The debate among economists around how the imposition of a national sales tax on Americans would affect inflation is not fully resolved. Different models generate different conclusions. And even within a single model, the presence of different assumptions will generate different results. For example, within the standard model, which holds that a tariff increase will cause the inflation rate to rise, there is the question of whether the short-run inflation expectation would remain anchored. If the inflation expectation stays anchored, then the inflation rise is transitory. And if the Fed confidently expected that result, then it could just let it play out.
I gravitate to inside-the-box thinking, which has been underrated since the Global Financial Crisis. But, for example, the fiscal theory of the price level would hold that any tax increase that tilts the path of the debt/GDP ratio lower is likely to generate disinflation – at least at some horizon. They say we inside the box thinkers have the sign wrong, although the the effect is small in any case.
But that is probably not what the debate around the inflationary effect – or not – around the Trump tariffs is really about. It seems to me that most people in markets accept the standard view that Trump’s proposed income tax cuts funded in large part with tariff increases would be somewhat inflationary and would provoke a response from the Fed that would raise the medium-term recession risk. We are not talking about an inflation spiral or an immediate collapse into recession. We are talking about something that would displease markets.
Here is what I take to be the source of confusion here, which is well summed up by a comment that incoming Treasury Secretary Scott Bessent has made. Bessent said a tariff raises the price of “one thing” but lowers income available to spend on the “other thing.” So, no inflation.
Fair point. A tariff increase taken in isolation would look like a contractionary supply-side shock, very similar to the oil shocks we used to worry about. It puts upward pressure on a select group of goods and / or services, but also results in the loss of what we might call discretionary income to spend on the goods and services not immediately affected. So, assuming no policy response to support real aggregate demand, their price goes down. Hence there is no general inflationary effect.
This does not mean that the tariff increase would be celebrated any more than any other national sales tax increase. But the damage is not done mostly through inflation. And yes, we get to this result without invoking that the Fed kills the inflation. On unchanged monetary “policy,” so to speak, the inflation rate goes up and then comes right back down. And it might even be plausible that the general price level goes up and then comes right back down. You may remember from back in the day folks used to say that oil shocks are “ultimately” deflationary. They had a point, although admittedly a debatable one. Models vary and assumptions matter.
The thing is, though, this is not really what is – or at least has been – on the table as a policy proposal. (I will get to that caveat below.) What Trump proposed during the campaign, and what fiscal policy analysts have been dutifully scoring out, is not a fiscal contraction delivered by what is effectively a sales tax increase. Trump has formally proposed an income tax cut funded in part with a national sales tax increase, as mentioned.
So, the answer to Bessent’s point is, yeah, but your guy is proposing to give them more money so they do have just as much to spend on the “other thing” as before. The price of your “one thing” goes up and there is just as much money as before to spend on your “other thing.” So, we get some inflation to start. And then the rest is mostly up to which assumptions in the standard model hold and how the Fed reacts. Goodness, Bessent sounds a bit like a dread academic. Ivory tower thinking!!
But there is a caveat. And the importance of that caveat underscores my main point here. There is far more debate around what Trump will do than there is in reality around what the expected effects of this or that “fiscal” program might be. Most people seem intuitively to accept the standard model – and those who don’t will be wondering what model the Fed will follow. We may be confident that what Trump proposed during the campaign would incline the Fed to deliver a period of weaker aggregate demand growth than would otherwise be appropriate.
Here’s where the caveat comes in. Recently, the incoming president has responded to the national security threat posed by Canada by suggesting that he would use an executive order on Day One to impose a 25% tariff on Canada and Mexico, along with a 10% tariff on China. He can do this unilaterally, because Congress has effectively given the president such authority in trade matters. But it does not follow from this that we get the (at least) offsetting income tax cuts implemented on the same day.
For those still interested in constitutionally restrained government, the income tax offsets would require Congress to act, and it is not clear how quickly Congress could deliver the incoming president’s income tax program, particularly while the GOP margin in the House is thinned out ahead of special elections that will have to be held to replace GOP reps that Trump is plucking to place in other parts of the government. So, if the question is how to think about the executive orders that Trump is talking about, then, yes, Bessent has a point. It is not remotely one that supports what Trump has recently proposed. But nor is it obviously wrong.
So far as I understand, then, the debate here is not really about economic models or even assumptions within them. The theorists quibble, but within markets and among policy makers there seems to be an acceptance effectively of the standard model. The question, as often in macro, is what people even mean by the question they propose. When they ask what the inflationary effect of a tariff might be, do they mean with or without a fiscal contraction?
