Time for a Texas Compromise?
Table of Contents
Author(s)
George R. Zodrow
Baker Institute Rice Faculty Scholar | Allyn R. and Gladys M. Cline Chair of EconomicsJohn W. Diamond
Edward A. and Hermena Hancock Kelly Fellow in Public Finance | Director, Center for Public FinanceThe Texas Legislature is currently wrestling with an always contentious issue – how best to reform the state tax structure. Dramatic increases over time in the share of K-12 education financed by local property taxes, coupled with widespread (although by no means universal) dislike of the “Robin Hood” system of redistribution, have created significant political pressure for reductions in school property taxes and the use of an alternative method of supplementing education spending in poorer school districts. Thus, it seems inevitable that more revenues for education finance will be raised at the state level, although the means by which this goal will be achieved is by no means clear.
There are, of course, a wide variety of tax reform options, most of which are analyzed at length in Zodrow (An Economic Evaluation of Alternative Sources of Tax Revenue for the State of Texas, available at the Texas School Finance project website, http://www.tlc.state.tx.us/roadmap/tsfp/reports.htm). Recent weeks have seen bills proposing property tax reductions passed by both the Texas House and Senate, coupled with a sales tax increase and a new payroll tax/franchise tax option in the case of the House bill, and a statewide property tax and a new “business activities tax” or BAT under the Senate bill. The BAT would be a broad-based business tax which would allow “expensing” or immediate deductions of all capital investments, as opposed to the multi-year depreciation deductions typically allowed under the franchise tax it would replace. The obvious differences in these proposals suggest that Texas legislators are far apart in their approaches to solving the education finance dilemma. However, we believe that these differences, although significant, are far from insuperable, and that in fact the potential exists for an excellent outcome – a “Texas compromise” that would be fashioned from the best elements of the competing proposals.
The starting point for the compromise is the fact that the key features of both the House and Senate approaches reflect a recognition that the state tax structure should limit the extent to which it discourages capital investment. The high degree of mobility of investment capital, both across the states and internationally, is an increasingly critical feature of the fiscal landscape in all states. The implication of such mobility is that, in the long run, imposing taxes on mobile capital in excess of the benefits of public services received is ultimately counterproductive from the viewpoint of Texas residents. Such taxes reduce investment within the state, ultimately lowering wages and land rents in Texas by roughly the full amount of the tax, while simultaneously reducing the productive efficiency of Texas businesses by reducing the capital intensity of their production methods. Because labor is generally less mobile than capital, this reasoning suggests that state economies should primarily utilize taxes on labor rather than on mobile capital, even if the goal of policy makers is to improve the well being of Texas wage earners. Although this argument is subject to a number of qualifications, it suggests that taxes on businesses in excess of their benefits from public services should be approached with great caution.
For this reason, the Senate proposal to tax property at the state level is at best only a partial solution to the state’s tax problems. It does have the advantages of (1) reducing distortions of investment location decisions across jurisdictions within the state, (2) reducing large discrepancies in school property tax bases and thus ameliorating school funding inequities, and (3) providing a source of state revenues to finance redistributive educational expenditures. But these revenues will still, in large part, come from the taxation of mobile capital, and thus have all of the deleterious effects described above. Accordingly, state-level property taxation should, as in the Senate bill, reflect an overall lower rate and be accompanied by some new method of finance that primarily taxes labor income.
All of the remaining tax reform proposals currently under active discussion are at least in principle consistent with this reasoning. (We do not include a state income tax in this group of active proposals. A state income tax would include counterproductive taxation of mobile capital under its corporate income tax, and its personal component would, at least under current rules, effectively be a non-deductible tax for most Texas residents. Given these problems, coupled with longstanding and fierce political opposition, a state income tax is not a viable option.) Taxes on consumption are, roughly speaking, very similar from an economic perspective to taxes on labor income, which suggests that the proposed increase in the sales tax rate is a good reform option. Unfortunately, the sales tax in Texas is not the uniform tax on consumption that is required for this argument to be valid. Instead, roughly half of the base of the sales tax consists of business purchases, converting the tax into a haphazard tax on business income, and much consumption, especially of many consumer services, is not subject to tax. The resulting sales tax is highly distortionary and inequitable. Although expansion of the sales tax base to include more consumption including consumer services is desirable, an increase in its already relatively high tax rate is not an attractive option, especially since such a high rate would only further encourage tax-avoiding purchases over the Internet.
