THE COST OF A GASOLINE TAX INCREASE TO THE WASHINGTON STATE ECONOMY: A GENERAL EQUILIBRIUM ANALYSIS

Transcription

1 THE COST OF A GASOLINE TAX INCREASE TO THE WASHINGTON STATE ECONOMY: A GENERAL EQUILIBRIUM ANALYSIS Mukund P. Upadhyaya Research Associate David W. Holland Professor of Economics Department of Agricultural Economics Washington State University INTRODUCTION As a preface to our paper, the reader should understand the issues that are involved with the public choice of maintaining or enhancing public infrastructure (education, highways, port facilities, etc.). The infrastructure generates a public good that is enjoyed by a wide range of consumers and firms across the economy; but at the same time, the taxes necessary to provide the infrastructure impose a cost on the economy. In the case of highway spending, one side of the issue is the benefits that result from improved highway infrastructure. The qualitative benefits of improved highway infrastructure include reduced congestion and pollution costs, and a more efficient flow of goods. The more efficient flow of goods, for example, may increase the competitive position of the state in other markets. In terms of a perceived congestion problem, for example, the general externalities (spillover) costs involve the costs that each person using the highways has on others (i.e., the increased journey to work time and the increased delivery time for goods). When the externalities are perceived to be large, there is general argument for market intervention in some form. Typically this takes the form of general or specific taxes that produce government expenditures on public capital that are expected to alleviate the problem. In Washington, revenues generated from the motor fuels tax are used by the state, cities, and counties for the construction and maintenance of highways and streets. The intent of the gasoline tax to make those who use the public roadways share in the cost of providing them, and fuel consumption is assumed to be an approximate indicator of the benefits received. It is clear that states face a balance between expenditures on public infrastructure and the negative economic effects of the taxes that finance the public spending. In the case of added transportation infrastructure, policy makers must address the trade-offs between increased transportation infrastructure (lower costs of transportation, reduced congestion costs, and other related externalities) versus the cost of increased gasoline taxes on gasoline producers and gasoline consumers. The latter is an important aspect since an increase in a specific tax can result in additional costs to some sectors as well as households which may lead to reallocative Upadhyaya, Mukund P. and David W. Holland. Northwest Journal of Business and Economics, Center for Economic and Business Research, College of Business and Economics, Western Washington University, 1996 Edition,

2 Northwest Journal of Business and Economics, economic behavior throughout the state s economy. For example, in the case of a gasoline tax increase, gasoline producers will have to adjust to a lower net of tax producer price while households and sectors using gasoline as an input to their consumption and production processes will have to adjust to a higher gross of tax consumer price. The purpose of this paper is to examine the economic effects of a change in the Washington state gasoline tax through simulation with a computable general equilibrium (CGE) model. While neither the qualitative nor quantitative benefits of increased highway spending are addressed, we can examine the important price, income, and inter-sectoral adjustments that would be expected to stem from an increase in the gasoline tax. It is to this side of the policy issue that we now turn. The gasoline tax is the fourth largest source of state revenue in Washington after the retail sales and business and occupation (B&O) taxes and state property tax. The motor fuel tax is imposed on each gallon of gasoline sold, distributed or used by distributors in the state. The purpose of the gasoline tax is to make users of public roadways pay their share in the cost of providing roadway services. The gasoline tax was first imposed in Washington in 1921 at a rate of 1 cent a gallon. According to the 18th amendment to the State Constitution in 1944, gasoline tax revenues must be used only to finance the construction and maintenance of highways and streets. In 1949, the gasoline tax rate was 6.5 cents a gallon. The tax rate was allowed to fluctuate between 12 cents and 16 cents per gallon from 1978 to The tax rate was increased to 22 cents in 1990 and to 23 cents a gallon in Table 1 shows gasoline tax collections, percentage changes in revenues, and the gasoline tax percentage of all state taxes from 1986 to Gasoline tax revenues increased by percent from 1986 to 1990 (not shown in the table). But, the relative share of gasoline tax revenue in the total state taxes declined from 7.55 percent in 1986 to 6.75 percent in In 1995, The Transportation Commission proposed to increase the gasoline tax by 12.5 cents a gallon. The existing state gasoline tax rate was 23 cents a gallon. The Commission s proposal would increase the tax to 35.5 cents a gallon. The tax increase was justified by the need to relieve gridlock on Puget Sound highways. TABLE 1 GASOLINE TAX COLLECTIONS, PERCENTAGE CHANGES, AND PERCENT OF ALL STATE TAXES, FISCAL YEARS Fiscal Collections % of All Year ($000) %Change State Taxes , , , , , Source: Tax Statistics 1990, Revenue Research Report, Department of Revenue, State of Washington, May 1991.

