Introduction
President Trump believes that his tariffs will help reindustrialize America. That might happen, but so far not. Employment in US manufacturing has been falling since he took office – having plateaued during both Biden and Trump-I administrations (punctuated by the massive Covid shock).
But here’s the thing. POTUS’s tariffs will almost surely benefit Canadian and Mexican industry. Not their auto sectors – those will be hurt (see my last Factful Friday for why). But the rest will win (see below).
How could that be? The headlines shout about how Donald Trump is doing horrible things to Mexico and Canada. Prime Minister Carney is livid. President Sheinbaum is threatening retaliation. Surely new tariffs couldn’t be good for their economies.
The President himself claims he is striking back and extracting vengeance against the foreign globalist elite, including Canadians and Mexicans, who have been stealing American factory jobs. As POTUS put it when announcing his April-2 tariffs:
“For decades, our country has been looted, pillaged, raped and plundered by nations near and far, both friend and foe alike. … foreign cheaters have ransacked our factories, and foreign scavengers have torn apart our once beautiful American dream. … In 2016, I declared I am your voice. Today, I add: I am your warrior. I am your justice. And for those who have been wronged and betrayed, I am your retribution.”
– Donald Trump 2 April 2025.
This Factful Friday lays out the simple, but often overlooked, economics of how charging different tariffs on imports from different countries (i.e. discriminatory tariffs) can have unexpected, and unintended effects. Helping Mexican and Canadian industry will be one of those.
Canada and Mexico get a sweetheart deal on US tariffs
There has been much sound-and-fury during the President’s war on the global trade system. That’s on purpose.
Tariff announcements are the point; implementation is a detail to POTUS.
President Trump’s goal, in my view (as I laid it out in my May 2025 eBook, The Great Trade Hack) is to demonstrate to his angry base that he is carrying out his promise to hit back at those thieving foreigners, as the quote above said he would.
Just take the 23 July 2025 deal with Japan. Along with the usual hyperbole we have come to expect, he, tellingly, said:
Japan will pay reciprocal tariffs to the United States of 15% … Finally, Japan is treating America fairly.
– President Trump 22 July 2025.
In search of retribution, he has implemented an unprecedented rise in US tariffs – but not for all trade partners.
To set the stage for a gander at the tariffs, let’s have a look at which US trade partners matter the most. Fortunately, and here simplicity is all happy, there are only four or five biggies. Four if you limit it to double digit import shares; five if you want to cover at least 2/3rd of the imports.
Tariffs on the Big-5 trade partners (EU, Mexico, China, Canada, Japan)
When talking tariffs, the devil is always in the detail; you can read all about in one of my favourite trade policy books, Why Politicians Lie About Trade, by Dmitry Grozoubinski (2025).
In the case at hand, we have to take account of the various tariffs that: 1) were in place before POTUS declared war on world trade upon re-assuming office in January 2025, 2) he has put on since, and 3) all the exemptions he has granted to various countries and on various products. Here is a pretty good narrative of the bold announcements and frequent flipflops (ReedSmith Trump 2.0 Tariff Tracker) for those of you who care for that sort of thing.
The key concept to get a grasp on is the so-called “effective tariff rate”, or ETR for cognoscenti, where “effective” really means “trade-weighted average”. It is simple. You take the total tariffs paid on imports from, say, China, and you divide it by the total imports from China. Plainly this is an average tariff rate.
It boosts intuition to write the same thing a little differently to let us see the average tariff percent paid and the share of imports on which the average rate was charged. That is:
For example, China got some really high tariff rates – like 25% on most goods held over from Trump-I, and an extra 20% from the Fentanyl-justified tariffs, and 10% from the 2 April tariff salvo.
An additional complication comes from the fact that some US imports are exempted. For example, POTUS exempted energy goods, and semiconductors, and some electronics (so far). This meant that some countries, like Saudi Arabia which exports most energy products to the US, pay tariffs on almost no tariffs. But few of China’s exports are covered by these exemptions.
The table below give the effective tariff rates calculated by Fitch along with some details of the average tariff on non-exempted goods and the share that is non-exempted.
This table makes is clear that Canada and Mexico have really much lower effective tariff rates.
- Canada has only 9% of its exports to the US hit by tariffs, and the average rate charged is 21%.
- For Mexico, the numbers are higher at 17% and 25%.
This difference reflects the very different composition of the goods these two send to the US. Canada is more of an energy exporter than Mexico, for example.
Here is a chart that shows the current levels and rise since January 2025.
Why do Canada and Mexico get special treatment? A Trumpian trade dilemma
POTUS wants to look strong, look really MAGA, and look like he’s finally standing up for middle America by punishing foreigners who, in his own words, “looted, pillaged, raped and plundered” America. And for reasons that have never been clear to me, he believes tariffs are the superpower that can achieve that. The uncomfortable thing for him is that reality trumps the wishes of economically illiterate populists. He himself realises that his base actually likes buying low-cost manufactured goods, including the imported ones.
