Grade inflation has long been acknowledged as a severe problem in academia. Academic work that would have gotten a B a generation ago now gets an A. From 1990 to 2020, college grade point averages (GPAs) increased by 21.5 percent. At some colleges, such as Yale, As and A minuses account for 80 percent of grades.

Harvard’s brute-force strategy to fight grade inflation isn’t without precedent. It falls in a long line of attempts to curb the practice. The Trump administration, for example, included a proposal to combat grade inflation in its recent compact, but the federal government lacks the authority, capacity, or trust to lead such an effort.

The problem past efforts have run into isn’t a lack of awareness or even a lack of will. It’s that no single institution can act alone without putting itself at a competitive disadvantage. But in an insightful piece, David Eubanks identified the key player that could solve this coordination problem: accreditors. To see why accreditors are uniquely positioned to reverse the trend, we first need to understand what’s driving grade inflation and why previous efforts to combat it have failed.

What’s Driving Grade Inflation?

There is a significant incentive for all actors in academia to accept artificially boosted grades.

Students are the most willing to accept grade inflation. And who can blame them? The boss who offers more money for less work would have many takers among his employees. Students are no different and will, as a general rule, gladly accept higher grades and a lower workload.

Another contributing factor to student demands for higher grades is the consumer mindset. When students pay tens of thousands of dollars for their education, they feel they should get As, regardless of performance. Even students who prefer learning over grades can be swayed to play it safe. After all, it is far less risky to choose the easy‑A course than one taught by a tough grader who will ensure students learn the material but may award lower grades. Those lower grades can reduce a student’s chances of landing a job interview or gaining admission to graduate school.

The second group of people driving grade inflation is professors. As an old Soviet-era joke goes, “we pretend to work, and they pretend to pay us.” A similar dynamic is at play with grading. Professors give students high grades, and students reward the professors with high student evaluations. They also benefit from fewer grade disputes since few students with an A will dispute their grade.

The third group driving grade inflation is graduate school admissions committees. Undergraduate GPAs are one of the primary tools committees use to make their decisions. But this means that every college and professor who wants to see their students accepted into graduate school is better off inflating grades.

How to Reverse the Trend

There has been no shortage of initiatives to combat grade inflation. Princeton tried to cap A grades, as Harvard is doing now, from 2004 to 2014, before scrapping the plan. Harvey Mansfield famously started giving students two grades: their official grade for their transcript and the grade they would have gotten without grade inflation. None of these policies has solved the problem. In part, that failure owes to a misdiagnosis of what’s driving it.

Efforts that presume grade inflation is a failure at the individual level, which can simply be remedied with calls for stiffer spines, are doomed to failure. Grade inflation is structural and incentive-driven, and it is therefore futile for an individual professor, department, or even a college to try to fight it. That is why Harvard’s new plan will fail just like Princeton’s.

The key to stopping grade inflation is solving a coordination problem—a prisoner’s dilemma. If all professors stopped inflating grades, we’d all be better off. Yet no professor or college wants to be the only one to stop inflating grades. Thus, the key to stopping grade inflation is solving a coordination problem. And that’s where an accreditor’s ability to apply a policy to many colleges at once sets them apart.

For example, the largest accreditor, the Higher Learning Commission (HLC), accredits 950 colleges, nearly a quarter of all federal aid-eligible colleges in the country. If HLC adopted contextualized transcripts, or adopted the abandoned Princeton cap on As, or a new Harvard cap, a critical mass of colleges would be committed to ending grade inflation, allowing other colleges and accreditors to follow suit.

Harvard should be applauded for trying to fight against grade inflation, but its efforts are doomed to fail because it cannot solve the coordination problem on its own. Accreditors in general, and HLC in particular, have the best chance of ending the scourge of grade inflation because they are the only ones who can solve the coordination problem.

The first is the Billionaire Tax Act, a onetime, 5% levy on the accumulated net worth of the state’s richest residents.

Lesser known is the Retirement and Personal Savings Protection Act, which would draw constitutional lines around what Sacramento can and cannot tax, prohibiting new levies on retirement accounts, personal savings and individually owned assets and banning retroactive taxation.

Everyone with even just a little bit of money set aside — not just the California billionaires targeted by the wealth tax — should understand what these two measures represent.

Start with the Billionaire Tax Act.

The gap between what it promises and what it would deliver is stark.

Joshua Rauh of Stanford University has run the numbers with his Hoover Institution colleagues, and the results cast doubt on the prospect of any revenue gain whatsoever.
Proponents claim the tax would raise $100 billion.

Rauh’s team found that billionaires have already been voting with their feet: Larry Ellison left California in 2020, and six others, including Google cofounders Larry Page and Sergey Brin, departed between the proposal’s announcement and Dec. 31, 2025 — the day before the liability would take effect.

These departures alone reduce the measure’s supposed tax revenue by nearly 40% before a single dollar is collected. Once migration patterns uncovered in the academic literature are applied to quieter departures, expected revenue falls to only $40 billion.

Now, factor in the normal state taxes that will no longer be collected from departing billionaires. Rauh’s team calculates that by shrinking the existing tax base, the measure’s “net present value” is at least a $25 billion loss for California.

Then there is the retroactivity problem.

The proposal aims to tax billionaires based on residency and conduct that reaches back to Jan. 1, long before any vote was cast.

Individuals who believe they lawfully established residency elsewhere might have to fight California in court for years (at the expense of the remaining taxpayers), based on details as arbitrary as where these billionaires kept their pets or held club memberships.

The “onetime” framing of the tax deserves equal skepticism.

As Rauh points out, the measure includes a constitutional authorization to lift California’s cap on taxation of intangible personal property.

Once that legal infrastructure exists, future wealth taxes can be imposed at any rate, at any threshold, at any time. It is, in other words, a permanent new power for the state.

The Billionaire Tax Act is so erratic and its precedent so problematic that it practically begs Californians to pay attention to the second ballot measure.

All Americans’ savings should be safe from such confiscation based on three clear principles.

First, fairness: When a worker sets aside after-tax income to invest for retirement, the resulting balance is not untapped revenue.

To treat this savings as a fresh tax base is to tax the same dollar twice.

Second, stability: A tax system that reaches into asset values rather than income flows is inherently volatile.

A founder whose stock drops 40% in a downturn still owes wealth tax on last year’s greater valuation. An ordinary saver whose 401(k) is taxed would face the same absurdity.

Third, and most urgent, is California’s own track record. According to the state’s nonpartisan Legislative Analyst’s Office, state spending is poised to grow by nearly 70% between 2019 and the coming fiscal year, drastically outpacing a significant revenue hike over the period.

The result is a cumulative deficit exceeding $50 billion over the next two years, a hole entirely of Sacramento’s own making, unrelated to Washington.

Trusting politicians with that spending record to stop at taxing billionaires is reckless and naive. When the wealth tax inevitably fails to deliver, the state will look for the next available pool of assets.

Nonbillionaires who remain after California’s billionaires depart will be the likely targets, and their retirement savings could be the new tax base.

As Rauh wrote earlier this month in his ongoing exploration of the proposals, “While approximately 0.001% of California households are billionaires, approximately 62% have retirement accounts.”

If this prediction sounds farfetched due to federal protections — or if you think billionaires will always be treated differently than normal savers who fill retirement accounts over a lifetime — consider what California already does to health savings accounts.

Federal law treats HSA contributions and earnings as tax-exempt. But under California’s tax engineering, the interest, dividends and capital gains are treated as ordinary income, affecting roughly 4.5 million residents.

These people are not billionaires or millionaires.

Politicians simply decided this was revenue the state was entitled to tax.

Doing the same with 401(k)s and IRAs would not require new principles, just the same willingness.

A wealth tax on billionaires is the first step, and it puts the retirement savings of ordinary Californians at risk. The HSA precedent suggests that the threat is real.

The Retirement and Personal Savings Protection Act would erect constitutional barriers against exactly that kind of expansion.

The importance of judicial independence was recently on display in Learning Resources v. Trump, where two of the three justices appointed by President Trump voted to strike down his tariffs as not authorized under the International Emergency Economic Powers Act. Even though he had chosen the justices, he could not fire them. Two months later, when Trump attended oral arguments in the birthright citizenship case, it again starkly demonstrated the lack of power he (or anyone else) has over the Supreme Court. He could glare all he wanted, but the justices knew that he could neither shorten their tenure nor cut their salaries. That is because the Constitution guarantees that all federal judges “shall hold their Offices during good Behaviour” and that their salaries “shall not be diminished during their Continuance in Office.”

Life tenure and salary protection were granted to federal judges because the Constitution’s framers knew that judges would sometimes be called on to make unpopular decisions. In Federalist 78, Alexander Hamilton wrote that life tenure would contribute to an “independent spirit in the judges which must be essential to the faithful performance of so arduous a duty.” Life tenure was thus intended to be a “barrier to the encroachments and oppressions of the representative body.” If judges instead had to depend on periodic reappointment, there would be a “danger of an improper complaisance” to the appointing branch of government. Such a system would “be fatal” to judges’ “necessary independence.” And in Federalist 79, Hamilton similarly explained that with a guaranteed salary, a judge “may then be sure of the ground upon which he stands, and can never be deterred from his duty by the apprehension of being placed in a less eligible situation.”

But despite these important benefits, our system of life tenure on the Supreme Court has led to unintended and harmful consequences. As lifespans have increased, the typical tenure of a Supreme Court justice has lengthened. As the Brennan Center has noted, “On average, justices today sit on the bench for more than a decade longer than their predecessors did as recently as the 1960s.” That means vacancies on the high court now arise roughly every three years rather than every two. And as the ideological balance of power on the court has become more important to both parties, qualified potential nominees are now treated as disqualified if they are over 55 because they are not as likely to serve “long enough.” But perhaps most concerning, justices now unabashedly time their retirements for when the presidency and Senate are controlled by their preferred party. Partisans on both sides urge justices to retire not because of infirmity but because of the political moment, such as recent campaigns urging Justices Stephen Breyer and Samuel Alito to retire. (Breyer retired in 2022.)

As a result of this system, the ideological balance of power on the court shifts not predictably due to elections, but unpredictably due to the randomness of justices dying, such as Justice Ruth Bader Ginsburg, or being forced to retire due to health, such as Justice Thurgood Marshall. A system in which constitutional lawyers obsessively follow the justices’ medical updates and the justices likely follow Senate polling averages is not a healthy one.

The question is: Can a new system be designed that fixes the randomness and unseemliness of the current system while retaining the vital benefits of Supreme Court independence? The answer is yes. That system is one in which justices know on the day they are appointed they will serve for 18 years on the Supreme Court and also for life afterward on a federal court of appeals.

First, as many others have convincingly argued, a single 18-year term on the Supreme Court would have many significant benefits. It would ensure predictability of appointments, yielding two appointments per presidential term. It would end strategically timed retirements (and with them the politicized campaigns urging retirements). It would end the unpredictable seismic swings in the balance of the court caused by justices’ deaths and infirmities. And it would allow appointing presidents to be agnostic as to the age of nominees, knowing that all appointees will serve for the same amount of time whether they are 40 or 60.

If Supreme Court justices served for a single guaranteed term of 18 years and then left the judiciary entirely, ineligible for reappointment, they would still be protected from premature firing or salary diminishment. But that would not necessarily be enough to ensure their complete independence. Life tenure in the judiciary ensures that judges need not fear punishment or ostracization after their tenure as judges has concluded. If judges could not rely on guaranteed income and authority for life, they might legitimately fear that certain decisions could harm their future economic prospects or their future ability to attain positions of similarly fulfilling and consequential work.

Thus, life tenure in the judiciary is important, but that does not mean we must guarantee life tenure on the Supreme Court. The vast majority of Article III judges in American history have served and will serve on some court below the Supreme Court, and life tenure on those lower courts has proven sufficient to provide them security and independence. The ideal system is one in which those appointed to the Supreme Court serve 18 years at the pinnacle of our judicial system and then serve for as long as they wish just one level lower, still protected from political punishment and still using their talents to decide weighty and important cases for as long as they wish.

With 18-year Supreme Court terms combined with subsequent life tenure on a court of appeals, we can ensure regular, predictable turnover on the Supreme Court without sacrificing the high court’s vital independence.

It’s clear from their public statements that both men are motivated by a genuine concern over the inability of the Secret Service to consistently and effectively protect the current sitting president. But the historical record shows such failures are inevitable regardless of where the Secret Service has existed on the government’s organization chart.