3 Northwest Journal of Business and Economics, While the proposed gasoline tax increase will generate increased tax revenues and increased spending on construction, it will also impose increased costs on Washington firms and households. The overall objective of this paper is to measure the economic cost of gasoline tax increase on output, income. and employment in the state economy using a Washington State Computable General Equilibrium model. While in the multi-sector tradition of input-output models, the CGE model departs from the demand driven assumptions of input-output analysis and presents the economy in terms of conventional neoclassical supply and demand relationships. The model assumes that the stock of private capital and public capital is fixed, but the factors of production (labor and private capital) are free to adjust across sectors in response to changes in economic opportunities. The analysis presented in this paper is best described as one of comparative statics where the economy is subjected to some shock (a tax increase), adjusts to new costs and opportunities generated by the shock to a new equilibrium, and the new equilibrium is compared to the benchmark equilibrium. General equilibrium tax policy analysis has a clear welfare implication only in revenue neutral scenarios, where government activity is held fixed in real terms, and the direct incidence of taxes is shifted from one set of economic factors to another. If government activity is allowed to change during equilibrium adjustment, as is the case with a gasoline tax increase, the difficulty of valuing different sizes of government precludes comparative welfare judgments (Ballard, Fullerton, Shoven, and Whalley, 1985). Any value that households and firms might place on the public goods and services generated by the gas tax is not accounted for in the model used for this paper. If it were possible to make estimates of the value placed on the public goods and services gained because of a gasoline tax increase, one could judge the welfare change under the increase. Suffice it to say that such a judgment is beyond the scope of this paper. Instead, we propose to present an analysis of the likely cost to the Washington economy of the tax increase as a standalone analysis. No welfare judgments are implied. PREVIOUS APPLICATIONS OF CGE MODELS TO TAXATION The first application of a general equilibrium model to tax policy was by Harberger (1962). Harberger studied the incidence and efficiency effects of the taxation of income from capital. Shoven and Whalley (1973) provided the existence proof of a general equilibrium in the presence of taxes. By 1984, Shoven and Whalley (1984) reviewed 18 studies using CGE models applied to tax and trade policy issues. In a well known recent study, Melo, Stanton, and Tarr (1989) used a 12-sector CGE model of the United States economy to examine the effects of a 25 percent import tax on imported crude oil and a 15 percent excise tax on petroleum products. They examined tax impacts on cost, employment, and welfare in the U.S. economy. Regional CGE models have been slower to develop because of a shortage of the necessary data at the regional level. Morgan, Mutti, and Partridge (1989) developed a model incorporating six regions in the U.S. to assess the long-run effects of state-local and federal tax policies on output

