An acquired taste for public goodies

Charlene Nelson is on a mission to limit government's ability to spend your money.

Nelson is the founder, co-chair and spokeswoman of Protect and Empower the People (PEP), a newly formed group in North Dakota bent on curbing government's appetite for taxpayers' money, because she believes "without a doubt" that government is overspending.

Total government outlays are increasing in North Dakota, Nelson said, "but I don't see a huge outcry from citizens for additional services." Indeed, given stagnant or declining state population for the last decade, she added, "the need for government services is decreasing. But government leaders don't see that."

So Nelson founded PEP in the spring of last year after being approached by "two dozen people at least ... of all political stripes and professions" who said they "were getting eaten alive by taxes, especially property taxes." The group's central focus is to get a proposal on the November 2004 ballot that would require all state and local taxing jurisdictions to get 60 percent approval in a voter referendum if they wanted to raise taxes. The intent is not necessarily to strangle government budgets, Nelson said, but to make the process of increasing taxes more deliberate "so [that] whatever decisions are made are done with general [taxpayer] interests in mind and not just small vocal interests."

The group must get about 26,000 signatures by August, and by early last November it had "a few thousand," Nelson said. But she added that "80 percent to 90 percent of the people we approach are willing to support it."

The same rhetorical theme is evident across the district: Government at all levels is spending more money—too much tax money—for services of dubious value to most taxpayers, the argument goes. Much of the discussion stems from widespread state budget shortfalls, which some states have seen three years running. The sentiment for "no new taxes" helped elect new governors in Minnesota and Wisconsin; in the latter state, a Democratic challenger beat a Republican incumbent, with both running no-tax-increase campaigns. In both states, the gubernatorial winner faced a multibillion dollar state budget shortfall while having one revenue arm tied behind his back.

The inference that government is overspending today is almost taken as fact—at least cumulatively, if not for every program. Blame is widely dispersed to all parties. Tax-and-spend liberals. Pork-barrel Republicans. Unfunded mandates. Fat-cat lobbyists and "everyone else's" special interests. According to a Montana state senator, "It is an incredibly dismal situation."

In a republic by the people, for the people, it's also arguable that elected bodies are simply giving the public what it's asking for. If spending reflects the desires of the average citizen, as some contend, the lesson to be learned about supposed government overspending is more complex than is often portrayed.

Whether government is spending too much money is an issue that poses more questions than answers. Some contend that the role of government and government spending "depends on your philosophical viewpoint," as another Montana state senator put it. Others contend that government spending is an economic issue, where returns on public investments must be weighed against returns on private investments.

This issue of the fedgazette looks at a trail of related questions concerning government spending. First it identifies trends in government spending compared with personal income over the last two and a half decades. Next, it looks at public opinion polls and voter referenda for insights regarding the public's appetite for government services and its willingness to pay for those public goods. Lastly, it looks at how district states have dealt with serious budget shortfalls and whether the resulting budget solutions suggest a changing pattern in both government spending and the public's desire for such goods.

Well, so whadya find out?

The fedgazette analyzed the combined expenditures of state and local government in each district state over a 23-year period (1977 to 2000 because these data are the most recent and comprehensive) using data from the federal bureaus of Economic Analysis and the U.S. Census.

Source: U.S. Census Bureau

Growth in local and state expenditures were then compared (on a per capita basis) with each state's growth in per capita personal income over this same 23-year period. The fedgazette then sought to understand the ebb and flow of public spending to uncover funding priorities over time in each district state.

Source: U.S. Census Bureau

The central finding of this research is fairly simple: Combined spending by state and local governments from 1977 to 2000 rose slightly faster than income growth. But increases in government spending were not consistent among district states or for individual categories of public goods. (See charts for summary data and additional findings for district states and major spending categories. [xls])

Total local and state spending. District states fell in the same ballpark as the national average when it came to total growth in local and state government expenditures compared with income growth. Michigan, North Dakota and Wisconsin were above the U.S. average, Montana was the same, and Minnesota and South Dakota came in below the national average.

