Stock

Scott Lincicome and Alfredo Carrillo Obregon

Buried in yesterday’s Census Bureau release of the latest US international trade data for full-year 2024 was a nice reminder that, for all the talk of tariffs and “de-globalization,” American consumers and companies aren’t having any of it—at least not yet. In fact, inflation-adjusted US goods imports hit an all-time high last year (Figure 1).

As shown in Figure 2, moreover, this surge was fueled by large increases in imports of consumer goods and capital equipment, the latter of which are used by American manufacturers to make other stuff:

Of course, these data come before the 10 percent tariffs that President Trump just imposed on all Chinese imports, as well as the tariffs he’s threatened on imports from Canada, Mexico, Europe, and elsewhere—taxes that could forcibly reverse the above trends. In the meantime, however, the data are a nice reminder of Americans’ revealed preferences for imports and that while American politicians might be souring on trade, American companies and consumers most definitely aren’t.

Walter Olson

In a remark that might seem to sum up Congress’s current approach to its oversight role, Sen. Thom Tillis (R‑NC) acknowledged the other day that the Trump administration’s opening moves to cut spending and do away with agencies without congressional approval were in some cases not lawful, but said, “Nobody should bellyache about that.” In particular, he said, “That runs afoul of the Constitution in the strictest sense. But it’s not uncommon for presidents to flex a little bit on where they can spend and where they can stop spending.” (I briefly summarized the ongoing lawbreaking spree in this space on Monday, and further likely illegalities have come to light since then.)

One group of thinkers who were given to bellyache when officials acted in unconstitutional ways were the framers of the Constitution, who had in recent memory the “long train of abuses and usurpations” committed in the name of the British crown. To guard against a repeat, they provided in Article II, Section 1, that the president take the following oath: “I do solemnly swear (or affirm) that I will faithfully execute the office of President of the United States, and will to the best of my ability, preserve, protect and defend the Constitution of the United States.” In Article VI, along with declaring that the Constitution “shall be the supreme law of the land,” they provided that Sen. Tillis, along with all his legislative colleagues, “be bound by oath or affirmation to support this Constitution.” 

A lot of good that seems to have done. 

These days both presidents and lawmakers tend to follow what I’ve called the goalie theory of constitutional enforcement, in which their team gets to take as many shots at the opponents as it can manage without bothering much about which may be constitutional and which not, “and the courts as goalies then block what shots they can.” If you buy into this theory, it can even seem mysterious and self-defeating for lawmakers or presidents to get all agonized about whether what they’re signing is constitutional. Don’t they know that’s the job of the courts? 

If the Framers adhered to the judiciary-as-goalie theory, it’s not easy to see why they would have made a point of binding the other two branches with oaths to support the Constitution or what they expected the oaths to accomplish. The fact is that taking a step as president that one knows violates the Constitution, or voting as a member of Congress for a bill one knows to do so, looks very much like a violation of one’s oath. 

This may shed some light on one argument heard in some quarters lately, which is that the incoming administration, influenced perhaps by some of its techie supporters, has decided to “move fast and break things.” 

We can all recall instances of tech innovations that ignored possible legal obstacles and then proved their worth. Could sharing apps or social media have gotten off the ground without seeming to disregard some laws relating to unlicensed local business practices or the use of copyrighted material? The theory, when it works, is that the industry or service newly created is so useful and valuable that the law will bend to accommodate it, heeding the clamor of satisfied users who don’t want to go back to the old ways. 

Perhaps those moving fast and breaking things in the new administration believe that what they are doing won’t be reversed, even should the law catch up with them. The public will be delighted at the changes, a grateful Congress will pass new bills retroactively blessing the irregular measures taken, and so forth. Another, not quite as sunny, possibility is that the unlawful moves will translate into facts on the ground that simply cannot be reversed whether or not the public winds up happy. 

Federal prosecutors and FBI agents fired contrary to civil service laws may eventually prevail in court, but many will have found other jobs in the interim and not care to come back. Improperly deported persons will in many cases settle into life abroad and not return, even if a court a year from now says they can. A nonprofit that misses payroll, because its research or social service grant was cut off without warning, may simply close its doors and never return to claim the grant. 

It’s worth remembering, however, one crucial difference between business start-ups and incoming public officials when it comes to moving fast and breaking things: the start-ups were never sworn to a public oath to put the law first. Thus, an ambitious delivery business may tell its drivers to double-park when they feel the need, reasoning that the income from conducting speedy business will more than pay for the occasional ticket. Officials are in a different station entirely. Their oath reminds them that respecting the Constitution, and with it the laws enacted thereunder, is very much their responsibility, even if they could more speedily accomplish some worthwhile objectives otherwise. 

If we’re not going to take the oath of office seriously, maybe it’s time to amend the Constitution so as to formally do away with it. 

Chris Edwards

With today’s massive budget deficits, the Trump administration and Congress should look for savings in every federal department. That includes the massive $859 billion Department of Defense (DOD) and the huge $365 billion Department of Veterans Affairs (VA).

The Congressional Budget Office released charts last week with an overview of spending in the two departments.

This chart breaks down DOD spending into six components in constant dollars. Note the huge operations and maintenance costs of the Iraq and Afghanistan wars, and notice how this part of spending remained elevated even after the wars ended.

And note the sharp decline in spending on procurement and personnel during the 1990s. We always hear that the “military-industrial complex” has unbeatable lobbying power. Members of Congress will go to the mat defending spending on bases and weapons systems in their districts.

Yet in the 1990s, Congress wanted to cut military spending, and it did, sharply. The lesson is that Congress can cut any spending any time it wants. Members do not have to be slaves of parochial interests and campaign contributions. Today, we need them to legislate in the national interest by cutting spending across the federal budget to avert a fiscal disaster.

This chart shows the other financial cost of national defense—the budget of the VA. Here are a few interesting points from the CBO study:

“After experiencing very little growth from 1980 to 1999, VA’s budget has more than quadrupled in real terms since 1999.”
“VA’s funding has grown since 2000, though the number of veterans has been declining.”
“Because of policy changes, veterans who served after 9/11 (known as Gulf War II veterans) are more likely to get benefits, whether they served in combat or not.”