During the century-plus the Secret Service resided within the Treasury Department, there were presidential deaths at the hands of assassins (McKinley and Kennedy) or near-death episodes (Reagan). There’s no reason to believe that moving the Secret Service somewhere else in the Executive Branch will deter future assassins from making an attempt on a president’s life, and even less reason to believe that an organizational shuffling will ensure improved Secret Service performance in presidential protection.

There is, however, ample reason to believe that if the Moskowitz-Fry proposal were to become law, it would usher in a new age of presidentially directed spying and political repression unlike any seen in the history of the Republic, but which has clear historical parallels reaching back to the Roman Empire and its Praetorian Guard.

The reason lies not in the bill’s surface rationale but in what its sponsors appear not to have considered: the Secret Service is not merely a presidential protection agency. Far from it.

Under 18 U.S.C. § 3056, the Secret Service has jurisdiction over counterfeiting, financial fraud, wire fraud, access device fraud, and computer crimes. The financial crimes mandate has been a core Secret Service function since the agency’s founding in 1865.

Move the Secret Service into the Executive Office of the President (EOP) and you move its financial crimes jurisdiction there with it. Under the current administration’s own legal framework, that distinction is not incidental — it is the ballgame.

In April, Trump’s Office of Legal Counsel (OLC) issued a 52-page opinion concluding that the Presidential Records Act is unconstitutional in its entirety, and that the president “need not further comply with its dictates.” The legal architecture underlying that opinion holds that the EOP is the president’s constitutionally protected domain, not a space subject to ordinary congressional regulation or oversight. Congress created the Department of Homeland Security; it can regulate DHS. The EOP is different in kind: it is where presidential authority is most plenary and congressional reach is most attenuated.

A future OLC opinion — and this OLC has demonstrated it will reach for maximal presidential authority claims — could argue that the Secret Service’s director, as an EOP officer, serves entirely at the president’s pleasure without Senate confirmation and that the president may direct Secret Service financial crimes investigative priorities as an exercise of inherent Article II supervisory authority that no statute can override. The bill, as drafted, doesn’t even require Senate confirmation for the Service’s director — making that argument easier, not harder, to sustain.

Consider what that means against the backdrop of where this administration already is.

In September, President Trump signed National Security Presidential Memorandum 7, directing federal law enforcement to investigate Americans based on ideological markers — “anti-Americanism,” “anti-capitalism,” “anti-Christianity” — that describe protected First Amendment activity, not criminal conduct.

The administration’s FY2027 budget request confirmed that a multi-agency NSPM‑7 Joint Mission Center is now operational, drawing personnel from 10 federal agencies, specifically tasked with integrating “intelligence, operational support, and financial analysis” against its designated targets. What’s more, the administration has already named specific organizations — the Open Society Foundations, the Southern Poverty Law Center, the Ford Foundation — for investigation and possible prosecution and has directed the Internal Revenue Service to treat their tax-exempt status as a live question.

That is the financial targeting apparatus already in motion. The Secret Service’s financial crimes jurisdiction is precisely the enforcement tool most relevant to it. Place the Secret Service inside the presidency under a theory of plenary presidential authority over that space, and the president gains direct, unmediated supervisory control over the agency best positioned to investigate the financial networks of his administration’s designated political enemies.

Congressman Moskowitz, to his credit, joined the April 2026 House Judiciary Democrats’ amicus brief arguing that Trump’s executive orders targeting law firms for their advocacy clients violate the First Amendment and separation of powers. He grasps, in that context, that an administration willing to weaponize executive authority against political opponents is one whose control over law enforcement cannot be expanded without consequence. Yet his proposed legislation does not reflect that same understanding.

The bill would also dramatically reduce congressional oversight that the Secret Service so plainly needs. EOP components routinely invoke executive privilege to resist congressional subpoenas and document demands. Moving an agency with documented accountability failures into the portion of the executive branch most resistant to scrutiny is not a reform — it is an immunity grant.

None of this impugns Moskowitz’s or Fry’s motives. Both men are responding to real operational Secret Service failures. But the test of institutional reform is not whether the sponsor’s intentions are honorable — it is whether the authority being transferred will be used honorably by every future occupant of the office receiving it.

Congress should not respond by moving a federal law enforcement agency with sweeping jurisdiction directly into the president’s institutional inner circle. Doing so would not solve the Secret Service’s operational problems. It would instead create the conditions for a future president to exercise direct political control over one of the federal government’s most powerful investigative tools while shielding that control from meaningful oversight.

It is a blueprint for presidential political policing.

Those goals were always unachievable without targeting legal immigrants, so the administration has undermined every option to come or stay legally. The latest evidence against the claim came late last week, when DHS said it would require most immigrants to leave the United States to attempt to obtain their green cards through immigrant visa processing abroad.

After much outcry from businesses, DHS attempted to downplay the memo’s importance, particularly for skilled foreign workers. Yet the department has not withdrawn or amended the memo in any way. All immigrants will still be expected to cite unusual circumstances that rebut the presumption that they must leave the country. This will set up denials, loss of status and arrests at green card appointments and interviews.

In 1952, Congress created a process for people who entered legally with a temporary status to adjust that status to legal permanent residence without leaving the country. Since the 1980s, most legal permanent residents have received their green cards in the U.S. A majority are spouses of U.S. citizens or skilled employees of U.S. businesses. DHS wants to end this adjustment of status process and either arrest anyone denied under the new policy or take credit for the self-deportations that result.

As brazen as this new policy is, it is nothing new. Since Day 1 of this administration, DHS has attempted to cut off legal paths and arrest people when they lose their status.

The department’s initial focus was to cut off ways for immigrants to avoid the deportation dragnet. Over the last year, DHS has stripped more than 2 million legal immigrants of their parole and temporary protected status. It effectively banned asylum — even for immigrants who entered legally — reducing grant rates to just 1 in 14 applicants.

But DHS soon realized that many immigrants were still evading arrest and deportation through the green card process. In response, DHS has cut approvals of green cards in half over the past year. This caused the expiration of many immigrants’ underlying temporary status — whether a visa, parole, TPS or otherwise — setting them up for arrest.

For humanitarian categories — asylees, refugees, trafficking and crime victims — the agency quit processing almost completely. DHS has also more than quadrupled arrests of Cuban immigrants since Trump took office, even though Cubans have a special law enabling them to receive green cards after one year in the U.S.

In December, after DHS shut down refugee green card processing, it sent agents to Minnesota to arrest refugees who entered legally from abroad and still had refugee status. The justification? Failing to receive a green card within one year of entry as the law requires — even though DHS was not processing refugee green cards.

This latest policy move goes after the nonhumanitarian categories — employer- and family-sponsored immigrants — that are not already totally suspended. With DHS apparently concerned about business groups’ opposition, the main targets will likely be close relatives of U.S. citizens who are adjusting their status.

DHS is still allowing immigrants to apply for green cards, but it is saying it will only grant them to those with exceptional cases. Then, as their temporary statuses expire, it may deny them and arrest them at the same time. DHS is already arresting green card applicants, but it has been forced to release them when it finds that they are eligible. That’s the “problem” it wants to fix.

DHS is telling people they can just apply from abroad. But in addition to the family separation and costs, leaving will often cause applicants to lose their chance to return. If an employer-sponsored immigrant has to leave their job for months, their employer may be forced to fire them, costing them a job and the right to stay. In another convoluted legal twist, if someone whose status expired more than 180 days before applying entered legally, they are eligible to get a green card inside the U.S. But if they leave the country, they are barred from a visa for 3 years. If the lapse in status was over a year, they are barred for 10 years.

DHS knows all this — and would treat every one of these outcomes as a win. As significantly, DHS knows that Trump has banned 40 nationalities from receiving immigrant visas, and the State Department has suspended processing for immigrants from 75 countries. Combining these two lists means that more than 90 nationalities — half of legal immigrants — cannot undergo the green card process abroad. Despite downplaying the memo’s effect, DHS’ press comments have specifically suggested these nationalities will be forced out.

The new wall of red tape for legal immigrants isn’t some anomalous departure for the Trump administration. Rather, the bureaucratic hoops are an essential component of mass deportation. DHS needs to “illegalize” people — that is, transform people from legal to illegal — to meet its impossible deportation goals. This is also one reason why the administration has sought to end birthright citizenship. It knows that citizenship for U.S.-born children of immigrants protects those children from deportation.

Mass deportation and draconian limits on legal immigration are two sides of the same coin. So many people are here illegally solely because historically the U.S. government has made legal immigration impossible. If we want to solve the illegal immigration problem, legal immigration is the solution. But clearly DHS wants the problem more than the solution.

The crackdown was national and brutal. According to Amnesty International: “Immediately after the military crackdown, the Chinese authorities began to hunt down those involved in the demonstrations. Thousands of people were detained, tortured, imprisoned, or executed after unfair trials charged with ‘counter-revolutionary’ crimes.” Millions of Chinese, anyone suspected of having liberal sensibilities, were purged from the CCP. The desire for liberty remained widespread but had to be disguised.

Of course, the terrible death toll, likely in the thousands, was actually modest for the PRC. The only good news of Tiananmen was that the new China is more repressive than murderous, as during the 27 years when Mao Zedong effectively ruled the country. His madcap tenure, highlighted by consolidation of power, revenge against opponents, various ideological campaigns (Anti-Rightist Movement, Anti-Deviation Right Struggle, and others), agricultural collectivization and backyard industrialization (Great Leap Forward), and combination political purge/​personal revenge/​civil war (Great Proletarian Cultural Revolution), cost tens of millions of lives, a staggering toll making him the most prolific mass murderer in history.

Unfortunately, Mao’s image and influence continue to beset China. Xi Jinping, general secretary of the Chinese Communist Party and president of the Chinese government, fears instability of the sort so characteristic of Maoism. However, in other ways, he is a devoted Maoist, having steadily reinforced the role of the CCP and his control over it. In almost every way, China has become less free since he took control in 2012. Political controls are much tighter today than in the years after Mao’s death.

There is no easy strategy for outsiders, including Uncle Sam, to force Beijing to treat its people with dignity and respect their liberties. Repression is essential to the CCP’s control. Few authoritarian governments have voluntarily dismantled themselves. The Eastern European states generally went peacefully, but only because they were unable to hang onto control. Even Mikhail Gorbachev, Ronald Reagan’s essential partner in ending the Cold War, hoped to reform, not end, communism.

Moreover, Washington has a plethora of contentious issues with the PRC. Controversies include trade and investment, technology and supply, maritime and allied security, international pressure and domestic meddling, and more. How to balance such competing concerns and manage contradictions among them remains contentious. President Donald Trump emphasizes economic and commercial concerns.

Although there is no easy response to Chinese repression, Americans should support the liberty and dignity of the Chinese people when confronting that country. Human rights are more than a political issue for America. It is vital for 1.4 billion Chinese, especially average folks who simply want a better life for themselves and their families and friends.

This reality was dramatically highlighted when 68-year-old Dong Guangping recently escaped the PRC to South Korea, making the 30-hour trip on an inflatable boat. Reported the New York Times, he “was found in an 11-foot-long, light gray rubber boat attached to a 9.9‑horsepower motor.” It was his fourth attempt to escape frequent detention and what otherwise amounted to an open-air prison for a quarter century. Detailed the Times:

Currently held by South Korea, he hopes to join his family in Canada, which, a decade ago, offered him asylum. Although his story is dramatic, he is not the first Chinese dissident to flee by sea. Three years ago, Kwon Pyong escaped the PRC by jet ski. Although barely half Dong’s age, he also suffered greatly for refusing to kowtow to the latest Red Emperor ensconced in Beijing. Reported the Times, after his escape, “Mr. Kwon, 36 and an ethnic Korean, had mocked China’s powerful leader and criticized how the ruling Communist Party was persecuting hundreds of pro-democracy activists at home and abroad. In response, he said, he faced an exit ban and years of detention, prison and surveillance.” (Such attempts at liberation are not as rare as one might imagine. Russian freedom-seekers have similarly sought freedom, seeking to reach America via water.)

Americans are suffering through contentious political times at home. We should never forget the liberties that we enjoy, which are routinely denied to so many people in so many other lands, such as China. Dong Guangping understands the value of freedom. So does Kyong Pyong. We should remember them and their oppressed countrymen as another anniversary of Tiananmen Square approaches.

