The GOP presidential candidate said he would ban immigrants from sending money home to Mexico.

Donald Trump’s proposal to force Mexico to pay for a Wall guarding against the flux of immigrants into the U.S. made news this week, and rightfully so. Trump’s idea would be to curtail the ability of banks, credit unions, and wire transmission companies to send money abroad — a sharp departure from policy and law whose bipartisan aim has been to bring remittances to all countries into the financial mainstream and out from the shadowy illegal word of people moving cash in suitcases.

Encouraging remittances to go through the financial system benefits everyone: it enhances the ability to combat terrorist finance and money-laundering, it reduces crime in both the U.S. and abroad, it increases economic growth in the U.S. and overseas, and it provides for greater competition and market incentives to allow people to keep more of their hard-earned money to use as they see fit. Moving in the opposite direction would be a major mistake.

This is a big issue that affects a lot more people than one might think – more than just sending money to Mexico. In America today, more than 40 million people were born in other countries, a record number. This translates into just more than 1 in 8 Americans, a sharp increase from 1970 when fewer than 1 in 25 Americans were foreign born. Thus, it is not surprising that many people perceive America to have more foreign-born people than any time in their lifetime. However, that is not the case for the lifetime of America. Between the Civil War and the 1920s, America had as high — or higher — share of foreign born as we do today.

Remittances are not a new phenomenon. Most American families likely sent remittances at some point whenever their family first immigrated. My great-grand father sent money back to what is today the Czech Republic so that his wife and their children (including my grandmother) could come and join him and escape what became the Second World War. Today, remittance flows go toward the new generation of American immigrants and the children of those immigrants. More than $120 billion was sent abroad in 2012 according to the Pew Center and while it is true that Mexico received the largest amount at just under $23 billion, the rest of the top 5 countries may surprise you: China ($13 billion), India ($12 billion), Philippines ($10.5 billion), and Nigeria ($6 billion). And old habits remain as Germany ($2.5 billion) and France ($2 billion) are still among the top 15 countries that receive remittances from the United States.

This money comes in lots of small chunks, which can make sending it expensive. The typical new migrant worker sends money home around 14 times a year, which corresponds to once a month plus Mother’s Day and Christmas. These are usually small sums (less than $300) and represent an extraordinary level of savings given the worker’s income. The money goes through both the formal banking system including banks, credit unions, and wire transmitters who eventually use banks like Western Union and MoneyGram. Some goes through informal means, including “viajeros” who are people that literally carry cash in suitcases on planes that are often breaking the law and outside of the standard anti-money laundering and terrorist finance enforcement system. Why would anyone want to encourage that?

The idea of using this flow of funds to try to implement other policy objectives, such as border control, would be a sharp departure from current practice. The Patriot Act and subsequent federal law governing remittances in financial laws like the Dodd-Frank Act were never intended to be used to threaten to cut off the flow of migrant worker remittances. These laws were intended to track and crack down on the flow of money laundering or support for illegal and terrorist organizations while at the same time providing consumer protections to workers who are sending hard-earned cash back home to their parents, grandparents, and children. In fact, the bipartisan goal of policy concerning remittances has been to encourage the flow of money to come into the official system and to discourage the flow of funds through the underground network.

In 2004, then Federal Reserve Governor Ben Bernanke made clear that, “The Federal Reserve is attempting to support banks’ efforts to better serve immigrant populations, with remittances and other money transfers being a key area of interest.” House Financial Services Chairman Mike Oxley (R-OH) told President Bush’s then-Treasury Secretary John Snow, “Remittances between established and emerging economies foster growth in both types of economies simultaneously. I will be interested in hearing your views on how unnecessary costs can be eliminated in this area.” When Senator Paul S. Sarbanes (D-MD) introduced legislation that became the basis for today’s law that covers remittances, he had the simple goal to “increase transparency, competition and efficiency in the remittance market, while helping to bring more Americans into the financial mainstream.”

The longstanding bipartisan support for bringing remittances into the financial mainstream is based on the fact that most immigrants, regardless of whether they are U.S. citizens, legal residents, or undocumented, send remittances. A system that tried to assert proof of citizenship or legal status upon wiring money overseas would be burdensome, costly, and ineffective at best and if effective, it would simply drive more money into illegal transmission schemes while increasing crime here in the U.S. and abroad. Imagine if an entire community knew that someone would be walking through their immigrant neighborhood with a suitcase full of tens of thousands of dollars in cash.

