In response to the crisis, Congress and the Federal Reserve have provided cities and states with hundreds of billions of dollars in aid. But there are calls for more from the Fed chief, lobby groups such as the National League of Cities, and Democrats and some Republicans on Capitol Hill.

News articles are whipping up fears of an apocalypse unless Congress passes another state-local aid package. Politico claims that states and cities are “slashing” services with “severe” cuts, while Education Week worries about “draconian” cuts to schools. The New York Times says that the virus is “ravaging” state budgets, while Bloomberg worries about California’s looming “budget disaster.”

Such scare stories were common during the Great Recession a decade ago. But we can look back and see that the overall budget adjustments at the time were not that severe, particularly for local governments, which I examine here.

A recent study by the Council of Foreign Relations in support of more state-local aid gets the history wrong with regard to local governments: “The deep economic recession of December 2007 to June 2009 and slow recovery put many subnational budgets in unusually dire straits … The situation was particularly bleak at the local level, where many balance sheets were battered by the collapse of home values and property tax revenues.”

In fact, nationwide local tax revenues did not fall during the Great Recession, as BEA data from Table 3.21 shows in the chart. Property tax revenues—which account for about 72 percent of local tax revenues—remained robust. Revenues rose during the 2000s, flat-lined for a few years during the recession, then started growing again. Even though nationwide home prices dropped more than 20 percent from 2007 to 2011, local governments are slow to adjust valuations which smooths property tax revenues over time.

We will likely see a similar pattern this time. Housing prices do not appear to be falling much, so property tax revenues should remain steady. Zillow expects housing prices to fall 2 percent this year, while Real​tor​.com expects they will rise 1 percent.

However, prices may decline substantially for commercial properties in some areas, thus suppressing property tax revenues for some governments. And some big cities such as New York may face deep financial problems as people and businesses flee to the suburbs and other states permanently in response to the virus, harsh shutdown laws, high taxes, and urban civil disorder.

For most local governments, however, tax revenues should stay quite strong, as during the last recession. If governments hold spending flat for a couple of years, they should be fine. News stories often call flat spending a “cut.” Let’s say a city passed a budget pre-virus proposing to increase spending 5 percent from $100 million to $105 million, but now with the recession the city prudently decides to withdraw the $5 million increase. That is not slashing or draconian, but it is often portrayed as such.

Anyway, many cities are making sensible and modest budget adjustments. The City of Boston had planned to increase its budget $154 million or 4.4 percent in 2021, but with the recession, it has trimmed $65 million from the increase by freezing hiring, cutting police overtime, and other adjustments. San Antonio’s budget will be roughly flat next year as it trims such items as street improvement projects and economic development incentives, while imposing a hiring freeze, employee furloughs, and cuts to executive pay.

State and local governments are not subdivisions of the federal government. They have the power to adjust their spending, taxes, and borrowing to handle the recession. They can also speed the return to economic growth by removing regulatory barriers to entrepreneurs who want to start businesses and create jobs. More businesses and jobs mean more revenue for governments.

The more aid that Congress provides, the less incentive for states and cities to improve operational efficiencies and to deregulate to spur broad-based economic growth.

In the continuing aftermath of George Floyd’s death at the hands of Minnesota police, qualified immunity has come to the forefront. By protecting police officers from liability, even when they violate people’s constitutional rights, this doctrine has become the cornerstone of our near-zero accountability policy for law enforcement. Shamefully, the Supreme Court recently declined the perfect opportunity to address the mess that it made by inventing this doctrine. But Congress has also turned its attention to the issue, and there are now several pending legislative proposals to abolish qualified immunity. And that development has compelled the few proponents of the doctrine to actually put forward arguments in its defense.

Some of these arguments may sound reasonable, but they all rest on misunderstandings or misrepresentations about what qualified immunity actually is, and how it works. I’ve already discussed some of these points in this recent post, as well as on Unlawful Shield’s FAQs about qualified immunity. But I thought it would be helpful to collect and rebut in one place all of the most common arguments that I’ve heard come up in recent congressional hearings and other public commentary.

