Modeling the macroeconomic requirements of policy reforms

Policy, R.eah, and Extenal Affairs 1 WORKING PAPERS Macroeconomic Adjustment and Growth Country Economics Department The World Bank May 1 990 WPS 417 Modeling the Macroeconomic Requirements of Policy Reforms William Easterly, E. C. Hwa, Piyabha Kongsamut, and Jan Zizek The macroeconomic requiremnents of policy reforms can be de- termined through an extension of the Bank's RMSM model to include fiscal and monetary variables and behavioral functions. MMc PmMq. Runn. u~Bmum &MW . CminpIm do*MOM MEB Wa P.Pin to &M~M 69 bW of we& a U NW wq _=w *a e apu d_ E Wm on_ BasNW __ l *F eema Wrp "*W_ db ** s ibdlm a, sm w andmd be uind mad nad a -l. Meindiy bd. nkamn, and inb.i - on wdeiWm Thep *Ahl mm be .mm~bes to dhe Wan Bki Dan San d of D~esm Au mqmm. um prm (nmmwc  Polky, Reseah. an.Ext.ma Affairs Macroeconomic Adjustment and Growth This paper- a product of the Macroeconomic Adjustment and Growth Division, Countrv Economics Department - is part of a larger effort in PRE to develop a macroeconomic projection model that will improve and extend the Bank's RIMSM model. Copies are available iree from the World Bank, 1818 H Street NW, Washington DC 2W33. Please contact Raquel Luz, room N 1 1-057, extension 34303 (82 pages with figures and tables). Easterly, Hwa, Kongsamut, and Zizek assess the detail on the tax and banking systems. They usc macroeconomic policy regime required for the extended model to discuss macre conomic structural economic reform in trade and financial adjustments needed for policy reforri, being policy. To do so, they develop a macroecon- considered in Colombia for trade (reduced tariff omic model for deriving the appropriate stance rates and relaxed import quotas), finance (re- on macroeconomic policies needed to support duced reserve requiremerits and forced invest- structural reform measures in trade and finance. ment requirements for the banking system), and reducing inflation. The macroeconomic projection model of Colombia they use is an extension of the Bank's AU the policy reforms require fiscal adjust- Revised Mlinimum Standard Model (RMSM) ment to be consistent with available financing. currently in use. They enrich the traditional For example, reduced tariff rates mean lost RMSM by adding fiscal and monetary identities revenues, which must be compensated by as well as behavioral functions for the following revenue increases elsewhere or spending cuEs. variables: private consumption, private invest- The incipient increase in import demand because ment, money demand, demand for quasi money, of reduced tariffs requires more currency depre- export supply, and import demand. ciation, although not as much as the increase in the tariff rate. Quota reduction also requires They illustrate how Lhe basic model func- some currency depreciation. tions by comparing three simulations to a base case run, increasing lhe targets for three vari- All scenarios require reeucing public ables: the real exchange rate, the real intcrest investunent if no other fiscal measures are taken. rate, and the inflation rate. Since reducing public investment lowers growth, these simulations dramatize the need to pursue All three simulations increase the Tinhance- fiscal reforms that compensate for the adverse able fiscal deficit, but in differcnt ways. A fiscal effects of trade or financial liberalization target of rcal exchange rate appreciation in- without reducing urgent social spending or creases it by making more foreign credit avail- investments in public infrastructure. able to the public sector. Higher targets for real interest and inflation rates increase it by making The model does not try to capture the more domestic credit available. favorable effects of reforrn on efficiency and growth, which other evidence suggests would be They then extend the model to take into large enough to raise growth in the long run account the external financing constraint facing even if there were an ill-advised reduction in the government and the economy, and by adding public investment. The PRE Working Paper Series disseminates the findings of work under way in the Bwk's Policy. Rnewch and External Affaurs Compeix. An objective of the series is to get these ftndings out quickly, even if preuentations xe aes than fully polisheL The findings, interpretations. and conclusions in these papers do not necessarily mepat official Bank policy. Produced at the PRE Dissemination Canter  TABLZ OF CONTENTS Page I. INTRODUCTION ...................................................... PART I: THE BASIC MODEL ................................................. 