Thought of another way, if I went to the bank to send money to my mother who lives in France part of the year, how would I prove that I’m a citizen? My driver’s license alone is not proof of legal status. Would I need to bring my passport? What if, like the majority of Americans, (62% according to the State Department) I don’t have a valid passport? Would I have to bring my birth certificate to the local Western Union? I guess the one positive thing from such a system is that it would help stop the email scams asking for money from a Nigerian Prince….

Aaron Klein is a fellow at the Brookings Institution and served as Deputy Assistant Secretary of the US Treasury Department from 2009 to 2012. He also serves as an unpaid member of the Clinton campaign’s Infrastructure Finance Working Group; he has not served as an advisor on any banking or finance issues.

Editor’s note: This piece originally appeared on Fortune.

Editors’ Note: Donald Trump has exposed the tension between democracy and liberal values—similar to the Arab Spring, writes Shadi Hamid. This piece originally appeared on The Atlantic.

When I was living in the Middle East, politics always felt existential, in a way that I suppose I could never fully understand. After all, I could always leave (as my relatives in Egypt were fond of reminding me). But it was easy enough to sense it. Here, in the era of Arab revolt, elections really had consequences. Politics wasn’t about policy; it was about a battle over the very meaning and purpose of the nation-state. These were the things that mattered more than anything else, in part because they were impossible to measure or quantify.

The primary divide in most Arab countries was between Islamists and non-Islamists. The latter, especially those of a more secular bent, feared that Islamist rule, however “democratic” it might be, would alter the nature of their countries beyond recognition. It wouldn’t just affect their governments or their laws, but how they lived, what they wore, and how they raised their sons and daughters.

Perhaps more than at any other time, millions of Americans are getting a sense, however mild in comparison, of what it might feel like to lose your country—or at least think about losing your country—because of what people decide to do in the privacy of the voting booth. It still remains (somewhat) unlikely that Donald Trump, the now presumptive Republican nominee, can win a general election. Regardless of the final outcome, however, the billionaire’s rise offers up a powerful—and frightening—reminder that liberal democracy, even where it’s most entrenched, is a fragile thing.

* * *

When I hear my friends debating how, exactly, so many of their fellow citizens could support someone like Trump, it reminds me a bit of Egypt. In my forthcoming book, I relay a telling conversation I had four years ago, which has stayed with me since. A few days after the country’s first post-revolutionary elections concluded in January 2012, I visited my great aunt in her extravagant flat in the posh Cairo suburb of Heliopolis. She was in a state of shock, but worse than that was the confusion. It was one thing for the Muslim Brotherhood, long Egypt’s largest opposition group, to win close to 40 percent of the vote, but how could 28 percent of Egyptians vote for ultraconservative Salafi parties, which believed in the strict implementation of Islamic law?

Like most Egyptians, she personally knew Brotherhood members even if she didn’t quite like them, but she hadn’t had much experience with Salafis and seemed totally unaware that they had extended their reach deep into Egyptian society. She realized, perhaps for the first time, that the country she had thought was hers for the better part of 70 years would never quite be the same. It hadn’t really even been hers to begin with.

What if voters don’t want to be liberal and vote accordingly?

What my aunt feared was that Egypt would become an “illiberal democracy,” a term popularized by Fareed Zakaria in his 2003 book The Future of Freedom, but one that’s still difficult for Americans to fundamentally relate to. In the American experience, democracy and liberalism seemed to go hand in hand, to such an extent that democracy really just became shorthand for “liberal democracy.”

As Richard Youngs writes in his excellent study of non-Western democracy, liberalism and democracy have historically been “rival notions and not bedfellows.” Liberalism is about non-negotiable personal rights and freedoms. Democracy, while requiring some basic protection of rights to allow for meaningful competition, is more about popular sovereignty, popular will, and accountability and responsiveness to the voting public. Which, of course, raises the question: What if voters don’t want to be liberal and vote accordingly?

* * *

When the stakes are high, there is more to lose, and if there is more to lose, those on the losing end of a ballot box have powerful incentives to play “spoiler.” Fortunately, in the post-Civil War United States, the stakes have never reached what political scientist Barry Weingast calls the “threshold” at which citizens decide to defend themselves through extra-constitutional means, including by appealing for the military to take sides. This, in part, is why (good) constitutions are so important: They lower the stakes, reassuring citizens that even if their preferred party loses the election, it’s still just that—an election.