The combined spending of the DOD and VA in 2025 will be $1.2 trillion. That will be 17 percent of total federal spending and 20 percent of spending aside from interest costs.

These two departments are ripe for the types of efficiency reforms that Elon Musk and his DOGE team are promoting. Regarding the DOD, for example, policymakers should close excess foreign military bases and thin the department’s bloated civilian bureaucracy of 795,000 employees.

Congress needs to step up to the reform plate and start assembling lists of programs to cut and eliminate. The CBO describes dozens of possible DOD and VA reforms here, and Cato discusses wide-ranging budget reforms in its recent DOGE report here.

# # #

Note: the components of the first chart add to $848 billion in 2025, which differs a bit from the $859 billion for DOD in CBO’s January Outlook.

Justin Logan

President Donald Trump rarely bores. At a much-awaited press conference with Israeli Prime Minister Benjamin Netanyahu on February 4, Trump was expected to discuss the prospects for the ceasefire between Israel and Hamas, and potentially US policy on Iran. Would he take up the idea that the war would likely resume soon? Would he commit to a hardline policy on Iran?

Trump changed the subject entirely, shocking hawks, doves, and anybody watching with his big idea:

“The US will take over the Gaza Strip and we will do a job with it too. We’ll own it and be responsible for dismantling all of the dangerous unexploded bombs and other weapons on the site, level the site and get rid of the destroyed buildings, level it out. Create an economic development that will supply unlimited numbers of jobs and housing for the people of the area … do a real job, do something different.”

Asked whether he saw a long-term US presence in Gaza, Trump doubled down:

“I do see a long-term ownership position, and I see it bringing great stability to that part of the Middle East and maybe the entire Middle East…. Everybody I have spoken to loves the idea of the United States owning that piece of land, developing and creating thousands of jobs with something that will be magnificent in a really magnificent area that nobody would know…. I don’t want to be cute. I don’t want to be a wise guy. But the Riviera of the Middle East, this could be something that could be so … magnificent.”

A few thoughts.

First, I really don’t want to put Trump on the couch, but there’s a weird psychological move happening. At bottom, the conflict between Israelis and Palestinians is a political problem, not a real estate development deal. Trump’s language here—“level it out,” “create jobs,” “ownership position”—evades the very real political problems underpinning the conflict, sounding instead as if he were wearing a hard hat overlooking a desolate rail yard outside Newark. Is Trump using the familiarity heuristic to render an intractable problem something familiar? I’ll leave that to the shrinks.

More practically, it is hard to see this plan going anywhere. It would require huge numbers of US troops on the ground to try to maintain order in the Gaza Strip. Does political support for this seem to be forthcoming? What would happen to the political support when US forces inevitably start facing attacks from Hamas?

Second, such a plan would involve somehow moving almost 2 million Palestinians away from their homes while construction firms haul away rubble and construct new buildings. Where would they go? The Arab states had made clear their opposition to this idea before last night—they fear Palestinians will never be allowed back and Israel will simply take the land over permanently. They reiterated their opposition after the press conference.

The Israeli right shares the Arab states’ analysis but views their bug as a feature: Israel’s expansionist Finance Minister Bezalel Smotrich hailed the plan as “The true response to October 7. Those responsible for the horrific massacre on our land will face the permanent loss of their own.” There’s zero support from anybody who could help to transfer Palestinians from Gaza.

But to zoom out a bit from the particulars: Why are we still involved in this process at all? What is the US interest in taking another ride on the hopeless carousel of destruction that is the Israeli-Palestinian peace process?

Israel has turned Gaza into rubble—Trump’s Middle East envoy Steve Witkoff visited the Strip and described it as “uninhabitable… What was inescapable is that there is almost nothing left of Gaza.” And yet the Jewish state is not close to achieving the war’s two strategic aims: the eradication of Hamas and the release of all the hostages held in Gaza. As Jon Hoffman and I note in a recent piece,

“Hamas remains the dominant political and military force inside the enclave, and former Secretary of State Antony Blinken recently said Hamas has recruited almost as many fighters as it has lost in 15 months of war. The Jerusalem Post has similarly reported that Palestinian Islamic Jihad and Hamas combined are back to more than 20,000 fighters. As for the hostages, eight were rescued by Israeli soldiers, but more than 100 were released by Hamas because of diplomacy.”

So why should this be a US project at all? As George Kennan lamented in 1977,

“We have allowed ourselves to be maneuvered into a position where each of the two parties believes it can use us for its own ends, where each has the impression that it is primarily through us that its desiderata can be achieved, with the result that we are always the first to be blamed, no matter whose ox is gored; and all this in a situation where we actually have very little influence with either party. Seldom, surely, can a great power have got itself into a more unsound and unnecessary position.”

But to try to brighten things up a bit, Trump’s vision of a developed, peaceful Gaza next to Israel changed the subject from resuming the war, let alone broadening it to Iran. He expressed sympathy for people on both sides, as well as a desire to get out of this mess and create a future better than the one Israelis and Palestinians face today. But leapfrogging the very real, and potentially intractable, political problems sets a dubious foundation for doing anything more than changing the subject. Which, in fairness, might have been the best anyone could hope for.

Alex Nowrasteh

The Trump administration issued an executive order (EO) on January 20, 2025, to create a process to designate some international criminal organizations and drug cartels as foreign terrorist organizations (FTOs). There have been no updates to the State Department’s list of FTOs yet. 

To be designated an FTO, the State Department’s Bureau of Counterterrorism (SDBCT) must demonstrate that the organization engages in terrorist activity according to the definition in 8 USC §1182(a)(3)(B), 22 U.S. Code § 2656f(d)(2), or that the organization retains the capability and intent to conduct terrorism. The definition in 22 U.S. Code § 2656f(d)(2) is clearer and closest to that commonly understood definition of “terrorism” by the public in scholars. The most relevant portion reads:

(2) the term “terrorism” means premeditated, politically motivated violence perpetrated against noncombatant targets by subnational groups or clandestine agents.

