally remunerated at below-market interest rates, sometimes substantially below market4; (iv) the Central Bank has been burdened with the servicing of the private external debt (the so-called "sucretization"), but the domestic counterpart of the debt is vastly insufficient to cover the actual servicing burden (see below). 4/ Although recently Central Bank credit lines have been moved in the direction of the market rates when issued, there are no adjustment mechanisms for outstanding loans, once they are approved. - 6 - 11. We can take into account these features by defining the observed nominal quasi-fiscal deficit as the sum of the potential deficit (as defined in (1)) and the various subsidies granted and taxes levied by the Central Bank on the other sectors of the economy. These features are resumed in eq. (2), which restates the nominal quasi-fiscal deficit in terms of the different subsidies that emerge as a consequence of the application of different interest rates: (2) QFDN = 10E + REG + i(CBB + D9 - CRO - CR '+ OLts) - (0D) - (i8-i)CRQ)]+ [(i - ife)CR - iOLs) + ei* (R *-LIAB*] + Cei SUCRET - EXCHCOM] M QFDN + SUBSP +SUBSP The first term in square brackets represents the quasi-fiscal result that would emerge in absence of subsidies, as defined in (1). The second term in square brackets represents the net subsidy from the non-financial public sector received by the central bank. It is composed of two termss first, as deposits of public entities are not remunerated, the quasi-fiscal surplus is increased by the amount corresponding to the interest forgiven; second, as the interest rate charged on government loans is lower than market rates, the quasi-fiscal deficit is increased by the corresponding amount. The third expression in square brackets, finally, represents the effect of interest rate subsidies on loans to the private sector (directly or through the financial sector). The quasi-fiscal deficit is increased by the lower-than-market interest rates charged on credits to the financial sector, but is reduced by the fact that no interest is paid on other non-monetary obligations (such as advanced import deposits, etc.). 12. The quasi-fiscal deficit is then completed by its external component. This is composed of two terms, relating to the public and privatized debt. The first term is given by the difference between interest received on reserves (R ) and interest paid on public external debt serviced by the central bank (LIAB ). This component can be interpreted as a subsidy to the - 7 - non-financial public sector. The second term is given by the difference between the payments on external "sucretized" debt of the private sector (SUCRET), and the counterpart funds available to the central bank through principal and interest on the stabilization bonds and through the exchange risk commission (EXCHCOM). Provisioning and Bad Loans 13. In addition to the income items considered above, it should be noted that appropriate loss provisions should be made each year (thus increasing the quasi-fiscal deficit) against the contingency that difficulties in recovering credits and investments may materialize. This may become particularly important for the central bank at times of financial distress, when contingent liabilities may run extremely high. For lack of information on the subject, however, no attempt at estimating the appropriate level of provisioning was made in this paper. Real vs. Nominal quasi-Fiscal Deficits 14. The definitions in (1) and (2), as discussed, are those of the nominal quasi-fiscal deficit. This may not be the most appropriate concept from an economic point of view, however. As the activities of the Central Bank involve primarily financial transactions, an additional problem emerges in the evaluation of its result, i.e. the evaluation of the real vs. the nominal results. Simply put, part of the income flow associate4 with a particular asset or liability represents a payment on account of capital. If the interest rate happens to be negative in real terms, for instance, the nominal flow associated with the activity is interpretable as a partial repayment of principal, leaving a negative result in real terms. Distinguishing between real and nominal results is particularly important when inflation is high, and consequently high nominal interest rates are dominated by accelerated repayment of principal. Depending on whether the interest- yielding assets are greater or smaller than liabilities, a positive net financial result might in fact be masking a decapitalization of the central bank. Thus, it is argued that the nominal result should be corrected for the - 8 - loss of value suffered in the period in consideration by the respective assets and liabilities. 