Donald Trump, or more specifically what he represents, calls some of these assumptions into question. Trump himself isn’t quite an Islamist, but he is a proponent of a kind of “illiberal democracy,” even if he himself may not be familiar with the term. Drawing on a wellspring of white nativism and machismo, candidate Trump has regularly made demeaning statements about entire groups of people, including African-Americans, Mexicans, and women. His commitment to the protections enshrined in U.S. constitution are questionable, at best, and if we assume the worst, downright frightening (the difficulty with Trump is that he’s not precise with words, so it’s sometimes hard to make sense of what he’s saying). He has expressed support for registering Muslims in a database, elaborating that they could “sign up at different places.” When a reporter asked how this was different from requiring Jews to register in Nazi Germany, Trump said “you tell me,” prompting The Atlantic’s David Graham to note that “it’s hard to remember a time when a supposedly mainstream candidate had no interest in differentiating ideas he’s endorsed from those of the Nazis.” Trump, for good measure, has also refused to disavow President Franklin D. Roosevelt’s internment of Japanese-Americans.

The U.S. Constitution includes robust civil-liberties protections, enshrined in the Bill of Rights. But these protections are not unlimited. Contrary to popular belief, majorities—if they’re large enough—can, in fact, do nearly anything they want, even in established democracies. It’s only really a question of how high the majoritarian bar is. In the United States, two-thirds of Congress and 75 percent of the states can amend or repeal articles of the Constitution. They could theoretically pass a constitutional amendment banning abortion. In countries like Egypt, Tunisia, and Turkey, where alcohol is currently legal and relatively easy to find, the issue of alcohol consumption is a touchstone for endless “what if” hypothesizing. Yet, Prohibition happened not in any of those countries but in America, with large majorities in the Senate and House of Representatives as well as 46 of 48 states backing the 18th Amendment (of course, banning alcohol in the U.S. wasn’t justified on primarily scriptural grounds, while in Muslim-majority countries, prohibition is seen as fulfilling an explicitly Quranic directive).

In other words, built-in constraints and constitutional “guarantees” aren’t enough on their own to preclude illiberal outcomes. What Americans really depend on, then, is a shared political culture and the ideas and ideals that undergird it. As James Fallows notes, “Liberal democracies like ours depend on rules but also on norms—on the assumption that you’ll go so far, but no further, to advance your political ends.” But all it apparently takes is one man with charisma and an unusually perceptive understanding of the human psyche to change that. There are norms against politicians suggesting that minorities should have special identification cards. There are norms against saying you want to kill the families of terrorists. There are norms against encouraging your supporters to use violence against their political opponents. It’s not entirely clear why you don’t do or say these things (because Trump clearly has), but you just don’t. The very fact that Trump has made such frightening comments on national television—without any corresponding “disqualification” or decline in popular support—has already undermined these longstanding norms.

The United States has had demagogues before, but they rarely make for viable presidential candidates. This is democracy’s blessing as well as its curse: that people you really don’t like—people who you think might threaten the Republic—can actually win. In the specific context of the Republican nomination, Trump opponents basically called for prioritizing good outcomes over democratic ones. They continued to search for possible paths to denying Trump the nomination, despite the fact that, barring acts of God, he was certain to win the popular vote and a plurality of delegates in the primaries.

Even if Trump reached the magic number of 1,237 delegates, which would normally settle the matter, there were those who still seemed intent on scouring the rulebooks, parliamentary procedure, and delegate details in the hope of averting disaster. Democratic norms, the thinking goes, are great in normal contexts, but sometimes the stakes are simply too high to let democratic outcomes stand. As the columnist Walter Shapiro wrote, “[W]ith the threat of the first takeover of a modern political party by an authoritarian who traffics in racism and exudes contempt for the First Amendment ... [t]here would be nothing anti-democratic about GOP leaders using every mechanism in their power to stop Trump.” Nate Silver pointed out that “technically [Republicans would] be able to deny Trump the nomination even if he had a delegate majority by changing the rules at the last minute.” They could still theoretically do something like this, even after Trump’s decisive victory in Indiana. The Republican Party is not a country, and the party can disregard the preferences of primary voters if it so chooses, but elite pacts and back-room negotiations would seem decidedly antiquated during an unusually populist moment in American politics.