That definition contrasts with that in 8 USC §1182(a)(3)(B), which is poorly written (like the rest of the Immigration and Nationality Act) and broad enough to include much criminal activity unrelated to terrorism. The SDBCT analyzes whether the group has committed attacks, planned attacks, prepared for attacks, funded possible future attacks, or retains the capability and intent of doing so.

Under the definition in 22 U.S. Code § 2656f(d)(2), drug cartels are not FTOs because they are criminal firms seeking monetary profits. They are not organizations motivated to conduct politically motivated violence unless it increases their profits. However, the definition under 8 USC §1182(a)(3)(B) is so broad that it could include criminal organizations motivated by profits rather than politics. That’s why the recent Trump administration order doesn’t mention 22 U.S. Code § 2656f(d)(2) and instead relies on a process laid out here that allows designating organizations as FTOs under the simpler, broader definition in 8 USC §1182(a)(3)(B).

The courts can review the designation of entities as FTOs, and those entities themselves can challenge the designation. Still, drug cartels aren’t likely to do so because they are criminal organizations. However, individuals prosecuted because of the new designation may be able to challenge it. The more likely outcome is that the courts will defer to the president’s judgment on national security.

Cartels and terrorists have different motivations, and this Trump order exploits ambiguities in current law to blur the two. When it designates cartels as FTOs, the government will rely heavily on the capabilities of the cartels and cite their extreme violence and interference with politics in other countries. Most cartel violence targets rival criminal organizations, civilians, and officials who are either corrupt or seek to prosecute them. But they rarely target symbolic targets unless doing so is in service of a wider profit-seeking goal. 

Cartels have territory, but it’s better understood as local monopolies enforced through violence rather than for territorial control and governance as ISIS attempted. Terrorists, on the other hand, seek political power or to affect politics. The difference between cartels attacking governments to increase their profits and terrorists attacking governments to take them over is the definitional difference between terrorists and normal criminals. The US government could today defund the cartels by legalizing drugs and immigration, not terrorists.

There are many reasons why the Trump administration is seeking to designate cartels as FTOs.

The first is to massively increase the US government’s powers to crack down on the border despite the low and falling apprehensions that preceded Trump’s election. Declaring that already existing criminals on the border are terrorists gives the president enormous military and other legal powers to combat them. The application of laws against providing material support to terrorists would increase scrutiny of the over $154 billion in annual immigrant remittances to locations with cartel activity, which could now include Latin America and the Caribbean. The result would be the confiscation of funds, higher transaction costs that would hit migrants and their families like a tax, and fewer financial institutions willing to facilitate such remittances. Economic sanctions to deter terrorism will provide President Trump with sweeping new powers to punish countries to the south that do not cooperate more with immigration enforcement.

Second, declaring cartels to be terrorists supports his other EOs that declared the border is (absurdly) being invaded and that there’s a national border emergency. An invasion requires an invader, and FTOs are closer to a military threat that may justify a declaration of invasion, whereas normal illegal immigration doesn’t provide a justification. After all, the presence of FTOs has justified other US military invasions overseas. 

Since there’s no actual invasion or national border emergency, re-designating criminal organizations that smuggle drugs and people across the border as FTOs is a rebranding that reinforces the other border enforcement actions. Trump recently delayed threatened tariffs on Canada in exchange for the Canadian government declaring drug cartels as FTOs, which bolsters his argument that drug cartels are terrorists by claiming that another country agrees, even though Canada only did so because of US government threats. 

Third, declaring drug cartels as FTOs could change the political and media narrative about terrorism on the border. Terrorist attacks on the border have not occurred despite decades of fear, and politicians can only cry wolf for so long before looking ridiculous. Zero Americans were murdered in attacks on US soil committed by foreign-born terrorists during the Biden administration, the first time a president hasn’t faced a deadly attack since before the Ford administration. 

Designating drug cartels as FTOs transforms cartel violence into terrorism, seemingly justifying these fears in a superficial and misleading way. Just a day after Trump issued his order, suspected cartel gunmen shot and robbed hikers near the border. Customs and Border Protection described the crime as “A U.S. citizen and a Canadian citizen were robbed and attacked by armed suspected Mexican cartel terrorists.” If every crime committed by cartel members is now a terrorist attack, then the amount of terrorism along the border will surge via a re-accounting of crimes that would have occurred anyway, and many will be fooled by this definitional mirage, at least for a while.

The fourth reason is to more harshly punish immigrants for illegally crossing the border. Unlawful entry is a low-level criminal offense, and only in some cases, but most illegal immigrants hire smugglers. Many of those smugglers are drug cartels or at least affiliated with them in some way. Giving money to an FTO is a serious crime that can result in a long prison sentence. Thus, redesignating cartels as FTOs transforms many illegal immigrants into material supporters of foreign terrorist organizations—defying common sense and the intent of lawmakers.

Fifth, designating cartels as FTOs will result in more asylum seekers losing their asylum claims. Trump has shut down asylum along the border by canceling the CBP One App and continuing the existing ban on seeking asylum at ports of entry. However, declaring a border emergency, invasion, and that drug cartels are FTOs gives more reasons for asylum seekers to claim that they are fleeing a well-founded fear of persecution. That’s a conundrum for the Trump administration that could result in more future asylum seekers or those awaiting a hearing to eventually win their claims. However, providing material support for terrorist organizations bars immigrants and asylum seekers. Since some asylum seekers paid smugglers who could be tied to cartels, the courts can safely deny them asylum because they materially supported an FTO.

Sixth, the administration can use anti-terrorism laws and the military to freeze assets more easily, prosecute smugglers, and attack with military force (the last may require an AUMF). The penalty for terrorism offenses is greater than for drug offenses. Anti-terrorism laws also allow for greater secondary sanctions on foreign banks and firms doing business with FTOs, while cartels and human smugglers only have their funds targeted by law enforcement. The increased regulatory compliance burden and potentially higher costs of doing business with cartels-cum-terrorists will reduce their access to the financial system. Furthermore, the bar for counterterrorism operations is lower than for law enforcement actions against cartels. 

