Daily pricing data reveal the slow-rolling impact of tariffs

Government statistics are typically the first stop to get a read on inflation. But private data-gathering efforts can sometimes reveal a level of detail that the official monthly indexes cannot.

As broad new tariffs roll out against U.S. trading partners, the Harvard Business School Pricing Lab has the detail needed to measure, in real time, the pass-through of tariffs to consumers. The HBS Pricing Lab “tariff tracker” monitors prices collected daily on more than 350,000 goods sold by the largest U.S. retailers. More importantly, the researchers have determined each good’s country of origin (either directly or, in a minority of ambiguous cases, using artificial intelligence).¹

These findings are focused explicitly on consumer goods—not a total picture of the U.S. consumption basket, but a deep sample of that portion of our basket most directly affected by tariffs. We discussed the latest findings with Harvard economist Alberto Cavallo, a founder and principal investigator for the HBS Pricing Lab. Cavallo also serves on the advisory council for the Center for Inflation Research at the Cleveland Fed.

The transcript below is edited for length and clarity; a video of our conversation is also available. We spoke on September 10, 2025; data below are current as of that date.

When you look at these granular, rapidly updated data, what do they tell you about how new tariffs are affecting the prices of goods in the U.S.?

They are, in fact, affecting the prices. But the reaction is gradual over time and modest in terms of size. What we’ve seen for the first six months of the tariffs—also past research that I’ve done on the [2018–2019] trade war—suggests we should expect a gradual pass-through instead of a one-time, big jump in the level of retail prices.

And there are many reasons why that is the case. It is a complex pricing decision for the firms. There’s a lot of uncertainty about the levels of those tariffs—whether they’ll be permanent or not, how they will impact each individual firm. And there are a lot of concerns also about how consumers will react. All this is preventing many of the adjustments from happening quickly.

That echoes what we have heard from many economists, that we wouldn’t expect to see the effects all at once.

That is correct. Now, you do find [bigger price increases already] in the places that are most likely to have some reactions. So, for example, we find it mostly on Chinese goods, which is one of the countries that has currently the highest tariff rate. We’ve seen an escalation of the tensions and no clear indication that there’s going to be a deal. So that makes the retailers more likely to pass this on. By the way, people know that the Chinese goods are getting affected significantly. That makes it easier for retailers to pass that on because consumers won’t get mad at them. We do find [inflation] more in [China-dominated] categories like household goods and electronics.

Overall, if you look at all the goods that we cover—which is a good representative sample of what you would find if you went to one of these very large retailers—the overall [price increases] are between 5 percent for imported goods and about 2.5 percent for domestic goods (Figure 1).

It’s important to think about the proper counterfactual here. This is not about the counterfactual being that prices would stay the same, or even that they would go up slightly. The pre-trend is different here.

Yes. If you focus on these kinds of goods [such as] electronics and household goods, they tend to have a mildly deflationary nature over time. These are goods that are launched at high prices and discounted over their life cycle. So, in normal times you would expect to see a little bit of a deflation. We can project these pre-tariff trends and look at the current level relative to that, and that raises the magnitude a bit. Instead of being a 4 percent increase in imported goods prices on average, we get 5 percent.

But you would expect more effect over time?

Yes. Put yourself in the shoes of a retailer. You know there’s going to be something there [but] you don’t know what that level should be. So, they tend to prefer to do this gradually over time and perhaps not overreact to any news that they see about the tariffs at this stage. I think many of them have become quite used to this idea that there are huge announcements and then they are pulled back. And then we get some news perhaps from the legal side. It’s all very uncertain at this stage, and it’s optimal to be gradual.

We can look at things like the import price index to see whether foreign producers are eating the costs of the tariffs. From what I understand, there’s not much sign of that.

No. We don’t have, ourselves, data on [prices at] the border, but if you look at the import price indices—pre-tariffs—produced by the BLS, they are not falling. In fact, they’ve actually increased a bit. So that suggests that foreign exporters are not lowering their export prices, and therefore, the U.S. is again bearing the cost. The fact that my current paper doesn’t show [more] pass-through to consumers suggests it’s U.S. firms at this stage who are eating the tariffs.

You also find that prices for domestically produced goods are increasing. What’s the economic theory, or the logic, behind the rise in prices across the board?

So that surprises some people and there are several reasons for that that are well known. One could be that they have imported inputs and they expect some of those imported inputs to face tariffs and therefore raise their costs. And then the second: Even if they don’t have imported inputs, they compete with imported goods. So, if your competitor is forced to raise prices, that gives you more pricing power and you can raise yours as well, even if your costs have not changed.

It’s a well-known fact that tariffs not only affect imported goods, they also affect domestic goods. You would, in principle, expect them to be affected to a lesser degree, and we do find that in the paper. And it’s precisely on the goods that appear to be competing a lot with imports. We can look at categories that have more than 50 percent imports, and that’s where you’ll find some quick and more significant reactions of domestic goods prices (Figure 2).

This category of “domestic-affected” products, could you just define that for us?

We call a domestic good “affected” if it is sold in a category that’s been directly targeted with tariffs or if it’s sold in a goods category where more than 50 percent of the goods are imported. In fact, during March the increases in those domestic-affected goods were just as high as what we saw for imported goods. Since then, imported goods have had more inflation than domestic-affected goods. But still the trend is higher for those domestic-affected than what it was before.