[T]here will no doubt be a temptation to defy or otherwise undermine a democratically elected Trump.

This particular debate in some ways mirrors arguments over the tensions between democracy and liberalism, a debate that will only intensify if Trump gains ground on Hillary Clinton in the coming months. It is probably time to err on the side of imagination, since party elites and pundits failed to imagine the unthinkable once already. What if Trump actually wins the presidency? How would we as Americans deal with an outcome that at least some of us see as a potential danger to our Constitution as well as our livelihoods?

If Donald Trump wins, he would have, whether we liked it or not, a democratic mandate. Once in power, he might moderate his rhetoric and policies (yet another data point in the debate over the “inclusion-moderation hypothesis”), rendering at least some of this discussion moot. Yet it’s also possible that, facing a growing terrorist threat and a sputtering economy, more and more Americans might, like their newly elected president, dispense with the norms of reasonable conduct and support extreme measures. Still, a President Trump would be a legitimate president, having been freely and fairly elected by enough Americans. He would be, as much as it pains me to say it, our president. Still, there will no doubt be a temptation to defy or otherwise undermine a democratically elected Trump. For those of us who study the Middle East, the idea of not respecting democratic outcomes is business as usual, but I never thought it would be up for debate in the United States.

* * *

“Deep state” is a phrase that’s used to describe the constellation of autonomous and self-perpetuating institutions, namely the judiciary, military, and security services, which operate outside the glare of the public and are immune to the electorate’s whims. This deep state, acting as the guardian of national identity, puts limits on what elected politicians can hope to accomplish. The deep state was responsible for four coups in Turkey, the most recent of which deposed the country’s first-ever democratically elected Islamist prime minister in 1997.

It would be difficult for Americans to think about their own government—or “regime”—in such terms. The U.S. military is subject to civilian control, while Supreme Court justices, though unelected and appointed to life terms, are nominated by the president and confirmed by the Senate. It is possible, however, to imagine a president so reckless as to activate state institutions against him or her, in a way that makes the notion of an American deep state more meaningful and relevant.

Former CIA Director Michael Hayden ignited some speculative debate when he said that the military “would refuse to act” if ordered by a President Trump to take actions that were clearly illegal, such as killing the families of terrorists. Moreover, he said, military commanders are “required not to follow an unlawful order.” Even short of flagrant illegality, the military can still do what it’s done, at times, with nearly every sitting president. Peter Feaver, a leading expert on civil-military relations, notes that “the historical record is replete with cases of the military shirking—withholding information and options, slow-rolling, end-runs to Congress and the media, inflating cost estimates, etc.—to thwart civilian policies they deem to be unwise.” Considering, however, that Trump would likely be more “unwise” than most past presidents, such tensions could intensify well beyond what America’s political system is accustomed to.

"[C]oup”...is not a word that Americans should ever get used to hearing in everyday political discourse.

One can also easily imagine left-of-center (and right-of-center) civil servants in the Departments of State and Defense working against the president from within to mitigate his effectiveness and even his authority. This would be good, insofar as Americans wouldn’t want their president doing things that were crazy, illegal, or both. But it would still raise difficult questions about democratic legitimacy and how far an elected president can pursue his preferred policies, especially when it comes to issues that aren’t clear-cut. If the military refused to obey orders, however justified their refusal, then it could very well erode norms against military intervention in domestic politics. In response to Hayden’s comments, host Bill Maher joked that the former CIA director was floating “a coup.” This is not a word that Americans should ever get used to hearing in everyday political discourse. The norm against “coups” is a powerful one, which explains why American analysts (if not the U.S. government) are generally uncomfortable with military coups in foreign countries. No one teaches us that military coups are bad. Rather, it’s something we absorb in the process of being American. It goes without saying, so it’s rarely said.

Recently, a few friends (who work on Middle East issues) and I had an interesting although ultimately frightening conversation, as Trump extended his delegate lead over Ted Cruz. Sometimes it’s useful to game out worst-case scenarios, however unlikely they might seem. We tried imagining a dystopian future and came up with internment camps, (threats of) military coups, and pro-Trump militias. Soon enough, the last didn’t seem nearly so farfetched, with volunteers offering to provide security at Trump rallies (for Trump supporters).