Drug cartels are vicious criminals, but they are profit-seeking criminals who are not motivated by political, economic, religious, or other social goals. There is a tendency to label the worst criminals as terrorists because crimes terrorize. That feeling is understandable, but that’s not the purpose of the anti-terrorism laws, and that’s certainly not consistent with any reasonable understanding of the real terrorist threat (small as it is). 

Congress has given the president too much authority to prosecute a war on terrorism without sufficient oversight by the legislature and the courts. By designating drug cartels as FTOs, the Trump administration unlocks new powers for itself, creates a new media narrative that could fool many, and reinforces the rest of its anti-immigration and border enforcement agenda. 

Jeffrey Miron

Since at least Aristotle, new technologies have raised fears of major job losses as businesses substitute machines for people. The industrial revolution led to the Luddite movement in early 19th-century England; 100 years later, Ford’s assembly line sparked fears that skilled artisans would no longer be needed.

A priori, the net impact of “labor-saving” technologies is ambiguous. In some cases, firms will substitute machines or software for people, but this substitution might have a minimal impact on employment. By making products cheaper, the use of cheap machinery might lower costs and therefore price, implying increased production and employment. 

After the introduction of the assembly line in Ford’s car factories, the cost of a Model T fell from $850 to $260. As the number of car owners grew, dealerships, service stations, garages, and repair shops expanded employment. In other cases, improved technology creates a demand for tasks that only people can perform well. For example, the advent of computers created the need for software engineers, UX designers, and cybersecurity specialists.

Overall, history pushes back against worst-case fears; technology has expanded dramatically (the wheel, locomotives, cars, computers, AI), yet unemployment rates show no secular increase.

Recent research provides a useful illustration that technology fears are probably misplaced:

“Worker shortages are especially salient in elderly care. In many countries, nursing homes experience persistent staff shortages and high levels of turnover while the elderly population and demand for caregiving grow. Robots have become increasingly common in service organizations, but they often prompt concerns about job replacement. Our research studies the effects of robots on labor and service quality in Japanese nursing homes and finds that robot adoption was positively associated with the number of caregivers and nurses. Additionally, certain robots were positively associated with the number of residents receiving care, nursing home revenue, and the length of a facility’s waitlist. Furthermore, the use of restraints and cases of bed sores decreased with robot adoption.”

Thus, while technological advancements may shift the nature of work, historical and empirical evidence suggests they often complement human labor rather than cause widespread unemployment.

This article appeared on Substack on February 4, 2025. Jonah Karafiol, a student at Harvard College, co-wrote this post.

Chris Edwards

The federal government is hurtling toward a fiscal crisis. Accumulated debt is reaching historic highs, borrowing rates are rising, and spiking inflation is a major risk. President Trump and Congress must focus on control of spending and deficits.

Given their plans for tax cuts and defense spending increases, Republicans need large budget cuts to prevent the government’s $2 trillion deficits from spiraling higher. Many Republicans want to cut spending, and Trump’s DOGE team talks about annual cuts of $1 trillion or more. But as the Wall Street Journal discusses today, other Republicans are nervous nellies, who need more convincing that cuts are both good economics and good politics.

Republican tensions are currently coming to a head with party leaders assembling a budget resolution that sets overall fiscal parameters for the year. What is a realistic spending cut goal for Republican reformers?

History can be a guide. Let’s look at three charts showing real-world experience in our federal government and Canada’s. Trump is hostile toward Canada these days, but there are fiscal lessons to be learned from our northern neighbor. Canada faced a debt crisis in the 1990s similar to ours today, and its politicians had the backbone to pursue large and sustained spending cuts.

Figure 1 shows federal government spending in the two countries as a percentage of gross domestic product (GDP).

The government grew in both countries from the 1960s to the early 1980s. The US had a string of big-spending presidents, and Canada had Pierre Trudeau, left-wing father of Justin Trudeau.
Both countries began restraining spending in the 1980s and cut in the 1990s. The US cut spending by about 4 percentage points of GDP, while Canada cut about 7 points.
The US abandoned restraint under George W. Bush after 2001, and spending has trended higher since, including spikes during the Great Recession and COVID-19.
Canada slashed spending and then held it down for another 15 years. Canada’s spending spike during the Great Recession was smaller than America’s.
Justin Trudeau’s government (2015–2025) reversed course and has grown spending, including a larger COVID-19 spending spike than under Trump-Biden.

Figure 2 shows spending without interest costs.

This figure shows spending that the politicians directly control—spending just on programs and excluding interest costs.
The US cut noninterest spending by about 3 points of GDP in the 1990s, while Canada cut about 5 points. The Canadian cuts were sustained much longer than the US cuts.
In the 1990s, Canada slashed spending on defense, welfare, business subsidies, aid to lower governments, and many other things. It privatized infrastructure and ended the related subsidies. I discuss Canada’s reforms here.
For the US, 85 percent of the cut amount between 1990 and 2000 was defense. Defense procurement fell from $81 billion in 1990 to just $52 billion in 2000 in nominal dollars.
The US balanced its budget 4 years in a row (1998–2001), while Canada balanced its budget 11 years in a row (1997–2007).

Figure 3 shows spending without interest or defense spending.

The US cut noninterest, nondefense spending in the late 1980s but not much in the 1990s.
Canada cut noninterest, nondefense spending far more than the US, reducing it by about 4 percentage points of GDP from the early 1990s to the end of the decade.

Both countries have shown that substantial spending cuts are possible. In Canada, the parliament cut a range of programs by about 5 percentage points of GDP. In the US, Congress cut almost 3 points of GDP, mainly defense spending, and that was despite the lobbying power of the “military industrial complex.”

Congress can cut any program any time it wants, and 3 percentage points of GDP or more in a decade is entirely doable. It turns out that 3 percentage points is about what we need to stabilize federal debt and hold off a spiraling budget crisis.