South Dakota was the only district state where per capita state and local public expenditures grew more slowly than per capita personal income over the 23-year period—though it just barely got under that bar. The other five district states saw local and state public expenditures grow slightly to modestly faster than income, with Wisconsin experiencing the fastest overall growth. More recently, however, government spending growth has slowed compared with income growth. From about 1995 to 2000, only Montana and Michigan saw local and state expenditures grow faster than income.

A cross section of states was also analyzed for the relationship, if any, between states' per capita income and government as a share of income at particular points in time (in other words, the relationship between average state income levels—be they high or low—and the percentage of that income that goes to government in any given year). There was little correlation, which indicates that state-specific characteristics affect government spending more than income levels. However, a correlation did appear when the relationship was analyzed over time. As states grew richer, spending on government increased proportionally.

K-12 education. Average state and local K-12 spending was slightly lower than average income growth with the exception of Michigan and Wisconsin, where K-12 spending was modestly higher than income growth. However, there was a considerable amount of fluctuation in comparative spending rates in all states over time.

Higher education. Average spending growth in higher education saw wide disparity among states. In Michigan, Montana and North Dakota, local and state expenditures for higher education grew moderately to much faster than income, while spending in the other three states was slightly to much slower. South Dakota's average rate of spending growth for higher education was easily the lowest among district states.

During the 1990s, both North Dakota and Minnesota saw significant declines in the growth rate of higher education spending compared with income. Montana's rate of spending growth in higher education skyrocketed compared with personal income between 1993 and 2000, apparently the result not of increased state appropriations, but much higher fees in the form of tuition—a trend also evident in other district states.

Health and welfare. Average local and state spending on health and welfare (which includes public hospitals and health programs, the state's portion of Medicaid not including federal transfers, and assistance to the poor) grew as fast or faster than average income in all states, but there has been significant fluctuation over the 23-year period. Three states experienced growth that was on par with or higher than income growth but were below the national average.

North Dakota, Minnesota and Montana saw average spending growth on health and welfare that was considerably faster than average income growth. However, average growth in health and welfare spending trended down strongly in most states during the last half of the 1990s, the likely result of changes in welfare programs in 1996 and before significant increases in Medicaid enrollments and rising health care expenditures, whose costs are also shared with the federal government.

Public safety. State and local spending on public safety (which includes corrections, the judicial system, and police and fire protection) had the highest average growth against income among all categories, with all district states posting moderate to high growth during the 23-year period, with Montana and Wisconsin leading the way. Montana and North Dakota have seen particularly strong spending growth rates in this category since the early to mid-1990s.

Transportation. For transportation (roads and bridges), local and state spending increased at a rate slower than personal income for all states. Wisconsin's highway spending growth rate was the highest among district states (virtually on par with income growth), but that masks a legacy of high per capita spending that the Dakotas and Montana have for their transportation systems.

Own-source revenue and the spending effect of federal transfers. Own-source revenue is that collected specifically through state and local taxes or fees. Put another way, it's the sum of all state and local revenue minus federal intergovernmental transfers. District states saw own-source revenue grow faster than personal income over the 23-year period, though there have been small declines in the growth rate since the mid-1980s.

At the same time, the Dakotas, Montana and (to a lesser degree) Michigan have seen strong growth overall in federal transfers since 1983, while growth of federal transfers to Minnesota and Wisconsin was significantly lower than growth in personal income or tax revenue from in-state sources. However, from 1995 to 2000, the rate of growth in federal transfers has been negative among all states nationwide and in four of six district states. Only Montana and South Dakota continued to see federal transfers increase over that period, with Montana's growing significantly.

(For more information on individual states and categories, see the charts, the underlying data, along with additional "income elasticity of demand" charts for individual spending categories.)

Un-surprised party

That central finding—local and state government spending has grown faster than personal income—did not come as a shock to many of the dozens of sources contacted for this project. One particularly common theme among district legislators involved government's penchant for simply piling new programs onto old, without much consideration for whether the old programs are needed.

