General equilibrium effects of investment incentives in Mexico

l wes O924 Policy Research WORKING PAPERS Public Economics Country Economics Department The World Bank June 1992 WPS 927 General Equilibrium Effects of Investment Incentives in Mexico Andrew Feltenstein and Anwar Shah In Mexico, reducing corporate taxes stimulates investment more than increasing the investment tax credit or the employment tax credit does. PolicyResearch WovingPapersdsseatethefindgs ofwork inprogres and encouragetheexchange ofideasaongBankstaffand all others interested in developmnent issues. These papers, distributed by theResearch Advisory Staff, carry the names of the authors, reflect onlytheirviews, andshouldbeused and cited accordingly.Thefindings,interpretations,andconclusionsaretheauthors'own.Theyshould not be atttibu-ed to the World Bank, its Board of Directors, its managenent, or any of its mernber countries.  Policy Research Public Economics WPS 927 This paper - a product of the Public Economics Division, Country Economics Department - is one of a series of discussion papers prepared for the research project "An Evaluation of Tax Incentives for Industrial and Technological Development" (RPO 675-10), funded by the Bank's Research Support Budget. Copies of this paper are available free from the World Bank, 1818 H Street NW, Washington DC 20433. Please contact Carlina Jones, room N 10-063, extension 37699 (June 1992,44 pages). Mexico has experimented with several tax investment tax credits in Mexico. Mexico had instruments designed to promote private capital high inflation and high nominal interest rates, formation. Among such initiatives were general with real interest rates negative for certain years and industry-specific tax credits, employment tax - so firms faced severe financing constraints. In credits, and corporate tax reductions. such a macroeconomic climate, firms see re- duced tax rates as improving their cash flow and Feltenstein and Shah examine the relative a signal of an improved public policy climate. efficacy of such instruments using a dynamic computable general equilibrium model. They In a period of economic uncertainty and carry out model simulations using three equal- decline, nonrefundable, unindexed tax credits on yield investment incentive scenarios: increases in new investments are less valuable than an investment tax credits, increases in employment immediate reduction in tax liability from both tax credits, and an equivalent reduction in the old and new capital. corporate tax rate. Finally, in an open economy, reducing the Of the three, they find that reducing corpo- tax rate increases the demand for all capital rate taxes is most effective at stimulating invest- rather than new capital alone - so the relative ment in Mexico. value of domestic capital rises. Accordingly, the public increases its holdings of domestic debt, Various explanations are plausible for why causing the price of domestic bonds to rise and reducing tax rates is superior to providing real interest rates to fall, stimulating investment. The Policy Research Working Paper Series disseminates the findings of workunder way in theBank. An objectiveof the series is to get these findings out quickly, even if presentations are less than fully polished. The findings, interpretations, and conclusions in these papers do not necessarily represent official Bank policy. Produced by the Policy Research Dissemination Center  General Equilibrium Effects of Investment Incentives in Mexico by Andrew Feltenstein and Anwar Shah Table of Contents 1. Introduction 1 2. Tax Incentives for Investmnent in Mexico 1 3. Model Specification 2 a. Production 3 b. Consunption 7 c. Transfer Payments and Govermment Financing 14 d. The Foreign Sector and Exchange Rate Determination 15 4. Simulation Results 17 a. Calibration 17 b. Counterfactual Simulations 22 Suninwaay and Condusion 28 Bibliography .9 Appendix Corporate Structure and Investment Incentives in Mexico 30  OENERAL EQUILIBRIUM EFFECTS OF INVESTMENT INCENTIVES IN MEXICO Andrew Feltenstein and Anwar Shah* 1. Introduction Public policy officials in Mexico have, over the past several decades, experimented with a number of tax instruments designed to promote private capital formation. Among such initiatives were general and industry specific tax credits, employment tax credits, and corporate rate reductions. This paper examines the relative efficacy of such tax instruments using a dynamic computable general equilibrium framework. The paper is organized as follows. Section 2 presents an outline of the tax policy environment for the corporate sector in Mexico. Section 3 presents model details. Section 4 highlights alternate tax incentives regimes and model simulation results. Finally, a concluding section provides a summary of the results. 