The Congressional Budget Office baseline shows budget deficits of about 6 percentage points of GDP over the next decade. Treasury Secretary Scott Bessent says that we should aim for a 3 percentage point deficit, and the Committee for a Responsible Budget calculates that reforms of about that size would stabilize federal debt held by the public at about 100 percent of GDP.

A 3 percent of GDP cut would amount to a 13 percent, or $1.3 trillion, spending cut in 2035. Such a cut can be achieved with reforms Cato has proposed here and here. These cuts are from the CBO baseline, and so with tax cuts and defense spending increases, we will need more cuts. Canada shows that sustained cuts of more than 3 points of GDP are possible.

One final point. Stronger GDP growth will shrink the relative burden of debt, so microeconomic reforms such as deregulation should accompany budget reforms. Alternatively, trade wars would shrink GDP and exacerbate the debt crisis. In the 1990s, Canada’s economy grew strongly after adopting large spending cuts, privatization, and free trade.

# # #

Canada’s fiscal data is here and the story of its reforms are here, here, here, and here.

US fiscal data is here.

A fiscal agenda for Congress is here.

A discussion of how spending cuts boost growth is here.

A discussion of how to avert a fiscal crisis is here.

Note that the charts are for federal spending and do not include state/​local or provincial/​local spending.

Brandan P. Buck

President Trump was elected in 2024 partly on his promise of ending “America’s endless wars.” The Trump administration says it doesn’t want new wars, boldly declaring that “[w]e will measure our success not only by the battles we win but also by the wars that we end—and perhaps most importantly, the wars we never get into.” While it is too early to judge his sincerity or ability to do so, in the early weeks of his second term it appears like the Trump administration is poised to breathe new life into America’s original “endless war,” that of the war on drugs.

The Trump administration has repeatedly floated the idea of using unilateral military force in Mexico, specifically launching Special Operations Forces (SOF) raids and airstrikes at the drug cartels. Framing the emerging sentiment was Tom Homan, who, during an appearance on Fox News, asserted that President Trump was prepared to “use [the] full might of the United States special operations to take ‘em out.” 

Elon Musk, who opined on X, added to the overheated rhetoric: “I doubt the cartels can be defeated without US Special Operations.” American venture capitalist and Palantir Technologies founder Joe Lonsdale confidently declared on X that “the same US tech & expertise that eliminated thousands of terrorists could overcome cartels and their allies in the Mexican government, root out the fraud and corruption, and eliminate the criminals.”

Adding to the drumbeat, Heritage Foundation scholars Robert Greenway, Andrés Martínez-Fernández, and Wilson Beaver have authored a paper titled “How the President Can Use the U.S. Military to Secure the Border With Mexico.” Among the issues covered in the report was the possibility of using US military force in Mexico, which its authors argue “may be necessary to prompt cooperation from a resistant Mexican government or otherwise contain the cartel threat.” 

The authors are more sensitive to the risks of unilateral military force and do not display the unmoored confidence of some proponents, recommending US-Mexico joint action instead. Still, they nevertheless underestimate the tradeoffs that would emerge should the US commit troops to a cross-border war with the Mexican drug cartels.

To their credit, the authors list numerous perils of unilateral military intervention in Mexico so thoroughly that one wonders how they can sustain their argument. First among the many problems is the lack of actionable intelligence on cartel members and their supply networks. Relatedly, they note that the nature of fentanyl production, unlike other drugs, presents a small visual signature, thereby impeding intelligence gathering. They also concede that actionable intelligence in Mexico is further hampered by corruption in the Mexican government. They note such labs are often located in crowded urban centers, so strikes against them would result in high civilian casualties. They cite that the cartels are estimated to hold approximately 160,000–185,000 well-armed members. They note that the cartel networks have so far withstood the “disruptive effect” of network degradation through the killing of high-profile members and that the cartels have “proven in the past their ability to restructure after fragmentation.” 

Perhaps most chillingly, they note the prospects of cartel reprisals against American citizens and businesses in Mexico and the United States. Finally, they admit that prolonged border militarization, much less a cross-border incursion, would risk other American strategic interests. Ultimately, the authors make a strong case against using unilateral military force in Mexico.

The authors argue that the United States government needs to plan and prepare for such an intervention anyway. This case, despite all the downsides, rests on a belief that the use of unilateral military action against the cartels could “be enough to galvanize the Mexican government into cooperation with the U.S.” Furthermore, they assert that even on its own, a unilateral American military force could inhibit supply chains, impede cartel networks, and create “deleterious, if temporary, effects on cartel trafficking activity and networks.”

The Heritage report is an exercise in contradiction and wish casting. As noted earlier, the authors rightly observe that fentanyl labs are smaller, more difficult to detect, and easier to replace than conventional drug production. Nevertheless, its authors conclude that “with sufficient intelligence and coordinated measures, the potential exists for well-targeted actions to disrupt more vulnerable links in fentanyl supply chains.” How will American technical and human intelligence assets overcome these hurdles? Furthermore, if even tactically successful, how will the degradation of the cartel networks occur fast enough to constitute a strategic success? We do not know because the authors do not tell us.

For the scrupulous reader, one is left to ponder how the US government will succeed where the Mexican government has failed. The authors assure us that the answer lies in heeding the “lessons learned from the experiences of the past failures by the Mexican government to defeat the cartels.” Those lessons, apparently, call for more of the same, like “targeting vulnerable links” in cartel supply chains and “kinetic action” [i.e., killing] of cartel leadership. If insanity is doing the same thing twice and expecting a different result, their proposal certainly qualifies. We should demand a higher bar for supporting new wars than waving off serious objections and engaging in sublime wishful thinking.

The authors’ geopolitical arguments are similarly lacking. Their claim that American military power might force the Mexican government into cooperation neglects the enduring pull of Mexican nationalism and latent anti-American sentiment. On this point, the report’s one historical example, the Pancho Villa Expedition, which they cite positively, is terrible. That expedition failed in its objective of capturing its intended target and significantly damaged bilateral relations for a generation. Rather than serving as a model, the expedition should serve as a warning.

