Tim Graf (TG): This is Street Signals, a weekly conversation about markets and macro, brought to you by State Street Markets. I'm your host, Tim Graf, head of Macro Strategy for Europe.

Each week, we talk about the latest insights from our award-winning research, as well as the current thinking from our strategists, traders, business leaders, clients, and other experts from financial markets. If you listen to us and like what you're hearing, please subscribe, leave us a good review. Get in touch, it all helps us to improve what we offer.

With that, here's what's on our minds this week.

TG: It's only been a few weeks since the “Liberation Day” tariffs were announced by the White House, and two weeks since some of the tariffs went into effect. But the breadth and size of US tariffs, particularly on Chinese imports, begs the question of whether we will soon start to see their impact on the prices paid by American consumers.

To answer that question, I can't think of anyone better to have on this week than Alberto Cavallo, a professor at Harvard Business School, our academic partner, and a co-founder of PriceStats, who captures inflation trends in close to real time using online prices from retailers around the world. Alberto has been at the forefront of research on this topic of tariffs and their pass-through to prices, and PriceStats is the perfect tool for us to capture these effects.

We talk a lot about what he's seen in the daily data, as well as some newly published groundbreaking research focused on trends in US goods prices over just the last few weeks.

Alberto Cavallo (AC): Hi, Tim.

TG: Hello. How's it going, Alberto?

AC: Good. How are you?

TG: Very well. Very busy these days, as I'm sure you are as well.

AC: Yeah. What happened with markets today? They went up again.

TG: I think some of it's a reversal of yesterday's weakness. Some of it, there's this headline from Bessent rumored to have said in a meeting that a deal with China will probably come. This is unsustainable. So all the things we're going to talk about are unsustainable apparently. So we'll see.

AC: Right. Okay.

TG: Well, Alberto, great to see you as always. Great to catch up on all the interesting work that you're doing. And of course, you have a new paper that we're going to talk a lot about, Tracking the Short-Run Price Impact of US Tariffs that you've written with your co-authors Paola Llamas and Franco Vazquez.

But of course, this is not your first set of work on tariffs. You had a paper a few years back that talked about the 2017 and 2018 tariff experience. And I wanted to start there, actually, as kind of laying the groundwork for that. And thinking about that experience, can you talk about what happened then basically and the work that you did using the data that you get from PriceStats, how that informed the thinking around tariff pass-through to inflation then?

AC: So yeah, we wrote that paper together with several co-authors, Brett Nyman, Jenny Tang and Gita Gopinath in 2019, 2020, essentially looking at what was happening with the trade war at the time, the first trade war. And we wanted to answer the question of who was paying for the tariffs. This was around early 2020, where our data set ended.

And we had these advantages that we could look at prices at the border, not the retail level. So essentially, we could distinguish between how much of the tariffs were being borne by importers and consumers. And we can also think about whether foreign exporters, in this case, Chinese exporters, were being affected or not. And we looked at prices.

So what did we find? We essentially found a lot of very quick and full pass-through at the border. And US firms being the ones that bore most of the tariff incidents. They were the ones paying much of the additional cost. This happened both when we looked at imports and exports. If you think of retaliation tariffs that the Chinese were imposing on US exporters. We noticed that in both cases, it was US firms who were having to lower their prices in the case of exporters or facing higher prices in the case of importers.

One of the lessons there was that the type of goods that were being affected mattered a lot. So if you looked at imports at the time in the US that were coming from China, there were mostly electronics and a lot of differentiated goods for which the US importers had little alternative suppliers at the time. The pressure was on them, on the importers, essentially to pay the tariff. If you looked at exports and the retaliation tariffs, the story was actually the opposite. The US at the time, and still today, exports mostly agricultural goods to China. And the Chinese could find alternative suppliers for those goods. So the burden again fell on US exporters, on US firms and US exporters in particular.

We also looked at the retail level. And there we found actually that even though the importers were facing these higher payments, these tariffs, they were not passing it all into the retail level. In fact, the retail pass through at the time was very limited. And bear in mind we were writing this paper a year and a half after the first tariffs had been imposed. And there were several lessons there that we learned.

First, the size of the tariffs mattered. But if you look at the first tariffs that were imposed, for example, on washing machines, they were relatively high at around 25 percent. And we did see pass through happening at the retail level quite quickly. And in fact, I think in that case, it was not just the size, but also how visible the tariffs were. Everybody knew the laundry machines and related goods were getting affected by tariffs. So it was relatively easier, I think, for many of these US firms to pass it on to consumers.