4 Northwest Journal of Business and Economics, and allocation of factors among sectors and regions. Fisher and Despotakis (1989) used a regional CGE model of California to study the impacts of a uniform tax on energy consumption (coal, gas, and electricity) and a severance tax on crude oil. Morgan, Mutti, and Rickman (1992) used a CGE model that included six regions of the U.S. and assessed regional growth and welfare in the regions with tax structures that imposed more of the burden on residents living outside the region. Waters (1994) developed a nine-sector CGE model for Oregon to examine the economic impact of the property tax reduction under Measure 5 on the Oregon State economy. Upadhyaya and Holland (1995) developed a Washington State CGE model to examine the impact of alternative tax policies on the Washington economy. THE CGE MODEL FOR WASHINGTON STATE The Washington State CGE model assumes optimizing behavior for all economic agents (consumers and producers) in the economy and traces the impact of gasoline tax changes through effects on prices, sales, employment, income, distribution of income, and tax revenues. The model simulates market economy in which quantities and prices adjust to clear product and factor markets in response to the increase in gasoline tax. There is general equilibrium feedback in production, income, and demand. The model represents the Washington economy in 12 sectors: Agriculture, Forestry, and Mining; Construction; Non-durable Manufacturing; Durable Manufacturing; Aerospace; Transportation, Communication, and Utilities; Wholesale Trade; Retail Trade; Finance, Insurance, and Real Estate; Business Service; Health Service; and Other Service. The Schematic view of the Washington CGE model can be seen in Figure 1. Value is added by inputs of primary factors (labor, capital, and proprietor s services) with a constant returns-to-scale Cobb-Douglas production function and combined with intermediate inputs to produce output in each sector (Figure 1). The demand for primary factors is derived from the first-order profit maximizing conditions so that the return paid to the factor in each sector equals the value of marginal product produced by the factor. Intermediate demand is calculated from sectoral output, using fixed technical coefficients. Producers are assumed to maximize profits by an optimal allocation of output between the Washington market and exports using a constant elasticity of transformation (CET) aggregation function (Figure 1). Export prices are assumed fixed except for the Aerospace sector which is assumed to face a downward-sloping export demand curve. Final demand (households, government, and investment) and intermediate demand is satisfied by a composite mix of both state-produced and imported commodities. The optimal mix of state produced commodities and the imported commodities is derived using a constant elasticity of substitution (CES) aggregation function (Armington function) to form a composite commodity (Figure 1). The model makes the small country assumption regarding imports, implying that prices for imports are fixed. 1 1 CES elasticities of 0.96 for agriculture; 2.15 for non-durable mfg.; and 2.87 for durable mfg.; and CET elasticities of 3.40, 2.90, and 2.90 for these sectors are based on the works of Melo and Tarr (1992), respectively. We assume

5 Northwest Journal of Business and Economics, Households are disaggregated into 3 income classes (low, medium, and high). Households are assumed to maximize utility subject to a budget constraint (Figure 1). Household demand is derived using a linear expenditure function (Stone-Geary utility function). 2 The sources of household income are the factor incomes earned in Washington factor markets, exogenous government transfers, and transfers of other property and labor income from outside the state. Factor incomes, net of payroll tax, corporate profit tax, depreciation, retained earnings (enterprise savings), and adjustments for incomes to out-of-state factors, are distributed to the different household income classes according to fixed institutional shares (Figure 1). Household disposable income is computed net of household residential property taxes and federal income taxes. Household savings (HHSAV, Figure 1) are modeled as constant proportion of household disposable incomes (after-tax income). State and local government expenditures are treated as endogenous and are assumed to be driven by state and local tax revenues. These expenditures are treated as a function of the total state and local revenue plus intergovernmental transfers. State and local tax revenues are generated from the payroll tax, the household direct taxes (residential property taxes), and indirect business taxes (the gasoline tax, business and occupation tax, sales tax, and business property tax). Investment demand is specified assuming a linear expenditure system. Total investment is assumed to be driven by regional savings. In the Washington model, investment demand adjusts to aggregate savings that consists of income-driven household savings and fixed external (outside-the-state) savings (EXOSAV, Figure 1). In the model, labor, capital, and proprietor s services are assumed fixed for the Washington economy (fixed in state total supply) but mobile between sectors. Returns to factors adjust until all factor markets are in equilibrium and all factors are fully employed. The length of run in the model is long-run. The gasoline tax is imposed on the value (price times quantity) of all gasoline transactions for regional absorption (state produced and consumed commodities and imports) and export. The gasoline tax directly affects the price of gasoline which, in turn, affects consumer, government, investment, and intermediate demand. Consumer demand is affected as households now face a higher gross of tax composite price with the tax increase. Producers using gasoline as an input are affected as the gross of tax price of gasoline increases which lowers value added price. 3 (The the same CES and CET elasticities for all other sectors (1.50 for construction; 0.50 for aerospace; 1.10 for transportation, communication, and utilities; wholesale trade; retail trade; and finance, insurance, and real estate; 0.40 for business service; health service; and other service). 2 The Linear Expenditure System (LES) allows for nonunitary price and income elasticities. For calibration of the LES the Frisch parameter for low, medium and high income households are 1.8, 1.16, and 1.40, respectively (Lluch, Powell and Williams, 1977). We assume the same income elasticities for all households (0.6 for agriculture; 1.05 for non-durable mfg.; for durable mfg.; 1.06 for aerospace; 1.05 for transportation, communication, and utility; wholesale trade; retail trade; finance, insurance and real estate; business service; health service; and other service). These elasticity estimates are based on the works of Melo and Tan (1992), and Lluch, Powell, and Williams (1977). 3 Value added price is the output market price, net of indirect business taxes and the cost of intermediate inputs.