Furthermore, the authors fail to consider the potential diplomatic fallout of damaging relations like undermining cooperation on illegal migration, countering Chinese influence in the Western Hemisphere, and, if recent interventions into the Middle East are any guide, causing an uptick in war-related displacement. Their consolation prize, that unilateral force against the cartels would yield results that would offset damaged relations, similarly rests on a faulty assumption of success. 

Rather, in a reality plagued by the numerous quagmires of the Global War on Terror, a long-term military campaign in Mexico would be as politically and logistically challenging, especially if Mexico resists or if cartel violence escalates in response.

While the Heritage Foundation report thankfully lacks the hubris and bravado of some administration officials and surrogates who have, in recent months, championed this cause, it nevertheless fails to account for the diplomatic fallout and possibility for strategic failure that would likely accompany US military action in Mexico. Despite the authors’ awareness of the pitfalls of such an endeavor, they nevertheless fail to account for how the US government would succeed where the Mexican government has failed. Furthermore, its authors pile numerous analytical leaps atop one another, positing that military success, itself an assumption, would outweigh whatever diplomatic fallout may occur. 

A more sober-minded analysis holds that neither is guaranteed. While the US is undoubtedly suffering through the misery of the fentanyl epidemic and Mexico is enduring the related horror of cartel violence, solutions to both must be based on sound reasoning. Possible solutions should not incur tradeoffs as damaging as the issues they are meant to solve. Heritage’s analysis of the problem, like the braggadocio emanating from some corners of the Trump administration, does neither.

Adam N. Michel

The final installment in our tax bootcamp series covers the international tax system. It starts by describing what multinational tax systems attempt to do, covers the three theoretical types of international tax systems, discusses how real-world rules fit the theory, and reviews the magnitude of profit shifting—the boogieman that drives aggressive international rulemaking. The last section briefly overviews the Organisation for Economic Co-operation and Development’s (OECD) two-pillar framework. 

This is the last in a four-part series covering everything you need to know about the US tax code ahead of the 2025 congressional tax debate. Or at least everything that could fit into six hours of lecture and discussion. Part one is on Tax Code 101, part two is on the Tax Cuts and Jobs Act, and part three is on the history of radical tax reforms. This fourth installment draws heavily from my more comprehensive report on the same topic.

What is an International Tax System For?

Corporate income taxes must collect revenue from sprawling multinational business networks that span dozens of countries. The core challenge is to determine which multinational profits are subject to tax, in which country, and at what tax rate.

In apportioning multinational income, international tax systems usually try to achieve four competing goals: 1) raise revenue, 2) attract global investment and jobs, 3) increase the competitiveness of home-grown firms doing business abroad, and 4) reduce complexity.

Policymakers must balance these four goals, which are often in tension. For example, the corporate tax can collect more revenue from domestic firms by applying increasingly costly rules to include more foreign income in the domestic tax base. However, more revenue may come at the expense of higher compliance costs and difficulties for domestic businesses competing abroad.

Paper profit shifting, whereby firms manipulate the rules to reduce taxes, often takes two forms: aggressive pricing agreements between subsidiaries and financial techniques, such as strategically allocating debt. However, determining the legitimate transactions from the rest can be difficult.

About two-thirds of profit shifting occurs through the transfer pricing system—a complex set of rules governing the transfers of assets and services between related parties in different taxing jurisdictions. For example, patents and their connected royalties can be moved to lower-tax countries. For a time, Starbucks famously held its roasting recipe in the Netherlands, accruing profits at a preferential tax rate. The second primary method of shifting profits involves borrowing in high-tax jurisdictions, where interest payments are often deductible, and lending from low-tax jurisdictions, where interest income is more lightly taxed.

The difficulty of policing these transactions is one reason many economists favor eliminating the corporate income tax. It is also among the most economically distortionary forms of raising revenue. The easiest way to cut the Gordian knot of international tax is to simply repeal the corporate income tax and rely on more stable, less destructive sources of revenue.

Three Theories of the International Tax Base

In principle, cross-border tax systems can levy taxes based on the taxpayer’s residence, the source of the profits, or the destination of the end consumer. Most corporate income taxes begin with the principle that profits should be taxed where the income is generated but then piece together a hodgepodge of other rules that incorporate other principles. The following subsections describe each tax base and its use in the US context.

Residence-based taxes

Residence-based or “worldwide” tax systems tax the income of resident entities irrespective of where the income is earned. This is often also called “capital export neutrality.” Under a worldwide system, a US-headquartered firm would pay the same corporate tax on income earned in the United States and income earned overseas.

If implemented fully without deferrals or other carveouts, a worldwide tax system can limit domestic firms’ profit-shifting incentives by applying the same tax rate no matter where profits are located. However, if domestic tax rates are higher than in other countries, such systems will disadvantage home-grown and domestically headquartered firms investing or otherwise competing abroad. 

The United States had a worldwide tax system before 2018, leading to more than 60 “inversions” of domestically headquartered firms moving their headquarters to lower tax countries without worldwide systems. These US-based firms competing abroad had to pay the US’s 39 percent federal and state combined corporate tax rate. In comparison, their competitors faced average tax rates 14 percentage points lower in the typical OECD country, with average combined tax rates of about 25 percent that same year. This system led to about $2.6 trillion in profits held overseas by US-based firms that could defer their domestic tax bill on some foreign profits.

Source-based taxes

Source-based or “territorial” tax systems only tax profits earned within the borders of the taxing jurisdiction. Often called “capital import neutrality,” territorial systems ensure that two investments in the same location face the same tax rate, regardless of where the investor is located.

Territorial systems are implemented through a “participation exemption” for all or a portion of foreign-earned capital gains and dividend income. Most OECD countries have complete territorial systems that exempt 100 percent of both types of income. In 2018, the United States switched to a territorial system with an exemption for foreign dividends received (but not for capital gains). Only Chile, Colombia, and Mexico still have worldwide tax systems as of 2024.