But then, as the tariffs were applied to different goods, first, there were several sub-categories of Chinese goods like handbags, bicycles, and the first tariffs were only 10 percent. We did not see much of a pass-through until those tariffs were eventually raised to over 20 percent. So size, I think, and visibility and people knowing who's getting affected or not does seem to matter a lot for the pass-through to the consumers.

We also found at the time that many retailers found ways to adjust. So a lot of them did stockpiling of inventories before the tariffs, something that is probably also happening right now or happened during this past few months.

And we found areas of trade diversion. So some of the retailers that import directly were bringing goods from Vietnam, from Singapore, from other locations, trying to avoid the tariffs being imposed on China. But overall, we did find also a lot of evidence that suggested these US firms were reducing their margins.

And my impression at the time was that many of them believed that the tariffs were going to be temporary and that a deal would be reached. So there was uncertainty about that. And the more uncertainty there was, the more the belief was that they were temporary, the more that actually would delay some of that pass-through from reaching consumers.

TG: You've talked about a lot of things that we are going to apply to the present. But I think the question I've been asked the most by clients in talking about the PriceStats data, daily data that capture retail prices around the world, but we're specifically really focused on the US here, is when will we start to see the impact?

And you mentioned in that experience with washing machines, it was pretty quick, high-profile tariff, high-profile goods. But others, it took more time and it also took getting tariffs to a higher level for retailers, not importers, but retailers to pass that on. I'm curious if you have any sort of guidance on that this time around now.

Before we get into the new work, I wanted to just talk about that timing element and see if you had any thoughts of when should we start to see this in data more broadly?

AC: That's a great question. I think it's going to be perhaps sooner than what we saw in the first trade war, and I can mention the arguments for why that is the case. But I also don't think we're going to see an immediate impact, because as we proved in that paper, there are many dimensions to this.

It's a complex pricing decision that many of the firms are facing, and there are many reasons for them to wait until they have more clarity to make those decisions. So I think we're talking about months, several months, not a matter of weeks.

I'll tell you why I think it might be sooner than what we saw in the first trade war. So if you think of the results and what we learned from that first paper, size of the tariff rate was clearly important, and that's very intuitive. You get hit by 1 percent and you may wait, you may decide to absorb some of that depending on other reasons that you have.

But when we're talking about 120 percent, clearly that gives firms a lot more reasons to pass it on. We still don't have clarity on exactly what those will be, and I think that is delaying some of that pass through. But if the eventual tariff rates we end up getting are those levels, even above 20 percent, will suggest we are going to get more pass through at the consumer level.

There's also more sectors this time that may be affected because the tariffs are also affecting more countries, in particular retailers. They may decide that they're getting these higher costs across all categories, and that will increase their willingness to pass it on.

There's also, as I mentioned, the perception of whether this is temporary or not. I think there's still a tremendous amount of uncertainty, but there might be a growing sense that we're going to be in these type of tariffs back and forth and escalation for quite some time. And, in fact, if you think about what happened with the first trade war tariffs, the fact that Biden never removed them, looking back, many of these retailers and others may think that some of these at least will be there for quite a while, and that will increase their willingness to pass this on to consumers.

And finally, there's a fourth one that I wanted to mention. We just came out of a relatively high inflation environment, and firms are very used to thinking about price setting more frequently than before. So there's a lot of flexibility in pricing. And that could mean that this time when they get hit by higher costs, they're going to pass this on quickly because they've been exercising that mechanism of price setting a lot more often than in the past.

There's so much uncertainty, and the uncertainty is even greater at this time around. And in these current weeks, that I think is pushing for a lot of delays as well.

TG: Yeah. In doing a lot of meetings with you about this last year, one of the great learnings from it and thinking about that paper and what you highlighted in it was the trade diversion element you mentioned. Things that might have previously come from China now being routed through Vietnam or Cambodia or Mexico. That was clearly what happened then.

But of course, this is far more universal. But at the same time, it's not a common universal tariff rate. There are still differentiations.

So I'm wondering, thinking about this as an economist, do you think that impulse to substitute where you're getting your imports from and trade diversion from that perspective is still a mitigating factor or are just the rates too high now to think about that?

AC: No, I think it will be a mitigating factor, particularly because there's a growing sense now that some countries may get some exemptions or at least much lower tariff rates. Of course, after the “Liberation Day” announcements, it looked like it was going to apply everywhere, and that shuts down the trade diversion channel. But I do think there's still going to be an element to that.

