Political markets, property tax referenda, and local school spending

For better or worse, fiscal decisions made through property tax referenda allow local political markets to work. Demand, supply, and voting process components of such markets are estimated for those Oregon K-12 school districts that held referenda between 1981 and 1986. Various attributes of the median voter were related to school spending, but supply decisions by school boards and administrators were also important. Large districts used state aid to substitute for local property tax revenues on nearly a one-for-one basis, while relying on reversion budgets (inadequate property tax bases and implicit threats of school closures) to extract greater-than-desired spending levels from the median voter.     

Tests of institutional versus non-institutional models of local public expenditure determination

Are institutional or non-institutional factors more important determinants of local public expenditure decisions? Employing data that allows us to directly identify the entire distribution of voter expenditure demands, we estimate models of deviations of actual expenditures from the expenditure demand of defensible benchmarks — namely, the median voter's expenditure demand and the mean of voters' expenditure demands. Three competing paradigms are employed to specify these models; namely, a non-institutional model, the reform government movement's explanation of how institutions affect local government decisions and a transactions-cost model of this same phenomenon. Predictions from these three competing models differ in many respects. Comparisons of results from estimating models of expenditure deviations are found to be generally more consistent with the non-institutional model than with either of the two institutional models. These results suggest that a competitive local public service market helps limit the ability of governments to tax and spend in excess of the demands of the median or mean voter.     

Tax Competition and Tax Coordination in a Median Voter Model

This paper analyzes the welfare effects of capital tax coordination in a simple model of fiscal competition where fiscal policy is subject to majority voting and households differ with respect to their labor and capital income. It turns out that a coordinated capital tax increase may raise or reduce welfare, depending on the relative magnitude of i) economic distortions induced by a labor tax and ii) political distortions resulting from the influence of the median voter on fiscal policy decisions. A negative welfare effect is more likely, the smaller the marginal excess burden of the labor tax and the smaller the ratio of the median voter's labor income to average labor income. We also use empirical estimates of the marginal excess burden of taxation to determine the welfare effects of tax coordination; it turns out that a negative welfare effect of coordinated tax increases may emerge in our model for empirically reasonable parameters.     

Is the Median Voter a Clear-Cut Winner?: Comparing theMedian Voter Theory and Competing Theories in Explaining LocalGovernment Spending

This paper attempts to test the power of the median model againstthe respective strength of other alternate models based onredistributive, political-institutional and interest group theoriesin explaining the demand for public spending in New York statecounties during 1990, 1980 and 1970. To execute the comparison ofthe performance of median voter model with that of each of thenonmedian voter ones, various nonnested tests such as J and JAtests, N-tilde, W and encompassing tests have been employed.Results of the study show that although the median voter model hasa marginal edge over the rival models based on the alternativetheories, it may not be relied upon solely when many otherinstitutional, redistributive and interest group factors are alsorelevant for explaining public spending. The results of this studydiffer from those in Congleton and Bennett (1995). We do not findthat interest group models are substantially weaker than the medianvoter model.     

A distributional theory of government growth

This article presents a closed form solution for time-consistent taxation and public spending in a dynamic game between government and median voter. Extending Meltzer and Richard’s static analysis of government size the article offers a theory of growth of government. At low stages of economic development the median voter, identified as a relatively poor worker, prefers to have no or only small redistributive taxation in order to foster savings. Through this channel he expects improvements of his labor productivity and wage. At higher stages of development, however, when capital is relatively abundant and prospects of further labor productivity gains through capital accumulation are smaller, the incentive to tax and redistribute income rises. Yet, in line with previous work on growth and infrastructure spending the median voter prefers a constant share of productive public spending at all times. Hence, government growth is solely driven by an expanding welfare state.     

School Finance Reform's Impact on New Jersey's State Spending Priorities

The combination of school finance reform, voter opposition to higher taxes, and rising costs forced the state of New Jersey to reorient its spending priorities. This article presents an analysis of budgetary data for the period FY 1990 to FY 1996, which clearly indicates that: 1) state resources were shifted from direct state services to state aid; 2) even though the 1991 sales and income tax hikes were revoked, the state's tax structure was more progressive in 1996 than in 1990; 3) Governor Florio's attempts to level down per pupil expenditures by reducing payments to wealthy school districts were largely stymied; 4) the proportion of state resources allocated to public education was lower in FY 1996 than the year preceding school finance reform; 5) because of education's reduced budget share, efforts to level up per pupil expenditures were severely circumscribed; and 6) state aid was diverted from homestead rebates to municipal aid.     