6 Northwest Journal of Business and Economics,

7 Northwest Journal of Business and Economics, value added price for producers is then used to derive their demand for factors of production). Gasoline producers face a fall in their producer price since the tax increase lowers their net of tax price. Thus, the gasoline tax shock affects both supply and demand sides of the economy through a change in the net of tax and gross of tax prices faced by producers and consumers respectively. It may be noted that the impact of a gasoline tax increase in Washington will differ from the neighboring states of Oregon and Idaho. Washington has a major gasoline producing industry while such is not the case in Oregon and Idaho. Thus, a tax increase will increase the cost to firms and households in Oregon and Idaho, but there is no state based refining industry to be affected. Such is not the case with Washington. Market equilibrium requires that the total demand for factors (labor, capital, and proprietary services) must equal the supply of factors; and market demand for commodities must equal composite commodity supply. We expect value added price (output price less taxes and cost of intermediate input) to decrease with the increase in the gross-of-tax price of gasoline especially in those sectors where gasoline is an important input. These value added prices are used in deriving first order conditions to determine the demand for the primary factors of production. Factors are paid according to their marginal contribution to the total production. The sectors which are directly or indirectly hurt by the tax increase are less able to compete for factors of production. The market clearing price of all three factors is expected to decline so that less damaged sectors can absorb the released factors and maintain full employment in the economy. The net of tax producer price and export price are expected to fall in the gasoline producing sector where the tax increase is levied. A decrease in net of tax export price is expected to cause export sales of gasoline to decline. In addition, households are expected to receive less income because returns to the use of their factors are expected to decline. A decline in the state household income has a further adverse effect on state output through a reduction in household consumption. The one sector that is expected to gain from the tax increase is construction, where increased demand is generated through spending of larger gasoline tax revenues. Other sectors that use little gasoline as input will have relatively low intermediate input cost change and may find it profitable, with lower cost primary factors, to expand output as well. On balance, new equilibrium output, income, consumption, tax revenues, and other major macro variables are expected to change modestly from benchmark equilibrium values as all economic agents respond to the increase in gasoline tax. SOURCES OF DATA A Social Accounting Matrix (SAM) of the Washington economy was constructed to organize the data required to build the CGE model. A SAM is an extension of Input-Output (10) accounts. The SAM is a square matrix with each row representing receipt accounts and each column