But there are also multiple ways in which democracy can often threaten liberty and human welfare. These dangers include the tyranny of the majority and widespread voter ignorance. Democracy can also be a threat to its own perpetuation, by bringing to power authoritarian political movements. These are all longstanding problems. But recent events demonstrate their continuing—and in some cases growing—significance. Liberals need to acknowledge their gravity and more aggressively pursue various potential solutions. These include limiting and decentralizing government and possibly measures to make it more difficult for illiberal anti-democratic movements to take power.

Democratic Tyranny

It is tempting to dismiss the idea of democracy devolving into tyranny or injustice as a contradiction in terms. People often use “democratic” as a synonym for “good,” and “undemocratic” for “bad.” Whatever the linguistic merits of this usage, it is not analytically helpful. If anything good is by definition also compatible with unconstrained democracy, and anything democratic is by definition also good, then democracy ceases to be an analytically useful concept. Better to define democracy as a political system governed by majoritarian political processes. Such processes can make both good decisions and bad ones.

A slightly more sophisticated basis for dismissing the possibility that democracy can be bad is the idea—advanced by a few political theorists, such as Ian Shapiro—that there is no objective external basis for evaluating democratic decisions. If the voters and their representatives support a particular policy, who are we to say it’s wrong?

Such moral relativism has a range of well-known flaws. Here I will merely note that, if we have no objective basis for evaluating the justice of democratically enacted policies, we also have no basis for concluding that democracy is superior to dictatorship, oligarchy, or theocracy. If there are metrics—such as liberty, equality, and human welfare—by which we can conclude that democracy is superior to these other systems, then those very same standards can be used to evaluate democracy’s own output, and to conclude that democratic government may need to be constrained or amended in various ways.

Once we recognize that democratic governments are not inherently just or immune to critical evaluation, it becomes clear that they may be prone to systematic flaws. The most obvious is the “tyranny of the majority.” If democratic governments represent the will of a majority of the population, that majority might sometimes oppress minorities. There are many obvious historical examples from around the world, including the oppression of ethnic, racial, religious, and other groups. This danger is once again a serious menace in many democratic nations, thanks in part to the rise of ethno-nationalist movements—like Donald Trump’s MAGA movement and similar ones in various European countries—that seek to mobilize ethnic majorities in their respective countries by highlighting the supposed threat posed by minority groups and immigrants. Nationalists have a long history of persecuting and oppressing minority groups, and today’s nationalist movements are much like their predecessors in that regard.

The Problem of Political Ignorance

Voter ignorance is a second way in which democracy often menaces liberal values. As described in many studies, including my book Democracy and Political Ignorance: Why Smaller Government is Smarter, overwhelming evidence shows that most voters know little about government and public policy. Majorities are often ignorant even of such basics as the names of the three branches of government, how the national government spends its money (voters in many nations massively underestimate how much is spent on entitlement programs, while greatly overestimating foreign aid), and which government officials are responsible for which issues.

Such behavior is actually perfectly rational for most of the public. If your only reason for following politics is to be a better voter, that turns out to not be much of an incentive at all, because there is so little chance that your vote will make a difference to the outcome of an election (about 1 in 60 million in a U.S. presidential race, for example, though there are variations depending on the state). For most people, therefore, it is rational to devote very little time to learning about politics, and instead focus on other activities.

Of course there are people who learn political information for reasons other than becoming better voters. Just as there are sports fans who love to follow their favorite teams even though they cannot influence the outcomes of games, there are also “political fans” who enjoy following political issues, and cheering for their favorite candidates, parties, or ideologies.

There is nothing wrong with being a political fan. But if you are seeking out political information for the purpose of enhancing your fan experience, that objective is often inimical to the goal of seeking out the truth. Much like sports fans, political fans tend to evaluate new information in a highly biased way. They overvalue anything that supports their preexisting views, and undervalue or ignore new data that cuts against them, even to the extent of misinterpreting simple data that they could easily assess correctly in other contexts. Moreover, those most interested in political issues are also particularly prone to discuss politics only with others who agree with their views and to follow politics only through like-minded media. This problem may well be even worse today, in the age of social media and a fragmented internet, than in some previous eras.

Thus, we have a serious two-level problem of political ignorance. Most voters are rationally ignorant, knowing little about politics and government. The minority, who are much more knowledgeable, are mostly political fans—highly biased in their selection of information sources and their evaluation of what they learn. In combination, these problems predictably lead to the election of political leaders and parties that pursue a wide range of badly flawed policies, including many that threaten liberty and other liberal values.

Voter ignorance is exacerbated by the enormous size, scope, and complexity of modern government. In most advanced democracies, government spending accounts for a third or more of GDP, and the state regulates almost every form of human activity. Even relatively knowledgeable voters cannot effectively monitor more than a small fraction of these policies and their effects.

The harm caused by political ignorance also exacerbates the problem of tyranny of the majority. Much oppression of minorities is itself the result of ignorance and bias. For example, ethnic discrimination and xenophobic hostility to immigration are often rooted in the ignorant belief that the economy is a zero-sum game, in which gains for one group can only come at the expense of others—ignoring the reality of growth, innovation, and mutually beneficial gains from exchange.

When I first started writing about political ignorance over twenty-five years ago, many scholars argued that voter knowledge levels are not a significant problem, because voters who know very little about government and public policy can still do a good job thanks to information shortcuts, the “miracle of aggregation,” and other workarounds. Such optimism is far less prevalent today, thanks to the rise of Donald Trump and other similar right-wing populist leaders exploiting political ignorance to their advantage. But Trump and his ilk are just particularly egregious examples of a problem that long predated them.

As I and other critics have long argued, information shortcuts have serious shortcomings. Among other things, they often require preexisting knowledge to use effectively. For example, the most common information shortcut is “retrospective voting,” rewarding or punishing incumbent politicians based on whether things went well or badly during their terms. As explained in chapter four of Democracy and Political Ignorance, voters often reward or punish officeholders for things they didn’t cause (most notably short-term economic trends, but also things like droughts and even sports team victories), while ignoring some things that politicians are in fact responsible for.

The 2024 U.S. presidential election showed that shortcuts are even less effective than I previously thought. Key swing voters overlooked the dangerous—and obvious—menace to liberal democracy posed by Donald Trump’s efforts to use force and fraud to overturn his defeat in the 2020 election. Instead, they focused on punishing Democrats for the less severe problem of inflationary price increases. In the process, they failed to recognize that Trump’s own policies—most notably tariffs, mass deportation, and exclusion of immigrant workers—would predictably increase prices rather than lower them, as in fact went on to happen.

“Miracle of aggregation” theories hold that errors caused by voter ignorance matter little because ignorance-induced voter errors in one direction (e.g., in favor of Republicans) are offset by those in the other (e.g., in favor of Democrats), allowing more knowledgeable voters to determine the outcome. Alternatively, voters collectively might have greater knowledge than they do individually, and the electorate as a whole can make effective use of that aggregate wisdom. As discussed more fully in Democracy and Political Ignorance, this happy outcome can only occur if voter errors are randomly distributed or other highly improbable circumstances arise. In the real world, even a slight nonrandom ignorance-induced bias can make errors virtually certain.

Bringing Illiberal Authoritarians to Power

In worst-case but all-too-plausible scenarios, democracy can be the cause of its own demise by bringing to power illiberal authoritarians. This famously happened in the case of Nazi Germany. More recently, authoritarians have come to power by democratic means and then proceeded to subvert democracy in countries like Russia, Venezuela, Turkey, Nicaragua, and elsewhere. In Hungary, the authoritarian Viktor Orbán was elected by democratic means, used government power to suppress opposition, and stayed in power for sixteen years, until he was finally thrown out by an opposition landslide so great that it turned his rigged electoral map against him. Even in the United States—one of the world’s longest and most deeply established democracies—voters elected Donald Trump despite his visible authoritarian tendencies, and then reelected him in 2024, even after he incited violence to try to stay in power after losing the previous election.

Trump probably won’t be able to completely subvert America’s liberal institutions, which are better-entrenched than those of Hungary, Russia, or Venezuela. But the very fact that he twice got the opportunity to try highlights a serious weakness of democracy.

Possible Solutions

The traditional solution to the danger of tyranny of the majority is to impose constitutional limits on government power, and to ban various types of invidious discrimination. These approaches have great merit, and have achieved much, including the abolition of Jim Crow racial segregation in the United States. But they cannot always cope with the full range of majoritarian tyranny. Most obviously, current constitutional rules, as interpreted by U.S. courts and those in many other countries, often don’t do nearly enough to curb majoritarian oppression of immigrants. The enormous size and scope of the modern state also makes it very difficult to curb these dangers, as it multiplies opportunities for oppression and discrimination.

There is no easy way to “fix” political ignorance. Experience shows that we cannot rely on public education to increase voter knowledge significantly, as knowledge levels have stagnated even as education attainment greatly increased over the last several decades. I assess a range of other possible options in my 2023 article on “Top-Down and Bottom-Up Solutions to the Problem of Political Ignorance,” and in Democracy and Political Ignorance. 

I believe the best approach is to make fewer decisions at the ballot box and more by “voting with your feet,” where incentives to seek out information and use it wisely are better. People can vote with their feet by choosing which governments to live under, based on their policies, and by making decisions in the private sector. As I discuss in various works, such as Free to Move: Foot Voting, Migration, and Political Freedom, we can empower people to make more decisions through foot voting by limiting and decentralizing government power, and by breaking down barriers to both domestic and international migration. These measures can increase the range of options available to foot voters and reduce moving costs. Decentralizing power to lower levels of government and—in many cases—to the private sector can empower foot voters to choose between a wide range of options without having to move long distances. We can also reduce the information burden on rationally ignorant voters by cutting back on the scope and complexity of government functions, thereby making it easier for voters to keep track of them.

In “Top-Down and Bottom-Up Solutions,” I suggest we should give more consideration to the possibility of simply paying voters to increase their levels of political knowledge, thereby altering the structure of incentives that leads to rational ignorance. For example, philanthropists or nonprofit groups could create a “Voter Achievement Test” that tests basic political knowledge, make it available to anyone who wants to take it, and give monetary awards to anyone who scores above a certain level, awarding perhaps $500 or $1,000 each. But any effective approach will take time, and there may be no one fix that is sufficient by itself. We likely need a combination of several strategies.

Constitutional constraints on government power can also help contain would-be authoritarians. But, given the opportunity, they can sometimes break through the constraints. A common strategy for doing so is the abuse of emergency powers, utilized by Vladimir Putin in Russia, Hugo Chávez in Venezuela, Benito Mussolini in Italy, and others; and most famously by the Nazis, with the 1933 Enabling Act. Legislators would do well to strictly limit emergency powers, and courts should not defer to the executive’s claims that the emergency situation required to trigger their use actually exists; they should demand proof.

Liberals should also give more consideration to mechanisms by which illiberal authoritarians can be barred from power in the first place. Barring candidates and parties from running for office based on their illiberal ideologies alone is a dangerous tool, since it could be abused to suppress opposition parties more generally. On the other hand, there is less danger in barring officeholding by people with a demonstrated record of dangerous, illiberal, and anti-democratic actions. For example, several post-communist Eastern European nations enacted “lustration” laws barring from office some former officials of their communist dictatorships, particularly former agents of the secret police. These laws have helped prevent democratic backsliding, and they have not become a menace to democracy themselves. Had Russia enacted such a law, it might have avoided the horrific regime of ex-KGB Lieutenant Colonel Vladimir Putin.

The United States, for its part, would have done well to enforce Section 3 of the Fourteenth Amendment, which barred future officeholding by public officials who engaged in “insurrection” against the United States, a provision originally aimed at Confederate insurrectionists after the Civil War. Elsewhere, I have argued that the Supreme Court was wrong to rule that Section 3 cannot be applied to Donald Trump because it supposedly could not be enforced without additional congressional legislation. A future Congress would do well to enact such enforcement legislation.

The above is just a preliminary overview of possible tools for mitigating the danger that democracy often poses to liberal values and—in some cases—to its own perpetuation. There is room for disagreement over exactly which solutions are best, and the optimal approach for some nations may well differ from that which is best for others. But the beginning of wisdom is to recognize that these are serious dangers indeed. Liberalism cannot do without democracy. But it also cannot survive a democracy with too few constraints on its power.

There has been much outcry, but little progress, however, toward the administration’s goal of eliminating overregulation in general or of reining in the CFPB in particular. The president alone can’t scale back or eliminate the CFPB. Congress created this overpowered agency, and only lawmakers can fix it. Curbing the bureau’s frequent abuse of its subpoena power is one good place for them to start.

