State and Local Spending Isn't the Problem

Governments nationwide need productivity gains

For the third straight year, state and local governments across the country are dealing with weak revenues, rising expenses, and the prospect of enormous budget deficits. In some cases, the problem has grown to a staggering size: Texas had to wrestle with a $10 billion deficit, while California is still struggling with a shortfall as large as $29 billion.

State legislators and governors have been stuck between two vocal groups. On one side are those who would raise taxes to save jobs and services. On the other are those who think that the deficits stem from a decade-long spending binge and that cutting jobs is precisely the way to go.

But both sides are wrong: There's no evidence that state and local spending is out of control. By BusinessWeek's estimates, employment in the state and local sector outside of public schools and the criminal justice system -- including such mundane jobs as garbage collection and recreation -- has grown only 8% since 1993. That compares with a 19% rise in the private sector over the same stretch. Pay for state and local workers has trailed that of private counterparts.

The only real increases in spending and employment have come in fields voters really wanted: education, criminal justice, and medical care. Unfortunately, those are precisely the areas where productivity is hard to measure and even harder to increase, even given all the advances in technology. As a result, while many states cut income tax rates in the 1990s, the combined state and local tax burden, including property and sales taxes, is near an all-time high.

Consider pay first. Unlike the 1980s, the '90s were a good time to be working for a private company rather than the government. Inflation-adjusted compensation in the public sector -- including both salary and benefits -- rose merely 6% from 1993 to 2002, compared with a 9% gain for private-sector jobs.

There's no sign of any pay boom, even for unionized groups such as teachers, police, and firefighters. Real median weekly earnings for these workers was more or less unchanged from 1993 to 2002, compared with a rise of 3% for workers in general. Similarly, benefits rose faster in the private sector than in the public sector.

What about public employees in a city such as New York, known for its strong unions? Surprisingly, a recent study from the Citizens Budget Commission estimates that from fiscal years 2000 to 2004, wages and salaries for New York municipal workers will rise at a 3.5% annual pace. That's not much faster than the 3% increase in wages and salaries for private-sector workers nationally over the last year. Moreover, benefits are 29% of compensation, in line with the private sector, and health-care costs are rising at about the same rate as in the private sector. The one big difference is pension costs, which have soared in New York -- but that is partly due to the fall in the stock market, cutting pension returns.

Nationally, there's no sign of any big expansion in state and local hiring, either, outside of public schools and criminal justice. From 1993 to 2002, public education systems added the equivalent of 1.2 million full-time positions, almost 60% of total hiring by state and local governments. A further 20% came from hiring additional police, judges, and correctional officers. Politicians did what constituents wanted: educated their kids and kept their streets safe.

State and local governments have recently begun cutting jobs (except in education), something not done in the previous recession. According to the latest bls data, noneducation state and local employment appears to have peaked in late 2002.

Nevertheless, the public sector has not been able to improve productivity as fast as the private sector -- and certainly not fast enough to hold down fiscal pressures. The bls does not provide productivity figures for state and local government, but there's a way to make a rough estimate. The output of state and local government can be thought of as the services that they provide, garbage collection being one example. As the population rises, more services are needed. If government workers are becoming more productive, then state and local employment should increase more slowly than the population.

That's what happened -- but the gain in productivity wasn't sufficient. From 1993 to 2000, the total U.S. population rose by about 9%, while public-sector workers, again not counting education and criminal justice, rose by 4%. That means "output" per worker rose by roughly 5%. By comparison, output per worker in the typical private-sector service industry rose by about 20% from 1993 to 2000.

There are productivity hurdles in education and criminal justice that may be hard to overcome. For example, during the past decade, the growth of education employees far outpaced enrollment growth. Pupil-teacher ratios have plunged sharply, from 17.4 in 1993 to 15.1 in 2001, the last year available.

Unfortunately, this increase in education workers per pupil was not accompanied by a similar rise in accomplishment. The mathematics performance of 12th-graders, as measured by the government's National Assessment of Educational Progress, was about the same in 2000 as it was in 1992. The performance of fourth- and eighth-graders inched up a bit, by about 3%.

Boosting this performance is possible but not easy. For example, it makes sense that school choice -- allowing parents to pick their children's school -- should put pressure on existing schools to improve their performance. In practice, competition seems to work, but the impact is not overwhelming. A recent study from the National Bureau of Economic Research shows that competition from charter schools can produce a 1% gain in achievement at traditional schools. That's passing, but hardly an A.

Improving the performance of the criminal justice system is even trickier. While police, court, and correctional jobs were soaring, crime was plunging, by 16% nationally. Under normal circumstances, that would make such spending a natural candidate for cutbacks or productivity improvements. Indeed, earlier this year, Oregon started laying off state troopers, letting people out of jail early, and closing courtrooms one day a week around the state in order to save money.

Across most of the country, shrinking the public-safety workforce seems shortsighted in a world where terrorist attacks are a real threat. And public support remains strong for anticrime measures: California, despite its budget woes, is still planning to spend $220 million expanding San Quentin State Prison's death row.

The question, then, is how to handle the fiscal problems of the states and cities, especially if the economy remains sluggish. In historical perspective, it's hard to justify any big increase in state and local taxes. In 2002, the state and local tax burden -- including levies on personal income, property, and sales -- was 9.3% of gross domestic product, far above the 40-year average of 8.7%, and just a hairbreadth below the record of 9.5% set in 2000. By comparison, the federal income tax is only 8.1% of gdp.

The bottom line: States and cities have been spending on the things that people want. The U.S. has made education a national priority, public safety is critical, and no one is ready yet to ration medical care.

The key to U.S. economic growth in the 1990s was the enormous increase in productivity across many industries. Unfortunately, productivity gains in education, crime-fighting, and health care are hard to come by. But that, in the end, may be the only way for states and cities to make permanent headway against their fiscal crunch.

By Michael J. Mandel in New York

    Before it's here, it's on the Bloomberg Terminal.