Honey, I Shrunk the Government
Don’t listen to the budget hawks: in relevant terms, Washington state government has been shrinking for years
Washington is simply not a high-tax state. Nor is it a profligate spender. These are facts. But why let mere facts stop you from warning about the economic dangers of our allegedly out-of-control state budget?
“Main Street businesses bore the brunt of last year’s legislative session,” laments Association of Washington Businesses president Kris Johnson in a recent op-ed. Citing a “$52.8 billion operating budget” that he says included “more than $1 billion in tax increases,” Mr. Johnson sounds a familiar alarm:
“The higher taxes came despite a surge of $5.6 billion in new revenue coming into the state’s coffers as a result of the growing economy, and they raised serious questions about the long-term sustainability of Washington’s budget. The final budget amounted to an 18.3% growth in spending, one of the highest growth rates in decades.”
So, does our current budget really reflect “one of the highest growth rates in decades?” Um… I dunno. Mr. Johnson doesn’t bother to show his math—is he comparing year to year, biennium to biennium, quadrennium to quadrennium (yes, we now technically draft four-year state budgets)? For the life of me, I can’t get his numbers to add up. But regardless, this sort of cherrypicked budget porn actually tells us little if anything about the relative size and impact of state government.
State budgets are complex documents that can easily obscure all sorts of meaningful facts and trends. For example, much of the “surge” of new revenue in the current budget comes from a complicated public school “levy swap” that lowered local school levies while increasing dollars raised and spent directly by the state. By shifting local taxes to state taxes, this policy change shows up in the state budget as a substantial increase in state public school expenditures while, on its own, not raising total spending on public schools by a single dime. That’s why they call it a “swap.” Likewise, Mr. Johnson doesn’t specify whether the “18.3% growth in spending” he bemoans is in the Near General Fund or Total operating budget. If it’s the latter, much of that growth could be attributed to Medicaid expansion and other state-administered federally-funded programs that have no real budget impact at all.
Furthermore, these top-line budget numbers are a terribly misleading way to represent the growth of state government, because they inherently fail to account for growth in population and inflation, or growth in demand for government services as a whole. To be clear, there is a perfectly legitimate debate to be had over the proper size and scope of state government. But if we want an informed debate, we need to be looking, not at the state budget, but at combined state and local spending and tax revenue as a percentage of personal income—that is, the size of state and local government as a percentage of the state economy as a whole.
The chart above was copy and pasted directly from the pages of the Washington State Office of Financial Management (OFM) website, and while it only includes data through 2017, it simply does not show the unsustainable trend that Mr. Johnson implies. In fact, what we see here is a pretty substantial multi-decade decline in total state and local government spending from a peak of $224 per $1,000 of personal income in 1993 to $177 in 2017—somewhat below the 50-state average. If government expenditures relative to personal income are up in the latest budget (and I don’t have the data to say one way or the other), the percent increase would be exaggerated by it coming on top of an 18-year low.
Take another look at that chart and tell me that state and local government spending is out of control. You can’t. And for a very good reason: state and local tax revenues have been shrinking too.
This chart also comes directly from OFM, and it tracks a two-decade decline in state and local taxes as a percentage of personal income from 11.1% in 1997 to 9.4% in 2017—just a touch above the 40-year low. That this tax chart tells a similar story to the expenditure chart should not come as much of a surprise, because one thing pays for the other. It is literally impossible to grow state and local government relative to the economy as a whole when your effective tax rate is steadily shrinking over time.
Now, I know what some of you are thinking: Lies, damn lies, and statistics! I’m talking taxes and expenditures as a percentage of personal income in an effort to obscure some inconvenient truth, right?
Well, no. Read the academic literature and you will find that the most common metric used for comparing year-over-year growth in any particular government or for comparing the relative size and growth of different governments, is taxation and spending as a percentage of the economy as a whole. This is the gold standard. That’s why at the federal level, we always talk about taxes, spending, deficits, and debt as a percentage of GDP. (I use state personal income rather than state GDP because those are the charts OFM provides, but GDP would produce similar results.)
And if you’re worried that my inclusion of local taxes muddies the picture (Mr. Johnson was only talking about state spending, after all), well, you’re kinda right. When you break out just state taxes, the decline per $1,000 of personal income becomes even starker:
The data on this graph are over a decade old, but if you were to update it to the latest available personal income numbers ($468.4 billion) and the respective state tax collections ($23.8 billion), 2018 would be literally off the chart: just $50.81 in state taxes per $1,000 of personal income. At only 5.08%, you’d have to go back before 1960 (and probably way before) to find an effective Washington state tax rate this low.