The CFPB issues subpoenas, known as civil investigative demands (CIDs), that require recipients to turn over documents, testimony and other evidence. The bureau has used CIDs with little justification, even when it has no reason to suspect a violation of the law. Not only does the bureau’s CID overreach violate Americans’ privacy rights, but it also makes financial services more expensive for everyone. The burden falls heaviest on small and mid-size firms, limiting growth and competition.

Problems with the bureau’s CIDs are well documented. The bureau has targeted disfavored businesses or used CIDs for market research. Investigations drag on for months without the bureau ever explaining what the target has done wrong. Responding to a CID can cost a small business thousands of hours of employee time and tens of thousands of dollars, and larger firms may bear more than a million dollars in costs. Too often, these burdens are way out of proportion to any benefit to consumers.

One small-dollar lender in Arizona, for example, endured a series of CFPB subpoenas over almost three years, ultimately closing about a third of its locations and letting workers go. The firm reported receiving few complaints from customers or state regulators, and the investigation found no violations. The company’s chief executive officer, it should be noted, had previously criticized a CFPB regulatory proposal.

The bureau’s defenders argue that complaints about the bureau’s CIDs should be discounted because people are bound to complain about being investigated. But it is more likely that the complaints are understated because regulated firms are reluctant to antagonize their regulator.

What encourages the bureau’s overreach? Lawmakers gave in to the tempting idea that an agency could be insulated from lobbyists’ influence and regulatory skepticism without losing accountability. The CFPB’s designers funded it from the Federal Reserve’s (the Fed’s) earnings instead of congressional appropriations, seeking to limit financial firms’ influence over the bureau. And the bureau is headed by a sole director rather than by a commission, meaning that the CFPB acts with less deliberation. The commission structure was rejected because it presumably might often “produce a majority that is consistently skeptical about regulation”.

The bureau’s subpoena powers also allow CID targets to petition the bureau for relief from a CID, but the petitions are decided by the bureau’s director, not a neutral third party. The CFPB’s governing statute lacks clear language limiting investigators’ discretion, so judges too often defer to the bureau in court.

Lawmakers have noticed the bureau’s dysfunction. In 2024 and again in 2025, legislators offered a bill, the Civil Investigative Demand Reform Act (CIDRA), to address these problems. The bill would require the bureau to set out the specific facts that justify issuing a subpoena. It would also give subpoena targets the right to petition for relief from burdensome or duplicative subpoenas.

The CIDRA enjoys bipartisan support: After all, that agencies can treat some Americans unfairly is out of step with our constitutional republic. And it could serve as a model for curbing investigative overreach by other agencies, such as the Federal Trade Commission (FTC), which has begun to use its broad CID authority to punish free speech.

Making agencies accountable for misusing their investigative powers is a crucial aspect of deregulation. For this, legal changes are essential. Can executive-branch measures, such as firing staff or reducing an agency’s physical footprint, help? Some reformers posit that an understaffed agency will focus on targeting real violations and concrete harm, reducing the agency’s effects on legitimate firms’ productive activities. Yet, there are no rewards in the public sector for precision. Overworked staff might also rebel, perhaps by cutting corners and focusing on easy wins, targeting smaller firms that cannot afford to fight back. And the president’s progress in cutting staff and budgets can easily be reversed by his successor.

The bureau’s abuse of its subpoena powers is real. Ironically, the agency that was created to ensure that banks and other financial services treat consumers fairly has abandoned fairness in its own investigations. And attempts to cut the CFPB down to size by slashing its budget and staff are probably futile, if not counterproductive, in the long run.

Evicting CFPB from its headquarters is a great headline grabber, but it’s not going to end the agency’s overreach. If elected officials are serious about that task, they should take a look at their own handiwork. They created the agency. They’ll have to tame it, too.

Lawmakers have also proposed, as a backdoor form of rent control, that the Legislature restrict the use of algorithmic pricing software that helps landlords determine what their properties are worth.

Rent control in any form would ultimately hurt the very people it intends to help.

The Massachusetts ballot measure would cap annual rent increases to either 5 percent or the annual increase in the consumer price index, whichever is lower. But this benchmark has historically failed to track the actual costs of maintaining older housing stock.

Between 2000 and 2024, rent inflation averaged about a full percentage point per year above headline CPI — 3.6 percent versus 2.6 percent — exceeding CPI in 18 of those 25 years, and growing by a cumulative 140 percent against CPI’s 88 percent. Landlord operating costs have grown faster still: the National Multifamily Housing Council reported that property insurance alone rose 26 percent in 2023. A cap pegged to the CPI or 5 percent, whichever is lower, would force rents to grow more slowly than historically — and more slowly than the costs they must cover.

A building whose rents cannot keep pace with rising costs is a building whose owner will defer repairs, skip renovations, and eventually convert the property to condominiums or other uses that escape the cap. The pattern is documented across rent-controlled cities from New York to San Francisco, where decades of regulation left a disproportionate share of the controlled stock in poor physical condition relative to comparable market-rate units.

Rent controls generate arbitrary redistribution of resources. The Massachusetts cap would apply regardless of tenant income or need: A high-earning Back Bay tenant would receive the same protection as a low-income family in Chelsea, and — in dollar terms — a larger benefit, because the cap is a percentage of a higher base rent. The transfer of resources would run from landlords to incumbent tenants — not from landlords to whomever is most in need.

The ballot measure would exempt new construction for its first 10 years, as well as owner-occupied buildings of four units or fewer. But for all other rental housing, there would be no vacancy decontrol: When a tenant moves out, the new tenant would inherit the old rent rather than paying market rate, offering landlords no relief from the unit’s artificially discounted price.

The costs of rent control would also spill beyond the controlled sector. Deteriorating rent-controlled buildings would impose costs on surrounding property, making the neighborhood less attractive and dragging down values for uncontrolled units in the area.

A study found that ending Cambridge’s rent control in 1995 added about $1.8 billion to the value of the city’s housing stock over the following decade — and that more than half of that gain consisted of spillover effects for never-controlled units. In other words, the majority of the cost of rent control had been borne by property owners that the regulation had never directly touched.

Supporters of the Massachusetts ballot measure argue that its 10-year new-construction exemption would preserve the incentive to build new rental housing. It would not. Multifamily developers underwrite buildings on horizons of 20 to 40 years. That is why Governor Maura Healey has reported that six developers lost their funding after the ballot question cleared its signature threshold to move forward, with thousands of potential units at stake.

If Massachusetts wants to lower rents, the most important policy question is how to build more housing, not how to ration the housing the state already has. The tools to accomplish this are familiar and well-studied: Loosen zoning restrictions, raise allowable floor-to-area ratios, end single-family-only districts, legalize accessory dwelling units, and shorten permitting. Healey’s housing agenda already points in this direction, and the experience of cities like Austin — where permissive zoning enabled a surge of new construction and a meaningful drop in rents — suggests that these policies work.

Massachusetts voters rejected rent control in 1994. They should reject it again.

The president’s tariffs now stand 0–5 in court. They should be 0–6 before summer is over. \

The president imposed this latest round of tariffs arguing that deficits in the U.S. trade balance — and the current account more broadly — amounted to “large and serious U.S. balance-of-payments deficits” under Section 122. These new tariffs were always a stopgap: by law, Section 122 tariffs are temporary (unless affirmed by Congress) and set to expire in late July.

Yet Congress designed Section 122 for a different purpose than the one the president claims they address, one directly related to the monetary system of fixed exchange rates that existed at the time of the statute’s creation.

Under that system, a “balance-of-payments deficit” — not to be confused, as the president appears to have done, with a “trade deficit” — referred to a situation in which foreign governments holding dollars sought to convert them to gold, leading to a decrease in U.S. reserves. Section 122 provided a temporary, tariff-based remedy for this problem.

The fixed-rate monetary system ceased to exist half a century ago, after it was abandoned by the United States and other major economies in 1973 and overhauled in 1976. Section 122 became an afterthought soon after it became law. Only in 1984 did the Senate Finance Committee suggest invoking the statute to deal with the growing U.S. trade deficit, yet the Reagan administration rejected this idea because foreign investment offset the deficit and, thus, the country did not need to spend reserves to finance it.

Not much has changed since. While the U.S. trade deficit is higher today than in prior decades, the U.S. investment surplus has also increased. Dollars held by foreigners are no longer used to redeem gold reserves but to invest in U.S. financial assets or U.S. businesses. There is also no evidence that the U.S. is failing to attract such investments — the U.S. financial account reached a record surplus in 2025 — or that it is using up its monetary reserves to manage the dollar’s exchange rate.

In short, there is no “balance-of-payments deficit” that justifies invoking Section 122.

At worst, the administration’s rationale for invoking Section 122 would transform the statute from a tool to address rare, specific circumstances to a mechanism for imposing across-the-board tariffs at any time. A president could, as Trump did in February, point to the nation’s trade deficit to impose tariffs while ignoring the nation’s surplus in investment. If courts validate this reading on appeal and Congress fails to reform the statute, the president would have the power to impose sweeping tariffs with very few constraints.

The Constitution is unambiguous: the power to impose tariffs and duties — as well as to regulate international commerce — belongs to Congress. For nearly 150 years, lawmakers exercised this power directly. After the disaster of the protectionist 1930 Smoot-Hawley tariffs, a slow, decadeslong abdication followed: statute by statute, Congress handed that authority to the executive branch, each delegation sold as a narrow measure. The Trump administration’s erratic and costly tariffs have exposed the danger of constitutional imbalance.

The courts have done their job to correct these failures — multiple times. But court victories only go so far. Judges can strike down specific tariffs, but they cannot fix the underlying laws that give presidents the tools to keep trying. As long as those statutes remain on the books — or without serious modifications, such as requiring congressional approval before tariffs can be implemented — the ever-protectionist Trump administration will simply reach for the next legal hook, which is exactly what it has done.

Americans are not fooled. Polling shows they disapprove of the administration’s tariff policies by a wide margin: 60% oppose them, including 39% who strongly disapprove. More than three-quarters say tariffs are bad for the cost of living, and 70% say they hurt Americans’ standard of living. U.S. businesses paid more than $8 billion in Section 122 tariffs in March alone.

Congress should be listening, especially during an election year. It wrote these laws, expanded them over decades, and only it can truly rein in Trump’s tariff abuses. For now, though, it seems content to leave this work to the courts.

And while the administration’s losing record in court is good for importers, it’s a mark of shame for Congress.

Arizona’s massive drop in SNAP numbers isn’t because the state is catching fraud. The numbers point to the changed payment error calculation.

Across the country, SNAP recipients declined by 9% between July 2025 and February 2026. Louisiana, the state with the second-largest decline, dropped 20%, less than half the percentage decrease of Arizona.

So, what happened?

It starts with the “One Big Beautiful Bill” (H.R. 1) that President Trump signed into law on July 4, 2025. The bill expanded SNAP work requirements, changed eligibility rules, shifted more administrative costs onto states and tied future state costs to payment error rates.

But every state faced the same law. Arizona still became the outlier.

Is fraud behind Arizona’s drop in SNAP benefits?

An easy hypothesis is fraud. State Sen. John Kavanagh claimed “our massive decrease is caused by the fact that we had massive fraud in the system.”

But the idea that Arizona suddenly uncovered fraud in one out of every two beneficiaries doesn’t pass the smell test. SNAP’s Arizona administrator, the Department of Economic Security (DES), isn’t claiming any massive uncovering of fraud.

DES Director Michael Wisehart offered another possibility: Arizona simply moved faster than other states in implementing the new federal rules. Wisehart claims that in the coming months, other states will see a similarly steep drop in SNAP beneficiaries.

Maybe.

How do payment errors affect SNAP payments?

A less discussed hypothesis is that H.R. 1 changed the financial incentives of states in a way that exposed Arizona’s historic underinvestment in SNAP administration.

Prior to H.R. 1, the federal government covered 100% of SNAP food benefits and 50% of SNAP administrative costs. H.R. 1 reduced the federal administrative contribution from 50% to 25%, and it put a strict 6% threshold for payment error rates.

Importantly, “payment errors” aren’t a measure of fraudulent applications. Payment error rates are overpayments and underpayments issued by the state. Administrative mistakes and paperwork errors all contribute to the calculation.

Who pays when the SNAP error rate is above 6%?

