The Quarterly Journal of Economics

Natalia is a research economist and Emma is a professor of economics at the University of Virginia.

Their work has been published in top journals, including Science, the Quarterly Journal of Economics and American Economic Journals.

The research has been featured in the BBC, the New York Times, the Wall Street Journal, The Economist, The Atlantic, Good On Paper Podcasts, and NPR's Planet Money, among others.

You know, small, small publications.

The best evidence I've seen on this is a study of 5,000 customer support agents.

This was published last year in the Quarterly Journal of Economics.

When they got an AI assistant, the number of problems they solved per hour went up 15% on average, but the gains weren't spread evenly.

The newest and least experienced people got faster and their work got better.

In layman's terms this means that the additional price burden faced by consumers was larger than the tariff itself at least in the year or two that followed.

The complete pass-through of the 2018 tariffs onto consumers was also evidenced in reports from the American Economic Association and the Quarterly Journal of Economics.

In effect, this meant that tariffs were just a sneakier, more convoluted way to tax consumers without having to actually call it a tax.

In total, the 2018-2019 tariffs imposed around $80 billion worth of indirect taxes on Americans, which amounted to one of the largest tax increases in decades.

But this being said, I think there are very good reasons to believe why this particular form of reallocation from agriculture to manufacturing in particular in the first phase increases overall productivity.

And the two reasons that I think are very plausible to me is, first, there's a very large literature, starting with Arthur Lewis' early Nobel Prize-winning work on what's called the agricultural productivity gap.

This idea, or this fact, really, that workers might just be less productive in agriculture than in other parts of the economy.

And so as you start allocating labor from agriculture into the industries, that has direct productivity effects through gains from reallocation.

If we get this wrong, this could be very bad news indeed.

I think so.

To be honest, I can't really tell you much yet about what such optimal policy mix should look like because we are just so early in the type of research that I think could fill in the gaps here.

It's an exciting time to do this type of research.

There is some interesting research that helps us explain why this happens.

A 2022 study by economists Christine Exley and Judd Kessler published in the Quarterly Journal of Economics examined gender differences in self-promotion.

They found that women were less likely than men to promote their own performance, even when their actual performance was comparable.

Think about that.

Illusion two, the promotion illusion.

In 2019, three economists, Alan Benson, Danielle Lee, and Kelly Hsu, published a paper in the Quarterly Journal of Economics.

They studied roughly 40,000 sales workers across 131 firms, and they tracked two things.

who got promoted into management, and what happened to their teams afterward.

You can read it, and the words that you use can describe the same exact thing very differently.

And to be clear, in the paper, you cite this study from the Quarterly Journal of Economics, so something from last year, which was looking at customer support agents.

And so the folks...

Very specific study, the most experienced folks were able to have some performance gains, but the less experienced and lower skilled workers were able to greatly improve their performance and skills, right? So what organizations should think about is the extent

Yes.

David Autor and his colleagues published a landmark study recently in the Quarterly Journal of Economics.

They analyzed census data and occupational classifications over decades.

Their central finding is astounding.

Whatever taste is, right? And so there's a way to potentially...

Well, one is like AI could eventually have its own taste that we might value, but you can design a review system that focuses on concerns, atomic concerns, without making a judgment about, well, is this a big enough contribution for the QJE? Only Larry Katz can make that decision.

Right.

So it's distinguishing between a technical review and a judgment.