TG: So, reciprocal tariffs and “Liberation Day,” this is when they were all rolled out. And within a couple of days of that, actually, I started to see your name a little bit more on Twitter and in the media as well. I think it's all out there now. So, I was wondering if you could talk about why that was.

Why was your work brought up in that context? Can you talk a little bit about that?

AC: Yeah. So, after they announced the reciprocal tariffs, which are not really reciprocal, they are not, just to be clear, they're not matching what the other countries are tariffing or the tariffs that are being applied on the US. They appear to be designed specifically to reduce the bilateral trade deficits that the US has.

When they tried to explain how they had come up with these numbers, the government posted an explanation that had a formula in it, and then to justify some of the parameters in the formula, they cited one of my papers, actually the paper I was just talking about in 2021. But it's not clear to me how they actually used it. If you read the text, they seem to be using our paper as a way to justify a low sensitivity of prices to the tariff rates, which is important for their formula and the logic of the formula.

But as I pointed out on Twitter, and some of my co-authors have written about this, if they had actually looked for evidence of that particular sensitivity of prices to tariff rates in our paper, our results suggested that that sensitivity was much higher, which would have made these tariff rates that they were calculating about a quarter of what came out from those numbers.

There are several things we can learn here. One is that, like I said, these are not reciprocal tariffs. There's a logic apparently to their formula, which by the way is not in our paper, so it's not our formula, just to be clear. And the logic seems to be, what is the change in the tariff rate that I need in order to reduce imports enough so that we close completely the trade deficit we have with another country? If you look at the formula itself, there is some logic to it. Essentially, what they did is, let's calculate the change in the tariff rate. Let's think about how prices will react to that change, and then also think about how quantities of imports will react to that price change. That calculation had to be equal to the trade deficit so that it completely reduces it. That's the logic.

And the formula has lots of problems that it doesn't take into account what happens with exports. There's going to be retaliation, obviously, affecting the ability of the US to export goods, and that affects naturally the trade deficit. It's thought of as a simple bilateral relationship, but of course, there are many other categories, countries, and things to care about and worry about.

Even if you believe that particular formula and what they were trying to do with it, you needed to find some sort of evidence to assume that there's a certain sensitivity of prices to tariffs, and a certain sensitivity of imports to prices. That's where they used our paper for one of those elasticities or sensitivities. I made the point that if they had actually used the numbers we have in our paper, those estimates for tariff rates would have been tremendously lower. They would have been a fourth of what they came up with.

TG: But of course, that wouldn't have grabbed nearly as much attention or eyeballs, I suspect.

AC: No, but if you look at what actually happened, they applied these estimates, and the tariff rates were actually so high that they ended up dividing them by two. So perhaps finding more sensible sensitivities and parameters would have ended up giving them numbers that are more connected to what they may end up in practice doing.

Of course, people tell me, no, if these are negotiation strategies, there are reasons for this, and I understand that. But I think it's misguided even on that front. But in any case, I think there's very hard to justify it in terms of the formula or even going further, the numbers that they pulled from our paper.

TG: Yeah. Well, let's talk about the new paper now. We've led up to this. Again, the paper is called Tracking the Short-Run Price Impact of US Tariffs. It is now out on Alberto's website, and I suspect it will be widely discussed in the coming days, and it all leverages the PriceStats data that we've talked about many times on this podcast.

And so, as a very simple opening question, what tariff-related effects are we starting to see in the data?

AC: So we are doing this in two different ways. One is looking specifically at the microdata. For some retailers, in the paper, we can actually link this to their country of origin of individual goods. And we are finding some relatively quick adjustments to the announcements of tariffs. Now, the size of those adjustments is still small in magnitude, relative to the size of the announced tariff rates.

So this tells me two things.

One is that there is flexibility in the pricing, and some of these firms are ready to adjust quickly, but there's still tremendous uncertainty about what are the actual tariff rates going to be, and that leads to very small adjustments overall. There is some surprisingly quick reactions.

We also found that it's not just imported goods that are being affected, but many domestic goods that compete with these imported goods are also seeing some price increases. And that is actually consistent with what the economics literature has found from the first trade war. When you are selling a good that is competing with an imported good that is now becoming more expensive, you actually can raise your price as well, because you are expecting some of that demand to shift to your own goods.

It may also signal that some of these domestic firms are concerned about their supply chains, and they may have some inputs that are going to be either directly or indirectly affected as well. So that leads to price increases of domestic goods. And frankly, it could also be that the retailers are not sure what is going to get affected or not, in which categories, and you end up having these pricing strategies that allow them to spread cost of the tariffs, or perceived cost of future tariffs, into a wider set of categories.