Under the new rules, if a state is above the 6% mistake ceiling, the financial burden shifts from the federal government to the state. If Arizona fails to meet this new standard, it could result in annual liabilities between $100 million and $300 million. Unfortunately, we’ve only been below 6% twice in the last 10 years (2018, 5.99% and 2019, 5.24%). DES currently estimates an error rate of 10.45% for fiscal year 2025.

That means we’re facing a financial cliff. And that in turn might explain the drop in SNAP benefits. To steer clear of a huge financial liability, the state must reduce its error rate. That means proceeding very cautiously. It means proceeding slowly. Very slowly.

Compounding the problem is Arizona’s lean SNAP administrative operation. That’s a challenge that has only become more challenging because between July 2024 and July 2025, DES ran out of specially allocated federal grant, and the number of DES employees reviewing SNAP applications fell from 1,370 to 880.

What other factors affect SNAP application processing?

The final result of a system with a reduced staff that proceeds slowly and cautiously? Fewer SNAP applications processed. In December 2025, DES reported 54,000 pending new and renewal applications, including 18,000 older than 30 days. The Arizona ombudsman’s office received more complaints in the last six months of 2025 than it did in all of 2024.

And that could be how we ended up with far fewer SNAP recipients.

One way to address this problem is to simply throw more state money at the system. Gov. Katie Hobbs’s proposed budget includes roughly $61 million new dollars to DES. That’s unlikely to be the endpoint. The state legislature’s budget purposes going the opposite way – a 5% cut. But even if Hobbs’s proposal is accepted, it largely just makes up for the federal government’s impending cut to its administration contribution and does little to fix any large problem.

Another option is to spend more now to save money in the future by avoiding federal penalties. Specifically, invest in improved DES technology to administer SNAP (and other state-administered security programs). Arizona’s SNAP payment system currently runs on 45-year-old technology. That seems… suboptimal. Neither of us were born 45 years ago.

How can AI improve SNAP error rates?

And now might be an especially ripe time to invest in a new technology given recent breakthrough in artificial intelligence technology. Gov. Hobbs recently announced a “first-of-its-kind” artificial intelligence program for the state’s Medicaid program. Perhaps SNAP is another good candidate for AI innovation.

This might sound like an overly simplistic fix – a “deus ex machina,” like pulling a rabbit out of a hat. But with federal requirements going up, and an already lean budget running flat, the only way we forestall the complete evisceration of the state’s SNAP system is through planning, innovation, and perhaps … a little magic.

Of course, none of this answers the bigger policy and moral questions: Are we OK with 12% of Arizonans receiving SNAP assistance? 6%? And are we OK as a state with the removal of 213,000 children from SNAP in a matter of months? For analysts, this is often a discussion of large numbers and administrative procedures. But behind every number, and behind every policy decision, there’s a real human being.

Governor Spanberger signed into law the Employee Childcare Assistance Program to provide state matching funds for employer contributions toward employees’ childcare costs. She then signed a separate bill creating a childcare access calculation to estimate the annual state funding needed to support the new system.

The problem with childcare is not a lack of state intervention. It is, if anything, too much. Childcare is one of the most regulated activities in modern American life. I’ve seen it up close.

At one point, the small in-home daycare that looked after one of my children was cited for failing to include a child’s last name on a milk bottle, writing only the first name. The child in question had a rare first name (imagine something along the lines of Xanthippe) that no other child in the daycare—or perhaps in all of Virginia—shared. In context, following the rule was less of a safeguard than an exercise in box-checking.

That’s the sort of hair-splitting regulation daycares are saddled with. But no lawmaker wants to be on record for “making daycares less safe.” Instead, they opt to merely throw taxpayer dollars at the problem of high childcare costs. Pumping in money without cutting regulations or barriers to entry risks higher childcare prices for working Virginia parents.

This, sadly, seems to be a popular approach these days. From Zohran Mamdani advancing taxpayer-funded “free” childcare in New York City to Alexandria Ocasio-Cortez advocating a Universal Childcare Bill, there is no shortage of enthusiasm for promoting tax dollars as the alleged solution to high childcare costs. Hillary Clinton’s recent New York Times essay on “how to fix affordability” exemplifies this mindset.

And it’s not just Democrats: several states, including red and purple ones, have rolled out “Tri-Share” childcare cost-sharing programs that, while certainly cheaper than universal childcare, still rely on public funds

The playbook goes like this: promise affordability, enact flashy but ultimately ineffective subsidies, and then hope the voters don’t notice that nothing has been achieved.

But voters aren’t blind. If Spanberger wants to improve her lackluster approval numbers and actually deliver on affordability, she would be well advised to look at other states’ track record in subsidizing childcare without attendant reform.

Consider Michigan. Five years into the state’s failing $3.4 million “Tri-Share” childcare program, the state’s legislators are considering yet more subsidies to “tackle Michigan’s childcare crisis.” Michigan’s program has sparked criticism for both failing to increase the supply of care and tying childcare benefits to one’s employer (just as Virginia’s new program would do). If you want to know how government meddling tying a particular service to employment works out, just look at how tying healthcare benefits to employment has contributed to soaring healthcare costs.

Subsidies don’t work. And even minor regulations can combine to create real consequences for affordability.

Such requirements, like the overly specific bottle-labeling rule, might individually sound reasonable. Collectively, they become a dense web of compliance obligations that providers must manage: death by a thousand cuts.

Some regulations are especially damaging. Zoning rules can restrict home-based daycares, staff-to-child ratios cap how many children each worker can supervise, and licensing and education requirements limit the number of available caregivers. Together, these policies constrain childcare supply, raising costs and making affordable care harder for parents to access. Research suggests that requiring lead teachers to have a high school diploma may increase infant childcare costs by as much as 46 percent.

Virginia ranks 24th among the states in terms of the burden of its childcare regulations. Its ranking is lowered by unusually onerous child-to-staff ratio rules and education requirements, areas in which the state ranks 38th and 36th, respectively. Idaho, which ranks first in childcare freedom, has much lower infant and under-five child mortality rates than Virginia, showing that a less restrictive regulatory approach can coexist with strong child safety outcomes. Childcare in Idaho is also significantly more affordable than childcare in Virginia: there, childcare consumes a much smaller share of median income for a single parent.

It’s further proof that real change isn’t going to come from piling on subsidies, but unleashing supply, and that it can be done without compromising children’s safety. If Abigail Spanberger and Virginia lawmakers are truly serious about making childcare affordable, they need to stop skirting what drives the problem and go straight at it: cut the red tape.

That conclusion does not follow.

Adults deserve truthful information about the products they consume. Cannabis is no exception. Heavy cannabis use — particularly among adolescents, frequent users and people with certain psychiatric vulnerabilities — is associated with elevated risks of anxiety, paranoia, psychotic episodes and other mental health problems. Consumers should receive clear and accurate information about those potential harms.

But there is an important difference between information that evolves through legal accountability and information imposed through politicized bureaucratic mandates.

If dispensaries or manufacturers make unsupported medical claims, conceal known risks or fail to disclose foreseeable dangers, consumers should have legal recourse. Civil liability and tort law can encourage warnings and disclosures to evolve organically as evidence emerges.

That process is imperfect, but it is also flexible. It allows standards to adapt to new information rather than locking them into bureaucratic mandates that often outlast the science behind them.

Sweeping warning-label mandates imposed by federal agencies or state public health bureaucracies often become rigid, one-size-fits-all and detached from evolving evidence. They are also vulnerable to political and special-interest pleading.

More importantly, once government agencies acquire authority over what adults may hear, buy or consume “for their own good,” the temptation to shift from informing consumers to restricting them becomes hard to resist.

We have seen this pattern repeatedly in drug policy.

Cannabis itself was prohibited for decades amid exaggerated claims untethered from serious scientific analysis. More recently, public health authorities often presented opioids in starkly moralistic terms that blurred the distinction between medical use, misuse, dependence and addiction. The result was not better-informed patients, but rather a climate of fear that transformed the doctor-patient relationship and left many legitimate pain patients stigmatized, undertreated or abandoned.

Adults should retain the freedom to make choices, even when those choices involve risk.

Risks associated with cannabis are already being cited by some commentators as evidence that legalizing it was a mistake. But the mere existence of risk has never been sufficient reason to abandon adult choice. Alcohol carries well-established dangers involving liver disease, violence, accidents and mental health harms. Yet we do not conclude that adults must therefore be prohibited from drinking. Psychiatric medications carry black-box warnings for suicidality in younger patients, dependency risks and difficult withdrawal syndromes. Yet we still allow adults, in consultation with physicians, to decide whether the benefits outweigh the harms.

Cannabis should be treated similarly.

Consumers deserve honest information, not propaganda from either side. They should not be told cannabis is harmless, nor should they be subjected to a new generation of “reefer madness” exaggerations dressed up as public health concern.

There is also reason for caution when interpreting claims about causation. Associations between cannabis use and mental illness are real, but they are often complex. Many studies show correlations, particularly among heavy users and those predisposed to psychiatric illness. That is not the same thing as proving cannabis directly causes schizophrenia or other psychiatric disorders in otherwise healthy individuals.

Investigators also face a persistent “chicken or egg” problem. People experiencing early symptoms of anxiety, depression, trauma or psychosis may turn to cannabis in an attempt to self-medicate distress long before receiving a formal diagnosis. That makes it difficult in many cases to determine whether cannabis precipitated the illness, worsened an underlying vulnerability or was simply used by people already struggling with emerging psychiatric conditions.

That uncertainty is precisely why caution is warranted before converting evolving scientific associations into rigid narratives or one-size-fits-all warning mandates.

A free society does not require the government to sanitize life of risk. It requires truthful information, personal responsibility and legal accountability for fraud or concealment.

If the Illinois lawsuit prompts cannabis companies to provide clearer and more honest information about potential risks, that may ultimately benefit consumers. But if it becomes another vehicle for moral panic, exaggerated claims or backdoor prohibitionism, it will repeat mistakes the country has already made too many times before.

She should choose the Republican primary. And you should too if you’re an independent.

I say this not because I’m a Republican. And not because I want your help in returning the Arizona Republican Party to one of principled statesmen (e.g. Doug Ducey, Jeff Flake, John McCain, Jon Kyl) instead of its current composition of carnival barkers, opportunists and idolators (although help is very much needed).

Rather, I say this because the Republican Party primary is objectively more interesting.

What makes the Republican primary more interesting?
Arizona doesn’t have a U.S. Senate race on the ballot for the first time since 2014. The three main statewide offices – governor, attorney general, secretary of state – are all occupied by Democrats (Katie Hobbs, Kris Mayes and Adrian Fontes) who are all seeking a second term and are unobstructed by primary challengers.

On the Republican side of the ledger, things are quite saucy. The Republican primary for attorney general pits current state Senate President Warren Petersen against longtime electoral aspirant Rodney Glassman. Political insiders might think that Petersen will easily win. After all, Petersen wields a position of significant political power and has never lost an election. Glassman has, in the last eight years alone, unsuccessfully vied for Corporation Commission, Maricopa County attorney, Maricopa County assessor and attorney general.

But no normal Arizonan can name a single state legislator, even the senate president, and Glassman has been on the ballot so many times that many Arizonans probably assume he’s an incumbent. Added to that, Glassman has raised his usual eye-popping amount of money: $4 million on hand as of the last financial report, compared with $1.5 million for Petersen and $2.6 million for Mayes.

Why is the Arizona attorney general race a hot topic?
Put this all together and you have a close race. A January 2026 poll by Center for Excellence had Petersen at 16%, Glassman at 15%, and the rest undecided. A late-February poll from Noble Predictive Insights found Petersen at 13% and Glassman at 12%.

The campaign rhetoric reflects the hot competition. Glassman never misses an opportunity to mock Petersen’s lack of legal experience (“Petersen has never been a full-time attorney, prosecuted a criminal, or ran a law office. He only became an attorney 28 months ago”) or to accuse Petersen of legislative pork projects and funding the “insane” attorney general (Mayes).

Petersen has called Glassman “shamelessly dishonest” and other casual insults. But most of Petersen’s dirty work has been done by campaign supporters like former Maricopa County sheriff Joe Arpaio, who recently wrote, “Unlike ‘Fraudney,’ Warren (Petersen) did not switch parties multiple times, work for Raul Grijalva … (or) get accused of plagiarism … (Glassman) is a desperate political opportunist with no principles, no consistency, and no business pretending to be conservative.”

