Franklin & Marshall College Franklin & Marshall College

    • u-h-9af179e6f4f3-jpg
    • u-h-20ccfc70467a-jpg
    • u-h-c8018e0f0ae5-jpg

The Great Tax Vote Myth

March 18, 2003

by Dr. G. Terry Madonna and Dr. Michael Young

Politics is a problematic world. Unknowns abound, randomness intrudes, and uncertainty prevails.  So, it is some comfort to know that there are a few verities--a few rock solid certainties--about how voters think and behave.

And most reliable of all those certainties is the common knowledge that voters hate taxes, and vote to defeat politicians who raise their taxes.

Or maybe not!

A close look at the record suggests that voters' reaction to tax increases is neither simple nor certain. In fact, leading academic findings, together with supporting empirical data from Pennsylvania, produces a startling conclusion: governors who propose tax increases are not necessarily punished by voters, nor in fact are legislators who vote for those proposals.

That is not the conventional wisdom.  In Pennsylvania, the conventional wisdom holds that politicians who support big tax hikes--defined here as income or sales tax increases--are likely to end up electoral road kills in subsequent elections.

But this widely held assumption does not stand up under strict scrutiny. Neither in Pennsylvania nor in other states is their much evidence that supporting tax increases evokes harsh voter retribution.

Some brief background on past tax hikes is helpful. Pennsylvania's legislature has increased the state income tax rate twice in the last two decades--in 1983 and 1991, and each time the sales tax was left alone.

In 1983, the income tax was raised from 2.2 percent to 2.45 percent. In 1991, it was raised from 2.1percent to 3.1 percent. The tax hikes in each case were occasioned by major recessions--the two worst since the Great Depression.

Both tax hikes were contentious and hard fought, and the legislature was unable to meet its constitutionally mandated June 30th deadline for budget adoption. Crisis ensued.

In 1983, the budget was finally adopted on July 20 while the budget debate in 1991 was not settled until early August.

But now the surprise: neither the contentiousness of the debate nor the tax vote itself proved injurious to incumbent legislators seeking reelection.

Here's the data:

 In 1984, subsequent to the 1983 income tax increase, 187 House incumbents sought reelection; four lost in the primary and none in the general election, for a stunning 98 percent reelection rate.

The 1992 incumbent reelection rate following the 1991 income tax increase was similarly spectacular.  Of the 174 incumbents who ran for reelection in 1992, seven were defeated; the reelection rate was 96 percent.

The situation was similar in the Senate, where one half of the Senate's 50 seats are in play each even-year election.  In 1984, subsequent to the 1983 tax increase, only two incumbents lost, while no incumbent lost in 1992, the election held immediately after the 1991 tax increase.

Closer study of the anecdotal evidence from individual House and Senate races confirms the statistical data.  The tax hike vote figured in only a few of the losses and was decisive perhaps in no more than one or two.  For the vast proportion of legislators, casting a tax vote was not injurious to their political health.

Nor in fact is there any reason to believe that governors who proposed income tax proposals are politically compromised in doing so. Going back some 30 years, three governors have proposed either a new income tax or a major income tax increase: Milton Shapp in 1971; Dick Thornburgh in 1983; and Bob Casey in 1991.

Shapp was subsequently reelected to a second term by a huge margin. Thornburgh, ineligible for reelection, finished his term with enough momentum to become a major figure on the national scene. And Casey, also ineligible for another term, left office more popular than he entered it.

Beyond the direct experience in Pennsylvania, there is considerable academic evidence that voting for a tax increase does not necessarily bring with it political retribution.

The conclusions that can be drawn from the scholarly literature can be reduced to a couple of basic themes. The first, somewhat unsurprisingly, is that tax hikes do have political costs: raising major taxes like income or sales taxes do lose votes for a governor and his party; moreover, each new tax increase in a four year period has a cumulative effect on vote loss.

But--and this is a critical caveat--the actual vote losses are modest in most cases. One scholar has reported, for example, that governors who raise either the sales or income tax receive about 3 percent less popular vote in subsequent elections on average than governors who eschew taxation.

Other scholarly findings flesh out this picture. The rate of the tax being raised seems to matter most.

If voters feel a tax is already high, and it is raised even higher, voter retribution may be expected. That was the painful lesson learned in neighboring New Jersey by Gov. James Florio who raised that state's already perceived high income tax. Voter anger swept him and his party from office in the next election.

But raising a tax whose rate is already thought as low seems to stir minimal voter wrath. Clearly, voter perception of overall tax levels matters. Raising tax rates where existing rates are perceived to be low has very different consequences than from raising taxes already believed to be high.

And it also matters on what programs the revenues from new taxes are spent. If spending is used for unpopular programs, voter reaction may be harsh, but if it is for popular programs (education is one example), voter anger is often muted.

The relevance of these findings to the current debate over the state budget is compelling. Gov. Rendell is about to propose some significant tax increases--the first in 12 years, and possibly the largest in a generation.

Advocates as well as opponents of a tax increase will find many interesting arguments to buttress their respective positions. And there are good and important arguments on both sides.

In fact, we need and should have a full debate on both new taxes and the programs they would fund.

But one debate our legislators can skip is whether voting for taxes will earn them early retirement.  Legislators may decide not to vote for tax increases for many reasons, but fear of electoral defeat should not be one of them.

It doesn't happen.

Politically Uncorrected™ is published twice monthly. Dr. G. Terry Madonna is a Professor of Public Affairs at Franklin & Marshall College, and Dr. Michael Young is a former Professor of Politics and Public Affairs at Penn State University and Managing Partner at Michael Young Strategic Research. The opinions expressed in this article are solely those of the authors and do not necessarily reflect the opinions of any institution or organization with which they are affiliated. This article may be used in whole or part only with appropriate attribution. Copyright © 2003 Terry Madonna and Michael Young.