Back in the day, before Paul Krugman became a full-time pundit, he was an academic economist who published actual economics. He became famous by defending free trade and globalization. However, pundit Krugman has recently recanted some of his earlier views (not, though, by publishing papers in AER showing he was mistaken), and instead says that free trade has hurt lower-income American workers. Maybe that's why Trump won in 2016!
I think Krugman is demonstrable wrong both about voters and the supposed negative effects of trade. But let's put that aside and assume that Krugman is right.
On Krugman's new view, a view shared by Trump, Bernie Sanders and many soi-dissant (but not actually) progressives, globalization greatly increased the incomes of higher income people in the US or Europe, but lower-income people in the US and Europe saw their incomes stagnant. Instead, the gains from trade went largely to people in developing countries, who as a result have gotten much, much richer. So, on the Krugman-Sanders-Trump picture, globalization helped the very rich and the poor, but not the American/European working class.
Remember that the "American poor" are actually quite rich. An American living at the US poverty line is well within the top 20% of world income earners in PPP-adjusted income. Indeed, as Krugman himself pointed out back in 1996, Americans near the poverty line in 1996 were on par with middle-income Americans from the 1950s.
Consider what the typical egalitarian, prioritarian, or sufficientarian would say if the supposed distributional effects of globalization happened inside a single country. Imagine that everyone lives in one giant country with massive inequality and with many people living in extreme poverty. Suppose that a new economic development takes place--call it globalization or neoliberalism or whatever word you want--which then causes the top 12% of income earners to get much richer, the next 8% of income earners to stagnate, and the bottom 80% of income earners to get much, much richer. Suppose that as a result, the number of people living in extreme poverty in this country drops from 66% to about 9%. Suppose that as a result, income equality is reduced.
If that happened, the typical egalitarian would celebrate and call this a massive victory for social justice. They might have preferred that if one group had to stagnate, it was the very richest decile rather than the second richest, but oh well. The poor get priority. The important thing is that the bottom 80% are now much better off and inequality has fallen.
However, oddly, when these exact same growth outcomes occur, but people live in different countries, suddenly soi-dissant progressives get upset. What happened over the past 50 years, on the Krugman-Trump-Sanders picture, is that the top 12% of world income earners got much richer, the next 8% of world income earners stagnated, and the bottom 80% got much, much richer. However, the twist is that the top 20% live in the US, Canada, Europe, Japan, Australia, South Korea, Taiwan, and New Zealand, while the Botton 80% live somewhere else. Otherwise, it's the same distributional outcome.
Yet for some reason, when we specify the same results occur across countries rather than within a single country, people get upset. Weird.
Left-wing nationalism a la Sanders is still a right-wing view. It's a view that prioritizes the welfare of people in the same nation-state over real equality, real social justice, or alleviating severe poverty.
UPDATE: I modified this slightly, because I realized that I don't know what Krugman thinks about trade all-things-considered.