Arizona secretary of state race is equally spicy
The Republican primary for secretary of state is similarly up-for-grabs and similarly heated. The February poll from Noble Predictive Insights had state representative Alex Kolodin at 17% and former Arizona Republican Party chairwoman Gina Swoboda at 16%.

Neither has won a competitive general election before, and neither has raised large amounts of money. As of March 31, Kolodin had $300,000 on hand; Swoboda only $60,000 (the incumbent, Fontes, had nearly $800,000).

But what they lack in resources, they make up for with words. In a debate hosted last week by Arizona PBS and Ted Simons, Swoboda criticized Kolodin for never having worked in elections (“It’s one thing to sponsor a bill; it’s another to actually know how to run the software and the systems”) and said, “(Kolodin would) get destroyed in the general election (against Fontes).”

Swoboda has previously called Kolodin, “a disaster (and) an unethical jerk,” reminding voters that Kolodin was admonished by the Arizona State Bar Association for unethical behavior as a lawyer. As for Kolodin’s suitability for the job: “He does not respond well under pressure. (The secretary of state) job is nothing but pressure,” Swoboda recently said.

Kolodin has mocked Swoboda for dropping out of a campaign for Congressional District 1 after she raised very little money and commanded little support. He regularly calls her a Democrat, criticizing her for donating to Hillary Clinton in 2015 and having previously worked for Katie Hobbs at the Secretary of State’s Office. Kolodin’s ally, state Senator Jake Hoffman has accused Swoboda of allying with “radical Tucson progressives” and of negotiating suspiciously lucrative, but ill-defined personal contracts with the state legislature while chairwoman of the Arizona Republican Party.

Can’t leave out the Republican race for governor
According to most polls, the Republican primary for governor isn’t as close. But it still offers drama. In March, U.S. Rep. David Schweikert accused his opponent, Rep. Andy Biggs, of associating with anti-Semitic and extremist groups like the Proud Boys. Schweikert claims he’s the “actual conservative,” not somebody “running on resentment and grievances.”

Biggs consistently belittles Schweikert as “desperate,” and suggests that while he, Biggs, has direct access to Trump, the president wouldn’t even take a call from Schweikert. Biggs’s allies at Turning Point have hounded Schweikert online, pressing him to drop out of the primary.

Wait, Arizona has lizard people?
And the fun doesn’t stop with these three races. Recent debates between Tom Horne and Kimberly Yee for superintendent of public instruction included discussions of Rolex watches, alleged chaos at the superintendent’s office and allegations of “vanity campaigns.”

If I’ve convinced you at this point, and you’re registered in Maricopa County, go to Request.Maricopa.Vote. Again, this is only for independents (party not declared).

If I haven’t convinced you yet, then I’ll offer this parting shot: Teletubbies, hallucinations, psychedelic drugs and lizard-people have all been referenced in the Republican primary for State Legislative District 3.

Many of these doctors entered the U.S. on J‑1 visas, which allow foreign medical graduates to complete residency and fellowship training in U.S. hospitals but ordinarily require them to return home for two years afterward. A J‑1 visa waiver offers an alternative: Physicians can stay if they agree to work for three years in federally designated shortage areas, typically in primary care fields like psychiatry, internal medicine and pediatrics. After years of training, they planned to move directly into these roles serving underserved communities.

Instead, they now face an administrative deadline. If the government doesn’t approve their waiver applications in time, it will force them to leave the country and begin the two-year return requirement. The waiver serves as a bridge that lets them move from training into practice, typically by transitioning directly to H‑1B status. Government delays bring that bridge down.

President Trump’s new $100,000 H‑1B visa fee, up from a few thousand dollars previously, only raises the stakes and puts continued employment out of reach for many hospitals that need them most.

Clinics preparing to serve vulnerable populations will lose physicians they had already recruited, and patients will face longer waits or no care.

What looks like a physician shortage is, in fact, a policy failure.

The U.S. relies heavily on international medical graduates. Roughly one in four practicing physicians attended medical school outside the U.S. or Canada, and many serve in the very communities policymakers say are hardest to staff. Far from peripheral, these physicians are essential to the health system. In California, international medical graduates account for nearly 40 percent of primary care physicians.

Yet after completing residency training in U.S. hospitals — training substantially subsidized by Medicare — these physicians face a gauntlet of approvals from multiple federal agencies. The waiver program itself requires sign-off from the Department of Health and Human Services, the State Department, and U.S. Citizenship and Immigration Services. A delay at any step can derail the entire process.

Federal agencies have created exactly that situation. After processing slowed late last year and a backlog built, applications that once moved in weeks now sit for months, with little public explanation. Physicians who signed contracts and planned to start in July now sit in limbo.

This kind of fragility is the predictable result of layering bureaucracies on top of one another. Access to care depends on synchronized decisions across multiple government agencies, so even a simple slowdown can trigger a system-wide failure.

Even the structure of the waiver program reflects a deeper problem. In exchange for remaining in the country, physicians must agree to work in government-designated shortage areas. This approach assumes policymakers can effectively direct the distribution of medical professionals. But as with other forms of central planning in health care, it produces distortions and unintended consequences — a pattern even mainstream health policy research acknowledges.

Nowhere is that clearer than in the $100,000 visa fee employers must pay if these physicians are forced to leave and reapply. Wealthier hospital systems and high-revenue specialties may be able to absorb such costs. Rural hospitals and community clinics cannot. A policy ostensibly designed to protect American workers ends up depriving underserved communities of care.

State licensing boards, by recognizing only a narrow set of accredited residency programs — effectively granting a monopoly to a single U.S. accreditor — limit training positions and block alternative pathways to licensure, leaving physician supply under administrative control. Immigration barriers are simply another layer in a system that manufactures scarcity and then struggles to manage it.

The government may force a psychiatrist ready to serve a low-income community to leave the country. Bureaucratic delays may leave a rural clinic with an empty office. Patients who were already waiting will wait even longer.

Meanwhile, Canada is actively recruiting many of these same physicians and revising its immigration rules to attract them. The problem is a system that prevents trained doctors from practicing medicine.

Addressing this problem requires rethinking the premise that access to medical care should hinge on bureaucratic timelines and centralized workforce planning. If a physician has completed accredited training and is qualified to practice, the government should not stand between that physician and a willing patient.

Until that changes, policies will keep pushing doctors away, and patients will keep paying the price.

If the president wants to help himself and his party ahead of this year’s midterm elections, the most effective thing he can do is eliminate the tariffs. The evidence in favor of this move is overwhelming, and it comes from his own tenure.

As an obligatory reminder, tariffs are levied on American importers who pass the costs on to American businesses and consumers. Those insisting tariffs are “paid by foreigners” must now dispute not just history but the present.

The Cato Institute’s Scott Lincicome and colleagues reviewed a year of data from Trump’s tariff regime and found that “the higher costs from tariffs passed through to prices paid by Americans at a rate as high as 96 percent.”

Using daily price data from major U.S. retailers, economists from Harvard Business School found that the 2025 tariffs raised consumer prices almost immediately, with imported goods rising roughly twice as fast as domestic ones and adding nearly a full percentage point to the overall Consumer Price Index by October 2025.

This finding isn’t unique. My colleague Jack Salmon examined 56 quantitative studies produced over the last 30 years and found 19 showing tariffs raise prices and zero showing tariffs lower prices.

This reality has a real impact on Americans. The Tax Foundation put the cost of the tariffs at roughly $1,000 per American household in 2025, with another $700 coming in 2026 from the Section 232 and Section 122 levies, which were left unaffected by Supreme Court’s recent rebuke. It shows up in grocery bills, appliance prices, and clothing costs — routine purchases for working-class households.

The damage goes beyond prices. Salmon’s literature review finds 25 studies documenting negative effects of tariffs on productivity and economic output. None of those studies show positive effects. Across Chile, India, Indonesia, Brazil, Hungary, Canada, and the United States, the pattern is the same: Lower tariffs raise productivity; higher tariffs reduce it.

What about revenue? The Tax Foundation projects $956 billion from the remaining tariffs over a decade, falling to $697 billion once the economic damage, including the uncertainty and foreign retaliation, is counted. That’s a sign of a bad policy.

To be fair, some supporters of Trump’s tariffs were honest about their impact. Isn’t that the whole point? Raise prices and hurt the businesses reliant on foreign goods and inputs to help domestic manufacturers. We’re told that conceding to the working-class white voters who demand protectionism is worth the price.

The political results are now available for inspection too. A new CBS News poll shows that Trump’s approval is underwater with most voters, including white voters without college degrees, among whom his approval rating fell from 68 percent last year to 46 percent today. This is unsurprising. The supposed beneficiaries of economic nationalism are instead its most exposed victims.

It didn’t help that manufacturing employment, which was promised to boom, kept declining throughout 2025. And economic growth decelerated despite the major investment and energy around AI.

The tariffs also produced a final insult: They energized the very Washington insiders the president promised to defeat when he first entered the White House in 2017. When tariffs are numerous, arbitrary, and have an exemption process attached, every affected business must hire a lobbyist to survive.

Data from Lincicome and his coauthors show that “the number of registered clients for tariff-related lobbying increased by 218 percent in 2025 with respect to the previous year…. Meanwhile, trade-related lobbying expenditures reached more than $900 million in the first half of 2025 alone and were 28 percent higher than in the first half of 2024.”

All that lobbying pays off, as evidenced by how the global tariffs have become riddled with exemptions. It’s also good for lawyers. More than 2,000 importers have now rightfully filed suits (an expensive process) seeking refunds on over $160 billion in tariffs the Supreme Court ruled were illegally collected.

The swamp was not drained. It was fed. Small businesses, who usually do not have the luck or resources to access the right people in the administration, pay the full tariff while their larger competitors petition for relief.

The president still has time to change course. The economic case for dropping the tariffs is airtight. The political case is increasingly urgent.

Now, one could debate how soon and how reliably AI will create such bounty. One could also debate whether deflation (falling prices) with steady wages induced by AI-led productivity is a problem at all. Everything would become a lot more “affordable,” of course. There’s also an argument that higher real (after adjusting for inflation) interest rates are needed to induce savings and investment to build AI. Whether the Fed should act in anticipation of a productivity bonanza is another question.

But today the Fed faces essentially the opposite problem, a stagflationary shock that looks eerily like 1979. Inflation never really went away. It is now surging, thanks to tariffs and energy costs via a conflict with Iran. Should the central bank fight that inflation by raising rates, swiftly incurring President Donald Trump’s wrath and risking a weaker economy? Or should the Fed once again look through a price-level rise, hoping that the economy will stabilize at higher prices, and swiftly incurring the wrath of regular people already unhappy about today’s high prices?

Warsh, who was confirmed by the Senate last week as chair, has advocated that the Fed reduce its balance sheet by selling assets. Most economists believe reserves past $2 trillion or so have little long-run effect on the economy. But the Fed may be tempted to reduce reserves until bank operations are squeezed, in order to tighten without raising interest rates, a version of monetarist restraint. The United States tried a similar idea in 1980, imposing credit controls by tightening bank regulation. They produced financial chaos and stagnation.

Warsh wants to reexamine the Fed’s models and overall approach. This is wise. Inflation, peaking at 9 percent in June 2022, was a failure that the Fed has not satisfactorily accounted for. It was not an individual failure. The Fed’s actions were backed by consensus in and outside the central bank. It was a collective, conceptual, institutional failure. The models don’t work. The forecasts don’t work. But there is no off-the-shelf alternative. Nobody really knows how monetary policy works, and certainly not with the complex technocratic expertise that the Fed pretends. Other forecasts do not reliably outperform the Fed’s.

The Fed should instead act with more humility. Recognize the fog in which it is trying to steer the ship. Refrain from acting (again) based on forecasts and what-if analyses that have proved unreliable.

Bigger challenges lie ahead. Because the national debt is beyond 100 percent of gross domestic product, every percentage point that the Fed raises interest rates increases interest costs on the debt and thereby the deficit by 1 percent of GDP. Neither Congress nor the president will be happy about that.

Fiscal pressure on the central bank will mount. Today’s precedent for reduced Fed independence is the era from World War II to 1951, when it was obliged to hold down long-term rates for fiscal reasons, not 1972, when President Richard M. Nixon pressured the Fed for election-year ease.

Countries that run uncontrolled deficits soon face higher borrowing costs. The Fed will feel pressure to hold down those costs. “Moderate long-term interest rates” is, after all, part of the Fed’s legal mandate. With the precedents of massive bond buying in the 2010s and again in 2020, it will be hard to resist. Rising yields will also tempt financial repression. The Fed will be tempted to force banks, insurance companies and other institutions to hold Treasury debt, impose capital controls and so on.