2 II. DESCRIPTION OF THE MODEL .. 2 A. Boring accounting ............................................. 2 B. Behavioral equations, projection rules, and other stories ......7 C. Residual variables: a tale of two closures .................... 11 D. The Economics of the fiscal closure ........................... 14 III. HOW THE MODEL WORKS: ILLUSTRATIVE CASES . . 20 A. Real exchange rate appreciation ............................... 21 B. Higher real interest rate ..................................... 27 C. Increase in inflation ......................................... 30 D. Conclusion .............................. 33 PART II: KXTENSIOr- OF THE MODEL ....................................... 42 IV. EXTERNAL FINANCING CONSTRAINT ..................................... 43 V. MACRO REQUIREMENTS OF POLICY REFORMS .............. ............ 44 A. Policy simulation .: low ring import tariffs .................. 48 B. Policy simulation II: relaxing import quotas .................. 51 C. Policy simulation III: lowering reserve and forced investment ratio .............................................. 53 D. Policy simulation IV: reducing inflation ...................... 56 VI. CONCLUSION ................, .. . ................. 59 APPENDICES: I: Technical Implementation of the Model ......................... 61 II: Simulations of the Private Sector Closure of RMSM-X ........... 73 BIBLIOGRAPHY .................8.................. 82 We are grateful for comeonts from Bela Balassa, John Holsen, Vittorio Corbo, Jaume Ventura, Peter Hiovic, Lorenzo Perez, Paulo Leme, Carmen Rinehart and Luis Valdivieso. Eduardo Wallentin provided valuable help at an early stage of the project.  Table l: Consistency Accounting Matrix ........................ 4 Table 2: Model Simulation: Fiscal Closure .3.................. 3 Table 3: Simulation: Differences from Base Casi Fiscal Closure - Exchange Rate Appreciation .24 Table 4i Simulation: Differences from Base Case Fiscal Closure - High Interest Rate .28 Table 5: Simulation: Difference: from Base Case Fiscal Closure - High In'1stion . Table 6i Bats Case with Exchangs Rate Adjustment. 45 Table 7: Base Case without Exchange Rate Adjustment .46 Table B. Import Tax Rate With Exchang, Rate Adjustment. 4 Table 9s Quota Case With Exchange Kate Adjustment .52 Table 10: Forced Investment Rate With Exchange Rate Adjustment ....... 54 Table 11: Simulation: Differences from Base Case - Lower Inflation Case ...............57 Estimation results for money and quasimoney demand, unconstrained .66 Estimation results for money and quasimoney demand, constrained ..67 Estimations of import and export volume .68 Estimations of import and export volume for extended model .............. 69 Variable List for Estimatis on. 71 Table A2.l: Model Simulation: Private Closure .78 Table A2.2t Simulations Differences from Base Case Private Closure, Exchange Rate Appreciation .79 Table A2.3: Simulation: Differences from Base Case Private Closure, High Interest Rate .80 Table A2.4t Simulation: Differences from Base Case Private Closure, High Inflation ..................... 81 Figure lat External Finar.cing of Public Deficit .34 Figure lbs Domestic Financing of Public Deficit .35 Figure Ics Total Financing of Public Deficit .36 Fligure 2: Public Savvngs ............... 38 Figure 3s Public Investment .......39 Figure 4i Differences from Base Case ................ . 41 Filgure A2s Differences from Base Case .77  I. INTRODUCTION 1. The Colombia macro model presented in this paper is part of a family of macromodels being prepared by the Macroeconomic Adjustment and Growth Division in collaboration with colleagues in Operations for use in projections and evaluation of adjustment programs. These models represent a cont inuum of economic and programming compiexity. The projection models known as RMSM-X are intended to be the successor to the 3ank's RMSM (Revised Minimum Standard Model) model, with the main improvement being the inclusion of fiscal and monetary identities in addition to the traditional external balance and saving-investment identities. These models embody simple projection rules and are solved recursively to obtain whatever are defined to be the residual variables. 