15. This can be done by correcting expression (1) by the term -IrNW, which represents the loss of value of the net worth of the Central bank, to yield (3) QFDR - QFDN - WNW 16. Under the assumption that the real exchange rate (defined as the nominal exchange rate deflated by the domestic rate of inflation) does not change during the period, there will be no patrimonial gain or loss on account of inflation and the external debt. (3) can thus be re-written as (4) QFDR = OE + REG + (i-w)CBB - (1g-T)CR9 - (ift-)CR - ei*NFA + T(D9 + OL f -D9 + MB) 17. The (real) quasi-fiscal result is equal to the real interest rate charged on domestic loans, plus the net external interest payments, minus the loss in real value of the non-interest bearing liabilities. The last item in the last set of brackets is particularly important. .1MB represents in fact the (gross) inflation tax "levied" by the Central Bank on holders of domestic currency and on the compulsory reserves of the financial system. It has been argued by some that this definition of the inflation tax should be considered a financing item, rather than a real income item for the central bank. 'While the issue is primarily semantic, in the estimates shown below we identify separately each relevant component. - 9 - C. ESTIMATES FOR ECUADOR Main Determinants of the Quasi-Fiscal Result 18. In this section we present estimates of the quasi-fiscal deficit of the Central Bank of Ecuador for 1989, and projections for 1990, based on the monetary program for that year, and on the macroeconomic assumptions underlying it. As discussed in detail belnv, the economic result of the Central Bank is dominated by three basic itatures of its balance-sheet, which are partly the result of past policies and of the recent instability of the Ecuadoran economy. First, the Central Bank has long had a role of financial iatermediary, by far the most important one. However, interest rates on loans to the private sector have been systematically below market levels and, in addition, until recently were not adjusted in line with increases in the rate of inflation and the real rate of interest. 19. The second factor is provided by the assumption by the Central Bank of the foreign liabilities of the public and private sectors without a full counterpart on the asset side. This has been the case since 1984, when the domestic private sector was offered the opportunity of converting its external liabilities into domestic ones, with the Central Bank providing the credit for the operation. The terms of the credit, however, were extremely generous: (i) the interest rate was sharply negative in real terms; (ii) the amortization of principal was rescheduled several times; and (iii) the Central Bank assumed virtually all of the exchange risk, as the exchange commission provision resulted in a vastly under-adjustment of principal in surres. 20. Thirdly, the demand for money has been declining in the past few years. This has implied that, ceteris paribus, the Central Bank has been compelled to issue interest bearing internal debt to sterilize the excess supply of monetary base, in the process increasing its marginal cost of funds. 21. Against these shortcomings, it must also be recognized that the Central Bank Balance sheet is favorably affected by the fact that a large part of its liabilities are unremunerated. This includes compulsory reserves of - 10 - the banking system, and, in particular, deposit of public entities, which are very large, and receive no interest5. This feature reinforces the characterization of the Central Bank as an agency for the execution of transactions of a fiscal nature, but without the checks and balances provided by the budgetary process. Composition of.the Balance Sheet 22. Table 1 shows the composition of the balance sheet of the Banco Central del Ecuador for a selected number of years. Foreign currency obligations and assets are evaluated at the intervention exchange rate, rather than at the official exchange rate. The table reveals interesting trends. First, "net unclassified assetsu (a balancing item) account for over ninety percent of total assets. This line item reflects any past losses as well as the large external debt carried in the books of the Central Bank. It is also an indication of possible future losses: net unclassified assets do not generate income, while liabilities do. 23. The main counterpart of the net unclassified assets is of course the large external debt carried in the books of the Central Bank, amounting to about 40 percent of GDP. By contrast, net international reserves amounted, at end-1989 to little over 2 percent of GDP. The large external indebtedness is composed of: (i) rescheduled public debt (about half of total); (ii) the "sucretized" private debt (amounting to about one billion dollars, or twenty percent of total); (iii) and other obligations, taken over by the Central Bank either in its role of financial intermediary, or for other reasons.6 As will be discussed below, the Central Bank has very limited resources available for 5/ This despite the existence of legal provisions mandating remuneration of deposits, for instance, of the Social Security Institute. 