* * *

It is hard to imagine such things because, despite a long, low-intensity war on terrorism, America hasn’t faced a large-scale terrorist attack on the homeland since September 11, 2001. Democratic systems produce self-perpetuating norms, because they are accountable to a voting public. It’s this very responsiveness, though, that can be a source of vulnerability, if enough citizens, in the grip of fear, decide to prioritize “security” over liberty. As the legal scholar Christopher Kutz writes in the suggestively titled article “How Norms Die,” democracy can be “at the same time both fertile and toxic: fertile as a source of humanitarian values and institutions, but toxic to the very institutions it cultivates.”

This is something we can measure. As Daniel Bush observed, after analyzing Pew survey data from 2002 to 2014: “During each campaign season, respondents reported having a higher negative impression of Muslim Americans than in non-election years.” This is a bit more mild than the link between elections and religious riots in India. As the historian of religions Michael Cook notes, “There is no doubt that Hindu nationalist politicians believe that communal riots can get out the Hindu vote for them. ... Under the right conditions the communal riot is a winning [electoral] strategy.”

Norm shifting of an even more dangerous kind than India’s can happen rather quickly in countries where democracy is not yet consolidated. For example, millions of Egyptians who demanded freedom and democracy in 2011 turned seemingly against it in less than two and half years, supporting not just a return to authoritarian rule but the August 14, 2013 massacre of more than 800 protesters—what Human Rights Watch calls the “worst mass killing in [Egypt’s] modern history.”

The kinds of shifts that occur in established democracies are less nefarious, but they can happen just the same. Torture is a good example. Kutz calls the spread of global norms against torture “one of the most impressive successes of the post-war period.” Yet, in the United States, these norms began to erode after the attacks of September 11th. Soon enough, torture—or what some were now euphemistically calling “enhanced interrogation”—came to enjoy broad support among the American public. The lesson again is clear. However strong they may first appear, norms, particularly those relating to national security, are more fragile than we might like to think. Once their sanctity is undermined by authority figures (whether presidents or presidential candidates), others can judge that what was once considered shameful is now not just socially tolerated but also necessary, good, and just. This is why “political correctness”—even if it seems irritating and is sometimes abused to restrict reasonable debate—still represents a public good: It makes us think twice about saying things that might contribute to the erosion of liberal and democratic norms.

[N]orms, particularly those relating to national security, are more fragile than we might like to think.

We have now reached a point where current or former presidential candidates from both parties have flirted with the idea of internment camps (former Democratic candidate Wesley Clark has called for “segregating” radicalized Muslims who are “disloyal to the United States”). In a series of incidents that have received less attention, a Tennessee State Representative called for using state institutions, in this case the National Guard, to “round up” Syrian refugees. Meanwhile, the mayor of Roanoke, Virginia, called for suspending assistance to refugees, but went further in an official statement on government letterhead. “I’m reminded,” he wrote, “that President Franklin D. Roosevelt felt compelled to sequester Japanese foreign nationals after the bombing of Pearl Harbor, and it appears that the threat of harm to America from ISIS now is just as real and serious as that from our enemies then.”

No less than Supreme Court justice Antonin Scalia believed that it could happen here. On this, he is on strong ground, since it has, of course, already happened. In 1944, the Supreme Court upheld Roosevelt’s internment of Japanese-Americans in Korematsu v. United States. While Scalia said that the decision was “wrong,” he also issued a warning in his blunt style: “You are kidding yourself if you think the same thing will not happen again.”

The norm against internment has been undermined, even though Americans do not face anything close to the threat presented by the Nazis and Japan during World War II. Which raises the question of what a plurality, or even a majority, of Americans might be willing to support if they had to confront a threat that was truly existential. We Americans are not, today, at war, at least not in the normal sense. I hope to God that we never will be again. But we might be. And this is where Scalia’s words that day were perhaps most chilling, in part because he was right. Evoking the Latin expression inter arma enim silent leges, he reminded the audience that “in times of war, the laws fall silent.” All we will have then are the things we still believe in—our norms. But, by then, they might not be enough.

Editors’ Note: Though our allies could do more, the United States benefits not just from military spending, write Kathleen Hicks and Michael O’Hanlon. U.S. allies are not just an advantage for America; they also are the central characteristic of its global leadership. This post originally appeared on USA Today.

Donald Trump questions the value of America's alliances, and at one level it must be acknowledged that he has a point. There is not enough burden-sharing.