While territorial systems can be simpler because they disregard activity outside the territory, they can also create arbitrage opportunities. Thus, most territorial tax systems employ anti-abuse rules to protect the domestic tax base, such as controlled foreign corporation (CFC) rules and limits on intraparty payments.

The 2017 Tax Cuts and Jobs Act paired a federal corporate income tax rate reduction from 35 percent to 21 percent and the partial participation exemption with four new anti-abuse rules. The first two apply a carrot-and-stick approach to high-return intangible income, creating a minimum tax on foreign income and an offsetting subsidy for locating the income domestically. The second two place limits on intraparty payments. 

Global Intangible Low-Taxed Income (GILTI): a minimum tax between 10.5 percent and 13.125 percent (increasing in 2026 to 13.125 percent and 16.406 percent) on income that exceeds a 10 percent return. The GILTI rate is a range due to an 80 percent limit on foreign tax credits (FTC). GILTI is a new CFC income category layered on top of the existing “Subpart F” and expense allocation rules.
Foreign-Derived Intangible Income (FDII): a deduction for foreign export income connected to intangible assets held in the United States, creating a lower effective tax rate of 13.125 percent on eligible income (16.406 percent in 2026).
Base Erosion and Anti-Abuse Tax (BEAT): a minimum tax of 10 percent (12.5 percent in 2026) on income of affiliated foreign entities with gross receipts of $500 million or more. The tax applies if qualifying “base-erosion payments”—interest, rents, royalties, services, depreciation, and amortization—exceed 3 percent of total corporate deductions.
Interest Limitation: limits interest expense deductions to 30 percent of earnings before interest and taxes (EBIT). From 2018–2021, the limit was 30 percent of earnings before interest, taxes, depreciation, and amortization (EBITDA).

Destination-based taxes

Destination-based tax systems tax profits based on where the customers or end users are located instead of the location of production or headquarters. The destination principle is most commonly used in value-added taxes (VATs) on consumption, but also underlies digital services taxes (DSTs) and some features of anti-abuse rules, such as the BEAT in the United States and diverted profits taxes in the UK and Australia.

Instead of the US territorial system implemented in the Tax Cuts and Jobs Act, there was an original proposal for a novel destination-based corporate tax system. This would have been implemented through a “border adjustment” that would allow exporting firms a deduction for the cost of exports and would tax the value of imports at the corporate tax rate. This proposal was abandoned due to political concerns and uncertainty over some of the strong assumptions necessary to keep the tax from acting like an across-the-board tariff.

For more on the economics of destination-based income taxes, see “Reviewing the Case Against a Border-Adjusted Corporate Income Tax.”

Is Profit Shifting a Problem?

Implicit in the layers of complex and costly anti-abuse rules is that profit shifting—the problem the rules are intended to fix—is a big and growing problem, undermining the ability to raise sufficient revenue through the corporate income tax. However, the available data show that profit shifting to low-tax countries or “tax havens” is small and declining.

Measured correctly, corporate profits in tax havens amount to about 8 percent of total US corporate profits (or as much as 11 percent). Figure 1 from a longer Cato report shows two different measures of tax haven profits of US multinationals. The chart shows that the share of total US corporate income reported in tax havens grew modestly over time, and following the 2017 corporate tax cut, it fell to its lowest level in a decade. The data suggest a lower corporate tax rate is one of the most effective reforms to reduce profit shifting. The share of US corporate profits in tax havens declined 30 percent after the 2017 corporate tax cut. 

A significant share of corporate profits in low-tax countries (shown in Figure 1) are not shifted there artificially but are associated with real investments. However, even the artificially shifted profits are still associated with real economic benefits, like jobs and investment, in both tax havens and non-haven jurisdictions.

The OECD Two-Pillar Tax Cartel

In October 2020, the OECD outlined a “Two-Pillar” approach to remake the international tax system in close consultation with the Biden administration. The proposal is the latest in the OECD’s efforts to stop profit shifting. It is intended to raise multinational businesses’ effective tax rates and reallocate taxing rights away from countries like the United States to others.

President Trump distanced the US from the OECD agreement in one of his first Executive Orders. Pressure from the Trump administration will likely stall the deal’s progress. The most effective way to end the OECD’s global tax deal is for Congress to withdraw from the organization and cut its US funding.

Pillar One would reallocate an estimated $205 billion of large multinational corporate profits to countries where customers are located and away from where the firms have a physical and productive presence.[1] This would be done using a complicated formula based on a company’s sales, marketing, and distribution in each jurisdiction. The new tax is intended to replace a patchwork of DSTs that some countries currently charge large technology firms based on revenue and users in their country.

Like DSTs, Pillar One intentionally targets America’s most profitable firms. By one estimate, US companies account for 58 percent of profits redistributed under the new tax system. By redistributing the US tax base, the Joint Committee on Taxation (JCT) estimates Pillar One could reduce US revenue by between $1.4 billion and $100 billion annually. The pact would require a multilateral treaty and significant changes to domestic tax laws.

Pillar Two is a 15 percent global minimum tax on large multinationals. Like Pillar One, the Pillar Two rules primarily target American firms, which are estimated to earn nearly 40 percent of all in-scope multinational income, about the same amount as the next 10 countries’ shares combined. The JCT estimates Pillar Two could reduce US domestic tax revenues over 10 years by between $57 billion and $122 billion. Other estimates predict the tax will reduce US domestic investment by $22 billion annually.

Pillar Two includes a novel extraterritorial feature called the Undertaxed Profits Rule (UTPR), which allows countries to increase taxes on a business’s domestic subsidiary if a related entity in another jurisdiction pays a tax rate below 15 percent. Columbia University legal scholar David Schizer likens the UTPR to California being able to tax a resident of Virginia on income earned in Virginia simply because his daughter lives in California.

UTPR’s tax base shifts the global competition for international business investments from reducing economically beneficial tax rates to economically costly government subsidies. It does this by treating direct state subsidies more favorably than non-refundable tax credits and other tax benefits. The new OECD-built tax system favors China’s preferred model of economic competition and incentivizes a global shift to its less efficient and more corruption-prone system of governmental favoritism.

