Leprechauns, Effective Tariffs and Inflation
Yes, tariffs are hurting us. No, Trump wasn’t right.
Lately there’s been a lot of chatter that economists were wrong about Trump’s tariffs. Didn’t the economists predict that we would be stuck in a recession with runaway inflation by now? So maybe Trump was right after all?
Well, no, Trump isn’t right. If you look at what economists actually predicted — as opposed to what non-economists claimed that they predicted — much of it has come to pass.
And to prove that point today’s post is about where we are right now on tariffs and their effects.
But there is one thing that I believe almost all economists (including me) overlooked —the possible role of tax avoidance/evasion in limiting tariffs’ impact. (Avoidance is when it’s legal, even if it shouldn’t be; evasion is when it isn’t.)
First, let’s look at what economists actually predicted. While there were many flat assertions that tariffs would cause a recession, these claims weren’t based on serious economic models. Here’s what I said in my April conversation with Ezra Klein:
[O]rdinarily I would say that while tariffs are bad, they don’t cause recessions. It makes the economy less efficient. You turn to higher-cost domestic sources for stuff, instead of lower-cost foreign sources, and foreigners turn away from the stuff you can produce cheaply. But that’s a reduction in the economy’s efficiency, not a shortfall in demand.
What’s unique about this situation is that the protectionism is unpredictable and unstable. And it’s that uncertainty that is the recessionary force.
My statement was much more nuanced than the conventional wisdom about what economists were saying. And if you’re asking why the uncertainty hasn’t caused a recession (yet?), the obvious answer is that something else has been happening that has prevented a recession: An immense boom in AI investment.
Then what about inflation? Economic models do predict that tariffs will raise consumer prices unless the cost of the tariff is absorbed by the foreign exporters. But we know that foreigners are not, in fact, paying the tariffs. The Bureau of Labor Statistics, when it isn’t shuttered, produces estimates of import prices — the prices America pays for imported goods, exclusive of tariffs. The Bureau’s data shows that these pre-tariff prices of imported goods have not gone down since Trump started slapping tariffs on exporters -- including the penguins of uninhabited islands. So the numbers show that exporters are not paying for Trump’s tariffs.
Are consumers paying these tariffs? Yes, kind of. Consumer prices are starting to show significant increases due to the tariffs, but so far the price increases have been more modest than what many economists, myself included, expected. And it’s interesting to ask why the response so far has been muted.
Now, official consumer price data don’t separate out imported products from domestically produced products. And in any case we won’t be getting CPI updates until the shutdown ends. So that data source won’t be of any use in answering our question. However, the Digital Data Design Institute’s Pricing Lab calculates a tariff tracker using online prices. This is the kind of private data collection that needs to be done both during the shutdown and maybe even after, if the Trump administration succeeds in corrupting the Bureau of Labor Statistics. Notably, these data show that the prices of imported goods (orange line) have risen substantially more than the prices of domestically produced goods (blue line) since Trump began his tariff spree:
According to these estimates, retail prices of imports are up 4.4 percent since Trump took office, versus 2.8 percent for domestic goods.
But aren’t these relatively small numbers given the size of the tariff hikes? The widely cited estimates coming from the Yale Budget Lab say that the average effective tariff rate has gone from 2.4 percent before Trump came back in to more than 17 percent now, a roughly 15 point rise. So why aren’t retail prices of imports up by something closer to 15 percent, rather than 4.4 percent?
Part of the answer is that the the Pricing Lab data reflect retail prices, not the prices paid by importers. The prices retailers charge consumers for imported goods always include a substantial markup, reflecting transportation and distribution costs as well as some profit margin. The size of that markup varies a lot across goods, but as one shipping and logistics blog notes,
Most consumer products are marked up by over 100%, meaning the initial tariff impact might be relatively small compared to the final price consumers pay.
This means that even if effective tariff rates had really gone up by 15 points, we should expect retail prices of imports to rise around 6 or 7 percent, not 15.