This leaves the two new proposals currently being discussed by the Texas legislature – the payroll tax and the business activities tax (BAT). As suggested previously, the generic forms of these two taxes are quite similar from an economic perspective. In particular, because the BAT does not allow deductions for labor compensation, labor income is subject to tax, just as it would be under a payroll tax. Moreover, because the BAT allows immediate deduction of all capital expenditures (rather than the depreciation deductions allowed under an income tax), it does not tax normal returns to capital investment and is thus an extremely “investment friendly” tax. In addition, the BAT has several key advantages over the payroll tax. Most importantly, if structured properly to be a comprehensive business level tax, its relatively broad base is much less susceptible to tax avoidance in the form of disguised payments to labor that are not labeled “wages”; by comparison, a payroll tax would typically not reach such disguised labor compensation. The BAT also captures what economists call “economic rents” or above-normal returns to capital, the taxation of which generally is not subject to the criticisms of capital income taxation made above. Finally, a BAT would, over a long transitional period, tax the income earned by existing capital that would have been (or should have been) subject to the franchise tax under the current tax regime, generating revenues that would not be captured under a payroll tax.
Like any tax, the BAT has its disadvantages. Interestingly, however, the most important of these disadvantages is not shared by the payroll tax option – creating the potential for the “Texas compromise” that underlies our proposal. Specifically, the BAT is ineffective in taxing financial institutions (because financial transactions are not included in the tax base) and, in practice, such a tax is highly unlikely to be applied to nonprofit institutions and state and local governments. However, it would appear that political opposition to applying a payroll tax to bankers, doctors and university professors, and government employees would be less likely to present a serious obstacle to reform (although this is not entirely clear). Accordingly, the Texas compromise we envision consists of a BAT applied to all non-financial private businesses, coupled with a payroll tax applied to financial institutions, nonprofit organizations, and state and local governments – a combined business activities and payroll tax that one might call a BAPT.
The critical advantage of such an approach is that it would result in an extremely broad and comprehensive tax base, implying that the required revenues could be raised with a relatively low tax rate. A broad base coupled with a low rate is the mantra of public finance economists as it results in a tax system that is efficient in the sense of minimizing economic distortions, equitable as it applies uniformly to all taxpayers, and relatively simple in terms of compliance and enforcement. Base broadening and rate reduction was of course the organizing principle of the highly successful national Tax Reform Act of 1986, and Texas would do well to emulate that effort. In addition to creating a tax environment conducive to investment growth, the BAPT approach would have the advantages of being a fully deductible tax that would result in a stable stream of revenues to finance K-12 education in the state.
To a first approximation, the labor component of the BAT and all of the payroll tax would be borne by labor as reduced wages (although it should be noted that some of the tax would be borne by land owners and reductions in economic rents to highly profitable ventures, especially to the extent that labor is mobile across states). Concerns about the regressivity of the BAPT could be addressed by allowing a deduction under the BAT and an exemption under the payroll tax of, say, roughly $10,000 per full-time employee, so that the tax would have no impact on minimum wage workers and a moderate degree of progressivity. In the interests of maintaining a comprehensive base and a low rate, as well as avoiding a significant shift in tax burden away from upper income individuals relative to the current property tax regime, the caps on the payroll tax recommended in the House proposal should be avoided.