6/ It should be noted that large discrepancies exist between balance- sheet data in national currency for external debt and external debt statistics in foreign currency. In addition, neither set of statistics deal adequately with the issue of the interest arrears that have been incurred into since 1987. - 11 - the servicing of this debt--short of accumulation of external arrears--which has thus become a major source of externil deficit. 24. Table 1 also shows a dramatic decline in outstanding credit to the financial and private sectors, which has decreased from over 14 percent of GDP at end-1986, to less than 3 at mid-1990. This important shift is the reflection of a decreased financial intermediation role performed by the Central Bank in the past few years, Lnd of the strict and determined stance of monetary policy since 1988, in the pursuit of economic stabilization. Net credit to the nonfinancial public sector has also slightly decreased in the course of the last two years, as a consequence of the improved fiscal performance. TABLE 1 . Buoo Central De*I Eeader - Belance Sheet 1/ (Percent of GDP) Dec 198 Dec 1987 Dec 1988 Jun 1969 Dec 1989 Jun 1990 A. Net International reserves -0.80 -1.80 -2.80 -1.88 2.88 2.11 B. Domestic Credit 16.61 18.88 9.28 7.56 5.91 4.94 Credit to the Public Sector 1.58 8.42 8.82 2.78 2.09 1.87 Credit to the Financial System 12.01 9.84 4.84 8.78 8.04 2.61 Credit to the Private Sector 2.07 1.12 0.62 1.00 0.79 0.66 C. Other Net Unclassified Assets 41.16 50.38 66.16 48.77 48.27 44.80 C. Medium and I.t. External Debt 88.72 41.87 44.44 89.26 89.87 87.76 D. Non-monetary liabilities 12.86 18.09 11.71 9.88 9.05 8.49 Deposits of Public Sector 8.64 7.74 6.92 8.79 6.81 6.38 Deposits of Financial Sector 0.98 1.21 0.90 1.00 0.70 0.81 Bonos de Estabilixaci6n 0.38 0.23 0.46 0.98 0.89 0.78 Liabilities to priv. Sector 2.36 3.92 3.44 1.18 0.88 0.52 E. Monetary Base 6.88 7.49 6.99 6.85 5.79 6.10 Currency in Circulation 4.31 4.47 4.27 8.88 3.48 3.15 Reserves 2.68 8.02 2.72 2.52 2.80 1.95 Total Assets 65.96 62.46 63.14 68.00 54.51 51.35 Total Liabilities 55.98 62.46 83.14 6.00 54.51 51.35 Source: Banco Central del Ecuador; World Bank 1/ Evaluated at the intervention exchange rate of the corresponding date. - 12 - 25. The monetary base (defined as currency in circulation plus monetary deposits of financial institutions) has considerably fallen in the course of the past two years, from a peak of 7.5 percent of GDP at end-1987 to 5.1 percent of GDP in mid-1990. The fall is a reflection of the changes in the composition of financial assets of the public, brought about by higher inflation (which increases the opportunity cost of holding unremunerated demand deposits) and the beginning of a process of financial deepening, with the emergence of a number of new financial instruments (polizas de acumulaci6n , etc.). Nominal Quei-Fiscal Deficit in 1989 and 1990 26. Table 2 displays estimates of the net income accounts for 1989 and a projection for 1990, based on the monetary program of the Central Bank for the year, as well as on other assumptions on macroeconomic variables (inflation, exchange and interest rates) as contained in the Government's economic program. The quasi-fiscal deficit of the central bank amounted in 1989 to 8/.71.6 billion, or 1.4 percent of GDP. By comparison, the deficit of the consolidated nonfinancial public sector during the same period was 2.2 percent of GDP. For 1990, it is estimated that (based on the official assumptions of the government economic program) the deficit would amount to S/.163.7 billion, or 2.2 percent of GDP. Again, the targeted deficit for the consolidated nonfinancial public sector for the same period is 1.8 percent of GDP. 27. The large deficits for both years are accounted for by large losses on account of external operations and high operating costs, wbich are not sufficiently matched by net income on domestic operations. Tnterest received on domestic credit amounts to about 0.9 percent of GDP both in 1989 and 1990. In ex-post terms, this implies an average interest rate of 14.7 percent in 1989 and 17.8 percent in 1990. Credit to the private sector (direct and through the financial system) bore a somewhat higher interest rate (16 and 21.3 percent, respectively), which was nonetheless substantially negative in real terms in both years. On the liability side, the BCE paid out approximately S/17.7 billion in 1989 and is projected to pay 29.6 billion in 1990, chiefly on account of the service of stabilization bonds. Net interest - 13 - income is thus positive in both years, for 0.6 and 0.5 percent of GDP, respectively. Table 2 - Banco Central del Ecuador - Quasi-fiscal Result 1989 1990 Interest Interest Average Average Million Percent of Average Average Million Percent of amount