Category Archives: Labor Economics

Two Lines Diverge in a Labor Graph

I came across this graph in a paper* the other day, and I found it rather disturbing (yeah, economists get disturbed over odd things). It shows labor productivity and real hourly compensation from 1947 to 2010.


The disturbing part is that the two lines are different. Not the same. Distinct.

In fact, starting around 1970, the real hourly compensation line increases at a slower rate than productivity, until by 2010 the two differ considerably.

This might not seem like a big deal. Why get disturbed by two lines?

The trouble is that standard labor economic theory says that these two lines should be the same. Workers are supposed to be paid, on average, their marginal product of labor. It says so in all the textbooks. I’ve taught it in economics courses.

If productivity increases, then wages should increase by the same amount. That’s how standards of living improve. Productivity increases, wages increase to match, and people’s lives improve. That how it’s supposed to work.

This graph says this isn’t happening, and not just for one or two years. Productivity and wages have been diverging for decades in the U.S. Workers have been growing more and more productive, yet haven’t been rewarded for that.

So who’s benefited from this increased productivity?

Have consumers benefited through lower prices? Possibly some. But productivity increases only get passed on to consumers if competition forces firms to lower prices. That doesn’t seem to be happening.

Instead, corporate profits have been reaching record levels. On the one hand, that’s a good thing. Profits are good, and increased profits mean increased stock prices. Anyone who own stock benefits from that.

The problem is that the stock owners are benefiting at the expense of the workers. After all, it’s the workers who are causing that increased productivity (possibly through more or improved capital, but still). By all rights, they should be compensated for it. Why aren’t they?

That’s a good question, and it’s hard to come to any firm conclusions based on one report. However, the only explanation that comes to mind is that workers simply haven’t had the leverage and clout to demand that management pay them for their increased productivity. Union membership has declined pretty steadily during the same period that productivity and wages have diverged.

If workers organize and demand increased compensation, firms simply shut the U.S. factories down and move overseas, where wages are lower. Productivity is, too, but there’s often still a net benefit to the firms.

That’s great for firms and shareholders. It’s just not that great – nor fair – to the workers.

Decreasing labor clout. It seems to have made all the difference.

*”The compensation-productivity gap: a visual essay”, Fleck, Glaser, and Sprague, Monthly Labor Review, January 2011

More Workers Quitting Is Good News for Labor Markets

The labor market continues to look (slowly) better. The Labor Department announced today that new claims for unemployment benefits fell to 339,000 last week, the lowest level in more than four and a half years.

Labor data on job turnover also is looking positive.

In addition to reporting the unemployment rate, the inflation rate, and dozens of other important economic numbers, the Bureau of Labor Statistics also has a program called the Job Openings and Labor Turnover Survey, or JOLTS. The JOLTS program collects data once a month from stores, offices, factories, and other employers about their job openings, hiring, quits, and layoffs. This information can provide a lot of insight into what’s going on in the labor markets.

According to this survey, hiring and job openings have both been trending upward since the middle of 2009. Also, layoffs and discharges have been decreasing since then.

Even more importantly, the number of quits has been trending upward. This might sound like bad news, but it’s actually quite the opposite.

When more people are handing in letters of resignation, it either means that they’ve found a new job already, or else they feel confident that they’ll be able to do so quickly. Either way, it’s a vote of confidence in the economy. When things are terrible, workers don’t want to risk changing jobs. When things improve, they’re more likely to take chance.

More People Aged 65 and Over Working Than Ever

More older Americans are working than ever before. Of the people aged 65 and over, 17.5 percent are working, according to Bureau of Labor Statistics figures. There are about 42.3 million people in the U.S. aged 65 and over who are not in institutions. Of these, about 7.4 million are employed.

Up until 1985, the percentage of people 65 and over who were working had been trending downward. Before 1952, more than 24 percent of this age group was working. By 1982, this has decreased to less than 12 percent.

But since 1985, the percentage has been increasin. Several factors are driving this change. First, people are living longer, and are physically able to work longer on average. Second, many people are finding that they have not saved enough money for retirement. This has been exacerbated by two brutal stock market declines within 10 years, which decreased what retirement savings people had. A tough economy also has made it harder for older people to rely on children and family for support.

A Tale of Three Indicators: How Bad is the Labor Market Really?

It is the best of times. It is the worst of times. It it the worst of times but getting better.

It all depends on how you measure it.