The system also includes two new domestic taxes—the Qualified Domestic Minimum Top-Up Tax (QDMTT) and the Income Inclusion Rule (IIR)—to enforce the 15 percent tax on domestic income and firms. Countries have already begun to implement these domestic components of Pillar Two but have not yet begun enforcing the UTPR. 

Nicholas Anthony

Debanking is generally characterized as the sudden closure of a financial account. The causes of debanking, however, are just as important as the end result. And it’s for that reason that it may be time to refine how debanking is discussed.

When looking at different cases of account closures, there tend to be two types of debanking taking place. On one hand, there are the decisions of private businesses about how they wish to conduct their day-to-day operations. On the other hand, there are the decisions of government officials about how those private businesses should operate. For that reason, the policy conversation would likely benefit by distinguishing between what might be called operational debanking and governmental debanking (Table 1).

These distinctions are not perfect, but they are important for policymakers to recognize if the problems around debanking are to be fixed.

Operational Debanking

Operational debanking is what occurs when a financial institution chooses to close the account of a customer because it is no longer in the institution’s individual interest. It could be because the customer violated part of their contract with the institution or because the institution decided to move in another direction.

Let’s consider a few examples.

Bank of America made headlines in 2018 when it announced that it was stepping away from certain gun manufacturers. According to reporting at the time, this decision was made after discussions with employees and customers who had been affected by high-profile shootings. Notably, however, gun manufacturers were not cut off from the entire financial system. Bank of America only closed some accounts while it maintained accounts with others (e.g., Remington and Vista Outdoor Inc). Furthermore, although Sturm Ruger & Company had lost its account with Bank of America, it later moved to Wells Fargo.

A similar case occurred in 2019 when JPMorgan announced it would no longer finance private prisons and detention centers. Again, this decision appears to have been primarily a response to customers and other members of the public who protested at shareholder meetings and JPMorgan CEO Jamie Dimon’s apartment. And again, the private prisons were not cut off all at once. Instead, JPMorgan reduced its credit exposure over time. 

With that said, most cases of operational debanking are unlikely to make headlines. A less controversial (though likely more common) example occurs when a financial institution closes an account after repeated overdrafts or late payments. Put simply, the account holder didn’t pay for the service, so the institution stopped offering the service.

In short, operational debanking centers on business decisions made within each financial institution.

Governmental Debanking

Governmental debanking is what occurs when a financial institution is pressured by the government to close the account of a customer. Governmental debanking can occur in two forms. The first form occurs when the government explicitly instructs a financial institution to close an account. This instruction can be as casual as a letter or as formal as a court order. The second form of governmental debanking, however, is more abstract. It involves the use of laws and regulations to make it increasingly harder to serve customers.

Again, let’s consider a few examples.

The case of National Rifle Association (NRA) v. Vullo made headlines in 2018 after then-superintendent of the New York State Department of Financial Services, Maria T. Vullo, issued regulatory guidance instructing financial institutions to review relationships with the NRA and “take prompt actions” to manage the risks. Vullo then joined New York Governor Andrew Cuomo to say that the government “urges all insurance companies and banks … to join the companies that have already discontinued their arrangements with the NRA.”

In another example, the Federal Deposit Insurance Corporation sent private letters to instruct financial institutions to stop conducting cryptocurrency-related activity. Although some people may take comfort in that the agency only told these institutions to “pause all crypto asset-related activity,” financial institutions know all too well that government suggestions are rarely optional. Furthermore, the agency failed to provide a timeline or follow-up. So, in practice, these letters were effectively termination orders. 

Further back, another example occurred where companies sending money between the United States and Somalia quickly found themselves debanked in 2015 after “a broad U.S. crackdown on money laundering.” At the time, the Office of the Comptroller of the Currency ordered Merchant Bank to shut down these companies’ accounts unless it could “maintain sufficient transparency to reasonably ensure the legitimacy of the sources and uses of customer funds.” 

Unfortunately for the companies, the cost-benefit analysis was not in their favor. The government made it so costly to serve these customers that the bank had to shut down their accounts.

Finally, in one more example, financial institutions will often close an account after a customer incurs too many suspicious activity reports (SARs). The concept may raise eyebrows, but one of the most common reasons for filing a SAR is that a customer made a transaction close to $10,000. As the Bank Policy Institute recently explained, “When a bank files a structuring SAR on a customer, that customer generally becomes designated as “high risk” under banking agency guidance, and agency examiners begin asking the bank why the account remains at the bank.” JPMorgan CEO Jamie Dimon also recently made this point saying financial institutions can face hundreds of millions of dollars in fines if it later turns out someone was breaking the law and the accounts were not shut down. In other words, the government has created a system in which all the incentives push banks toward closing the account.

Taken together, these examples show that governmental debanking occurs when the government orders or otherwise forces financial institutions to close accounts.

Conclusion

Distinguishing between these two forms of debanking matters greatly. It is the difference between losing access to one bank versus losing access to every bank.

With that said, these distinctions are not perfect. The examples with Bank of America and National Rifle Association v. Vullo show that operational and governmental debanking can overlap. If nothing else, the confidentiality associated with the Bank Secrecy Act and supervisory reports makes it nearly impossible to confirm there is zero government involvement in operational debanking. Likewise, it is also nearly impossible to confirm there were no bank employees relieved that the government forced them to part with a customer they were not particularly fond of in cases of governmental debanking.

Despite this complexity, the distinction still matters. Stopping governmental debanking requires rolling back sweeping regulations and excessive government discretion. Stopping operational debanking, however, would mean imposing new restrictions on private businesses—creating further market distortions.

Congress should resist the temptation to impose new market distortions in response to operational debanking. Instead, Congress should focus on reining in the laws and regulations that have fueled governmental debanking.

Are you interested in learning more about debanking? Check out my recent working paper that explains the approach taken in the Fair Access to Banking Act. You can find that paper here.