Int. rate Sucres GDP amount Int. rate Sueres GOP (A) Domestic Result Interest-bearing assets 819,741 14.9 47,557 0.91 886,271 17.8 65,489 0.88 Credit to Private Sector 201,544 16.3 80,795 0.69 247,200 21.8 52,007 0.71 Credit to Public Sector 118,198 14.3 16,761 0.82 121,084 10.6 12,882 0.17 Interest-bearing liabilities 61,276 29.0 17,780 Stabilization Bonds 44,028 86.7 15,729 0.80 74,001 40.0 29,600 0.40 Other obligations 17,248 11.8 2,081 Net Interest Income 29,797 0.67 86,888 0.48 (9) External Result -80,932 -1.54 . -186,285 -1.85 Interest on Intern. Reserves 11,448 0.22 29,847 0.40 Interest due on External Debt 144,481 2.76 227,788 8.05 Net External Interest -182,988 -2.64 -197,985 -2.65 Other Income 52,051 0.99 59,700 0.80 Transaction Tax 83,879 0.64 85,000 0.47 Exchange Risk 18,672 0.80. 24,700 0.88 (C) Operating Costs and Other -20,484 -0.89 -25,420 -0.84 Operating Costs 80,665 0.56 48,420 0.58 Other Income 10,280 0.20 18,000 0.24 (0) Quasi-fiscel Result (A*B+C) -71,570 -1.87 -168,656 -2.19 Source: Banco Central Del Ecuador, World Bank 28. Large losses (at least on paper) are instead registered on account of external operations of the BCE. Interest on external reserves provided 0.2 percent of GDP in 1989, and is expected to provide 0.4 percent of GDP in 1990. Supplementing these revenues are the receipts foru the exchange transaction tax, which was particularly high in 1989, and from the exchange risk commission on the sucretized debt. These two sources provided 1 percent of GDP in 1989 and are expected to provide 0.8 percent of GDP in 1990. Against these revenues, however, stands the large interest bill due on the medium- and long-term debt in the books of the Central Bank, which.is estimated at 2.8 and 3.1 percent, respectively. Thus external operations provided an extremely negative influence on the accounts of the Central Bank. - 14 - 29. Finally, operating costs net of other income (from sale of books, commissions, fees, etc.) are estimated to add almost 0.4 percent of GDP in both years to the financing requirements of the BCE. The sum of these elements results in the figures discussed above. These figures differ from the accounts of the BCE (which are seldom published), as those reflect only external interest payments actually made, rather than due, thus incorporating a financing item (arrears) above, rather than below the line.7 Quasi-fiscal Results Decomposition of Subsidies 30. Table 3 provides, for 1989, a decomposition of the quasi-fiscal result into the subsidy and non-subsidy components. The domestic result from table 2 is decomposed into a "potential net income" and the net subsidy from the nonfinancial public sector and the private/financial sector. Potential income, as discussed in part B, is defined as the income that would result from charging market interest rated on all domestic assets, and remunerating all domestic non-monetary liabilities at the same rates. The subsidy from each broad economic agent (NFPS and private/financial sector) is then calculated as the sum of the interest differential on outstanding assets and liabilities. 31. The table reveals interesting facts. The potential income from domestic operations would be negative in absence of subsidies: as non-monetary liabilities towards the nonfinancial public sector are substantially greater than assets, the losses that would accrue on this account from remunerating NFPS deposits at market rates would, together with the service burden provided by stabilization bonds, far outweigh the positive gains from credit to the private/financial sector. In 1989, these factors would have resulted in a deficit of 0.3 percent of GDP on account of domestic operations.8 The fact that public sector deposits are not remunerated, however, and that there are 7/ No "regalias" were haded out in either 1989 or 1990, in contrast to previous years, when non-reimbursable grants had been used at times liberally. 8/ In Table 3 the interest received on account of sucretization bonds is included in part B, external result, rather than part A, as in table 2. - 15 - ample non-remunerated liabilities of the private/financial sector, resulted in the Central Bank receiving a subsidy of 0.8 percent of GDP (0.6 percent from the non-financial public sector, and 0.2 percent from the private/financial sectors). As a consequence, the domestic result is positive (0.5 percent of GDP) in 1989. The latter fact is particularly surprising, given the common wisdom that sees the central bank as a provider of subsidies to the private sector. It must be remembered, however, that 1989 represented an exceptional year in that it saw the net credit to the private sector decline to its lowest level (table 1). Similar calculations for earlier years would have probably shown a rather different picture, as credit was large, and interest rates extremely subsidized. Table 8. Decomposition of the BCE quasi-fiscal