It’s now been several weeks since Trump and his associates began a series of executive orders, announcements, social media postings, and interview responses on what one might, generously, call the administration’s tariff policy. By now, most commentators have realized that almost every component of the policy – the size of the tariffs on particular countries; the products on which they are, or are not, imposed; and the dates on which they will or will not be levied – has been developed in a way akin to what statisticians call a Markov process: the policy at time t depends on the advice that Trump received at time t-1. Thus, Scott Bessent and Howard Lutnick were able to take advantage of Peter Navarro’s momentary absence from the vicinity of the Oval Office to convince Trump to pause some of the tariffs. A few days later, after Trump spoke with Tim Cook, the head of Apple, he further paused tariffs on consumer electronics made in China. And a few days after that, following a conversation with the CEOs of three large retailers, Trump and his aides indicated that eventual tariffs on China would be much lower than the 145% figure Trump originally imposed.
Were there but world enough and time, one might imagine that this process could continue until, as interest group theory would have it, Trump would finally arrive at a tariff policy which would not provoke more than a small number of screams from the officials and business representatives to whom he listens. (I doubt strongly that U.S. consumers, whether MAGA or otherwise, belong to those groups.) In this sense, Trump, as an individual, would be playing a role that used to be played collectively by the U.S. Congress during the nearly two centuries when it was in charge of tariff policy. However, as numerous economists have pointed out, by the time that this political equilibrium would have been reached, both the U.S. and the global economy would have tanked, with layoffs, business failures, and, yes, higher prices. The question then becomes what factors, if any, can accelerate this process.
To start with, a few observations. First, as I’ve pointed out on numerous occasions, the fact that a policy is projected to fail, or has even already failed, is simply not something that matters very much for foreign policy making, at least in the United States. The classic example of projected failure is U.S. Vietnam policy in 1961 and again in 1965. In the first case, Kennedy decided to send advisers and not combat troops, even though he was told explicitly that in the absence of the latter, it was impossible to win. In the second case, the Under Secretary of State, George Ball, pointed out that the French had been unable to win a guerrilla war in Vietnam; the blustering response by McNamara, defending his advocacy of a 200,000 troop escalation, was the claim that because the new U.S. troops would prevent the guerrillas from fighting in large units, the South Vietnamese armed forces (whose patent incompetence, it should be remembered, had sparked calls for those U.S. troops) would therefore have a chance to engage them. As for actual failures, U.S. foreign policy is replete with cases of the continued pursuit of massively unsuccessful courses of action: from covert military operations in Albania and Tibet to overt ones in Cambodia and Afghanistan, as well as decades-long refusal to extend or restore diplomatic recognition to detested regimes (16 years for the USSR; 30 years for China; 54 years for Cuba; 45 years and counting for Iran). In this regard, to imagine that “Trump’s fantasy of trade dominance” will “soon … become unsustainable” is, shall we say, unlikely to be the case.
Second, even when policies are finally, grudgingly retreated from, those changes do not happen because of external pressure but, at least as a proximate mechanism, because of pressure from what psychologists would call the reference group: in this case, the set of persons to whom the leader looks for validation. Even the most autocratic and authoritarian leaders depend on key followers: as Hume put it,
...as Force is always on the side of the governed, the governors have nothing to support them but opinion. It is, therefore, on opinion only that government is founded; and this maxim extends to the most despotic and most military governments, as well as to the most free and most popular. The soldan of Egypt, or the emperor of Rome, might drive his harmless subjects, like brute beasts, against their sentiments and inclination. But he must, at least, have led his mamalukes or prætorian bands, like men, by their opinion.
But what is sauce for the goose is sauce for the gander: for the leader to appeal to key followers, he must be able to convince them that he is being realistic. If not, for example, when key advisers change their assessment of the situation, then, even though the initial dissidents may be fired (for example, McNamara over Vietnam, Rumsfeld over Iraq), the leader usually ends by adopting the policy they advocate. Criticisms from outside the inner circle, whether by foreign leaders or from domestic political opponents, are in most cases ignored; only when they are channelled by the reference group, as Clark Clifford did for business leaders, or Melvin Laird for his former colleagues in Congress, or as Bessent and Lutnick apparently did for bond market traders, are they listened to.