But there’s a further point: At least so far, the average tariff rate importers are actually paying — which we can measure simply by the ratio of tariff revenue to the value of imports — is running well below estimates of the average rate implied by Trump’s decrees.
The Financial Times’s Alphaville blog has a revealing chart of the divergence:
As you can see, the expected tariff revenue based on the official Trump tariff rates (the blue line) is much higher than the actual tariff revenue (pink line).
What’s going on here? One explanation is that Trump’s tariff decrees include multiple loopholes and exemptions that importers are exploiting. For example, as Joey Politano points out, the data centers driving up your electricity bills are able to import chips and equipment tariff free. And as Alphaville notes, these exemptions create big problems in estimating just how much Trump has actually raised tariffs. Researchers at the Yale Budget Lab and other institutions are doing their best to take account of these exemptions when estimating the overall tariff rate, but it’s easy to see how their estimates may be somewhat off. Also, buyers may be blunting the effects of the tariffs by shifting away from highly tariffed imports to exempt products.
And there’s a second explanation, which I’m surprised hasn’t received more attention: Deliberate misclassification of imports by corporations, shifting them from tariffed categories to tariff-exempt categories.
If asked what I’ve learned about international economics over the past dozen years or so that really changed by views, it would be just how large a role tax avoidance by multinational corporations plays in distorting international statistics. For example, back in 2016 Ireland, famously a tax haven, reported 26 percent economic growth the previous year. This obviously didn’t reflect anything real. Instead, it reflected a change in tax strategy by a small number of multinationals — possibly just Apple! — that caused them to fictitiously report that a large share of their profits were earned in Ireland. At the time I dubbed this “leprechaun economics.”
The point is that corporations are very good at finding ways to avoid paying taxes, especially when the rules are complicated and the tax collectors don’t have enough resources to track down their clever strategies. The Trump tariffs are complicated and are surely overwhelming customs officials. Moreover, the Customs and Border Protection Agency is currently spending a lot of its resources going after undocumented immigrants, or people they think look like undocumented immigrants.
So it’s a good bet that Trump’s tariffs have been significantly leprechaunized, helping to bring the de facto tariff rate to 10 rather than 17 percent.
And even that 10 percent rate hasn’t yet been fully passed on the consumers: businesses have temporary absorbed part of the cost of the Trump tariffs because they are reluctant to raise prices. But that reluctance will come to an end, probably sooner rather than later. At that point we’ll see substantially larger price increases.
So Trump wasn’t right. Once you look at the details, tariffs are having more or less exactly the effects economists would have predicted. And it’s likely that those effects will become more pronounced over the next several months.
Finally, to go back to my Ezra Klein interview, in many ways the level of the tariffs is less important than the huge uncertainty Trump’s policies have created. And that uncertainty hasn’t abated at all. If anything, policy just keeps getting more erratic.
MUSICAL CODA
The "leprechaun" framing captures something essential: the real story isn't the headline tariff rate but the cumulative distortions beneath it.
But the most perilous effect isn't what has already happened — it's what has been taken away: policy flexibility.
We've moved from a broad, uneven trail where mistakes could be corrected to a tightrope where even small errors have outsized consequences. Tariffs, fiscal expansion, and political pressure on the Fed don't guarantee collapse; they guarantee vanishing margin for error.
The Fed can't easily stimulate without reigniting inflation. Congress can't spend without deepening deficits. Immigration restrictions keep labor supply tight just when flexibility is most needed.
A single shock — geopolitical flare-up, energy price spike, investment pullback, or policy delay — could now destabilize what would have been manageable before.
The "leprechaun" tariffs might not sink the economy directly, but they're building a world where policy becomes brittle. The danger isn't the wobble; it's that one strong draft could tip the balance — and the rope has become too narrow for recovery.
It’s disappointing to see many large, successful companies act as poor corporate citizens. Apple, for example, has benefited enormously from America’s laws, infrastructure, and past public investments in technology, yet it claims Ireland as its home base to avoid paying U.S. taxes. They take from the system that enabled their success but give little back. If a person behaved that way, constantly taking without reciprocating, you’d eventually stop associating with them.