2. Tax Incentives for Investment in Mexiso Tax incentive regimes in Mexico have undergone significant changes over time. These are briefly discussed below: 1955-1972: Between 20% (for secondary industries) and 40% (for basic industries) corporate income of Mexican majority owned enterprises was exempted from corporate taxation for periods varying between five to ten years. The same industries also could receive, upon application, exemption from certain indirect taxes and import duties on capital goods imports. 1972-1979: Industries that were seen to promote decentralization and regional development were granted import duty relief varying from 50% to 100% and reduction in corporate tax liability ranging from 10% to 40% depending upon their location and type of activity. 1979-1986s The practice of import duty exemption was continued. In addition, tax incentives certificates (CEPROFIS) providing tax credit in the * This is one of a series of discussion papers prepared for the World Bank research project, "An Evaluation of Tax Incentives For Industrial and Technological Development". The project is directed by Anwar Shah of the Public Economics Division, Country Economics Department. The authors are grateful to Daniel Oks for helpful comments.  -2- range of 10-25%, depending upon location, and type and size of the industry, for investment in physical assets were introduced. These certificates were negotiable and could be used against any federal tax liability by the holder. 1986-Present: The tax incentive certificate scheme was significantly tightened and targeted to priority industries and preferred zones (See Appendix Table Al). The top tax credit rate for CEPROFI was raised to 40% of total physical investment in 1986. In addition Mexican-owned enterprises are eligible for employment tax credits up to 30% of three times the annual area minimum wage multiplied by the number of new jobs created. In addition, full expensing of the present value of capital consumption allowance& calculated using a 7.5% discount rate was allowed in non-metropolitan areas. In the metropolitan industrialized areas of Mexico City, D.F., Monterrey and Guadalajara, only 60% of the present value of depreciation allowances could be deducted in the first year. R&D investment tax credit at 15% for the purchase of technological research (20% for small and micro enterprises), and 20% for capital purchases by technological enterprises (30% for small and micro enterprises) are currently permissible. Further details regarding the corporate income taxation and foregone revenues due to tax incentives in Mexico is given in Appendix A. 3. Model Specification In this section we will develop the model we will use to analyze a variety of fiscal issues in Mexico. In particular, the model will be designed to look at the implications for revenues, sectoral investment, and the balance of payments of a number of different tax programs. We will consider investment tax credits, and employment tax credits. The model can be easily extended to incorporate accelerated depreciation allowances, tax holidays, and immediate full expensing. Our model will also permit experimentation with changes in the structure of indirect taxation as well as the personal income tax. The model we develop is intended to be a microeconomic optimizing structure that generates macroeconomic outputs. Since our aim is empirical implementation, much of the structure we incorporate is chosen because of the availability of data.  - 3 - We use a two period general equilibrium system in which all agent. have perfect foresight, and hence in period 1 correctly anticipate the price. of period 2. We need to specify the behavior of production, consumption, and government output, taxation, and deficit financing. We need also specify the exchange rate regime and the characteristics of the trade system. A oolution is found for both periods simultaneously, so that we will be determining outcomes for both years, and hence corresponding rates of change. a. Production There are 8 factors of production and 3 types of financial sA-et The.e are: 1-5. Capital types 9. Foreign bonds 6. Urban labor 10. Rural labor 7. Money 11. Land 8. Domestic bonds The five types of capital correspond to the five productive sectors, which do not include agriculture, that we will describe shortly. Each of these factors and financial assets is replicated in each period, so that we have, for example, period 1 capital and period 2 capital. Period 1 money will be the numeraire. Thus the model has 22 dimensions, or prices. An input-output matrix is used to determine intermediate and final production. This matrix is replicated in each of two years. Corresponding to each sector in the input-output matrix, value added is produced using capital and urban labor for the non-agricultural sectors, and land and rural labor in agriculture. The technology that produces this value added is sector-specific.1 Our data source for the input-output matrix is Matriz de Insumo-Prod'4cto Anno 198 (1988). Here a 72 sector matrix is derived which represents Mexico's technology for 1980. We have not attempted to update the matrix for the years which we will be analyzing. Since it is not our intention to work at this leve. of sectoral disaggregation, we have aggregated the technology to seven sectors 1The use of neo-classical value added functions "sitting above" an input- output matrix is common. The reader may wish to see Shoven and Whalley (1984) for articles that use this approach. An application and detailed description of functional forms is given in Feltenstein (1986).  - 4 - by adding corresponding rcvs and columns. The resulting sectors and the corresponding sectoro in the initial matrix are: Table 1. agaregate Input-Output Sectors Aggregate Sector Correspondino Disaaareaated Sectors 1. Agriculture 1-4 2. Manufacturing 5,7-61 3. Petroleum 6 4. Commerce 62-63 5. Transportation 64 6. Communications and services 65-72 7. Imports We denote the resulting input-output matrix by A.2 The specific formulation of the firm's problem is as follows. Lot yjKi, yIi be the inputs of capital and urban labor to the jth non-agricultural sector in period i. Let YGi be the outstanding stock of government infrastructure in period i. The production of value added is then given by va - vajj(Yii#YLi (1) Recall that capital is sector specific and there are two types of labor. In the case of agriculture, equation (1) takes the same form, except that land is substituted for capital and rural labor is substituted for urban labor. We are supposing that there is a single type of infrastructure, although extensions to sector specific infrastructure would present no problem. Infrastructure may be thought of, for example, as roads, communications, education, and so forth, and enters private production as an increase in productivity. It is assumed that sector j cost-minimizes with respect to capital and urban labor, in the case of a non-agricultural sector, and with respect to land and rural labor in the case of agriculture. Sector j pays value added taxes on inputs of capital and labor, given by tiJi, tjki, respectively, in period i. We 2A program that permits the user to arbitrarily aggregate particular rows and columns is available upon request from the author.  - 5 - assume that there are no taxes paid upon the use of land by agriculture, although agriculture is taxed on its use of labor.3 We will also suppose that the sector may be given an employment tax credit. This credit is given by a percentage rebate on the value of the firm's wage bill. Hence the effective price for labor paid by sector j is: pLii - (I + tLij - aij) PLi where aU is the employment tax credit given to sector j. Similarly, the effective price of capital for sector j is: PKij - (1 + tKij) PKij Thus if PK4j and PLii are the prices of capital and labor in period i, then the prices charged by enterprises, Pi, are given by (Pi} - va(P,YGi)(l + t)(I - A) , (2) where va(P,YGi) is the vector of cost-minimizing value-added per unit of output, subject to P - {PKjj, PUj) and Y0i, and t - {tKi, tLi)- Here we treat imports as a single product that is distinct from domestic production.4 Thus there is no value added by factors in imports. Rather, imports require foreign exchange, which is, in turn, produced by exports. We suppose that each type of sectoral capital is produced via a sector- specific investment technology that uses inputs of capital and labor to produce new capital. Investment is carried out by the private sector, and since the capital that is produced in one period becomes available only in the next period, the investor must pay for the input cost of its production in the current period, but will receive the revenue from that capital in the next 3The interpretation of these taxe is thus as a profit tax and a personal income tax that is withheld at the source. 4This assumption, due to Armington (1969), permits us to avoid problems of corner solutions, tha. is, solutions in which a good is either entirely domestically produced or entirely imported.  period. We will assume that investment i. entirely financed by domestic borrowing, so that the investor sells domestic bonds to pay his factors of produec- ton.S Accordingly, the investor equates the coot of borrowing, given by the interest rate, with the anticipated future returns on capital. The investor is affected by several fLocal parameters in maklng his decision. He receives an investment tax credit an well as a depreciation allowance. He also pays a capital, or profit tax, on the returns to hi. investment. Let us define the followlng notation. kim Investment tax credit in period 1 (percent). dL- Depreciation allowance in period L (percent).6 tki Profit (capital) tax rate (percent) CHi- The cost of producing the quantity Hi of capital in period i ri- The interest rate in period i. PrK, The return to capital in period i. Pmi- The price of money in period i. Suppose, then, that the rental price of capital in period L+1 is Py+1. If CHi is the cost-minimizing cost of producing the quantity of capital, Hi, then future debt obligations must be equal to the return on new capital. Hence: (1 - tk)pS H1 CHi(l - ki- dL) + r (3) where ri is the interest rate in period i, glven bys SWe assume that all foreign borrowing for lnvestment is carrled out by the government, so that, implicitly, the government is borrowlng for the private investor but the debt thereby incurred is publicly guaranteed. In terms of Mexico, this may be viewed as the situation existing after the financial collapse. 