deficit by subsidy, 1989 Sucree Percent of Million GDP A. Domestic Component Potentia Income from Net Credit to Public Sector -19345 -0.87 Net Credit to Financial and Private Sectors 17797 0.84 Stabilization Bonds -15729 -0.30 Tota I -17277 -0.88 Net Subsidy to 1/ Nonfinancial Public Sector -28803 -0.56 Financial and Private Sectors -10879 -0.21 Equals: Domestic Result 22405 0.48 B. External Result Subsidy to Public Sector 1/: 82176 1.57 Net interest on external reserves 11448 0.22 Interest on Public Debt 98825 1.79 Subsidy to Private Sector 1/: -10708 -0.20 Interest due on Sucretized debt 50807 0.97 Payments from Private sector 61510 1.17 Interest on sucretized bonds 9459 0.18 Exchange risk commission 18672 0.86 Transaction surtax 38379 0.64 Total external result -71473 -1.86 C. Operating costs and other Operating Costs -3086 -0.68 Other Revenues 10281 0.20 QUASI-FISCAL RESULT -69508 -1.33 Source: World Bank 1/ Negative sign means tax on recipient. - 16 - 32. Part B. of the table also shows that a large subsidy is provided to the non-financial public sector on account of external interest obligations. While the Central Bank earned 0.2 percent on its external reserves, it had (accrued) payments on public debt amounting to 1.8 percent of GDP, for which the Treasury provided no resources. In addition, the revenues accruing to the Central Bank from the service of the sucretization bonds (the interest on the bonds and the exchange risk commission) fell short of the (due) interest, implying a subsidy to the private sector of about 0.5 percent of GDP. Offsetting this, however, was a special foreign exchange transaction tax, which resulted in the private sector contributing a net subsidy of 0.2 percent to the Central Bank on account of external operations.9 33. This decomposition thus shows that the quasi-fiscal deficit of the Central Bank of 1.4 percent of GDP for 1989 was the result of numerous and contradictory cross-subsidiess the nonfinancial public sector was discharged of the obligation of servicing an important portion of the external debt, but at the same time was compelled to maintain large and unremunerated deposits with the central bank. The private sector received subsidies on account of the sucretization of the debt, and of less-than-market rate credit lines, but was also compelled to maintain unremunerated deposits and pay a special transaction tax. The end result is a large loss for the Central Bank, whose monetary consequences could only be avoided through the accumulation of arrears on external interest payment. 34. In addition, it should be noted that the coexistence of positive and negative subsidies implies that the relationships between the Central Bank and the private/financial sector becomes erratic and arbitrary, as the agents that receive positive subsidies are not necessarily the same who are compelled to maintain unremunerated deposits. This naturally increases the perception of the system as unfair and lacking in transparency. 9/ In practice, virtually no interest was paid out by the Central Bank on the sucretized debt, so that, in cash terms, the subsidy from the private/financial sector was considerably larger. - 17 - Deficit Estimatest Real 35. Table 4 displays estimates of the real quasi-fiscal result of the Central Bank of Ecuador. As discussed in part I, estimates of the real quasi- fiscal result correct the nominal quasi-fiscal result for the fact that, when inflation is non-zero, part of the interest rate can be interpreted as accelerated repayment of principal. If the interest rate is negative in real terms (or if it is set to zero, as in the case of unremunerated deposits, reserves, and currency issue), then the nominal interest payment results Table 4 - Banco Central Del Ecuador - Real Quasi-Fiscal Result Flows Stocks Million Percent Sucres of GDP (A) Domestic Operations Nominal Result 1/ 29797.0 0.57 Adjustment for Inflation: Beginning-of-period Interest-bearing assets 249098.0 Inflation lose 185257.5 2.56 Interest-bearing liabilities 4172.0 Inflation Gain 2265.4 0.04 Net Inflation loss 182992.1 2.64 Non-monetary, non-interest bearing liabi ities 230758.0 Inflation gain 126801.0 2.89 Real domestic quasi-fiscal result 2/ 22106.5 0.42 Monetary Bases 184417.0 Inflation tax 72918.4 1.89 Real domest ic quasi -fi sca l resu l 8/ 95095.0 1.81 (8) External Result 1/ -60982.4 -1.54 (C) Operating Costs and other 1/ -20484.4 -0.89 (D) Real quasi-fiscal result 2/ -79260.2 -1.51 (D') Real quasi-fiscal result 2/ -6271.8 -0.12 Source: BCE, World Bank 1/ From Table 2. ?