1. “We need qualified immunity so that police won’t hesitate when they have to make split-second, life-or-death decisions.”

It is certainly true that police officers have to make difficult, on-the-spot decisions under conditions of danger and uncertainty. So, the argument goes, it is both unfair and unwise for courts to second guess these decisions, and holding officers personally liable whenever they make the “wrong call” will deter them from carrying out their duties in the first place.

This argument is reasonable, but it has nothing to do with qualified immunity. Our underlying legal standards for determining whether a constitutional violation occurred in the first place are already highly deferential to on‐​the‐​spot police decision‐​making. The Supreme Court’s decision in Graham v. Connor sets out an “objective reasonableness” standard for excessive-force claims, which makes clear that courts cannot second guess on-the-spot policing decisions:

The “reasonableness” of a particular use of force must be judged from the perspective of a reasonable officer on the scene, rather than with the 20/20 vision of hindsight. The Fourth Amendment is not violated by an arrest based on probable cause, even though the wrong person is arrested, nor by the mistaken execution of a valid search warrant on the wrong premises. With respect to a claim of excessive force, the same standard of reasonableness at the moment applies: “Not every push or shove, even if it may later seem unnecessary in the peace of a judge’s chambers,” violates the Fourth Amendment. The calculus of reasonableness must embody allowance for the fact that police officers are often forced to make split-second judgments — in circumstances that are tense, uncertain, and rapidly evolving — about the amount of force that is necessary in a particular situation.

To clarify, this decision has nothing at all to do with qualified immunity. In Graham, the Court was simply explaining that unless an officer acts objectively unreasonable, under the circumstances known to them at the time, they haven’t violated the Fourth Amendment at all.

Keep in mind that qualified immunity only matters in the circumstance where (1) a civil rights plaintiff can show that the defendant did violate their constitutional rights, but (2) those rights were not “clearly established” at the time of the violation (i.e., there was no prior case involving nearly identical facts). So if a police officer didn’t commit a constitutional violation in the first place, then they don’t need qualified immunity to protect them. Qualified immunity is therefore entirely unnecessary to ensure that police can make quick, split-second decisions, because that protection is already baked into our Fourth Amendment jurisprudence. If we eliminated qualified immunity tomorrow, that protection would remain untouched.

Moreover, the very fact that officers do have to make difficult, split-second decisions underscores how the “clearly established law” standard rests on a bizarre legal fiction. After all, the way that standard works is that police cannot be held liable for their misconduct unless a previous judicial decision has already held that the particular misconduct they committed was unconstitutional. But this supposes that individual police officers are carefully reading the most recent judicial decisions by their relevant appellate court and comparing the exact fact patterns of those cases to scenarios they encounter on the street when making split-second decisions. This is, obviously, not how policing actually works. As one federal judge recently stated:

The Supreme Court’s obsession with the clearly established prong assumes that officers are routinely reading Supreme Court and Tenth Circuit opinions in their spare time, carefully comparing the facts in these qualified immunity cases with the circumstances they confront in their day-to-day police work. It is hard enough for the federal judiciary to embark on such an exercise, let alone likely that police officers are endeavoring to parse opinions. It is far more likely that, in their training and continuing education, police officers are taught general principles, and, in the intense atmosphere of an arrest, police officers rely on these general principles, rather than engaging in a detailed comparison of their situation with a previous Supreme Court or published Tenth Circuit case. It strains credulity to believe that a reasonable officer, as he is approaching a suspect to arrest, is thinking to himself: “Are the facts here anything like the facts in York v. City of Las Cruces?”

In other words, qualified immunity is not only unnecessary for ensuring that police have discretion to make on-the-spot decisions, but also practically incapable of providing the sort of protection that proponents claim is needed.