That’s the optimistic scenario. In the next crisis, 2020 will replay at a larger scale. The Treasury will want to borrow trillions for bailouts, stimulus and likely military investment. But bond investors will be skittish, having suffered a substantial loss due to inflation last time they lent to the U.S., and with still no plan for the government to start repaying debts. Inflation expectations are primed to jump, so inflation itself can break out even more quickly. Pressure for the Fed to monetize government debt will be immense.

There is a limit to how much the Fed should resist. If Congress and the president want massive money-printing to finance a crisis response, should the central bank really force the government to borrow at much greater cost, spend differently, sharply raise taxes, restructure debt or withdraw from a confrontation? Tax, spending and foreign policy, even if unwise, are far outside the Fed’s limited mandate.

I would rather see a Congress impose a strong price-stability mandate, with restrictions on bond-buying, but one that Congress suspends in times of crisis, as it did under the Troubled Asset Relief Program in 2008. Meanwhile, the Fed needs to face squarely its role in facilitating the fiscal blowout of 2020 and begin to think about resisting next time.

Many challenges lie ahead. But challenges are opportunities. Great leaders are forged by their wisdom in the face of adversity. And every time the president tweets his disapproval, Warsh’s reputation for independence will grow.

Even so, the summit told us a lot about the bilateral relationship and what to expect in the months ahead. Little of it is good news for the US side.

First, it’s clear that China has weathered President Donald Trump’s multi-term tariff assault and effectively fought the US to a stalemate. The 2020 “Phase One Deal” was tellingly unbalanced: China made a laundry list of promises — grandiose and detailed purchase commitments, greater market access, regulatory reforms, and more — while the US did almost nothing. This time around, the promised purchases are much smaller and vaguer, while possible future tariff reductions are mutual. China arguably isn’t the United States’ economic peer today, but you’d be forgiven for thinking that based on the summit’s deliverables.

China achieved this draw through several channels. The government responded to rounds of US tariffs with not only tit-for-tat levies on America’s exports, but also efforts to expand economic ties — via trade agreements, diplomacy, regulatory approvals and unilateral tariff reductions — with Europe, Asia, Latin America and Africa. Along with longstanding (and questionable) government support for domestic industries, Beijing bolstered its economy’s resiliency by stockpiling energy and food while investing in infrastructure to encourage trade with non-US markets.

Chinese exporters, meanwhile, sought alternative destinations for finished goods such as solar modules and EVs, as well as intermediate inputs incorporated into “Made-in-Not-China” products – often exported to the United States at lower tariff rates. As a result, declining Chinese exports to the US were more than offset by surging sales elsewhere, and China’s trade surplus exceeded $1.2 trillion last year — a new record. At the same time, Brazil, Argentina and other suppliers have displaced US products in China’s agricultural import mix.

Beijing also deployed asymmetric retaliation, most notably by limiting rare earth minerals and related magnets that US manufacturers need. As most economists will tell you (and as we learned a decade ago), export restrictions can be self-defeating in the longer term, but they cause angst and short-term pain for influential companies and defense contractors — angst that clearly got Washington’s attention last fall. Other asymmetric moves included regulatory and investment delays and denials for US exporters and service providers, most recently Beijing’s decision to block Meta Platform Inc.’s $2 billion acquisition of Singapore-based AI startup Manus.

Trump’s tariff onslaught certainly wasn’t costless for China, but last week’s summit shows that the pain has been sufficiently mitigated via a playbook that few other nations, if any, can copy.

The deals announced were also noteworthy and not for their terms, but for the market’s collective shrug. As usual with US-China summitry, the initial reporting was on the Three Bs: Boeings, Beans and Beef, with both sides announcing billions of dollars in new soybean and aircraft purchase commitments and regulatory relief for US beef and chicken exporters. Yet the market was underwhelmed. Boeing Co. shares declined after Trump said China ordered just 200 jets, well below the 500 investors were expecting and the 300 China promised to purchase in 2017, before trade hostilities first inflamed. Crop prices initially sagged before recovering some to levels still below pre-summit highs.

The market’s ambivalence is warranted, given past Chinese noncompliance. Data from the Peterson Institute show that China came nowhere close to meeting its Phase One commitments to purchase hundreds of billions of dollars in US manufactured and agricultural goods — and imported less than pre-trade war trends would predict. China met last year’s agreed upon soybean target, but that was well below levels hit in 2024 and earlier.

Investors also seem to understand that economic realities have, along with geopolitical thorns like Iran and Taiwan, left both capitals uninterested in escalation. In China, a humming export machine and frontier-level achievements in certain advanced industries mask an economy that’s still bogged down by a Japan-esque property bust, stifling public and private debt, depressed household spending, stagnant productivity, high youth unemployment, and serious demographic headwinds.

The US, meanwhile, is riding an AI wave and strong corporate earnings, but is also battling stubbornly high inflation, its own government debt problems, jittery bond markets, Iran-related supply chain disruptions and persistent consumer doldrums. Both sides still have plenty of weight to throw around, but neither is itching for a big fight. In a welcome — albeit narrow and late — change, they’re actively looking for ways to lower barriers to bilateral trade and investment.

This brings us to the summit’s final lesson: scattershot US trade policy has, coupled with an ill-conceived war in Iran, weakened Washington’s hand on China. Blanket tariffs and related bellicosity have alienated allies and pushed neutral countries like Vietnam closer to Beijing for good reason. Indeed, additional tariff reductions for low-value Chinese imports could make them more competitive than goods from alternative exporters like India, which is precisely the opposite of what US tariff policy was intended to accomplish. And the strategic value of tariffs has been undermined by a series of court rulings against the broadest levies – something US “trade deal” partners have surely noticed.

At home, chaotic tariffs have raised prices for food and other consumer goods while, as aluminum shows, increasing supply chain vulnerabilities and dampening investment for manufacturers. Pledged tariff reductions could have grumpy American consumers asking why we have taxes on “non-strategic” items no matter where they come from. The war with Iran has exacerbated the economic pain all while China’s trade surplus and geopolitical influence have expanded.

This doesn’t mean the bilateral relationship is in a good place overall or that, without Trump’s global tariffs, the US would’ve emerged from last week’s summit victorious — whatever that means. But in a wrestling match between two global heavyweights, every bit of leverage helps, and the Trump-Xi summit showed the perils of carelessly squandering it.

Among certificate programs, failures are further concentrated in specific fields, such as culinary services, where all programs are expected to fail, and cosmetology, where 92.5 percent are predicted to fail. Cosmetology schools have responded by launching an intense lobbying campaign to either keep this new rule from being implemented or get new legislation that exempts cosmetology from the rule.

Their lobbying should be ignored for three main reasons.

First, their claims can be quite misleading. Elizabeth Faye and Stacy Wells argue that the “beauty and wellness industry represents one of the last true pathways to middle-class prosperity.” But recall that a program only fails if its graduates earn less than a comparable worker with a high school diploma. As Preston Cooper notes, the typical graduate from a cosmetology program earns $27,000, an income that most would not consider a pathway to prosperity. They also claim that the regulations “threatens to close 92% of beauty and wellness schools.” This is not the case. Programs that fail the earnings test are not forced to close; they simply lose access to the student loan program. Moreover, as Rachel Fishman and Ewaoluwa Obatuase point out, it is possible for these programs to maintain access to Pell grants in some cases.

The second reason to ignore the lobbying by cosmetologists is that many cosmetology schools already forgo participating in the federal aid programs. For example, in Texas, there are 824 cosmetology and barber schools, and only 117 (14 percent) of these participate in the federal aid programs. The existence and persistence of cosmetology schools that do not participate in the federal aid programs provides definitive proof that taking away access to student loans will not mean the end of cosmetology. Indeed, scholars have found that “these institutions can and do survive, and even thrive, alongside their aid-eligible counterparts.”

Third, the main effect of federal aid for cosmetology programs is to increase prices. Research by Stephanie Riegg Cellini and Claudia Goldin compared cosmetology programs that participated in the federal aid program to those that didn’t and found that the programs that utilized federal aid charged more. For example, in Florida, the state with the most comprehensive data, cosmetology programs that were eligible for aid charged 82 percent more than similar programs that did not utilize federal aid, which was roughly equal to the amount of federal aid their students received. In other words, federal aid didn’t make cosmetology programs more affordable for students; it padded the bottom line of cosmetology schools.

The bottom line is that Congress and the Department of Education did the right thing in establishing the Do No Harm rule to limit access to student loans for programs with low earnings outcomes. The accountability this rule imposes is eminently defensible on principle, easy to meet in practice, and yet effective in weeding out many financially predatory programs.

Congress and the Department should ignore any special interests, including cosmetology schools that are seeking to escape meeting even this low accountability threshold.

But since when is resisting politicians the litmus test for a central banker? Grading Powell chiefly for defending Fed independence is like naming a goalkeeper man of the match because he defied the manager’s tactics and organised his defence better, while omitting the three soft goals he conceded. Independence matters because it helps undergird price stability. Yet on that ultimate test of keeping inflation low, Powell failed.

He certainly faced testing circumstances, including a once-in-a-century pandemic, America’s lurch to protectionism, and conflicts in Ukraine and Iran pushing up energy prices. All were supply-side shocks, squeezing output and raising prices to force difficult macroeconomic trade-offs.

He deserves praise, too, for the eventual course correction. Once Powell’s federal open market committee (FOMC) accepted it had misread the post-pandemic inflation, it squeezed broad money growth and lifted rates by 5.25 percentage points in 16 months. Inflation fell gradually to 2.3 per cent by spring 2025 without an unemployment spike. It might have drifted toward the Fed’s 2 per cent target but for Trump’s tariffs and foreign-policy adventurism.

But in truth, this was cleaning up a mess the Fed itself had made. Policy tightened meaningfully only in spring 2022, when the Fed’s preferred inflation gauge was already at 7 per cent. The worst inflation in four decades squeezed real wages, redistributed wealth arbitrarily and confused households confronting a permanently higher price level. It also fed the ignorance and anger now spawning misguided price-control proposals.

The Fed had mistaken clogged ports and labour disruptions for “supply shocks” it couldn’t affect, when industry was really hitting capacity constraints amid rampant demand fuelled by monetary excess. Powell’s Fed famously thought inflation was “transitory” and would quickly dissipate. In reality, total nominal spending was soaring, underpinned by loose money.

Yes, the US Congress’s vast borrowing didn’t help. But an independent central bank targeting inflation should, in theory, set monetary policy accounting for this. The truth is that, while genuine supply shocks exacerbated inflation’s peak, it was monetary policy accommodating Congress’s fiscal incontinence that drove almost all excess US inflation since 2021. This is very different from the UK, where our economy’s supply-side weakness has played a comparable role to macroeconomic largesse.

Worse, the Fed’s own framework rationalised the delayed response. Its 2020 review, overseen by Powell, was an artefact of the post-2008 world of low inflation, falling neutral rates, near-zero policy rates and endless fretting that monetary policy had been insufficiently expansionary. The Fed embraced “flexible average inflation targeting”, saying that after periods of below-target inflation it would likely aim for inflation “moderately above” 2 per cent for some time, while focusing on employment “shortfalls” from maximum employment.

In effect, the Fed made the inflation target more discretionary. It had no specified averaging period, no numerical definition of “moderately above”, and no binding rule compelling the FOMC to tighten when nominal spending surged. The Fed was fighting the last war, cheered on by policymakers who believed the post-Great Recession lesson was to “run the economy hot”. As late as November 2021, it was still treating inflation as largely transitory, expecting supply constraints to ease and inflation to fall back as the pandemic economy normalised.

So, yes, Powell clearly deserves praise for resisting Trump’s pressure and defending the Fed’s policymaking authority. But his job wasn’t just to preserve central bank independence, nor firefight high inflation. He helped write the fire code, allowed the fuel to ignite, and arrived far too late with the hose. Judging him just by the actions of a president who deemed him an enemy lets Powell off too easily.

Those subsequent harsh measures presumably include the Trump administration’s removal of the U.S. Space Command from Colorado to Alabama, the dismantling of Boulder’s National Center for Atmospheric Research, the freezing of hundreds of millions in federal spending on social services, the cancellation of transportation grants to the state, and the veto of federal funding for a Colorado pipeline.