"Government responds to new needs and technological changes, but it doesn't give up old, outdated, special-interest expenditures due to political clout of affected industries," said Dick Kelly, a Republican state senator in South Dakota, via e-mail. "Government, with its unilateral taxing authority, bows to pressure of small groups or to the desire of the bureaucracy to keep its job."

Jim Peterson is a Republican in the Montana House, as well as a farmer and rancher, and owner of a small rural mercantile store. He wrote, "There are no easy answers, but I do know once a new program is approved, it's almost impossible to, at a later date, when budgets are tight, to cut programs. ... People want the services, but as our tax base shrinks in a state like Montana, the revenue simply is not there."

Linda Nelson, a Montana Democratic state senator, echoed Peterson. "During good times, we have created many good programs that, once created, are very hard to discontinue. The public does, indeed, expect their needs to be met in an expedient manner once they have been introduced to a service. However, legislators, lobbyists and bureaucrats are all at least partly to blame for instigating new programs."

Beware of caveats

Readers should be aware of a single, overriding caveat to this research on government spending—namely, that there are a lot of smaller caveats to keep in mind.

Just as the nature of government spending is subject to considerable interpretation, and therefore debate, so too are data on government spending, because there are innumerable data points to choose from and many ways to interpret them in combination.

• For starters, the fedgazette chose to analyze combined state and local government expenditures, rather than state and local spending separately, because the mix of public goods purchased by state and local authorities is not only different among the states but also changes regularly. For example, each state assumes a slightly different share of K-12 education expenditure, which means the local share is also different in each state.
  
  States also occasionally change their share of funding. The state of Minnesota, for example, assumed 100 percent of K-12 funding under the recently ended Ventura administration, and Wisconsin had paid two-thirds of K-12 costs since the mid-1990s but repealed that responsibility in its latest budget.

• Figures on combined state and local spending also include federal payments made to (and spent by) state and local governments. But they do not include spending from federally run programs—like Medicare, Social Security, defense and farm payments—that occurs within states.
  
  Some might argue that including federal intergovernmental transfers as part of state and local spending skews the results, particularly for those states (like the Dakotas and Montana) that tend to receive proportionally more per capita. But in the long run, spending is spending, regardless of source, and it can be said that federal funding displaces or substitutes for at least some spending at the state and local levels. In other words, if federal funding was to suddenly vanish, state and local taxes would likely go up to fill the hole. One can only guess how much funding substitution is occurring, but including federal transfers in the analysis helps round out the full spending picture.

• Rather than use annual figures, which tend to have a lot of volatility, all spending and income comparisons use a five-year moving average. Under such a methodology, five-year periods (say, 1991 to 1995) for spending on government services and personal income are added and then divided by five to arrive at a "smoother" annual average. The five-year average for both categories is then divided by the average of the previous five-year period (in this case, 1986 to 1990) to generate a growth rate. Per capita figures for both income and spending were used to control for any impact that population changes might have had on these data.

• The central comparison used here—rate of spending growth against the rate of personal income growth—provides only one view of the spending prism. Other important measures of government spending over time include base spending and base income, because they provide a context for the growth path of both. Comparatively speaking, Minnesota is a high-spending state and Montana is a low-spending state in terms of per capita dollars spent on public goods.
  
  That generalization holds true even though Montana's spending-to-income growth was higher than Minnesota's over the 23-year period analyzed. The reason for this is the disparity in per capita income and government spending in the base year, 1977. Despite a higher rate of spending, Montana was far enough behind Minnesota that it didn't significantly close the annual spending gap.

• Lastly, none of the data presented here have much to say about either spending efficiency or the return on public investment (what we get for the money spent). These two important measures are beyond the scope of this project but are critical in answering that elusive question of whether government is spending too much, too little or just the right amount.

Research for these articles was contributed by Tobias Madden, regional economist; Rob Grunewald, regional economic analyst; research assistants Eva Lassemo and Shefali Mehta; and intern Naomi Cytron.