6Thie may be Lnterpreted as an accelerated depreclation allowance, since the firm is permitted to take the allowance in the current period, although the capital does not come on line until the next period.  - 7 - ri - I/PBi (4) where PBi is the price of a bond in period i.7 Thuo all mectors in the economy pay both income and profit taxes to the government, while certain sectors, in particular agriculture, may receive subsidies. Theme taxes are collected by the central government which uses them to finance its own expenditure activities. The government produces public goods using capital and labor am inputs to production. Theme good. are divided between thome used for development, repremented by capital expenditure., and thome which are represented by current expenditure, and which have no direct impact on privato output.8 The government'. target for the output of public goods im determined exogenously in each time period as a fraction of GDP. An attempt to model an optimizing government is thus not made. b. Conoumption There are two types of conmumers, representing rural and urban labor. We suppose that both consumer classes have the same demand pattern. for goods, and that their demands for the seven different types of good. are given by constant fractions of their incomes.9 Thuu urban and rural consumers differ only in terms of their initial wealth. The consumers maximize intertemporal utility functions, which have am arguments the levels of consumption and leisure in each of the two periods. We permit rural-urban migration in that rural workers can choose to become urban labor if the relative wage is favorable. The consumers maximize theme utllity functions subject to intertemporal budget constraints. The consumer 7This formulation of the investment tax credit is adapted form Auerbach and Hines (1988). 8Current spending may, via its impact on wages, the availability of capital, and the interest rate indirectly have very considerable impact upon private output. Feltenstein and Morris (1990) and Shah (1992) examine the impact on private output of spending on public infrastructure. 9The assumption of equal relative spending on different goods by both urban and rural consumers is probably inaccurate. There is, however, insufficient data, for us to estimate individual demand functions.  -8- saves by holding money, domestic, bonds, and possibly foreign currency. He requires money for transactions purposes, but his demand for money is sensitive to changes in the interest rate. The consumer receives income from his labor, from the rental on any capital or land that he owns, and from the interest payments on bonds that he has purchased. He may also receive direct transfer payments from the government. He pays sales taxes on the goods he consumes, as well as tariffs on imported goods. The consumer's bond holdin7a are also subject to a capital loss if the domestic interest rate falls. His maximization problem is thust max U(x) x - (xl,xLl,x2,xL2) (5) such thats (l+ti) Pixi+PLuixLui+PLrixLri+PMixMi+P;lixBi+SiPBFixBFi (5a) - PKi (1-6) 'K+PAiAO+PLuiuai + PLriLri+PMiXM(i_l)+r(i_l)XB(i-l) +PBixB(i-l) +eiPBFiXBF(i-lj)+TRi log PMixMi - a + b log (l+ti)Pixi - c log ri (5b) log PBixBi - log eiPBFixBFi - a + B (log ri - log eirFi) (SC) log (Lui/Lr) - a, + a2log {PLui - PLri}/'PLui + PLril (Sd) if PLui 2 PLri; otherwise log (Lui/Lri) = 0 (if the representative household i8 rural, otherwise labor holdings are constant) PB2xB2 = (1+t2)P2x2 (Se) where: Pi - price vector of consumption goods in period i. Xi - vector of consumption in period i. ti - vector of aales tax rates in period i. PLJd - price of urban labor in period i. Lui - holding of urban labor in perlod i. PLZi prlce of rural labor ln period i.  -9- -,i - holding of rural labor in period i. &2 - elasticity of rural/urban migration. Pyj - price of capital in period i. K - initial holding of capital. 6 - rate of depreciation of capital. xLi - consumption of leisure in period i. PMi - price of money in period i. Money in period 1 is the numeraire and hence has a price of 1. A decline in the relative price of money from one period to the next represents Inflation. =Mi - holdings of money in period i. PBi - discount price of a domestic bond in period i. ri - domestic interest rate in period i. xBi - quantity of domestic bonds purchased in period L.