But I think it's an important takeaway that we are already finding in the data some areas that this is not just about the prices of imported goods, but also about domestic goods as well. And even domestic goods in categories that are still not yet affected by the tariffs themselves or the announcements of the tariffs.

In the paper, we focus on this mass market, large retailers, a couple of them, for which we have data on country of origin. Sectors that are heavily represented in this data are household furniture and goods. Those mostly come from China, which is in fact where we found the most impact.

One interesting result we have is that we can build price indices for goods coming from different countries. We looked at what happened in particular with Chinese goods, Canadian goods, and Mexican goods, which were obviously the first three countries to be affected this year. We found that initially they all rose, all those prices rose, but the Chinese goods are the most persistent increases. We actually end up, after the April 2 “Liberation Day” announcements, we saw them increasing significantly, whereas goods from Canada and Mexico had an initial jump, and then they have mostly either stabilized or even come down in the case of Mexican goods.

So that, I think, also suggests that perhaps many of these retailers are expecting goods from these countries to be eventually exempted or facing much lower tariff rates than what is happening with China, where we see the escalation.

TG: One of the things I've noticed as well in looking at the US sector-level data, and I was looking at something like apparel as well as recreation and electronic products, which presumably the US also does not produce. It has to import quite a lot, and there are maybe not as heavy a tariffs as you're seeing as those placed on China, those goods subject to those tariffs. But I would expect to have seen stronger than normal inflation in those sectors, and really it's just kind of keeping pace with seasonal norms.

Do you have a sense of why other sectors beyond the household goods and furnishing sector are not seeing such tariff-related pressures, at least not yet?

AC: Yeah, so I think it has to do with this uncertainty about we don't really know if they're going to be affected or not. If you look at the PriceStats series that are produced by PriceStats, we're using data from a lot more retailers than what I have in the paper itself. The only sectoral series that is showing a new inflationary trend since early March is household goods and furniture, which is clearly going to be affected if the escalation of the trade route in China and the US continues.

This is important also for the question, the broader question on what will happen with inflation more generally. There are some sectors that are in fact more likely to have a lower inflation in the next few weeks. Particularly fuel is one of those sectors that has been putting downward pressure on the aggregate index for quite some time. And what's important, I think, to remember is that this may not be just a supply shock that increases costs in some sectors.

This can lead potentially to lower demand and the fear of recession that many people are talking about would actually put downward pressure on prices. And we are seeing some of that in fuel. Fuel prices have declined because oil prices have declined on the expectation that a recession is coming and global demand may actually fall. So in the next few months, I'm not expecting a very sudden increase in the inflation rate more.

I think we have to keep an eye on the fact that there is going to be some sectors affected by falling demand, potentially fuel being the main driver of that, pushing prices down, and then we're going to see some particular sub-sectors being more affected by the tariffs, like household goods and furniture, which are already there, and potentially some of the others that you mentioned, apparel or food, where we still don't see anything significant happening, but we may, as we get more clarity about which are the imported sectors that are actually going to be affected by such high tariffs.

TG: You mentioned supply chain pressures. And actually, just before we recorded this on Tuesday afternoon, there was a report about Scott Bessent telling a private meeting of investors that effectively the US and China have entered a trade embargo on each other, given how high the tariff rates are, and that that's unsustainable and it will be negotiated down, which is, of course, the great hope that markets are placing around this whole situation is that there will be negotiation and the worst effects of these will be mitigated.

But at least for the time being, these supply chain pressures give rise to the thoughts of the work we did during the pandemic and looking at goods availability in particular and your ability to measure the stockouts that were captured in online retailer data. And I'm wondering if you have any sense of what that looks like right now.

You mentioned some front-loading of demand and I suspect inventories are probably still sustainable for the short term. But do you have any sense if that's changing?

AC: Yeah, so we started looking at that. We don't see anything happening yet. I do think it's an important dimension in the next few months. If the US does apply such high tariffs, we're all focused on prices and the impact that will have on inflation. But another cost that US consumers may face is actually the fact that we may end up seeing fewer varieties of certain goods when we go to the supermarket or to the retailers. But we still have not seen much of a reduction in that availability.

I think that's, like you said, that it's too early for us to see anything because there are stocks of some of these goods. And frankly, many of these goods have actually not yet experienced the tariffs. So we may see some stockpiling even happening right now until those tariffs are applied.

