There is lots of talk from the Trump administration these days about how the U.S. is getting cheated on trade. In this context, they have done some cherry-picking of the data to emphasize high foreign tariffs, while conveniently ignoring high U.S. tariffs. For example, Trump will mention a 270% Canadian tariff on dairy products, without mentioning U.S. tariffs of up to 187% on sour cream. Or White House trade adviser Peter Navarro will mention EU auto tariffs of 10% and argue that those are much higher than the 2.5% tariffs for car imports to the U.S, but he won’t mention the 25% U.S. tariff on truck imports.

So what’s the reality of tariff levels? The cherry-picking approach emphasizes particular products where tariffs are high, and as can be seen in the examples above there are still a few of these “tariff peaks,” including some imposed by the U.S. But with so much variation on tariffs by product, an average tariff level is more informative. Unfortunately, it can be difficult to get an accurate picture of this (as discussed here). Here’s a sampling of a few countries and the EU (explanations to follow):

As you can see, the table has three categories of average tariffs. That’s because measuring tariff levels is not that easy and there are several different methodologies. The “simple average” in the first column takes all of the tariff levels set out in a country’s tariff schedules and averages them. But that figure can be misleading. For example, with some products, there might not be much trade even without a tariff in place, so counting a high tariff in the average is misleading.

The second and third columns — the first from the World Bank, and the second from the WTO and two other organizations, using different methodologies — look at tariffs applied on actually traded goods. But that can be misleading too, because a 100% tariff might eliminate all trade, and therefore a high, trade-retricting tariff would not show up in the figures.

The best approach I can see here is to provide both kinds of figures, which is what I’ve done in the table. Taking all of these tariff figures into account, it can be hard to come up with a precise ranking, but you can see that New Zealand and Australia are the low tariff leaders. The U.S., EU, Canada, Japan, and Switzerland come next, clustered closely together. Mexico has tariffs that are a bit higher. Then come China and Brazil with even higher tariffs.

One important point to note is that these are the generally applied tariffs, but some of these countries have FTAs with each other, under which special lower tariffs apply (the World Bank averages are the only ones that take into account lower FTA tariffs). For the U.S., that means NAFTA, under which almost all tariffs (dairy and peanuts, among others, excepted) are zero; the U.S. — Australia FTA; and various other FTAs.

The lower FTA tariffs lead to an important point. If you are on the Trump administration trade team, and you think foreign tariffs are too high, the solution is to negotiate trade agreements that lower them (in both directions). So far, unfortunately, that has not been their focus.