8 Northwest Journal of Business and Economics, representing expenditure accounts. The total of each row account equals the total of the corresponding column account which fulfills the conditions that total expenditure equals total receipts or that total supply equals total demand. The IMPLAN (Impact Analysis for Planning) system was used to build the IO accounts for the 1990 Washington economy. The IMPLAN household classification of income is as follows: low < $20,000, medium $20,000-40,000, and high > $40,000. The percent distribution of Washington households in these groups was percent in low, percent in medium, and percent in high income households. The IO accounts were supplemented by a variety of other sources of data to construct the SAM. Other sources of data were the Census Bureau of the U.S. Department of Commerce (1990, 1991), Quarterly Business Review (1991), Tax Statistics (1990), Washington State Department of Revenue, and statistics of Income Bulletin (1990, 1992). MODEL RESULTS Total motor fuels tax revenue in the baseline 1990 Washington economy was $498 million dollars. If there were no change in number of gallons consumed, the new tax rate (35.5 cents a gallon) would be expected to raise to $769 million in gas tax revenues, a 54.4 percent increase. In the CGE model, gasoline sales are included as part of sector 3 (the non-durable manufacturing sector). The gasoline tax increase was levied as a transaction tax on non-durable manufacturing, the behavior of which is assumed representative of the gasoline industry. It was also assumed that the additional gasoline tax revenues are spent by state government on construction projects (such as highway, and local street, and road projects). Therefore, state government demand for construction increases by the increase in gasoline tax revenues in the model. The impacts of changes in tax structure on selected endogenous variables are presented by comparing simulation results with the initial benchmark results. The changes in variables are shown in terms of percentage changes from base levels. In response to the gasoline tax increase, consumers in Washington reduced their composite gasoline consumption (Non-durable Mfg.. sector) by 0.89 percent (Table 2). But the consumption of the state produced commodity (-2.20 percent) declined at a higher percentage than the imported commodity (-0.41). The price of state-produced gasoline increased by 1.38 percent, while the gross price of imported gasoline increased only by 0.53 percent (not given in the table). Thus, it was economical for consumers to reduce consumption of state-produced gasoline commodity more than consumption of the imported commodity (Table 2). The total in-state demand for consumption of gasoline includes demands of households, government, and industry (intermediate). Households reduced their gasoline consumption (-0.7 percent) by responding to the combined effect of the increase in the composite price of gasoline and the decrease in their net income (Table 3) (the gross composite price of gasoline increased by 0.76 percent). Government demand declined by 0.17 percent in response to a decline in tax revenues other than the gasoline tax. Intermediate demand also declined because gross output declined for the economy as a whole. While Construction and Durable Manufacturing sectors

9 Northwest Journal of Business and Economics, increased their output and consumption of gasoline, other sectors decreased their consumption resulting in the net decline in intermediate demand for gasoline by 1.27 percent (Table 3). TABLE 2 PERCENTAGE CHANGES IN SUPPLY Total Output Produced by Washington Output Produced and Sold in Washington Composite Commodity Export Import Sector (1) (2) (2) + (4) (3) (4) Ag\For\Min Construction Non-durable Mfg Durable Mfg Aerospace Trans/Commu/Ut Wholesale Trade Retail Trade Fin\Ins\Re Est Business Service Health Service Other Service Total TABLE 3 PERCENTAGE CHANGES IN QUANTITY DEMANDED FOR COMPOSITE COMMODITIES (Regionally Produced and Consumed, and Imports) Sector Household Demand State Gov. Demand Investment Demand Intermediate Demand Ag\For\Min Construction Non-durable Mfg Durable Mfg Aerospace Trans\Commu\Ut Wholesale Trade Retail Trade Fin\Ins\Re Est Business Service Health Service Other Service Total

10 Northwest Journal of Business and Economics, The price elasticity of demand for gasoline was calculated as 1.13 implying that the demand for gasoline was elastic. We should note that this elasticity was based on the use of gasoline by households, government, and in-state industries. The gasoline tax was also imposed on the sales of exports, and export sales are not included in this calculation of the elasticity of demand. Because of the tax increase on gasoline exports, it became less attractive for state-produced gasoline to compete for sales to export. Although the world export price was unaffected by the increase in gasoline tax, the price, net of the gasoline tax, for in- state producers decreased by 0.53 percent. As a result, exports of this sector were estimated to decline by 6.02 percent. The combined effect of the decreased exports and decreased in-state sales resulted in an estimated decline in gasoline production of 4.56 percent (Table 2). Other than Non-durable Mfg., the Ag\For\Min sector was the most adversely affected. Output in this sector declined by 1.25 percent (Table 2). Because of the contraction of output, all three factors (labor, capital, and proprietor's services) were released from Ag\For\Min. Durable Mfg. and Construction sectors absorbed most of the factors of production released by more damaged sectors in the economy. Total output increased by 1.58 percent in Construction due to the large increase in construction demand by the state. Output increased in Durable Mfg. by an even larger percent (2.33 percent) because the increase in Durable manufacturing cost due to the gasoline tax increase was small relative to the cost increase in other sectors. Value added in Durable Mfg. and Construction sectors increased even though the total value added in the state declined (-0.27 percent) (Table 4). TABLE 4 PERCENTAGE CHANGES IN FACTOR PRICES, LABOR EMPLOYMENT, AND FACTOR EARNINGS Percent Change Item from Baseline Factor Prices Wage rate Rental rate Proper unit price Factor Incomes Labor income Capital income Proprietor income Value added As the state economy moved from the benchmark equilibrium to the counter factual (after tax increase) equilibrium, the wage rate of labor, the capital rental rate, and the return to proprietor services declined in order to permit all factors of production to again be fully employed (Table 4). The relative decrease in the rental rate of capital (-0.53 percent) compared with the decrease in the wage rate of labor (-0.18 percent) and unit return to proprietor (-0.47 percent) induced