Perhaps Governor Polis understood commuting Peters’s sentence as politically necessary. Perhaps he understood the President as sending a message – namely, compliance with presidential orders is the only way to end the budget pain that the President is inflicting on Colorado.

But the Peters commutation sends a second message – namely, if you break the law to advance the president’s political prospects and get put in prison, the President will turn off the spigot of federal spending in your state until you are released.

The President shouldn’t be in the business of distributing “get out of jail free” cards to election tamperers, and Governor Polis shouldn’t be helping the president hand them out.

The administration is requesting $70.5 billion for the most critical, high-value munitions, representing a 188% increase from last year’s budget.

In the hearings, we heard much concern expressed from members of Congress on both sides of the aisle. Concern about the rapid depletion of munitions for an unauthorized war in Iran. Concern about the fallout of draining our stockpiles to arm Ukraine. Concern about the decrepit state of the defense industrial base and its inability to respond nimbly to strategic demand.

All of these concerns are valid, but none are at the root of the problem. The United States is in a munitions crisis because both the executive and the legislative branches, regardless of which political party is in power, failed to exercise restraint in protecting the U.S. war chest.

As budget discussions heat up over the next few months, Congress shouldn’t blindly grant the gargantuan increase in funding for and production of critical munitions. Doing so would amount to an expensive band-aid. Instead, lawmakers should ensure reforms to protect America’s stockpiles from strategic folly are reflected in an appropriations package considered by Congress.

We must acknowledge where this problem started. Following Russia’s invasion of Ukraine in 2022, Congress and the Biden administration collectively supported the unprecedented use of a tool called Presidential Drawdown Authority (PDA) to expeditiously provide weapons to Ukraine from U.S. weapons stockpiles.

But PDA, by law, was intended to be used narrowly in cases of “unforeseen emergency” to provide support to a foreign country in order to protect the U.S. national interest. Even in such circumstances, the law does not allow unfettered access or a blank check to U.S. weapons. Prior to 2021, Congress maintained an annual cap on the authority at $100 million worth of weapons. Congress, in a series of decisions that reflected a severe lack of judgement, authorized three separate increases to the cap in supplemental appropriations bills for Ukraine. This dramatically expanded the authority, allowing up to $14.5 billion in drawdowns for Ukraine.

Congress and the Biden administration together dug a huge hole in the U.S. weapons arsenal by providing weapons support to Ukraine in this manner. And they did so with little regard for the long-term consequences, namely, how such rapid depletion without rapid replenishment would limit America’s readiness and ability to respond to conflict elsewhere in the world.

The Trump administration inherited U.S. weapons stockpiles at a deficit. The president himself acknowledged the significance of Biden’s massive strategic error. But he committed a serious error of his own in starting a war with Iran despite knowing the demand it would place on U.S. weapons stocks while still being in the hole from Ukraine.

The numbers from the Iran war thus far are staggering. A recent report from the Center for Strategic and International Studies (CSIS) estimates the expenditure of some of our most high-value weapons — like Tomahawk, JASSM, and Patriot missiles — to be in the thousands. The report notes this is likely a depletion of “more than half of pre-war inventory” for some of these weapons.

Such estimates help explain why the Trump administration’s budget includes such massive increases in quantities requested. Take Tomahawk missiles, for example. Last year the Navy procured a mere 55 Tomahawks. This year, the new target for procurement is 785, an increase of more than 1,000%. (It is worth noting that, even if the Trump administration were granted its budget request in full, it would take at least three to five years for these high-value munitions to be delivered.)

Pulling the thread on how we got into this munitions crisis tells us that we have a foresight problem. American leaders are happy to arm a foreign nation in need or to jump into a war, but they fail to see around the corner and consider what the United States is trading off or how we are limiting ourselves strategically.

Congress has the opportunity this budget cycle to learn from past mistakes. Congress can condition munitions funding on receiving a strategy and timeline from the administration on the war in Iran; impose limits on how many high-value munitions must be maintained and preserved for U.S. readiness at all times; and even proactively reform PDA to prevent it from ever again being used at a reckless scale, as it was in Ukraine.

The unfortunate reality is that political appetite for such needed reforms is extremely low in Congress. Despite the growing influence of America First foreign policy, it is still anathema in Washington to put sound defense strategy above political risk, defense lobbyists, and foreign influence.

This apathy could prove disastrous. Amid a munitions shortfall, the U.S. is balancing a hot war in Iran, a military campaign in the Western Hemisphere, a proxy war in Ukraine, and the looming threat of a war with China in the Indo-Pacific.

The United States is not invincible, and it cannot fight nor manage multiple conflicts at a time successfully. There is no quick fix that can make the U.S. munitions arsenal immune from scarcity, tradeoffs, and mathematical reality.

The budget request no doubt prioritizes critical munitions in quantity and dollars. However, unless Congress can reckon with the strategic failures that created and worsened the stockpile deficit and pass the necessary reforms to protect the war chest going forward, we will circle back to the same problem on the same balance sheet again soon.

Republicans have long portrayed themselves as defenders of free enterprise against government encroachment. To vote for Democrats would be to vote for socialism, they said. But the second Trump administration is going well beyond the Biden administration’s industrial-planning schemes by inserting the federal government directly into the corporate bedroom.

Many of the issues that Trump-style economic nationalism seeks to address are not trivial matters. China is a serious strategic competitor. Supply chains for semiconductors, critical minerals, and weapons systems are important to national security and economic resilience. And conservatives are right to be frustrated by a federal government that cannot permit projects quickly, buy weapons efficiently, or maintain a serious industrial base without drowning it in red tape.

But identifying a legitimate problem does not mean heavy-handed government intervention is the solution, regardless of which political party controls the White House. Having Washington become investor, creditor, regulator, customer, and corporate overseer — a situation that conservatives would have rightly considered anathema until Trump took over the Republican Party — is a recipe for cronyism and economic sclerosis, not security.

The Biden administration’s industrial policy was bad enough. Through the CHIPS and Science Act and the Inflation Reduction Act, subsidies, tax credits, loans, and mandates were employed to push capital toward politically favored industries. Many Republicans criticized that approach as industrial planning, and they were right. Industrial policy can sound good on paper, but as corporate America is finding out, making business decisions based on government incentives subject to change from one administration to the next can easily backfire.

But instead of merely rolling back Biden administration interventions, the Trump administration has built on them by having the federal government take ad hoc ownership in a slate of American companies. It has done so randomly, in an opaque manner, and without regard for congressional input or constitutional concerns.

The administration and its supporters also apparently don’t care that a future Democratic administration can use the precedents being set and the equity stakes it inherits to engage in mischief. Indeed, were it a Democratic administration trying to concoct a pseudo-government investment fund for the president to play with, the outcry from self-described conservatives and congressional Republicans would be deafening.

The Biden model was Washington telling companies where it wanted capital to go. The Trump model has Washington supplying the capital in exchange for the power, leverage, and control that come with government ownership. So, while subsidies distort investment, federal ownership creates a deeper problem by forcing firms to make decisions based on political rather than market considerations. 

Once the federal government owns part of a company, every regulatory, procurement, tax, trade, antitrust, and permitting decision involving that company or its rivals becomes suspect. Is Washington enforcing neutral rules, or protecting its investment? Is a competitor being treated fairly, or punished for lacking the right political sponsor? Is the taxpayer being protected, or is a favored firm being propped up because officials do not want their deal to fail?

A government that owns the players cannot be trusted to call the game fairly.

While a government investment can initially boost a company’s immediate prospects, taking a bite of the proverbial apple comes with consequences. Broadly speaking, firms may come to see the path to capital as running through Washington. As we almost witnessed with Spirit Airlines, ordinary commercial problems will be sold as a national security emergency. Bad investments will be packaged as strategic necessities. And executives (and their lobbyists) will discover that the right sales pitch is more important than the right business plan.

Traditional conservatives know this. But the Trump-rebranded right should understand that if the solution to every economic problem is tariffs, subsidies, loans, targeting tax breaks, and federal equity stakes, then it becomes impossible to critique the left’s interventions credibly.

They’ll just have a rival industrial policy.

Republicans can’t denounce the government “picking winners and losers” when Democrats do it and then call it “deal-making” when Trump does it. They can’t warn that Democratic industrial policy is socialism while claiming that Trump’s industrial policy represents “America First.” And they certainly cannot credibly defend free enterprise while cheering federal ownership of private firms.

Markets discipline error while politics rewards influence. And once the federal government starts owning companies it regulates, subsidizes, and contracts with, that discipline begins to disappear. Conservatives used to make this argument when criticizing the left — and some still do.

But those self-described conservatives who jettisoned whatever principles they had for political fealty, and the money and influence that can come with it, will have zero credibility when the other team gets the chance to open the gifts the Trump administration left for them.

There’s a box to check on your tax return if you are legally blind. The 32 million Americans who qualify can receive an automatic tax deduction of $2,000 if single and $1,600 if married. With Social Security, there’s a monthly income ceiling above which people can’t receive disability benefits. In 2026 a blind person could receive disability benefits while making up to $2,830 a month. For any other disabled person, the limit is $1,690—40% less.

There seem to be four reasons for the differential treatment. The first is historical oddity. The earliest blindness-related tax deductions had to do with 1940s-era carveouts for readers and guides that evolved into a personal exemption.

Second, there’s an assumption that blind people have certain unique expenses. They may need specialized equipment (including modernized reading devices and trained guide dogs), personal assistance with household tasks, alternative means of transportation and so on.

Third, there are economic disparities. Only half of blind American adults are employed, while those with jobs make 82 cents for every dollar earned by the working population at large.

The fourth explanation for discrimination in favor of blind people is the best: They had better lobbyists. The National Federation of the Blind has been a political force since 1940—decades before much of the rest of the modern disability rights movement exploded onto the scene in the 1970s. It has fought hard for economic benefits and a say in government affairs to this day, as has the American Council of the Blind.

I’ve used a wheelchair my entire life due to a disability I was born with. I rely on help from others for everything from laundry to housecleaning. I buy equipment ranging from shower benches to portable ramps. And that’s just me. Other disabled Americans pay for everything from ventilators to speaking devices to surgeries without receiving the tax deduction reserved for blind people.

As for economic disparities, blind people actually fare better than disabled Americans as a whole. Their employment rate is higher, and though full-time blind employees make a bit less than the average full-time disabled worker, earning rates tip back when part-timers are included.

I’m certainly grateful to disability rights organizations for their role in advancing opportunities for disabled Americans. But I’m frustrated by the persistence of special preferences in tax and disability benefits. They’re not for all of us, but only those who are the beneficiaries of that legacy of extra political access.

Squaring things up will require broader conversations about the costs of being disabled in America. Most disabled people have to pay our taxes, then search for public benefits separately. Most disabled people have to make hard trade-offs between employment and disability checks. There’s no reason “benefits cliffs” should imperil everyone with a disability except the blind.

Fair policy, rather than privileges for a few, should be disabled Americans’ common goal.

The pardon power is different from other constitutional powers. It is unchecked and absolute: There is no other branch of government that can interfere with a president’s pardons. The Constitution was created in the shadow of absolute royal power; it was designed to resist the corruption of such power. The division of government into three separate branches is meant to ensure a system of checks and balances on government power. From this perspective, the pardon power stands out: The unilateral pardon power looks like the last remaining remnant of royalism in the Constitution.

State governments can also issue pardons. Historically, the governor of each state had the power to pardon state crimes. But over time, that has changed. In many states, the power to pardon criminal offenders is no longer concentrated solely in the governor’s office. In some states, that power is split between a governor and a group of appointees. Why the change? Some people thought that a governor with absolute pardon powers didn’t have the time to make thoughtful, well-considered decisions. Others feared that a governor might hand out pardons to political supporters — or might be tempted to take a payment in exchange for a pardon. That is why the law of many states has moved away from a solo, absolute pardon power.

Today, however, the way the president exercises his pardon power is disturbing and unprecedented. The current resident of the Oval Office grants pardons much more often than previous presidents. He has pardoned disgraced politicians, many of his campaign supporters and even business associates of his own family.

Do Donald Trump’s pardons give us good reason to change the rules of the Constitution? Is it time to put the brakes on solo pardon power? I think there are lessons to be learned from the president’s abuse of office. The Constitution encourages cross-branch consensus in almost every other realm of federal power, so that different parts of government must work together to make important decisions. It’s time to change the rules on pardons and remove the last trace of absolute, unaccountable royal power from the Constitution.