2. “Eliminating qualified immunity would negatively impact recruitment and retention of police officers because they would fear being sued.”

There are many, many things wrong with this argument.

First, as I discussed above, even in the absence of qualified immunity, it is by no means easy to demonstrate that a police officer violated your constitutional rights. Our Fourth Amendment jurisprudence incorporates tremendous deference to police decision-making, and officers who genuinely make reasonable, good-faith decisions about arrests, use of force, etc., are already protected from being sued. The idea that eliminating qualified immunity would somehow lead to ruinous lawsuits anytime someone disagreed with an officer’s use of force is reckless, inaccurate fear-mongering.

Second, even today, police officers are nearly always indemnified for any settlements or judgments against them, meaning that their municipal employers, not the officers themselves, actually end up paying. Joanna Schwartz, a UCLA law professor and one of the foremost scholars of qualified immunity, demonstrated in a 2014 article called Police Indemnification that, in her study period, “governments paid approximately 99.98% of the dollars that plaintiffs recovered in lawsuits alleging civil rights violations by law enforcement.” In other words, even when plaintiffs do overcome qualified immunity, the individual police officers rarely pay a dime.

I have written elsewhere about how this practice of near-automatic indemnification is itself problematic, because it fails to provide for individualized accountability for officers who violate people’s rights. A better practice, as my colleague Clark Neily has also discussed, would be to take some portion of the money that municipalities already spend on civil rights judgments, and instead put that toward an insurance allowance for individual officers. That way, officers would have a clear, direct interest in making sure their own behavior conformed to constitutional standards — because unprofessional officers who routinely committed misconduct would see their premiums go up, and would eventually be priced out of the market.

Nevertheless, as things currently stand, individual officers are almost never required to pay anything, even when they are found to have committed a constitutional violation. The immediate effect of eliminating qualified immunity would not be to subject these officers to huge judgments, but rather to ensure that victims whose rights are violated have a remedy. 

Third, keep in mind that qualified immunity means law enforcement is held to a far lower standard of liability than basically any other profession. Imagine, for example, that doctors started arguing that they should no longer be liable for malpractice, unless some other doctor had already been held liable for committing the very same kind of malpractice. Nobody would take this argument seriously, and rightfully so. The reality is that doctors, lawyers, accountants, engineers, architects, and professionals of all sorts expect to face liability if they commit professional misconduct. And in every context except for law enforcement, we comfortably accept the premises that the potential for liability (1) justly compensates the victims of professional misconduct, (2) encourages professionals to conform their behavior to legal standards, and (3) doesn’t discourage people from entering the profession in the first place.

So, the relevant question should not be “would eliminating qualified immunity deter people from becoming police officers?” It should be “why should police officers, unlike every other profession in our society, have special protection against liability, even when they break the law?”

Fourth, the proponents of qualified immunity are profoundly mistaken if they think the doctrine is doing the law enforcement community any favors. To the contrary, qualified immunity has dramatically undermined trust and respect for police officers, by telling the public that members of law enforcement can routinely get away with egregious misconduct. There is ample evidence that policing is far more difficult and dangerous when officers lack community trust, which means that qualified immunity is making it harder, not easier, for police to do their jobs.

Indeed, the more thoughtful members of law enforcement have already recognized as much, and for that very reason (among others) have called for an end to qualified immunity. For example:

• The Law Enforcement Action Partnership (“LEAP”) has joined several of Cato’s amicus briefs calling for the Supreme Court to reconsider qualified immunity, and they recently put out a set of policy recommendations stating that “Congress should pass a law to ensure that the qualified immunity doctrine does not stop officers who break the law from being held legally accountable.” LEAP’s Executive Director, Major Neill Franklin (Ret.), explained that “[a]ccountability measures that show an agency is serious about respecting the rights of all of its residents help the police as much as they help the communities we serve. There’s no better way to restore community trust. And we cannot do our jobs without trust.”