Yes, Mr. Johnson is correct that state tax revenues are up. They are even up per capita in recent years. Yet despite this rise in total and per capita revenue, the collapse in state taxes as a percentage of personal income is having a devastating impact on the ability of the state to keep up with demand for public services.
One reason for this disconnect is the failure to understand that the cost of providing state and local government services at a constant level tends to rise faster than population plus inflation. As I’ve explained elsewhere, the Consumer Price Index is the wrong measure for tracking growth in the cost of government, as the highly-educated, labor-intensive services governments tend to provide (doctors, nurses, teachers, fire fighters, police, etc.) do not benefit from the same sort of productivity gains that globalization and technology have bestowed on economic sectors such as manufacturing. You can’t offshore firefighting to low-wage workers in China. You can’t put robocops on the beat. And you can’t squeeze productivity gains (as traditionally measured) out of our school teachers without substantially increasing class sizes.
Thus, top-line budget numbers aside, as the effective state tax rate declines, we should expect the effective size of state government to shrink with it. And that is exactly what see:
State workers are in fact the state’s largest expense, and in this chart we don’t see much recent growth in full time equivalent employment, period—let alone adjusted to population. From 1970 through 2000, state government employment largely tracked population growth, but even before the onset of the Great Recession, the gap had started to widen. By 2018, state government employment was less than 2.2% higher than it was in 2009, while the state’s population had grown by more than 12.9%. This failure of staffing to keep pace with the growth of relevant populations can be seen throughout state government, and is particularly prominent in a number of crucial services:
As the charts above illustrate, in childcare, elder care, and in higher education, the state is consistently failing to keep up with population growth. And if charts were available for housing, transportation, and many other services, they would paint a similar picture. But the state isn’t just failing to keep pace with its growing population. It’s also failing to keep pace with growing demand.
It turns out that the economic metric that most closely tracks the growth in demand for public services is in fact growth in personal income. This is because most of the services provided by government are commodities — and as our income increases, so does consumption. As cities, states, and nations grow more affluent their governments grow to consume a larger percentage of the economy as a whole, because wealthy economies demand both more and better government services. Think about it. Why do wealthy municipalities like Mercer Island spend more per capita on schools, parks, roads, police, and other services than less affluent communities? Because their taxpayers demand it!
Of course also, they can afford it. On the revenue side, percentage of personal income is the relevant measure of what we can afford to pay. For example, if you earn $50,000 this year and pay $5,000 in taxes, and then earn $100,000 next year and pay $10,000 in taxes, your tax bill will double but your so-called tax “burden” will not. In fact, even at a flat 10% tax rate, your true burden will have gone down. This is because the marginal utility of your 100,000th dollar (the benefit you derive from it) is less than that of your 50,000th dollar. Not only can you afford to pay more taxes the more you earn, you can afford to pay taxes at a higher marginal rate. That is why if our state tax system was efficient, it would levy taxes progressively.
Unfortunately, Washington has by far the most regressive tax structure in the nation—if you’re in the top 1% of earners, you pay less than 3% of your income in state and local taxes; if you’re in the bottom 20%, you pay a whopping 17.8%. It is in fact this highly unfair and inefficient tax structure that is at the root of all of our budgetary problems. Lacking an income tax, Washington relied on sales taxes for 58.5% of state tax revenue in 2018. No state relies on sales taxes more. Yet the retail sale of goods has been shrinking as portion of our overall economy for almost a century (for example, from 32% in 1959 to 28% in 2000), dragging taxes as a percentage of personal income down with it. In any meaningful sense, as the effective rate on aggregate personal income shrinks, the size of Washington’s government shrinks too, relative to the wants and needs of Washington’s people.
Washington’s budget isn’t growing too fast. It is growing too slowly.
In his op-ed, Mr. Johnson rhetorically asks, “How much longer can Washington afford to continue on this path?” And the answer is, not much—though not at all for the reasons he implies. It is not state spending that is the problem, but state revenue. Washington is suffering from a structural revenue deficit caused by an antiquated and regressive tax structure that locks in a falling effective aggregate tax rate over time. And there is simply no way to fix this deficit without taxing income or wealth.