2. At the next level of complexity, the models known as RMSM-XX will cover the same ground as RMSM-X, but will incorporate behavioral functions for the main macroeconomic variables. They will be estimated for countries where the data and structural characteristics permit. They will be solved simultaneously to obtain relative prices that clear goods markets and financial markets. Finally the highest level of complexity will be macro models knowv as MACOR that will be state-of-the-art macro-econometric models. 3. The model presented in this paper is in-between RMSM-X and RMSM-XX in that--unlike RMSM-X--it does incorporate behavioral functions for the main macroeconomic variables, namly private consumption, private investment, money deand, demand for quasi-money, export supply. and import demand. However, unlike 3MSM-XX but like RMSM-X, it is solved recursively for residual  -2- variables. The oser specifies 'target values' for the real exchangr rate, real interest rate, and the inflation rate. The model then determines t'.e public sector behavior consistent with attaining these targets. In an alternative solution, described in Appendix II, the mode'. also permits setting exogenously the public sector variables and determining the private sector behavior necessary to reach the policy targets. Finally, the model is generalized to take into account external financing constraints by making the real exchange rate endogenous. The model is used to derive the macro adjustment required by trade and financial liberalization. PART I: TEE BASIC MODEL LI. DESCRIPTION OF THE MODEL 4. The struLture of the model can be explained by looking at its three key ingredients: the accounting framework, the behavioral equations and projection rules for the variables, and the choice of residual variables to satisfy the accounting iden.ities. Some illustrative simulations will then be presented to show how the model works. A. Boring Accounting 5. The model uses the flow-of-funds accounting framework presented in Easterly (1989), which is consistent with the framework presented in RMSM-X examples such as Holsen (1389) and Schmidt-Hebbel et al. (1989). This framework reconciles historical data on income, expenditure, saving. investment, and financing flows of different actors in the economy. The  -3- accounting identities are then also used as part of the projection model to . ure that budget constraints are observed for each of the sectors. 6. In this application, we use the minimum number of sectors; the consol.dated public sector, the banking system,1 the nonfinancial --ivate sector, and the external sector. Other applications of RMSM-X may wish to consider breaking the public sector down into the budgetary and extra- budgetary accounts and distinguishing the banking system between the central bank and the rest (the latter will be done in Part II). In this application, the simplest possible sectoral breakdown seemed appropriate because additiona1 complexity will be introduced by specifying behavioral functions for the private sector. 7. The consistent macroeconomic accounts for 1988 are presented in the matrix in Table 1 on the next page (defin!.tions of the variables follow Table 1). A similar exercise was done for 1986 and 1981, but is not shown. The construction of this type of a matrix is detailed in Easterly (1989). Briefly, it reconciles current income and expenditure with capital expenditure and financing flows. The current income and expenditure is in the upper left- hand quarter of the matrix, with expenditure flows shown down and income flows shown across. The organizing principle of the matrix, and of the macro accourte in general, is that expenditure by one sector will be the income of another sector. In addition to the four rows and columns for the four sectors, there will be another row and column for the national accounts, which 1/ The 'banking system' is defined to include the following Colombian financial intermediaries: commercial banks, corPoraciones financieras (development finance corporations), corporaciones de ahorro v vivienda (savings and loan companies), and compaftias de financiamento commercial (finance companies).  1'~ ~ ~ ~ ~ ~ ~ ~~~~~~~~~~C-or O -IA liP7n ~q~W~Y a~3m? O%s I%..aT* II sittPI l 'a-ri J.l.A I I )14 3 44%) Ii~ ~ ~~~. -(Ps 161) I I I 1iCu.1 P u b I.,S ..l,,. I 5--0-- ---- --- ---- - - -------- --1 I - - -- - - --I - - - -- I- -- - - - - - - - - --- -- -- -- 14- 3 752t Ilfe 2 201111t. I" O"~~~~~~I I III I TI IvF A 0S-fI I I I I" -----I---------------I---- I -I------ ----- ----- --------- ----------