• Ronald Davis, Chair of the Legislative Committee of the National Organization for Black Law Enforcement Executives (“NOBLE”), and a retired chief of police, submitted testimony to the House Judiciary Committee, in which he said that “the first action step in reconstructing a new policing system is to take immediate steps in strengthening police accountability and building trust with communities.” To that end, he recommended that Congress “[e]nd the qualified immunity doctrine which prevents police from being held legally accountable when they break the law.”

3. “Qualified immunity is necessary to prevent frivolous lawsuits against police officers.”

Whether or not you think “frivolous civil-rights litigation” is a serious problem, it’s a problem that qualified immunity, by its very nature, is incapable of addressing.

There are basically two things we might mean by saying that a particular lawsuit is “frivolous.” First, it could mean that a lawsuit is not legally meritorious, meaning that the facts alleged, even if true, simply do not make out a constitutional violation at all. If that’s the case, then qualified immunity, by definition, is unnecessary to dismiss the lawsuit, because qualified immunity only matters when the defendant has committed an actual constitutional violation, but where a court nonetheless determines that the law wasn’t “clearly established.” If the underlying lawsuit is meritless, however, then it can be dismissed for “failure to state a claim upon which relief can be granted,” without any need to invoke qualified immunity.

Second, a “frivolous” lawsuit could be one that is not factually supported — in other words, maybe the facts alleged, if true, would make out a constitutional violation, but the plaintiff is either mistaken or lying about the facts. But in that case, qualified immunity does little to help in dismissing the case, because of course, plaintiffs could theoretically lie their way around qualified immunity as well, just by alleging facts that do happen to closely match the fact patterns of prior cases.

The tools that we use to address and deter frivolous litigation are entirely separate from qualified immunity. Heightened pleading standards require plaintiffs to make specific, factual, non-conclusory allegations showing that they are entitled to relief. Rule 11 of the Federal Rules of Civil Procedure requires attorneys to attest that they have a good-faith basis for the factual and legal arguments in all submitted pleadings, and it provides for sanctions if they fail to meet this standard. Depending on the particular subject matter and context, more stringent requirements may apply. Rule 9(b) of the Federal Rules of Civil Procedure imposes extra pleading requirements for alleging fraud; the “anti-SLAPP laws” enacted by many states allow for early dismissal of frivolous defamation claims; and the Prison Litigation Reform Act of 1996 limited the ability of prisoners to bring successive, non-meritorious lawsuits.

Assuming there is a problem with frivolous civil rights litigation, addressing it will require rules like these. But qualified immunity does basically nothing to stop “frivolous” lawsuits because, again, the doctrine only kicks in when the underlying lawsuit is meritorious. The idea that eliminated qualified immunity will result in a wave of frivolous litigation is more baseless fear-mongering — qualified immunity does nothing now to prevent such litigation, so we shouldn’t expect a major change in this regard if we abolish the doctrine.

4. “Qualified immunity protects police officers from the time and expense of litigation by quickly filtering out bad lawsuits.”

This is a somewhat more sophisticated version of the previous defense, but it’s still mistaken. The argument goes like this: “Yes, in theory, non-meritorious lawsuits should still end up failing, even in the absence of qualified immunity. But it will take substantial time and resources for officers to successfully defend themselves against these lawsuits. Therefore, even if qualified immunity ends up catching some meritorious cases as well, the doctrine is worth the cost, because it will let defendants quickly and easily dismiss the frivolous ones.”

Throughout the coronavirus pandemic, Democratic governors have accused the Trump administration of favoritism in determining how the federal government has allocated medical supplies and aid to the states. Critics claim that the process has been politicized, with the administration assisting Republican states while leaving many Democratic states to fend for themselves. A recent New York Times article outlines concerns that $16 billion in pandemic relief for farmers will be used as another tool by the Trump administration to lock up support from farmers and buy votes in key battleground states.