Why do you believe that you’re seeing so little effect, even some deflation, when you look at Mexico specifically, versus Canada or China?

Several reasons. One is that a lot of goods actually are exempted under USMCA [U.S.-Mexico-Canada Agreement] rules. Other economists have actually shown that many of these goods are even introduced at zero effective (tariff) rates. Mike Waugh at the Minneapolis Fed has actually shown that the Mexican [effective tariff] rate is [quite low]. In practice, even though we hear these huge announcements of tariff rates, what matters is how many of them are getting exempted. And in the case of Mexico, there’s a lot of that.

I also think the expectations about future trade deals could be affecting retailers. If you’re a retailer today selling goods from China, you might expect the tensions to be there a few months from now. If you’re thinking about Mexico, it looks like the U.S. is more ready to sign some sort of deal. The relationship seems to be going better. So that will make the pass-through lower right away (Figure 3).

Even though there would be some North American exemptions with regard to Canada, maybe the outlook is a little bit different.

That’s correct. So if you look at our results, Canada is kind of in the middle between China and Mexico. It’s true that Canada has taken a more aggressive stance [than Mexico] relative to the U.S. They imposed retaliation tariffs. So, perhaps the perception on the part of the retailers and manufacturers is that the tensions with Canada are going to remain longer than the ones with Mexico.

Where do you think all this is going to land?

It looks like every time there’s a deal that is announced, the rate ends up being 10 or 15 percent. So that looks like the baseline and I would expect that if more deals are signed, that’s the level we will end up with.

When we did the first analysis of the [2018–2019] trade war, only when the tariff rates were about 20 percent did we see significant and quick pass-through. So, if we end up with levels of 10 percent, what I think will happen is we will continue to see this very gradual pass-through. And most consumers may not immediately notice it even though we may end up a couple of years from now or perhaps even longer with prices that are about 10 percent, or maybe a bit less, higher than they would have been if the tariffs had not been in place.

And of course, the tariffs, I believe, create a lot of other problems as well.

What are the other concerns, beyond higher prices for consumers, that you think about long term?

Well, it could be an adjustment in terms of the economy doing worse down the road. Because there’s going to be, potentially, retaliation. Americans’ nominal incomes could come down. Even if prices don’t change, our purchasing power is falling in that sense.

I look at product variety a lot. You go to a supermarket here in America and you get goods from all over the world (Figure 4). Imagine that tariffs are imposed and the retailers don’t find it easy to pass it on to consumers. They may decide not to bring those goods and offer them anymore. There’s not going to be inflation in the data; it’s simply that those goods will disappear from the stores and, perhaps a year or two years from now, we will be offered less variety. And that is affecting our welfare.

One of the things I experienced as a student when I first came to America was the variety of goods in a supermarket. In Argentina, we had periods where we were completely closed to the rest of the world. We had imposed tariffs, and you would find a couple of varieties of certain goods. But you walk into a supermarket in America and you have goods from all over the world. That may be different down the road.

Another way firms can adjust is through quality. You may end up with goods that roughly cost the same. But if we are not able to measure the quality well, we won’t be able to adjust [prices] for it. And then we just have lower-quality goods moving forward.

We’re just six months into this tariff journey, and the bar for what’s a better or worse economic outcome keeps moving. Because one day we have 145 percent tariffs on China, then it’s only 35 percent, for example, people tend to react that it’s not going to be as bad as we thought it was going to be. But even a 5 percent increase in prices sustained over time, along with these other effects you’re talking about, that’s not insignificant.

That’s a fair point. It accumulates over time. Even though it is contained to certain categories right now, if we end up getting broad tariffs on all kinds of goods—and we see escalation because we get retaliation—then the final outcome could look much worse than how it looks right now.

And let’s not forget this is the short-run impact of the tariffs. We’re only six months in. It’s important to remember that somebody has to pay for these taxes and this revenue that the government is collecting.

There could be arguments that the U.S. can use its power as a large market to force the foreign exporters to reduce their prices more down the road. But for that, it is useful if you don’t apply a broad tariff that affects everyone—because you can threaten a particular country with saying, “I’m going to bring the goods from these other locations unless you lower your prices.” If the U.S. applies high tariffs to a lot of countries—and, by the way, shifts them around all the time—you won’t get that benefit.

So, we’ll see. I do think consumers will end up paying a big chunk of it eventually. The bad outcomes with these kinds of policies happen later. I’m from a country [Argentina] that started imposing a lot of tariffs. And it’s not like you see the negative effects right away. But if you look back 10, 20 years later, you realize the economy has become tremendously inefficient, that goods are sold at levels that are much higher—and not just historically because inflation has kicked in. When you control for quality, or you compare them to the prices that they are being sold for in other countries, you realize you end up paying a lot more.

In trade, protectionism tends to escalate. Once tariffs are in place, it’s hard for governments to take them out. That’s what I think a lot of economists have in mind when they worry about the long-run effects of this.

Endnotes

1 Most products can be matched to country of origin using Universal Product Codes (UPCs) and published information. The researchers rely on AI for approximately 14 percent of all products. Of all products in the sample, approximately 37.5 percent are manufactured in the U.S., 36 percent from China, 4 percent from India, 2.5 percent from Turkey, 2 percent from Vietnam and from Taiwan, 1.5 percent from Mexico, 1.5 percent from Canada, and 14 percent from all other nations.