Most starkly, the United States spends 3 percent of gross domestic product on its armed forces, while the rest of NATO averages 1.4 percent of GDP even after agreeing formally to a 2 percent target. And the consequences are natural—for example, at the peak of the Afghanistan war, the U.S. provided 100,000 troops to the mission while the rest of NATO managed only about 35,000.

[W]hatever the imperfections of America's allies, it is unconvincing to view them as a drain on the country.

But whatever the imperfections of America's allies, it is unconvincing to view them as a drain on the country. On balance, they are not just an advantage for America; they also are the central characteristic of its global leadership. Russia and China are blessed with the likes of Belarus and North Korea as their formal security partners. Meanwhile, America has nearly 60, most of them among the world's dominant technological and economic powers. Two-thirds of global GDP and two-thirds of global military spending is found within the broad, U.S.-led Western security coalition. This is unprecedented in the history of the world, especially in the absence of a clear agreed threat that would motivate such an alignment.

The stakes are very high. Trump is apparently willing to disband NATO as well as our key Asian alliances, and to withdraw from the Middle East as well—a "Trexit." At risk is a core principle of America's post-World War II strategy—that trying to stay out of others’ business did not work and, in fact, helped lead to the world wars. Trump in particular seems to reject the core elements of America’s strengths in the world market and international security system. With the tumult roiling Europe today, it is worth reviewing the basics on why our unparalleled alliance structure pays off.

Allies are not all alike. Although our Western European allies generally under-invest in common defense objectives, our Middle East partners and several in Asia do far better. The Middle East allies spend considerably more than the NATO members; South Korea comes in at about 2.5 percent, with Australia at 2 percent. Moreover, contrary to conventional wisdom, the U.S. does not squander huge sums of money basing troops on the territories of its wealthy allies. According to RAND analyses and our own estimates, it costs less than $10 billion a year to keep U.S. forces in key industrial nations. That amounts to less than 2 percent of the defense budget. And of that nearly $10 billion, half or more is paid by the host nations themselves.

Moreover, basing those U.S. forces abroad can be more efficient than keeping them at home, allowing a modestly smaller force to do the same job in some cases (as with the aircraft carrier based in Japan, for example)—saving substantially more than $10 billion annually, in fact.

Military might and spending are not the only measure of alliance contributions to security. It is worth remembering that many wealthy allies spend a lot more of their national wealth on development aid and refugee resettlement than does the United States—on average, in the range of 0.5 percent of GDP more than we do, which is real money even measured against the sizes of our military budgets. NATO allies also commit much more to United Nations peacekeeping missions than we do—roughly 10,000 troops, plus or minus, in recent decades. While they did not match us in Afghanistan, they collectively suffered more than 1,000 fatalities and have stuck with the mission for 15 long, hard years.

Those who question the basic value of our alliances or engagement overseas go too far.

Nor is it all about military power. European members of NATO are absorbing the greatest costs and risks in applying sanctions on Russia over its behavior in Ukraine. The U.S. tally is not even a 10th of the European Union's Russian trade. Europe was also collectively crucial in applying sanctions on Iran.

Perhaps this would have happened anyway, without NATO, in the world Trump wants to create. Perhaps not.

It would be a different matter if America had lots of allies that went around causing wars and then expecting U.S. GIs and taxpayers to bail their chestnuts out of the fire once they overreached.

But that is not the case. Since the creation of the U.S. post-World War II alliance system, those few interstate wars that have happened have largely been caused by neutral or adversarial states—North Korea attacking South Korea, Arab states attacking Israel (in the days before many of the former became security partners with the United States), Iraq attacking Iran and Kuwait, China and Vietnam throwing their weight around their neighborhoods in earlier Cold War decades, and Russian President Vladimir Putin redrawing borders in Europe (after decades of the Soviet Union drawing an Iron Curtain across Europe).

In those rare cases where U.S. allies did employ force, as with Pakistan attacking India in earlier eras, or the British and French acting in former colonial states, it was generally understood that America would not help them, and we didn’t.

There are valid critiques about alliance burden-sharing, many of which are universally shared by foreign policy practitioners. Even so, the verdict is simple: Those who question the basic value of our alliances or engagement overseas go too far. In so doing, they distort the big picture. On balance, America’s alliances help this country undergird a global security system that has dramatically reduced the prevalence of war between nations in modern times, while currently costing the country only 3 percent of GDP.

A Trexit would be a disaster for Americans. To paraphrase Trump himself, our international security posture is a very good investment—and one that the U.S. global system of alliances does much to make possible.