If you go back to our 2018 paper, you can actually see the higher imports of Chinese goods in anticipation of the tariffs and for several weeks after the tariffs have been already implemented because there are lags on some of these shipments and uncertainty about exactly when those will happen. So I think this is potentially something we might be talking about in the fall, depending on whether these tariffs were applied or not. But it is an important dimension that we'll keep an eye on.

TG: Do you have a sense of whether currency might be playing a part here? We've had a big sell off in the dollar the last couple of months. That hasn't really come against the Chinese renminbi, of course. That hasn't really strengthened against the dollar. But the dollar is a lot weaker against most other currencies.

Can you talk a little bit about how FX markets might be impacting what we're seeing in prices?

AC: No, I mean, it could actually make things worse for the US. But traditionally, exchange rate passed through into retail prices is not that quick. So if you look at the results in our first paper, the first trade war, we did look at the fact that the renminbi was depreciating and asked the question on whether that was actually helping reduce some of the prices that the importers in the US were facing at the border. And we found some areas that it helped, but only marginally.

So to give you a sense of magnitude for 20 percent tariff increases, a year and a half after the tariffs had been imposed, the exchange rate channel had lowered prices that importers paid by roughly two percents. So it was a relatively small magnitude overall. And that's something that you find in the academic literature a lot, that this exchange rate pass through is not as quick. So I don't think it's mattering right now for the prices that we're seeing in the data either, although I haven't tested it econometrically yet.

TG: Yeah. Last question is about inflation expectations, particularly in the context that you talked about of this being also a hit to demand, a supply shock, but then an eventual demand shock. And long run inflation expectations, I think, are certainly giving something along those lines as far as a message in that, they've been pretty subdued since the start of the year and have actually come off quite a bit since the start of the year.

And I'm just wondering and thinking about this in the context of the policy response from the Fed, who are really, I think, like all of us, just waiting, watching, as you mentioned, uncertainty is the key word here. We don't really know.

But do you think we should heed that message from lower long-run inflation expectations that has now been in place the last couple of months as being the ultimate outcome of this entire tariff episode?

AC: Yeah, I don't tend to place too much emphasis on long-term inflation expectations that come from the consumer side, for example.

TG: Yeah. Yes, actually, that should be clear. I was talking about markets because of course, yes, the Michigan survey has gone the other direction. Yeah, good point.

AC: In general, I think the problems we have with expectations is connected to the uncertainty. I would expect them to be more volatile, and therefore perhaps less informative for the Fed in making its decisions. You can make the argument that it's not anchored and we should be concerned about that. But in the very short run, reacting to this is probably not a good idea. I think the Fed has to wait a little longer to see what happens with both indicators of inflation and recession or unemployment and output that we'll get in the next few weeks. So I would be, if I were them, on a wait-and-see mode and not necessarily overreact in particular to our expectations of inflation, which will be quite volatile.

We did a lot of work on consumer inflation expectations and we found in the past, we found it to be closely linked to the actual experiences or the perception of the prices that people were actually paying for some of these goods at the grocery stores. And since there's still not much happening in sectors like food, for example, which is very important for expectations of inflation at the consumer level, or in fact, the other sector that is very important is fuel. And that has actually not increased and even decreased a bit in recent months. I don't think that's something that the Fed should necessarily worry much on the consumer side.

And on the market side, there's just an indication of a tremendous amount of volatility. It will be important to see what happens, obviously, with the Fed's decisions in light of the pressure it's getting from the Trump administration. But that's something again that I think we have to wait a few months to understand how it's actually playing out.

TG: Yeah, I think that is an entirely different podcast as well. And so we shall wait and see. That is the key phrase, I think, Alberto. You've been very generous with your time with us this week. And hopefully, you'll be as generous later this year when we start to see some of the impacts we get from waiting. And hopefully, it comes through in the data.

But I really do encourage everybody to have a look at Alberto's work. It is about as timely a work, I think, as you can get on the topic. Again, the paper is Tracking the Short-Run Price Impact of US Tariffs, published by Alberto Cavallo, his co-authors Paola Llamas, Franco Vasquez.

Alberto, thank you so much for joining me this week.

AC: Thank you, Tim.

TG: Thanks for listening to Street Signals. Clients can find this podcast and all of our research at our web portal, Insights. There, you'll be able to find all of our latest thinking on markets where we leverage our deep experience in research on investor behavior, inflation, media sentiment and risk, all of which goes into building an award-winning strategy product.

And again, if you like what you've heard, please subscribe wherever you get your podcasts and leave us a review. We'll see you next time.