A forecasting model for state expenditures

In this study, the median voter model is applied to the problem of predicting state public expenditures in four major categories: human services, elementary and secondary education, higher education, and health services. Demand functions for public services are derived within a formal utility-maximizing model. The resulting model is estimated using time-series data for a representative state and the model is used to predict state spending for fiscal year 1985. The results of the estimation are consistent with the predictions of the median voter model and the forecasts correspond closely to those made by the State Bureau of the Budget. In contrast, a naive autoregressive model of state spending performs poorly.     

Interrelationships among public spending preferences: A micro analysis

This paper examines a system of qualitative demand equations for public spending on education, welfare, housing, health, highways, and defense. The demand for public spending for a particular category is hypothesized to be a function of income, tax-price, private benefit measures, and tastes as well as demands for other public expenditure categories. Based on individual survey data, the conditional maximum likelihood estimates of the logit equations are obtained. The results indicate the significant role of tastes, private benefit variables, and tax-prices. In addition, positive rather than negative, demand interrelationships are found to exist. These findings question the validity of the median voter hypothesis as a model for public budget allocation.     

Expressive Voting and Government Redistribution: Testing Tullock's `Charity of the Uncharitable'

Models of expressive voting postulate that voters will`consume' ideological stances on issues by voting for them,even when they are against the voter's own narrow selfinterest, if the probability of being a decisive voter is low.When a voter is unlikely to sway the outcome, the odds that avoter will incur any real personal cost (a higher tax burden,for example) from her own expressive vote is small. We testand find support for Tullock's straightforward empiricalimplication of this model, that government welfare (transfer)payments are inversely related to the probability of being thedecisive voter.     

The Georgia Lottery and State Appropriations for Education: Substitution or Additional Funding?

This article examines the use of lottery proceeds for funding public education in Georgia, with specific focus on the state's effort to guard against fungibility of lottery proceeds. The Georgia lottery earmarks proceeds for education and Georgia is among the states requiring a high level of transparency at each stage of the budget and appropriations process. Based upon comparisons of spending before and after the lottery was put in place, we conclude that lottery spending has not been completely offset by substitution. Lottery funds appear to have stimulated additional spending in the target areas. Budget fungibility has been constrained by the transparency of the budget and appropriations process, gubernatorial commitment to supplement not supplant, the policy architecture of the lottery–for–education program, and a relatively strong state economy that renders substitution unnecessary.     

Public pensions and immigration policy in a democracy

The paper analyzes the link between the public pension system and the immigration policy. In a pay-as-you-go system, the incentives for immigration vary significantly between individuals at different lifetime periods. In the framework of an overlapping generations model, we show that the median voter's choice in general leads to inefficient levels of immigration. The median voter neglects the effects of the externalities within the pension system on other generations. An immigration policy that is not affected by the median voter's choice but instead is constitutionally determined will avoid welfare losses. The expected lifetime income of each generation can be increased by applying a rule of steady immigration.     

The median voter model across levels of government

The median voter model provides a method of aggregating individual voter's demands to obtain community demand. Although higher level governments are often assumed to be less responsive to voters than lower level governments, it remains an open question whether or not the median voter model, which has been found to apply to the lowest tier of the federalist system, extends to higher level governments. We use the Cox specification test to find that the ad hoc model dominates the median model for counties and states. The county results are sensitive to the time periods, budgetary structure, and degree of urbanization.     

The Polarization of American Environmental Policy: A Regression Discontinuity Analysis of Senate and House Votes, 1971–2013

The partisan polarization of environmental policy is an important development in American politics, but it remains unclear how much such polarization reflects voter preferences, as opposed to disagreements between partisan elites. We conduct a regression discontinuity analysis of all major environmental and energy votes in the Senate and the House, 1971–2013. In total, we have 368,974 individual roll call votes by senators and House Representatives. The causal effect of electing a Democrat instead of a Republican in close elections on pro‐environmental voting is large: when a Democrat wins, pro‐environmental voting increases by over 40 percentage points. Because of the quasi‐experimental research design, this difference cannot be attributed to the median voter's preferences. Next, we test hypotheses concerning possible explanations for the elite partisan conflict. The Democrat–Republican gap is the widest when fossil fuel interests make contributions exclusively to Republicans and when state‐level public opinion is polarized.     

