The Surprising Truth About Tariffs and Inflation: A Wonk's Perspective
This article is an intriguing experiment in economic communication. Initially, the plan was to delve into highly specialized economic topics on Substack, but the platform's diverse audience of over half a million subscribers (yes, it's impressive!) has prompted a reevaluation.
The challenge is to strike a balance between engaging the economically inclined and not overwhelming the average reader. So, the strategy is to occasionally publish wonkish posts, keeping them accessible but not promoting them via email. It's a blog-like approach, and we'll see how it fares.
Now, let's dive into the topic at hand: the relationship between tariffs and inflation. Many media outlets portray the lack of significant inflation since a certain event as an enigma, baffling economists. And, of course, certain political figures are quick to claim that the experts were all mistaken.
But a closer look at the data reveals a different story. The crucial insight is that effective tariff rates have increased far less than the eye-catching headline rates. This is due to substantial exemptions, such as Apple's tariff exemption in India, and companies taking advantage of existing exemptions that were previously not worth the effort due to lower tariffs.
Here's where it gets interesting: the Penn-Wharton budget model (source: https://budgetmodel.wharton.upenn.edu/issues/2025/12/23/effective-tariff-rates-and-revenues-updated-december-23-2025) illustrates how imports from Canada and Mexico have claimed exemption from tariffs under the USMCA.
So, how much inflation should we attribute to the actual increase in average tariff rates? The graph at the beginning of this article provides a quick estimation method. It displays total customs duties as a percentage of GDP, representing the tax burden from the tariff hikes. These duties rose from 0.3% of GDP pre-event to 1.1%, an increase of approximately 0.8 percentage points, which can be our initial estimate of the tariffs' inflationary impact.
But wait, there's a catch. This estimate might be conservative, as the tariffs have likely reduced imports, meaning their effect on prices could be greater than their revenue impact. Nevertheless, it's a starting point, suggesting an inflationary influence of around 0.8% or slightly more.
Now, let's examine the real-world inflation scenario. We'll approach this from two angles.
Firstly, the HBS Pricing Lab (https://www.pricinglab.org/tariff-tracker/?utmsource=substack&utmmedium=email), created by the team behind the Billion Prices Index, has been tracking the tariffs' effect on retail prices. Their estimates fluctuate, but they indicate that the CPI is 0.8 percentage points higher than it would have been without the tariffs.
Secondly, let's compare 2025 inflation with pre-event forecasts. In late 2024, economists surveyed by the Philadelphia Fed (https://www.philadelphiafed.org/-/media/FRBP/Assets/Surveys-And-Data/survey-of-professional-forecasters/2024/spfQ424.pdf) predicted 2.2% core PCE inflation for 2025, the Fed's preferred measure. While we await the final 2025 numbers, the Employ America nowcast (https://www.employamerica.org/core-cast/december-2025-core-cast-post-ppi/) estimates annual inflation at 3% for 2025, 0.8 percentage points above pre-event expectations.
Interestingly, both methods yield the same result—an increase of 0.8 percentage points—which aligns with the rise in customs duties as a share of GDP. The consistency is remarkable.
While there's room for debate on the finer details, the core message is clear: the limited impact of tariffs on inflation isn't a mystery. Some economists might have exaggerated this effect, and certain political factions are keen to exploit this, but the reality aligns with what we should have anticipated.
And this is the part most people miss: understanding the nuanced relationship between tariffs and inflation is crucial for informed decision-making. So, what's your take on this? Do you think the media and politicians should present a more nuanced view, or is this level of detail unnecessary for the general public?