Third, what is channelled, though, are not recommendations to scrap or scale back policy X because it has been criticized or because its anticipated consequences are negative. Instead, advice from the reference group typically involves a recommended pivot from one policy instrument to another. In the case of problematic military interventions, the advice was to replace U.S. troops with local armed forces, U.S. bombing, or diplomacy. In the case of unacceptable foreign leaders, the advice was to foment a coup d’etat. In the case of popular uprisings, the advice was to find a so-called third force who could substitute for the local military. And for economic policies, such as the Nixon Shock of 1971 and the first Reagan administration’s support of a strong dollar, the advice was to drop particularly criticized tools (an import surcharge, nonintervention on foreign exchanges) for multilateral accords (Smithsonian, Plaza).
Fourth, and of particular relevance to Trump’s tariffs, when leaders followed their reference group’s advice and switched policy instruments, they were constrained on reneging on their decision for a simple reason: within a short time after the switch, the earlier policy instrument was no longer available. Thus, once the president decided to start withdrawing U.S. troops (Vietnam, Iraq), he lost the political support to reverse that policy. In the case of Vietnam, once troops were withdrawn, Congress imposed a bombing halt, thereby taking bombing off the table when things later became dicey. As for foreign economic policy, once the dollar had been devalued on two occasions in just over a year, neither Nixon nor his advisers no longer considered that tool to be practical. In each of these cases, what made the policy change irreversible was not the good will, the public-spiritedness, or the honesty of U.S. leaders, but the fact that regardless of their predilections, earlier policy instruments were no longer at hand.
These observations have clear implications for what, if anything, can be done by those affected by Trump’s tariff policies to get him to renounce those policies in a lasting way. The first is to act in such a way as to get the attention of Trump’s reference group. Any port in a storm: whether an importer has to raise prices, lay off workers, or shut down; whether a foreign government imposes countervailing tariffs or advocates boycotts of U.S. goods; whether a financial adviser recommends investing abroad or supports lawsuits against the tariffs; whether a U.S. exporter is faced with bankruptcy or the loss of the family farm, is irrelevant, as long as their actions, current or anticipated, are clearly communicated to Bessent and company, as well as to every Republican in Congress. The more often these messages are conveyed, the more loudly and the more panicked, the more likely they are to serve as a springboard for a recommendation to switch policy instruments.
As regards that switch, the most readily available alternative instrument is some kind of global conclave on trade rules. (Bretton Woods Two, anybody? Perhaps the Mar al Lago Accords?) I say this not primarily because it would speak to Trump’s ego, but because it is the only policy instrument ready at hand that can be implemented relatively quickly. Even if one believed that leprechauns exist and that the Administration is actually negotiating dozens of bilateral accords, the time that these would take to conclude would put us well past the witching hour at which we begin to see empty shelves. By contrast, a global trading summit could credibly begin with a negotiating deadline prior to the expiration of which all announced tariffs would be suspended, thereby permitting all parties to maintain their threats.
Of course, even a successfully concluded global trade deal would, in itself, provide no assurances that Trump or, for that matter, future U.S. presidents would not pull the tariff lever. For that to be the case, Congress either has to retake its tariff power, or else one of the many lawsuits now being filed against the tariffs has to succeed at least enough for a judge to suspend the policy. The first of these alternatives is highly unlikely to occur, at least until the U.S. economy has suffered a major blow. The chances of the second I leave to specialists, but would observe that the Trump administration’s record of obeying judicial orders does not inspire confidence that it would simply acquiesce in the disappearance of what it once called the most “beautiful” policy instrument at the president’s disposal. Nonetheless, when one’s in a hole, the first thing is to stop digging.