Polls Missed Tax Measure Results

A new column by Jon Coupal, president of the Howard Jarvis Taxpayers Association, argues that polling data frequently underestimates voter support for fiscal reform measures in California. Coupal notes that the current election cycle has heightened skepticism toward surveys, a sentiment fueled by the recent admission from Wedgewood Polls that its data was entirely fabricated.

The issue extends beyond disreputable firms. Aggregators such as The New York Times, FiftyPlusOne, and VoteHub included Wedgewood’s invented results in their calculations, skewing the overall picture. National News Desk reported that the firm lacked a history, formal credentials, or real data for the majority of its releases, yet it was treated as a legitimate source.

Coupal points to a 2024 Des Moines Register poll conducted by J. Ann Selzer, which predicted Kamala Harris would lead Donald Trump in Iowa by three points. Trump won the state by 13.2 points, a discrepancy that forced Selzer into retirement and damaged the newspaper’s reputation.

In California, the trajectory of Proposition 13 in 1978 illustrates shifting voter attitudes. Early polls in 1978 showed support at 25% and opposition at 27%, with nearly half of voters undecided. By spring, support rose significantly as homeowners received property tax bills. The final Field Poll before the June 6 primary projected a margin of 57% to 60% for passage. The actual result was 64.8% to 35.2%.

Proposition 218 in 1996 provides a starker example of polling error. Final Field Poll data showed the tax reform measure at 36% support among likely voters. The measure passed with 56.55% support, defying the negative survey trends.

Most recently, polls for Proposition 5 missed the mark by about five percentage points. The Public Policy Institute of California reported a 49% to 50% split between yes and no votes just before the election. The measure failed with 55% of the vote against 45%.

Coupal suggests these discrepancies indicate that many voters conceal their fiscal conservatism or do not respond to surveys. He concludes that the only reliable measure of voter intent is the ballot cast on election day.