POTUS wants tariffs, but he doesn’t want to hurt his base too much. That’s his dilemma. He wants to look like he’s standing up for the middle class against thieving globalist elites (“Finally, Japan is treating America fairly.”). But his method – tariffs on goods – actually harms the middle class. As I pointed out in Chapter 4 my May 2025 eBook, The Great Trade Hack, tariffs raise prices and thus protect jobs of workers in goods-producing sectors, but less than 10% of the middle class work in those sectors. The other 90% simply see higher prices.
This is why he is always backing off bold threats.
- TACO should really stand for “Tariff Announcement Climb-down Obfuscated.”
And he is not doing this due to foreign pressure, or a concern about retaliation. He is worried about what his tariffs will do to his base.
In other words, the President loves tariffs but he is not looking to cut off the nose of the American middle class to spite their face – to paraphrase the old saying.
The Trumpian spin has convinced his base that tariffs are good for America, but they can’t help but notice the higher prices that tariffs lead to in Walmart and Amazon. Here’s a chart from my 11 July 2025 Factful Friday showing how prices spiked when POTUS spiked tariffs on goods from China from March 2025 (Fentanyl tariffs).
And so why does that mean North American imports get a sweetheart treatment?
In particular, US tariffs on imports from Canada and Mexico have to be kept rather low to avoid harming US manufacturing and raising middle class costs of living. The reason is “Factory North America” – a phrase I coined in the 2010s.
It captures the reality that American manufactured goods are not made in America, they are made in North America.
The production process is deeply integrated and spread across all three nations: US, Canada, and Mexico. Over decades, production chains have become so intertwined that US factories could not make much without parts and components from Canada and Mexico.
If the US had hit these with steep tariffs, the result would be industrial disruption, higher costs for American manufacturers, and increased prices for US consumers. Since all that would undermine the very people Trump’s protectionist policies are supposed to help, he has exempted imports from the North American partners as long as they comply with USMCA rules that assure that goods are actually made in North America (as opposed to actually made in, for example, China and relabelled).
The flipflops that revealed this reality
This isn’t pure speculation on my part. It played out in real time due to the lack of staff work behind President Trump’s tariff shout-outs.
In this case, he first announced a 25% tariff on all imports from Mexico and Canada. A couple of days later – after US manufacturing CEOs explained the reality of Factory North America to him – he exempted goods that were made in North America.
The full timeline of this rapid education of the US President on trade matters, can be seen in my recent Factful Friday.
You can also see his strong-but-not-wanting-to-hurt-his-base approach in the fact that he exempted energy imports from tariffs. Putting tariffs on those would have shown up as higher gasoline prices at the pump and higher heating and electricity bills. So he didn’t do energy imports.
Now we come to the part where we have to do a bit of economics. In particular, what is known colloquially as “tricky-tariff stuff,” and more technically as the economics of discriminatory protection.
Basic economics of tariff discrimination: How trade diversion fosters Mexican and Canadian manufacturing
Simplifying to clarify, let's suppose that only Mexico and China exported to the United States. And consider the situation in 2024, when Mexico was in a free trade agreement with the United States, which meant that Mexican exports to the United States paid no tariffs. The competition between Mexican exporters and their Chinese rivals played itself out inside the United States, with China paying some tariff and Mexico paying no tariffs.
Now, starting from this situation, what do you think happens when the US raises a tariff massively on China? The most direct effect, and this is the one that's talked about the most in the newsfeeds, is the way in which the tariff shifts the competition against Chinese exporters and in favor of US-based manufacturers.
The mechanism driving this direct, obvious effect is the usual one. The tariff on China raises prices inside the United States, which allows US-based manufacturers to sell their goods at higher prices without being undercut by China. US-based manufacturers like this, naturally. That's why it's called protection: it protects the domestic producers by allowing them to charge higher prices and sell more than they otherwise would have.
USMCA producers also benefit from the tariff-induced US price rises
But remember that Mexico doesn't pay any tariffs in this thought-experiment. And so anything that raises the price inside the United States is good for Mexican exporters. They see a higher price, and this because the higher tariff on China has hobbled their rival, they sell more as well.
So that's the first-pass analysis. US tariffs that hinder the rivals to Mexican and Canadian exporters inside the United States and that benefits Canada and Mexico.
Now, to get a bit closer to reality, let's redo this thought experiment, but this time we’ll assume that the tariff on China goes up by 50%, and the tariff on Mexico only goes up by 1%. As before, the big tariff on China will lead to a rise in prices inside the United States. Since Chinese exporters will end up absorbing at least some of the tariff, the US domestic price will rise, less than 50%, but still a good deal.