The treatment of multi-state firms would be a difficult issue under the BAPT, as it is under the current franchise tax. Because it is virtually impossible to calculate accurately the income of a multi-state firm attributable to its operations in various states, the states currently use formula apportionment to calculate business tax liability; under this approach, a state requires a firm to calculate its total profits and then apportions a fraction of that income to the state, where the fraction depends on the fraction of total firm gross sales, payroll and/or property located in the state. (The states use different formulas, with Texas currently using s “single factor” formula that depends solely on gross sales.) Under formula apportionment, the tax becomes primarily a tax on the factors in the apportionment formula. The base of the BAPT could also be allocated using formula apportionment. However, consistent with the taxation of labor rather than mobile capital, at least the labor component of the BAT base of multi-state firms should be apportioned by payrolls, in which case the state’s share of the labor component of a firm’s national tax base would equal its share of the firm’s payrolls. The remaining BAT base, which consists of above-normal returns and the income to existing capital, could then be apportioned according to property. Although apportionment according to gross sales has some appeal in the sense in that reduces the extent to which the BAPT is tax on in-state production, this approach should be avoided as it effectively converts the BAT to a highly inefficient gross receipts tax.
There are two main disadvantages to the BAPT proposal. The most often cited is that as a tax on labor, the BAPT creates a tax disincentive for employment and job creation. The quantitative significance of this argument should not be overstated. As noted above, the tax is most likely to be borne by labor as lower wages, which will simultaneously benefit from either lower sales taxes or lower property taxes. However, any remaining tax disincentives for employment and job creation must be minimized by strictly adhering to the principles of a comprehensive base and a low rate – pleas for special treatment should be firmly resisted. A second disadvantage is that non-profitable firms will have to pay BAPT. In this regard, the BAPT should be viewed as a tax that requires labor-intensive businesses to help finance – at a relatively low rate – some of the public services they enjoy, independent of profitability, just as the non-residential property tax must be paid by capital-intensive firms (and as the net assets component of the franchise tax must be paid under the current tax regime). Again, special exemptions should be firmly resisted, although provisions for delayed payment of tax by non- profitable firms merit consideration. On the other hand, provisions that allow businesses to choose between alternative tax systems, such as the House proposal that would allow a choice between a payroll tax and the franchise tax, create significant opportunities for tax avoidance and evasion and should not be included in the BAPT plan or any other proposal. Finally, the BAPT is sometimes criticized as a “hidden” tax, but the publicity that would surround the adoption of such a new tax suggests that this concern may be misplaced; if deemed to be a problem, however, businesses could be required to report to their employees the estimated BAPT paid on their wages to increase the visibility of the tax.
Finally, we would like to note that often-cited objections to the business activities tax based on the experience with a similar tax in Michigan are not especially well- founded. The Michigan “single business tax” has been made extremely complicated with a host of exemptions and special provisions. Moreover, comparisons of the tax burden under the Michigan tax, which treats capital very generously but does impose a tax on labor, with tax burdens under the corporate income taxes in other states, which are taxes only on capital income, are clearly highly misleading. In a study of the Michigan experience, Professor James Hines of the University of Michigan, whose work is often cited by opponents of the business activities tax, notes that the Michigan tax has had its problems. But his conclusion is unambiguous, as Hines argues that although the design of the current single business taxes is flawed, Michigan would be well advised to consider maintaining a reformed version of the tax rather than abandon it entirely in favor of the likely alternatives. Although Texas should certainly learn from the Michigan experience, we believe that an appropriately designed BAPT would be far superior to the existing Michigan tax, and indeed would give Texas a broad-based low-rate tax that would be an efficient, equitable, simple and stable source of revenues, resulting in one of the best state tax systems in the nation.
This material may be quoted or reproduced without prior permission, provided appropriate credit is given to the author and Rice University’s Baker Institute for Public Policy. The views expressed herein are those of the individual author(s), and do not necessarily represent the views of Rice University’s Baker Institute for Public Policy.