There’s a lot of interest these days in alternative economic measures, particularly with regard to employment. Part of this is driven by real problems with unemployment rates. Even more is driven by a people wanting indicators to tell a different story than they do. And some is driven by a desire for measures that provide different policy incentives. The news yesterday that the unemployment rate fell from 8.1 percent to 7.8 has only exacerbated the situation.


Probably the three most closely watched macroeconomic indicators in the U.S. are the change in real GDP, the consumer price index (CPI), and the unemployment rate.

When we say “unemployment rate”, we really mean the “U-3 unemployment rate”; there are (at least) five others we could consider. They range from the narrowest U-1 to the broadest U-6 rates.

Basically the U-3 unemployment rate is the percentage of the workforce that is actively looking for work but is not employed.It’s true that the U-3 rate doesn’t capture a lot of the economic pain out there. But it’s the standard measure that policy-makers have focused on for decades.

Politicians who want things to sound worse will throw around phrases like “15 percent underemployment”. This is valid but misleading. The U-6 unemployment rate – the broadest measure that also includes most discouraged workers and others – is at 14.7 percent these days. But the U-6 is not the standard definition. If you’re going to use it as your standard in bad times, you also need to use it in good, or else you’re talking apples and crabapples.

That said, all of the unemployment rates move very much in tandem. In fact, since 1994, the correlation between the U-3 and the U-6 rates is 0.996. They’re practically the same measure. Here’s a graph of the six unemployment measures over the past four years, courtesy of the Washington Post.

But even the broad U-6 unemployment rate doesn’t capture people who have permanently left the labor force because they feel things are hopeless. This is one reason some analysts have proposed using the Labor Force Participation Rate (LFPR) as a main indicator of labor economic health.

I addressed the problems with the LFPR in an earlier post, and there are many. Even though the LFPR does capture long-term discouraged workers, it introduces many other difficulties. The LFPR is mostly driven by long-term demographic changes. It doesn’t reflect the business cycle at all, as is clear from the graph below.

The LFPR thus is pretty much useless as a “how are we doing?” sort of measure.

If you’re really desperate for an alternative labor economics indicator that captures both the long-term unemployed as well as the business cycle, I propose yet a third measure: the Employment-Population Ratio.

The Employment-Population Ratio (EPR) is the percentage of the total (noninstitutional) population aged 16 years and over that is employed. It has the advantage of capturing both shorter-term cyclical unemployment and long-term unemployment of discouraged workers.

The EPR has problems, too, of course. There’s no perfect economic indicator (I addressed problems with GDP in an earlier post as well). Specifically, the EPR is a victim of many of the same demographic realities that the LFPR is. When you have a lot of people reaching retirement age, the EPR will fall. When you have a lot of women entering the work force, the EPR will rise.

But the EPR does capture the business cycle. And it has the added benefit – for those who want the economy to look worse – of making the economy look a little worse.

To be fair, the EPR is being pulled in different directions by different factors. Baby boomers are hitting retirement age and are leaving the labor force in increasing number for legitimate reasons. At the same time, the economy is (slowly) recovering. The net effect is that the EPR has been basically flat since late 2010.

Personally I still prefer the U-3 unemployment rate as my main measure of how the economy’s doing, though I always keep in mind that it has its flaws like any other indictor.

If I were looking for a secondary labor measure to provide supplemental information, I’d choose the Employment-Population Ratio.

But all things considered, just measuring the economy really can scare the dickens out of you.

Reduced Earnings for Men in America

The Brookings Institution does excellent research. They recently produced an interesting report on men’s earnings. In case anyone doesn’t have time to read the full report, I thought I’d post the graphs here. Even without narrative, they speak volumes.

This slideshow requires JavaScript.

Labor Day Thoughts: Vacation Days in the U.S. vs. Other Countries

On this Labor Day, some thoughts about labor.

It’s no secret that American employees are some of the hardest-working and dedicated in the world. Alternatively, it’s no secret that American employees are some of the most exploited in the world. It all depends on your perspective.

Vacation and holidays are a case in point. Americans may view European workers as spoiled and pampered, but how many Americans would turn down 30 days of paid vacations a year?

Good data on vacation days around the world isn’t the easiest to come by, but according to the World Tourism Organization (via, here is a comparison of average paid vacation days for selected countries: has similar numbers for some, but not all of these countries. Expedia also adds information about days earned vs. days taken.

Based on this (admittedly limited) data, U.S. workers do appear to have very little vacation time compared to the rest of the world.

This could be evidence of a superior American work ethic, of differences in leisure-consumption preferences, or of employers taking advantage of American workers. But that will have to wait for a future post.

I’m taking the rest of the day off.