From the perspective of Mexican manufacturers, the litmus test is whether the US domestic price goes up more than the higher tariff they now face (1%). If China is a major supplier to the US (it is as the pie chart showed), and the Chinese tariff hit is big (as we saw in the table above), then the US price is likely to rise more than 1%, so Mexican exporters will still win even though they are paying a higher tariff.
More generally, we can say that if the Mexican tariff hike is sufficiently smaller than the Chinese tariff hike, then Mexico gains from the US tariff.
As my father, Robert Baldwin, used to say, economics can be explained in words, charts, and math. (He was a professor of international economics at UCLA when I was born, and a professor at Harvard when my older sister, who also has a PhD in economics from MIT, was born. Apples, trees, …) Any of the three is good, but if you really want to master the concepts, you should study all three. The math is in the annex. I’ll do the diagrams in another Factful Friday once the crazy pace of tariff changes slows.
Summary and concluding remarks
As part of his war on the trade system, the US President has explicitly rejected the idea that America should charge the same tariff on everyone. It is easy to see how that must have seemed like a good idea to him. If you think only in straight lines and not very far ahead, you will not see that charging a higher tariff on China than on Canada and Mexico would have these unexpected, and surely unintended, consequences. But the economic logic is unassailable. If the US price rises more than the Mexican and Canadian tariffs, then Mexican and Canadian exporters to the US win from Trump’s tariffs. That means the tariffs will be pro-North America, not just pro-USA.
The way POTUS has sprayed on the tariffs will result in a deepening of North American integration, but not necessary in the “America First” way as he was hoping. Given the low tariffs on non-auto non-steel trade inside North America, it’ll be more like “North America First.”
By exempting USMCA-compliant goods, which includes most imports from Mexico and Canada (see my recent Factful Friday for estimates), the very large tariffs that the president is putting on the big non-North American manufacturing powers (China, the EU, Japan, Korea), he has protected manufacturers located anywhere in North America.
How much of the expansion will happen in Canada and Mexico?
The US has a number of chronic issues to tackle before the new tariffs can expand manufacturing employment significantly. These were on full display when the 2022 laws attempted to stimulate US manufacturing. See the chart as the start of this Factful Friday.
The CHIPS Act and Inflation Reduction Act (IRA) spurred major new investments in US manufacturing, especially in semiconductors and green tech. However, these efforts quickly hit a major obstacle: America has a shortage of workers who have the skills and the interest to work in the new facilities.
Despite billions in incentives, companies like TSMC and Intel faced delays due to workforce gaps, sometimes importing technicians or postponing production. Surveys show most manufacturers see hiring and retaining talent as their top challenge. Recognizing this, the Biden administration tied CHIPS funding to local workforce development plans, but progress was slow.
The IRA’s green manufacturing push has met similar friction. Ford’s EV battery joint venture in Kentucky, despite receiving billions in subsidies, faced delays in staffing and had to adjust training timelines. Industry leaders warn that modern manufacturing requires a different skillset than legacy jobs.
Workers need more digital literacy, ability to troubleshoot automated systems, and comply with safety regulations. As the Financial Times wrote in mid-2024, “The average American factory is not short of orders or capital. It’s short of qualified people.”
This paradox of abundant capital and scarce labor underscores a structural weakness in the US industrial revival. “It’s harder to get folks who want to come in and do these jobs that are required to make manufacturing run,” said Tim Gaus, a consultant at Deloitte. Manufacturing has become less desirable in part because today’s workforce is looking for a different work experience. A Deloitte report from April 2025 explores possible fixes while highlighting the problem.
While Canada shares some of the US woes, Mexico has had more success in expanding its ranks of factory-floor workers. With a national per capita income of just $14,000, jobs in manufacturing are considered highly paid and stable.
In my view, such an outcome would be a good way to reduce US reliance on Chinese manufacturing, but I suspect such an outcome would be filed in the Oval Office’s “unintended consequences” folder.
Closing remarks
Wouldn’t it just be the craziest thing in the world if Donald Trump’s tariffs ended up raising the cost-of-living for middle-class Americans while creating factory jobs in Mexico?
Oh, no wait. There is one thing that would be even more crazy. If the President’s other signature policy – deporting Mexican workers en masse – ended up making the US labour constraint tighter while loosening the Mexican labour constraint.
Just imagine the Atlantic Magazine article following a deported Mexican who ends up working in a Mexican factory that expanded because the sky-high US tariffs hobbled the Mexican firm’s Asian competitors.
And that’s it for another Factful Friday!
Annex: Math of discriminatory tariffs – a simple case
And then:
QED! By the way, in this simple setup, welfare effects on US trade partners turn entirely on the border price they receive. If that goes up, the partner gets and terms of trade gain and a trade creation gain.