/ Excluding inflation tax 8/ Including inflation tax - 18 - in a capital loss or gain for the Central Bank, depending on whether it is applied to an asset or a liability. Accounting for all of these losses and gains allows to get a more accurate assessment of the change in the new worth of the Central Bank than simple nominal changes.10 36. The data for 1989 show that there is no marked difference between nominal and real result, provided that one does not take into account the inflation tax. Non-monetary domestic assets and liabilities, at the beginning of 1989, were in fact roughly equivalent, so that, while the Central Bank suffered a capital loss on its outstanding assets equivalent to 2.6 percent of GDP on account of inflation, it also registered a gain on its outstanding non- monetary liabilities equivalent to 2.4 percent of GDP. On this account, thus, the nominal result is worsened only by a modest 0.2 percent of GDP. 37. As argued, however, the Central Bank, in its capacity of issuer of currency and given its power of setting reserve requirements, has an increased source of revenue at its disposal, the inflation tax, i.e. the loss in real value of base money. This provides a substantial gain: in 1989, the real value of monetary liabilities of the central bank was reduced by 1.4 percent of GDP11. While it is arguable whether the inflation tax should be considered a revenue or a form of financing of the Central Bank deficit, the table shows that the real deficit would have been substantially reduced (to almost zero) if the tax were considered a revenue. 101 The Central Bank could in fact be registering a positive result, if nominal revenues exceed nominal outlays, at the same time that the value of net worth, deflated by some price index, might be declining rather sharply. It should also be noted that the nonfinancial public sector accounts in Ecuador do not make any allowance for a similar real correction to interest rate flows to and from the public sector. 11/ It should be noted that there is a difference between the inflation tax that is levied by the Central Bank (the concept used here), and the inflation tax that can actually be collected through monetary emission. The public is in fact free to reduce its real holdings of money if it chooses to do so. - 19 - D. CONCLUSIONS 38. The preceding discussion shows that the quasi-fiscal deficits already represent a sizable additional burden on the nation's savings. The Central Bank of Ecuador, beyond its monetary policy functions, has had, and continues to have, an important fiscal role, and is at the center of a complex maze of subsidies to and from the nonfinancial public sector and the private/financial sector. Estimates for 1989 and projections for 1990 show that the deficits have been substantial, of the same magnitude as the deficits of the consolidated public sector. Their macroeconomic consequences have been lessened by the fact that part of the payments on external debt have not been made; nonetheless, these figures reveal the existence of a serious problem. 39. In principle, the Central Bank of Ecuador need not run any deficit, if its only function were that of regulating monetary aggregates. Reducing the deficits will thus require that the function thae it is currently performing that are extraneous to its role be discontinued. There is no reason why the Central Bank should continue to play a predominant, or even marked, role in financial intermediation. Economic efficiency is enhanced if financial resources can flow towards the investments with better returns, rather than being administratively directed towards arbitrarily determined priority subsectors. In the context of the progressive liberalization of interest rates and the development of more modern financial institutions in Ecuador in the recent years, it appears appropriate that the Central Bank proceed to eliminate its direct intermediation role. This would imply that directed credit lines to the pricate sector should be discontinued, and that the financial surplus generated by other public institutions would find its uses through the intermediation of teh financial markets. Alternatively deposits of public institutions would be remunerated, if it were necessary to use them as a means of monetary policy. 40. The implementation of these recommendations might lead to greater Central Bank deficits; but these should be recognized as part of the overall public sector deficit. The Central Government should provide for a sufficiently large operational surplus in the accounts of the nonfinancial - 20 - public sector, so as to compensate for these losses. In addition, the servicing of the public external debt, should also be transferred to the Central Government, or sufficient resources should be provided for its servicing to the Central Bank. - 21 - References Barbone, Luca and Paul Beckerman (1989): 'Inflation, Monetary Policy and Quasi-Fiscal Deficits: A Simple Model, with an Application to Argentina", Unpublished Paper, March. Piekartz, Julio (1987)2 "El Ddficit Cuasifiscal del Banco Central*, paper presented at the Seminar on Central Bank quasi-fiscal operations, Brazilia, August. World Bank (1989)t Argentina? Reforms for Price Stability and Growth, Washington D.C., April