11 Northwest Journal of Business and Economics, producers in Durable Mfg. and Construction sectors to utilize relatively more capital compared to the other two factors. With the decrease in factor prices, total factor income declined. Factor income was distributed according to fixed factor ownership shares for each income class. High income households experienced a greater decline (-0.23 percent) in income compared to the other two household groups because they received a greater portion of their income from returns to capital. Since returns to capital declined at a higher rate, the relative income of high income households also declined at a higher rate than the income of medium and low income households (Table 5). The total household real consumption of all goods and services in the state declined by 0.19 percent (Table 5). Real consumption of low and medium income households declined by more than their disposable income, but this was not true of high income households (Table 5.). This behavior reflects the consumption bundle of the respective households, the demand structure of each household group, and the relative price changes for the composite commodities. State government demand for construction increased by percent, but government demand for goods and services in all other sectors declined by 0.17 percent reflecting the reduction in value based tax revenues. In total, state government composite demand increased by 3.14 percent because of the increased spending on the construction sector (Table 6). TABLE 5 PERCENTAGE CHANGES IN AGGREGATE HOUSEHOLD INCOME, REAL CONSUMPTION, AND SAVINGS HOUSEHOLD CATEGORIES Household Categories Med Item Low HH HHHigh HH Total Gross Income Disp. Income Consumption Savings Total real investment and intermediate demand for composite goods and services in the state declined by 0.13 percent and 0.23 percent, respectively (Table 3). The decrease in household, investment, and intermediate demand was not offset by the overall increase in government demand resulting in the decline in total composite commodity demand (a combination of state produced and imported commodities) by 0.05 percent (Table 2). Total tax revenues in the state increased by 2.31 percent mainly because of the large increase in gasoline tax revenues (Table 6). As mentioned earlier, the gasoline tax revenues were targeted to increase from $498 million to $769 million, an increase of 271 million (54.4 percent). The gasoline tax was imposed on the value (quantity times price) of composite commodity absorption and exports. After the gasoline tax was levied and the state economy adjusted to the tax shock,

12 Northwest Journal of Business and Economics, gasoline tax revenues increased only by $266 million, $5 million less than the targeted increase in revenues (about 2 percent less). The number of gallons of gasoline consumed in the economy decreased by an estimated 13.1 million gallons, a decline of 0.61 percent. There were mainly two reasons for this. First, the gasoline tax directly raised the gross of tax price of gasoline to all buyers. Final as well as intermediate demand for gasoline declined. Second, with less returns to factors, household incomes declined. Consequently, the consumption of gasoline further declined with the decrease in household income. In the gasoline sector where commodity gross price went up, total quantity consumed decreased. Producers net export price directly declined, and the quantity of sales to export markets substantially decreased. Therefore, the decline in the both total composite commodity absorption and the gross export sales after the tax shock led gasoline tax revenues to increase less than the earlier target. TABLE 6 PERCENTAGE CHANGES IN STATE GOVERNMENT REVENUES AND DEMAND FOR COMMODITIES Percent Change Item from Baseline Govt. demand for construction Govt. demand for other sectors Govt. total demand 3.14 Total tax revenue 2.31 Payroll taxes Corporate income tax None B&O tax Bus. property tax Specific sales tax (Gasoline) Specific sales tax (Alcohol & Tobacco) General sales tax 0.01 HH property tax HH income tax None Govt. sales revenue Other state tax revenues also declined after the economy adjusted to the gasoline tax shock in the state. For example, the business and occupation (B&O) tax is imposed on gross sales of output, and tax revenues declined (-0.08 percent) with the decrease in gross output sales receipts (Table 6). Although the substantial increase in gasoline tax (54.00 percent) has a large negative effect on the output and consumption of gasoline, it has marginal effects on the aggregate economy. Total output for the Washington economy decreases just by 0.09 percent, and total real household consumption decreases just by 0.19 percent. These results are very similar to the results found by Fisher and Despotakis (p 156) for California, and Hudson and Jorgenson (pp

