To predict tax revenue, look at corporate earnings

In the complex task of building a state budget, much rides on the accuracy of its fiscal crystal ball: its forecast of how much tax revenue will come in to fund services during the year ahead. Forecasting errors have increased since 2001 due to revenue volatility, such as wider swings in personal income and consumer spending.

New research from Texas McCombs offers states a tool that could boost the accuracy of their forecasts: the earnings growth of tax-paying public companies. The study, “Do Accounting Earnings Provide Useful Information for State Tax Forecasts?” is published in Review of Accounting Studies.

That could be good news for the more than 40 states that require balanced budgets. When they underestimate projected revenue, they may needlessly reduce spending, raise taxes, delay needed projects, and make other adjustments that affect millions of residents.

“When state tax revenue forecasts are not accurate, there are real consequences,” says Braden Williams, associate professor at Texas McCombs. “They are not simply a procedural formality.”

Williams co-authored the study with Lillian Mills, dean and professor of accounting at McCombs, and Anthony Welsch of the University of Chicago.

Corporate earnings are key

Currently, most states base their revenue forecasts on a combination of past revenue collections and macroeconomic projections, such as gross domestic product and unemployment rates. Forecasting teams are composed mostly of economists and statisticians with little accounting expertise.

They’re missing another ready information source, Williams says. Corporate earnings growth figures are publicly available, released quarterly, and audited.

What difference could such data make to revenue forecasts? To find out, the researchers analyzed the aggregate earnings growth of major industries in the U.S., such as mining, gas, technology, financial services, and health care. They linked that information to the size of the industries’ footprints in each state, coming up with a composite, state-specific earnings growth measure.

Finally, they plugged the information into states’ previous forecasts to get a hindsight view of the difference it could have made.

Overall, they found that it made forecasts more accurate. Including corporate earnings alongside traditional economic measures allowed them to explain up to 86% more of the variation in actual revenue collections.

The effect was most powerful for states with diverse industries. Smaller states with fewer dominant sectors—such as oil in Wyoming—already had the information they needed. By contrast, Williams says, “If their economy was balanced across a lot of different industries, then this earnings growth measure would have been very useful to them.”

Predicting personal income

Corporate earnings didn’t just enhance forecasting of corporate tax revenue. They also better predicted revenue from sales and personal income taxes. In dollar terms, in fact, it improved personal income tax forecasts even more than corporate ones.

“The degree of improvement is similar to corporate income tax, but the magnitude of the personal income tax base is so much bigger,” says Mills. “When you look at total revenues in aggregate, that’s where the biggest improvement was coming through.”

The study showed several reasons corporate earnings growth spilled over into other kinds of tax collections.

• Parts of corporate profits get returned to owners in the form of capital gains or stock dividends.
• Earnings growth can affect wage growth if companies share profits with employees.
• Business investments and purchases show up in sales taxes.

Besides helping to better predict inflows at the start of the year, the tool can help states avoid slashing programs later, Williams adds. “If states had used earnings growth in their forecasts and come up with better ones, there would have been fewer times they had to make midsession cuts to their budgets.

“It doesn’t have to be that way. There is publicly available information that’s being left on the table when the forecasts are made.”