As the article notes, the federal government distributed $28 billion in trade relief to pay farmers for the negative effects of increased tariffs in 2018 and 2019. Critics allege that the payments were excessive, estimated to be one-and-a-half to 33 times more than farmers lost because of economic damage from the trade war, and are a ploy to maintain support for President Trump among farmers, a group that is favorable to Trump and plays an important role in swing states.

Critics also argue that the trade relief payments were distributed mainly to states favorable to Trump—primarily southern states. A report by the Democratic staff of the Senate agriculture committee found that, in 2019, Georgia, Mississippi, Alabama, Tennessee, and Arkansas received the highest amounts of trade relief in the first round of payments. The concern is that the coronavirus relief will be no different and, so far, of the $500,000 in coronavirus relief distributed to farmers, the important swing state Wisconsin has received the third-highest share.

Because the pandemic aid has yet to be fully disbursed it is too early to know if there will be favoritism. But history suggests that the politicization of disaster relief is nothing new. In the summer issue of Regulation, Steven Horwitz and E. Frank Stephenson summarize the research on how politics has influenced the allocation of disaster aid going back to the New Deal. The findings are clear: relief funds, disaster declarations, and the allocation of medical supplies have frequently been used by presidents and legislators to reward supporters and buy votes.

The distribution of loans and relief during the New Deal is a prime example of how political factors can dictate how funds are steered. Economic historian Gavin Wright explored why western states received a large amount of aid relative to the South, which received a comparatively small amount. He found that the variability of Democratic voting in western states played an important role in influencing where the Roosevelt administration sent funds, implying that the administration sought to buy votes in battleground states while the South was less important because it consistently voted Democratic. Relevant to coronavirus farm relief, additional research also found that New Deal aid to farmers in Alabama counties was positively correlated with Roosevelt’s vote share.

As the federal government’s role in responding to natural disasters has increased, the allocation of Federal Emergency Management Agency (FEMA) spending has also been politicized. Multiple studies have found that presidential disaster declarations, which make regions eligible for FEMA disaster relief, increased in election years and that disaster declarations were more frequent in swing states. States with more representation on subcommittees overseeing FEMA also received more disaster relief. One of the studies, by Thomas Garrett and Russell Sobel, concludes that from 1991 to 1999, “nearly half of all FEMA disaster relief is explained by political influence rather than actual need.”

And separate research of agricultural disaster payments for areas with crop failures, droughts, floods, and other disasters found that states represented on Congressional committees overseeing agriculture and agricultural appropriations received more disaster relief. As much as 30% of agricultural disaster relief is estimated to be based on political influence instead of actual crop losses.

Most relevant to the current pandemic, research found that H1N1 (swine flu) vaccine allocations were politicized. First available in October 2009, doses of the vaccine were distributed by the Department of Health and Human Services which is overseen by the House Committee on Energy and Commerce. A study by Matt Ryan indicates that states received an additional 60,000 doses for each Democratic member the state had on the Committee.

Horwitz and Stephenson summarize the cause of this politicization nicely:

These outcomes are not the product of particular personalities or partisan affiliations, but of the incentives created by the institutional structure of politics. Politicized disaster relief is nothing more than politics as usual.

It is too early to identify whether the Trump administration’s decisions on how to allocate medical supplies, aid, and agricultural relief are truly political. But it should be no surprise if they are. At this point, the politicization of disaster relief is an American tradition.

Cities across the country are struggling to make their debt payments on municipal stadiums in an era of canceled events, report Sebastian Pellejero and Heather Gillers in the Wall Street Journal:

Public officials have borrowed billions of dollars to build stadiums for major teams. Since 2000, more than 40% of almost $17 billion in tax-exempt municipal bonds sold to finance major-league stadiums were backed by levies on hotels and rental cars—making tourism taxes the predominant means of public stadium finance, according to the Brookings Institution.

