In the wake of Hurricane Sandy and the havoc it wrought on New York, Sen. Chuck Schumer (D-N.Y.) and Gov. Andrew Cuomo (D) said they would ask the federal government to cover at least 90 percent—and perhaps all—of the cleanup and recovery costs. New Jersey Gov. Chris Christie (R) asked the same for the Garden State. “This was not a New York disaster, or a New Jersey disaster or a Connecticut disaster, but a national disaster, and FEMA and the federal government should be providing help to the region to the full extent they can,” Schumer declared on November 1. “I will continue to push the feds to reimburse the city and state for the full costs of repair and recovery for all aspects of the disaster.”

These reflexive calls for Washington to pick up the tab underscore one of the greatest shifts of power in American politics during the last four decades: the transition from state and local autonomy to federal subsidy and control. This centralization of government was made possible largely by grants-in-aid, money provided by the federal government to state and local governments or private parties. They have become the third largest category in the federal budget, trailing only Social Security and national defense. 

According to the Congressional Research Service, there were 1,724 of these grants in fiscal year 2011, paying for things such as bridges, teachers, Medicaid, farm subsidies, and abstinence programs. The total cost of these federal grants was $515 billion, up 160 percent in real terms since the beginning of the 1990s and nearly 60 percent since 2000. After the adoption of the American Recovery and Reinvestment Act, a.k.a. the 2009 stimulus bill, grant spending increased by 16 percent in 2009 and 11 percent in 2010—the highest annual spikes in history. 

Grants are not merely a substantial part of the federal budget. They have become like a drug for the states. The federal share of total state spending rose from 25.7 percent in 2001 to 34.1 percent in 2011. State and local governments drink up roughly 80 percent of total federal grant spending, with the remainder going mostly to nonprofit organizations providing services at the state and local levels. In a February 2011 Tax and Budget Bulletin, Chris Edwards of the Cato Institute found that in 2010 the federal government was micromanaging primary education with 109 grant programs costing taxpayers a total of $86.5 billion. Yet under the Constitution, K−12 education is a state and local activity. It was federal funding that allowed states to increase overall spending dramatically during the last two decades, including during the recent economic downturn. 

The Government Accountability Office, Washington’s fiscal watchdog, has been pointing out chronic problems with federal grants, such as duplications, inefficiency, and waste, since at least 1975. The Office of Management and Budget regularly notes that grant programs have poorer performance, on average, than other federal programs. The transfers often come saddled with one-size-fits-all rules, which reduce healthy policy diversity and the incentives for states to find innovative solutions to common problems. 

Take homeland security grants. After September 11, 2001, the federal government dramatically expanded its “first responder” programs, essentially appropriating federal funds to pay for the salaries and equipment of local police and fire departments. Because the funding formula gives insufficient weight to areas of high risk, money aimed at fighting terrorism wound up paying for $200,000 in surveillance cameras in Dillingham, Alaska (population: 2,400), and $1.5 million worth of decontamination tents in Grand Forks County, North Dakota. 

Thanks to all the dollars up for grabs, grants designed to increase security wind up increasing lobbying activity instead. On March 4, 2004, just after the homeland security grants were first announced, 3,000 state officials flew into Washington to lobby Congress for first-responder money. Hard on their heels came a March 16 influx of firefighters from around the country. 

Data from the transparency watchdog OpenSecrets.org show that the International Association of Fire Fighters—the largest firefighter group in the country, with more than 30,000 members nationwide—has increased its lobbying spending significantly, from $60,000 in 2000 to $416,000 in 2011, with the biggest growth after 2005. Firefighting is no longer a local issue.

Instead of lobbying for this massive growth in federal spending, states and local communities should reassert responsibility for most of their preparedness efforts. If police officers feel they need more equipment to do their jobs, or firefighters need more training and gym memberships, they should turn to their local and state officials for funding.

Federalizing first-responder programs accentuates the public choice problem that already plagues the political process. A congressman from Wyoming has no incentive to admit that his state is not a likely terrorist target; on the contrary, that would mean turning down “free” money, which almost no successful politician ever does. By contrast, when first-responder programs are the responsibility of cities, counties, and states, legislators have an incentive to more accurately assess risk and the potential benefits from additional spending.

What is true for homeland security is also true for most other federal grants to the states, including anti-poverty spending. In fact, there is growing evidence that grants ostensibly aimed at helping the poor or boosting the economy are used by lawmakers as electoral currency to gain or reward constituent support. In a working paper, Vanderbilt University’s John Hudak argues that the president and his subordinates strategically direct federal funding to electorally competitive states and that the executive branch delivers more money to swing states than all other states combined.

This strategy, while cynical, is a winning one. A May 2012 American Political Science Review paper by Boston University economists Douglas L. Kriner and Andrew Reeves finds evidence that “voters reward incumbent presidents (or their party’s nominee) for increased federal spending in their communities.” This phenomenon is heightened in battleground states. 

My own research on the way stimulus spending was allocated showed that the party affiliation of a congressional district’s representative was a factor. Districts represented by Democrats got more money than districts represented by Republicans, even controlling for variables such as state capital, income, and unemployment. 

This system of federal aid should be scaled back and ultimately terminated. With today’s massive deficits, the federal government can no longer afford to pick up state and local tabs.

For a model of how to reform federal grants, Cato’s Edwards suggests looking to Canada. Our neighbors to the north cut their national debt as a share of GDP from 68 percent in the mid-1990s to just 34 percent today. One key part of that reform was cutting federal aid to provincial and local governments while consolidating the remaining aid into three large block grants. 

Over all, just 38 percent of government spending in Canada is done at the national level, compared to 71 percent in the United States. Canada has no federal department of education, for example. Far from hurting children, Canada’s localized approach to education has correlated with test scores that are typically higher than those of American kids. Some provinces have made great strides in school choice and other innovations. 

This could happen in the United States too. The federal government should warn states that it is turning off the grant spigot and then do it. When bills come due for regular, predicable expenditures such as education, states and localities should figure out how to make ends meet. If states stop depending on the feds for bailouts and start saving for a rainy day, they will finally be ready for the next hurricane.