13 Northwest Journal of Business and Economics, ) for the national economy. Both studies looked at the impacts of increases in energy taxes on energy consumption and aggregate private consumption and output in the economy. CONCLUSION A CGE model of the Washington State economy is used to examine the economic impact of the proposed increase of 12.5 cents a gallon in gasoline tax. The baseline gasoline tax rate is 23 cents a gallon. With the implementation of the Transportation commission s proposal, the tax rate would increase to 35.5 cents a gallon. Previous analyses of such a tax increase have assumed a perfectly inelastic demand for gasoline. We extend that analyses to look at the response of Washington producers and consumers to the tax increase. Model results indicate that gasoline producers would experience a substantial decline in output. Output in this sector declines (-4.56 percent) because the sales to regional (-2.20 percent) and export (-6.02 percent) markets substantially decrease. The regional demand for state produced gasoline decreases because of increases in its gross price and decreases in regional income. Sales to the export markets decreases because producers face a reduced export price net of tax. State-wide, the price elasticity of demand for gasoline is estimated to be 1.13, implying that the demand for gasoline is slightly elastic. The Ag\For\Min sector is very negatively affected by the gasoline tax increase. The decline in the sales of this sector to regional and export markets leads to a 1.25 percent decrease in regional output. The increase in intermediate input costs for this sector leads to a large decrease in value added price that results in a cutback in the use of factors of production and output. Durable Mfg. and Construction sectors are the most benefited sectors in the economy. The output in the Durable Mfg. sector expands because this sector is little damaged in its input cost structure and is able to increase sales to export markets (2.6 percent). Construction is the most favored sector in the economy because all gasoline revenues generated in the economy are spent on purchases from this sector. Output in this sector increases because the state government demand for construction is raised by nearly 24 percent. Total construction increases by 1.58 percent. With increased output, the demand for all factors (labor, capital, and proprietor s services) increases in construction. The gasoline tax generally has negative effects on gasoline producers, households as well as most but not all businesses in the state. It is possible that the increased economic efficiency and household utility from better public transportation capital would more than offset the reduced real household consumption and income indicated in this study. Alternatively, the tax increase is estimated to reduce state value added by 0.27 of a percent. So the question becomes, is the increase in transportation infrastructure likely to be sufficient to offset this economic cost? The answer to this question is a logical next step for research in CGE analysis of tax policy. A weakness of the analysis presented in this paper is absence of any treatment of public capital (like transportation infrastructure) as it affects the productivity of the Washington economy. In

14 Northwest Journal of Business and Economics, somewhat the same vein, the model treats the private capital base of the economy as fixed. A more dynamic model would allow both the private capital base and the public capital base to change as a function of private sector investment and public sector investment, respectively. Nevertheless, we believe the analysis reviewed in this paper represents an important step in the economic analysis of tax policy in Washington. Previous work has been limited to a partial equilibrium of single industries with very little or no attention to the rest of economy or state level analysis with macro revenue forecasting models that are not able to capture the allocative behavior of firms or households to tax shocks. The CGE model provides a convenient middle ground. It presents the individual portions of economy in the familiar neoclassical paradigm, and at the same time permits an economy-wide perspective in assessing the consequences of a given policy shock.