The borrowers envisioned the sports facilities as a form of economic development that would attract fans from near and far, raising cities’ national profile and boosting their revenue beyond what was needed to pay back the bonds. The pandemic has turned that calculus on its head, crushing tourism proceeds and turning stadiums into a strain on city budgets.

But that calculus was never very sound, as I’ve written here before. Academic studies have consistently found few if any economic benefits of subsidies for stadiums, arenas, convention centers, and the like.


Several Cato studies over the years have looked at the absurd economic claims of stadium advocates. In “Sports Pork: The Costly Relationship between Major League Sports and Government,” Raymond Keating finds:

The lone beneficiaries of sports subsidies are team owners and players. The existence of what economists call the “substitution effect” (in terms of the stadium game, leisure dollars will be spent one way or another whether a stadium exists or not), the dubiousness of the Keynesian multiplier, the offsetting impact of a negative multiplier, the inefficiency of government, and the negatives of higher taxes all argue against government sports subsidies. Indeed, the results of studies on changes in the economy resulting from the presence of stadiums, arenas, and sports teams show no positive economic impact from professional sports – or a possible negative effect.

In Regulation magazine (.pdf), Dennis Coates and Brad Humphreys found that the economic literature on stadium subsidies comes to consistent conclusions:

The evidence suggests that attracting a professional sports franchise to a city and building that franchise a new stadium or arena will have no effect on the growth rate of real per capita income and may reduce the level of real per capita income in that city.

And in “Caught Stealing: Debunking the Economic Case for D.C. Baseball,” Coates and Humphreys looked specifically at the economics of the new baseball stadium in Washington, D.C., and found similar results:

Our conclusion, and that of nearly all academic economists studying this issue, is that professional sports generally have little, if any, positive effect on a city’s economy. The net economic impact of professional sports in Washington, D.C., and the 36 other cities that hosted professional sports teams over nearly 30 years, was a reduction in real per capita income over the entire metropolitan area.

In an updated study from the Mercatus Center at George Mason University, Coates finds similar results:

• Professional sports can have some impact on the economy. Looking at all the sports variables, including presence of franchises, arrival and departure of clubs in a metropolitan area, and stadium and arena construction, the study finds that the presence of a franchise is a statistically significant factor in explaining personal income per capita, wage and salary disbursements, and wages per job.
• But this impact tends to be negative. Individual coefficients, such as stadium or arena construction, sometimes have no impact, but frequently indicate harmful effects of sports on per capita income, wage and salary disbursements, and wages per job.

Michael Farren of Mercatus summed up, “Furthermore, peer-reviewed academic research consistently shows that public financing for professional sports stadiums is a poor way to spur economic development or accomplish ‘downtown revitalization.’ ”

The pandemic and the event cancellations it has generated have seriously disrupted the financial calculations that cities made in building stadiums at taxpayers’ expense. But they were never a good bargain.

Presidential candidate Joe Biden said that he does not favor defunding police but instead “conditioning federal aid to police based on whether or not they meet certain basic standards of decency and honorableness.” Biden is proposing to increase federal aid for state-local police forces.

The federal government currently spends relatively little on funding state and local police. Total federal aid to the states for the “administration of justice” in 2019 was $5.2 billion, a portion of which goes to police forces for programs such as COPS. Federal aid is a small share of total state-local police funding, but it is a tail that wags the dog in terms of federal control.

The chart below shows aid in this category since 1990 in actual or nominal dollars. (Data is from Table 12.3). The jump in spending in the 1990s stems from the high-profile 1994 crime bill.

Aid to the states for state-local police should be zeroed out for the same reasons it should be ended for other state-local activities, as discussed in this study. One problem is that federal politicians make big mistakes and then impose those big mistakes on the entire country. It is much safer if we let states and cities try to improve their policies with various approaches and then learn from each other’s successes and failures.