Ideological change and the economics of voting behavior in the US, 1920–2008

This paper tests the proposition that voters advance a more liberal agenda in prosperous times and turn more conservative in dire economic times. A reference-dependent utility model suggests that, with income growth, the relative demand for public goods increases and the median voter is more likely to vote Democrat. With slowing income growth, the median voter derives increased marginal utility from personal income—making taxation more painful—and is more likely to vote Republican. Ordinary and instrumented analyses of a new time series for the US median voter are encouraging of this income growth model. This work links voting behavior to economic business cycles and shows that ideological change is endogenous to income growth rates.     

Do differences in presidential economic advisers matter?

Using data on members on the Council of Economic Advisors as well as US Treasury secretaries and OMB directors from 1952 through 2005, I investigate the effect of economic advisors’ educational and employment backgrounds on the time series performance of several policy variables. Ivy League advisors appear to raise non-defense government spending, although the size of the impact differs by president. While voter preferences appear to matter for a wider variety of policy variables (changes in federal regulation and marginal tax rates), the share of Ivy League advisors is at least as important as voter preferences in explaining non-defense spending.     

Governmental behavior in representative democracy: a synthesis of the theoretical literature

What is the impact of bureaucratic behaviors on tax rates? What is preferable: a bureaucrat or a partisan politician? Does a flypaper effect occur when lobby groups influence public policies? This article tries to answer these questions by analyzing and comparing six different models of public choice: (1) the median voter model, (2) the Leviathan model, (3) the slack-maximizing model, (4) a general model of bureaucracy, (5) the partisan politicians model and (6) a general model of lobbying. Among several results, it is shown that the median voter’s income and tax share almost always play a significant role in the design of the public good/tax package, even if the government does not explicitly try to maximize the median voter’s utility. The article also provides a synthesis of the positive analysis of the flypaper effect.     

Spending Preferences of Public Sector Officials: Survey Evidence from Finland

We examine the determinants that shape the spending preferences of public sector officials on several budgetary appropriations. Following Niskanen's budget‐maximizing theory, we test whether these officials prefer larger budgetary appropriations rather than less. We measure their preferences to increase their own bureau's appropriations and compare those against their preferences for other bureaus' appropriations. The empirical evidence is gathered via a mail survey targeting high‐level officials from different ministries in Finland. The analysis of the responses suggests that Niskanen's theory is in part supported.     

Voter Responses to Referenda Seeking to Exceed Revenue Limits

Following voter approval of California's Proposition 13, a body of research has attempted to explain referenda outcomes aimed at restricting public sector tax and spending authority. Evaluating the determinants of voter attitudes toward referenda to exceed spending/revenue limits, however, has received little attention. Using data for all Wisconsin school districts, we examine the outcomes of referenda to exceed revenue caps. A key finding of our research is that when voters support bonding for capital expenditures, they tend also to approve referenda to override revenue limits. This enables school districts to cover new operating costs often associated with new capital projects.     

Federal tax reform and the financing of state and local governments

Although the Reagan Administration tax reform proposals would reduce federal income tax liabilities for most taxpayers, federal tax reform would also create strong pressures on state and local governments to cut taxes and public services. These pressures would arise primarily because itemizers would no longer be able to deduct state and local taxes in determining their federal income tax liabilities. In New York City and Boston, it is likely that the Administration's tax reform would induce cuts in spending that range from 2.5 to 7.5 percent. While the elimination of state and local tax deductibility may promote allocative efficiency in the provision of local public goods, the cost would be a decline in the degree of redistribution through the state and local public sector, and a reduction in local public services for the poor.     

Does restricting choice in referenda enable governments to spend more?

If school boards represent the preferences of the median voter, referendums to approve school expenditures should not constrain school expenditures. If school boards would choose expenditures larger than the median voter’s preference, referendums that restrict voters to approving or disapproving a school board’s recommended expenditure level will result in expenditures larger than the median voter would prefer. However, Florida used a unique referendum system which guaranteed the median voter’s most preferred outcome. Using Florida education expenditures as a benchmark, the evidence suggests that spending was slightly higher in restricted choice referendum states than in Florida.