The same GDP reports that have shown such progress recently also show that states and local governments have subtracted from the pace of real economic growth in three of the past four quarters, and it doesn't take much detective work to find out why. That same year of strengthening growth coincides with five consecutive quarters of weak state tax-revenue growth, including an outright year-over-year decline in the second quarter of 2016. Combined, state tax revenues were unable to clear the pace of inflation in the past year, their worst performance since the Great Recession.
How can state taxes be in recession when the rest of the economy is in expansion? This has been the biggest puzzle for policymakers this year, and the leading cause of subnational budget weakness throughout the country. The growing disconnect is twofold and attributable to both structural and one-time factors.
The first source of frustration for revenue estimators and policymakers is temporary and has to do with prices. This is particularly relevant to sales taxes, which have been the largest underperformer for most states over the past year and a half. Price levels impact sales taxes because sales taxes are levied as a percentage of overall taxable value. Thus, even though consumers may have more money in their pockets and are buying a greater number of goods, the value of those goods may not be growing in line with expectations. In fact, the taxable value may be declining.
Typically, at this point in the business cycle, low unemployment and rising wages push prices -- and therefore tax collections -- higher. However, the extraordinary strength of the U.S. dollar and the energy bust are holding prices well below expansionary levels. Headline consumer prices are increasing in line with the business cycle at around 1.7 percent annually, but the components of the Consumer Price Index tell a more nuanced story. Looking solely at goods prices, which make up the lion's share of all taxable sales, we actually still see deflationary pressures weakening sales-tax collections.
Take homebuilding as an example. Housing construction is a major driver of sales taxes because of the large durable-goods purchases that go into building and furnishing a new home. Last year the number of housing starts increased significantly, which should have resulted in stronger sales-tax collections. However, the prices of most goods used in new-home construction declined significantly. Lumber prices, for example, were down by more than 10 percent at one point in 2016. So even though more housing units were being built, the sales-tax collections from each unit were much less than the year before.
With inflation expectations rising, this dynamic will only prove temporary. However, a second and more structural consideration will continue to give revenue estimators fits for some time to come.
Over time, state tax revenues have generally grown more disconnected from the underlying economies upon which they're levied. A study we performed several years ago found that state tax revenues in the first decade of the 2000s had become three times more volatile than the underlying economy. This was orders of magnitude larger than anything states had experienced before, and it led to state tax revenues underperforming nominal GDP for the first time since records have been kept.
This dynamic has progressed in line with two growing influences on state taxation over the past few decades. First, states have become increasingly reliant on highly progressive personal income taxes. This isn't necessarily a bad thing -- an efficient tax structure should have at least some progressivity -- but a side effect is increased volatility. By making their tax structures more progressive, states are relying on a smaller set of higher-income taxpayers for a larger portion of their revenues. This set of taxpayers typically has some of the most volatile incomes. A taxpayer in the top 1 percent of the income distribution can make $10 million one year and very easily lose $10 million the next. It is no anomaly then that some of the states with the biggest income-tax surprises are also those with some of the most progressive tax codes.
Adding to this disconnect is the growing number of targeted tax incentives being put to use by state and local governments for economic development. This isn't necessarily a bad thing either, as these incentives can be effective at spurring local growth. But these rarely tracked incentives can also create a scenario in which the fastest-growing parts of an economy are growing tax-free. This distorts the relationship between broad-based economic statistics and tax revenues, making life very difficult on revenue estimators.
As a result, revenues become more unpredictable and policymakers have to become more cautious. Unfortunately, it looks like that dynamic is here to stay.