15 REFERENCES Armington, P.S. A Theory of Demand for Products Distinguished by Place of Production. International Monetary Fund Staff Papers, 16(1969): Ballard, C., Fullerton, D., Shoven, J., and Whalley, J. A General Equilibrium Model for Tax Policy Evaluation. (Chicago: The University of Chicago Press), Brooke, A., Kendrick, D., and Meeraus, A. GAMS: A User s Guide. (California: The Scientific Press), Dervis, K., De Melo, J., and Robinson, S. General Equilibrium Models for Development Policy: A World Bank Research Publication. (Cambridge: Cambridge University Press), Fisher, A.C. and Despotakis, K.A. Energy Taxes and Economic Performance, a Regional General Equilibrium Analysis. Energy Economics, Butterworth & Co (Publishers) Ltd., (April 1989): Harberger, A.C. The Incidence of the Corporation Income Tax. Journal of Political Economy, 70 (June 1962): Hudson, E.A. and Jorgenson, D.W. U.S. Energy Policy and Economic Growth, Bell Journal of Economics, Vol. 2 (1974): James, L.M. Washington State and Local Taxes: Initial Impact on Business. Institute for Public Policy and Management, University of Washington, Seattle, Lluch, C, Powell, A, and Williams, R. Patterns in Household Demand and Savings. (Oxford: Oxford University Press), Melo, J. de and Tarr, D. A General Equilibrium Analysis of US Foreign Trade Policy. (Cambridge: The MIT press), Melo. J. de, Stanton, M., and Tarr, D. Revenue-Raising Taxes: General Equilibrium Evaluation of Alternative Taxation in U.S. Petroleum Industries. Journal of Policy Modeling, 11(3) (1989): Minnesota Implan Group IMPLAN: Database Documentation Morgan, W., Mutti, J., and Partridge, M. A Regional General Equilibrium Model of the United States: Tax Effects on Factor Movements and Regional Production. The Review of Economics and Statistics, Vol. LXXI (November 1989): Morgan, W., Mutti, J., and Rickman, D. Tax Exporting, Regional Economic Growth, and Welfare. University Station, Laramie, WY 82071, Upadhyaya, Mukund P. and David W. Holland. Northwest Journal of Business and Economics, Center for Economic and Business Research, College of Business and Economics, Western Washington University, 1996 Edition,

16 Northwest Journal of Business and Economics, Rafts, J. Different Macroclosures of the Original Johansen Model and Their Impact on Policy Evaluation. Journal of Policy Making, 4(1) (1982): Sen, A. Neoclassical and Neo-Keynesian Theories of Distribution. Journal of Economic Record, 39 (March 1963): Shoven, J.B. and Whalley, J. General Equilibrium with Taxes: A Computation Procedure and An Existence Proof. Review of Economic Studies, 40 (October 1973): Shoven, J.B. and Whalley, J. Applied General Equilibrium Models of Taxation and International Trade: An Introduction and Survey. Journal of Economic Literature, 22 (September 1984): Strauss, R.P. A Study of Alternative Tax Structures for the State of Washington. Center for Financial Management, Carnegie-Mellon University, Upadhyaya, M.P. A Computable General Equilibrium Analysis of Alternative Tax Policies in the State of Washington, Ph.D. Dissertation, Department of Agricultural Economics, Washington State University, Upadhyaya, M.P. and Holland, D.W. Replacement of the Business and Occupation Tax with State Personal Income Tax: A Computable General Equilibrium Analysis. A.E. 95-7, Department of Agricultural Economics, Washington State University, Pullman, U.S. Department of Justice. Statistics of Income Bulletins. IRS, Washington, D.C., 1990, U.S. Department of Commerce. State Government Tax Collections. Economics and Statistics Administration, Bureau of the Census, 1990, Washington State Department of Revenue. Tax Reference Manual: Information on State and Local Taxes in Washington State. Olympia, Washington State Department of Revenue. Tax Statistics Olympia, Washington State Department of Revenue. Quarterly Business Review: Calendar Year Olympia, Waters, E.C. Tax and Budget Policy in Oregon: A Computable General Equilibrium Modeling Approach, Ph.D. Dissertation, Department of Agricultural Economics, Oregon State University, 1994.