The High Cost of Low Skills: Why Unskilled Workers are Punished in the United States

>> Eric Hanushek: I wanna talk a little bit about the pandemic, which is going to come home to a number of you who have been in school during the pandemic and have suffered from the disruptions that the pandemic caused. And what I wanna do is put this in economic terms.

I have been part of the governing board for our national assessment, which measures student learning over time. And they have come out over the last year with a couple studies that say, well, the pandemic was really important. The pandemic caused a nine point drop in the NAEP scores and everybody said something.

Nine points? Is that important? In fact, there were two of our leading national newspapers that had articles that said, no problem, nine points, there was only nine points, why should we be concerned about that? So that's the first part of the talk today, is to say nine points is a big number.

You don't realize how that single digit number is big. And we'll talk a little bit about that in economic terms. Then we'll move in a little bit to what's being done in schools and how schools are dealing with the pandemic and the learning losses from the pandemic. And then we'll try to open it up for some discussion and answer any questions you have on schools and policies today.

So I'm gonna start with a small interlude of economics and economic research, because that motivates the fact that what you learn is really important. And everybody in this room obviously understands that learning is important. That's why you've been in school for so long. But what I'm gonna do is try to put this into actual monetary terms for you.

And that allows us to then assess what the losses during the pandemic mean for individuals and for the nation. And then we'll talk a little bit about some policy options at the end. So the place to start is that skills pay off in the labor market. We know that, you know that intuitively.

There have been extensive analyses. For those of you who have done an economics major, you've seen that there's a lot of work that's been done on relating skills of people, particularly the amount of schooling that people get to earnings in the future, okay? What I'm gonna do is try to refine that a little bit.

And I'm gonna take some data that's across country in different countries that's rather unique, because what the PIAAC data, which is Program and International Assessment of Adult Competencies, it always takes a little time to figure out what these acronyms mean. It's done by the OECD, the club of rich nations, and they went around to 30 different countries and got a random sample of adults, working age adults, found out about their labor market histories and their personal backgrounds, and actually gave them a series of math and reading tests to these adults.

So what that does is allow us to relate measured skills, math scores or reading scores to earnings of people. And there's a standard economics way of doing this that dates back to a professor named Jacob Mintzer, who said, well, we can just look at the log of earnings as a function of years of schooling and experience and gender, and we get a pretty good picture.

Well, we don't get a perfect picture because years of schooling is a pretty crude measure of what people know, as you know from just your own classes, but if you look around. But if you add in measures of the performance, actual measured skills, like math skills of people, then you get a different picture.

You get a picture that looks like this. Now, what this picture is, is the 30 countries or so, I think it's 30, maybe it's 32 that participated in this PIAAC sample. The dark lines don't mean anything other than there were two different surveys at two different times that they gave the surveys.

But the vertical score here is what is the return to having more skills through a standard deviation of skills? And the thing that you ought to know is that the US is near the top of that chart of among the countries. Nobody ever objects to the fact that Greece is at the bottom, that's understandable.

And you go across countries and then you get to the United States, and it's near the top, which says that the US rewards skills more than almost all other nations or all other developed nations. So if you play that backwards, it says the US punishes the lack of skills more than almost every nation in the world.

And that's the story here that is relevant to the pandemic, is that the US punishes people that don't get the same skills, okay? We're gonna come back to that in terms of the picture of pandemic losses in the nine points in a minute. But before I do that, there's another element to skills, and that is that the nation as a whole really does better if it has a more skilled population.

For 30 years now, economists have been trying to figure out why do some nations grow faster than others, and have done a lot of data analysis, empirical work. And almost always they get a picture that relates the years of schooling of somebody in a country, or the average years of schooling in a country to the growth rates in the country.

It turns out that this isn't a very good way to do it for obvious reasons, right? A year of schooling in Peru is not the same as a year of schooling in Japan, okay? But there's been a lot of that, and we'll come back, recently it's been possible to incorporate information about the measured skills, the math and reading and science of people in different nations.

So you get a picture like this. This is the historic picture, and it sort of says years of schooling, conditional years of schooling, I'll come back to conditional in a second. And growth rates, long term growth rates. These are average increases in per capita GDP over the period 1960 to 2000.

So over a long period of time, what happens? And what you see is that countries are sort of arrayed in this vertical way. And some of them, starting out with Venezuela on the bottom and getting up to Taiwan and Korea and singapore at the top, it seems sensible.

The only other thing that I wanna point out is that there are some puzzles in this picture, you probably haven't figured out the puzzles yet. I've got all the nations arrayed up here. One of the puzzles is all the East Asian countries are above the line. They're doing better than you'd expect.

This is the East Asian growth miracle that we all know about. And then there's another group of countries at the bottom, which is what we call the Latin American growth puzzle because they're all below the line. What's going on? Well, I've already told you the answer, right? You learn a lot more per year of schooling in East Asia, on average, over this period than you do in Latin America over this period.

So if we change this, I didn't tell you what conditional meant, by the way. The only thing other behind the scenes here is that we measure the initial income level of countries. So the 1960 GDP per capita. And why would we do that? Why would we do that for this?

If you start with high 1960 GDP per capita, what do you expect that to do to growth rates? It would slow the growth rates because you have to invent new things in order to grow. If you start behind, what do you do? You just copy what's there, right?

But once we take into account, this picture takes that into account that some countries can just copy what other countries do. But this isn't the right picture at all. The right picture is this picture. This is the same picture, but it's now conditional test scores, measures of performance against long term growth rates.

And test scores here is very simple, you probably know, some of you have heard of the PISA Test. The PISA Test is by the OECD again, which takes a set of math problems, translates them into local languages and marches them around the world. And then you can see pretty easily which countries have better math skills than other countries.

And so if you use that information about what people actually know as opposed to how long they've sat in a class, you get this picture, which has a couple features that are important. One is it explains the differences in growth rates a lot better than the prior picture.

The prior picture, in technical terms, had an r squared explained variance of one-quarter. It could explain a quarter of the differences, the variation across countries in growth rates. This picture can explain three-quarters of the variation in growth rates across countries, again with only taking into account the initial income levels, but the long term growth rates.

That's the first thing you should know that this explains most of the variation growth rates. The second thing you should know that isn't apparent here, but will be in a second, is that that's a very steep line, that skills really have a strong impact on growth rates. And I'll come back to that, because that's important for talking about the cost of the pandemic, learning losses where skills actually went down as opposed to went up over time, okay?

The other thing that's kind of interesting, that is my interpretation of what all of you are getting out of the schooling you've had or are having. One of the big things that you get out of schooling is not learning a few equations in some engineering class. It's being adaptable to change, being able to adjust the new things.

And you can actually see that internationally. Same set of, oops, a little bit out of sequencer. What I wanna go back to, the fact that earlier I showed you years of schooling against growth rates in this positive line. If I just add in measures of what they know by these math scores, you get a perfectly flat line that years of schooling don't add anything on average to growth rates, once you account for what people actually learned when they were in school, okay?

Then the other thing is that the adaptability. If I have average growth rates on the horizontal axis, and the returns to schooling on the vertical axis, you see that in general, the returns to schooling are higher in places where there's more growth. And what is growth? Growth is changing the way we do things, improvements in productivity, adapting to different situations.

And so one of the background pictures here is that the US rewards skills a lot because there's a lot of change in our economy, and there's a lot of growth that people are adjusting to.