Remember that the 1994 crime bill was imposed on the country by a Democratic president and a Democratic Congress but is now viewed as a big mistake by the Democrats. Democrat Joe Biden helped write the 1994 crime bill and voted in favor, but he now apologizes for it. Let’s not make another big federal mistake.

Since the beginning of his presidential campaign, Donald Trump has prioritized imposing punitive tariffs on China. In a 2016 Republican presidential debate he said, “I’m totally open to a tariff. If they don’t treat us fairly, hey, their whole trade is tariffed. You can’t deal in China without tariffs. They do it to us, we don’t it. It’s not fair trade.”

True to his word, the Trump administration soon implemented steep tariffs on steel and aluminum, washing machines, and solar panels—specifically targeting China with an escalation of his trade war. Tariffs are bad economic policy—costing ordinary Americans hundreds if not thousands of dollars each year—and lead to predictable retaliation and escalation. In response to Trump’s tariffs on Chinese products, Beijing implemented high tariffs on US agricultural products. While previous executives might have heeded the advice of experts and advisors, Trump has not availed himself of their counsel. That a single political figure can implement such shortsighted, damaging policy is unconscionable.

As a result of the Chinese retaliation, the US agricultural export market faced serious challenges. In an effort to mitigate the self-inflicted damages President Trump directed the USDA to provide direct relief payments to the American farmers hit hardest by the trade war. This blatant attempt to buy off key constituents generated a predictable backlash. Now, government watchdogs are investigating the bailouts following allegations of improper management and allocation. Congress played no role; instead, unilateral actions of the executive branch directly impacted millions of Americans and spent billions of taxpayer dollars.

When the coronavirus pandemic hit, Congress acted rapidly to allocate stimulus and relief funding, including funds earmarked for the agricultural sector. However, the New York Times reports that:

Months before an election in which some farm states are major battlegrounds, Democrats and other critics of the administration’s agriculture policies are expressing concern that the new subsidies, provided by Congress with bipartisan backing, could be doled out to ensure President Trump continues to win the backing of one of his key voting blocs.

Between the trade war with China and the coronavirus pandemic, the Trump administration has spent $47 billion bailing out farmers. Time and time again the executive branch uses overly broad powers to implement bad policy and hand out political favors. The decisions on policy are unfortunate; but the blatant disregard for democratic decision-making is frightening.

Some protestors are calling for defunding the police in the wake of George Floyd’s death and other incidents of police misconduct. However, just 16 percent of Americans favor cutting police budgets, notes Emily Ekins. More people favor police department reforms such as better training and increased accountability.

That said, police budgets should be examined, particularly since policing costs vary widely across the country. Floridians, for example, should be asking why their police services cost twice as much per capita as police services in Indiana.

There should also be more focus on federal police spending. That includes spending on an alphabet soup of agencies such as the FBI, ATF, DEA, and CBP. This spending has risen much faster than state-local police spending in recent decades. That is a big concern if you believe in federalism and decentralized government.

As part of the national income accounts, the Bureau of Economic Analysis publishes data on police spending by the federal and state-local governments (Table 3.16). The data are charted below as a percent of GDP from 1960 to 2018. Spending on state-local police rose as a share of GDP from 1990 to 2010 but has declined since then. Spending as a share of GDP is about one-third higher today than in the 1980s.

Meanwhile, federal police spending has skyrocketed from 0.05 percent of GDP in the 1980s, on average, to 0.26 percent over the past decade. As a share of GDP, federal police spending has quadrupled since the 1980s and doubled since the 1990s. This mainly includes spending on federal functions such as the FBI, but also includes relatively small amounts for grants to state-local police agencies.

The BEA does not provide a breakdown of its measure of federal police spending, but you can see the components in federal budget accounts (function 751). There have been large spending increases in Customs and Border Protection, the Drug Enforcement Agency, the Secret Service, the Bureau of Alcohol, Tobacco, Firearms, and Explosives, and other federal policing agencies.