A client asks pointedly what role the dollar plays in my rough inference from Fed research that a 20% tariff newly applied to China, Canada, Mexico and Europe might lift the 4-quarter rate of US inflation by about a couple percentage points for about a year, followed by a gradual fade.
The short answer to that question is the dollar plays no role, or an unknown role. When I circulated that FEDS notes piece from February 28, I tried to be clear that I was acting more as a reporter than as an analyst. Trying to run the econometrics involved in that sort of work – or even commenting critically on the econometrics of others – would be far beyond my competence.
Instead, I just pointed out that the piece considers tariffs against China only and that we might brazenly multiply the estimated effects by four to figure out what the effect of a 20% tariff on a trade bloc four times as large might be. That is rookie commentary for sure. It is possible that it is easier for Americans to substitute away from, say, European exports, than to substitute away from Chinese imports. In that case, multiplying by four would be too much. On the other hand, there are nonlinearities going the other way too. The larger the range of countries whose products are tariffed, the lesser the opportunity to substitute away from anyone. So, four might be too small. This is definitely not rocket science. It is more a case of look at this, and this is sort of a fun point in addition: they considered only China. And the Fed piece does not make explicit their assumptions about the dollar.
It is probably prudent to limit my own interventions here to points I think I understand. First, Treasury Secretary Scott Bessent is wrong to suggest that tariffs are not inflationary because the more money we spend on one product the less money we have to spend on another, making it a wash. That is seemingly not helpful as an empirical matter. Tariffs seem to cause the general price level to rise permanently. But in any event, it is beside the point, if the idea is to cut income taxes to offset the tariff hikes. The price of the one good goes up, Scott, but you indeed have as much money to spend on the other good because the government, you know, you, cut income taxes. Happy to help where I can! Admittedly it is complicated, which is why people who want to get this right do complicated calculations, rather than just reasoning ex-navel. But I would not rely on folks who just know this sort of thing.
Second, at the end of the day, we all know, it is not really in dispute, that the inflationary effects of the tariffs will be reversed by economic weakness, which is the point. In Bessent’s counterfactual world, the weakness comes on its own. In a more realistic world, the weakness comes because of a hit to confidence, such as we are seeing, or because the Fed reacts. Whether the work is done mostly by the Fed or mostly by confidence is an interesting debate. As you will have noticed, the markets are acting like confidence will do the trick. I am not inclined to fight that right now, but there is a lot going on.
Not technician, but there is now economic resistance above
Source: Federal Reserve, Bloomberg, FH calculations
Official data are to February. Current value is estimated to this morning at 10:30, based on movements in nominal exchange rates since the February observation.
Ok, blah blah, but what about the actual question?
On the specific question of whether a higher dollar might bail us out here, I think we can confidently venture something. A stronger dollar acts in part as a favorable supply-side shock (although it also has demand implications), just as higher tariffs are an unfavorable supply-side shocks (with more complicated demand implications). So, I can see why the client in touch with me asks about this relevant issue.
My own view on this would emphasize two points. First, the real trade-weighted dollar, whose value I estimate to 10:30 this morning, has been trending higher for the past 12 years. As a rough rule of thumb, that rally may have been reducing US inflation by perhaps an average of 20 basis points for the past eight years or so, given the lags involved. For the dollar’s effect to intensify, we might need that rally to quicken or at least to sustain itself if there are non-linearities involved here, as there probably are. That seems unlikely.
Second, the point raised above is a bit slippery if the task is actually to answer the client’s question, which is what is the shock-minus-control offset from a dollar rally induced by tariffs. To say it will be smaller than what we have already seen may be a cop out. So, let me offer something a bit more substantial.
I have been pushing the view for the last month or so that the dollar’s sustained rally over the past 12 years has finally taken it to a point of vulnerability. The net external debt has risen dramatically, in large part because valuation effects have reversed from dampening to explosive, which was bullish while happening, but less constructive as a departure point from here. And closely related, the current account deficit is now a large enough share of GDP, that its widening from here, as is implied by where the dollar is now, would face some risk of going self-reinforcing. I have set out some of the numbers related to this awkwardness in recent notes. But it should not really be a surprise qualitatively. The dollar is high, don’t expect it to accelerate higher from here. Relatedly, if the tariffs are scaled back, the dollar probably falls from here.
To repeat, how this plays out for rates is complicated, because there are a lot of things going on. But the high-conviction point is simpler. Tariffs = weaker GDP growth. There is no obvious get around.
