David R. Henderson  

Another Thought About Why U.S. Wages Haven't Risen Much

Leland Yeager, RIP... Three contrarian opinions...

Regular reader Ryan P. Long, in response to my post a few days ago, "Krugman on a Wage Puzzle," writes:

I've quite enjoyed your recent blog posts. The one on Paul Krugman's idea about wage stagnation was particularly interesting to me. Both his post and yours had me thinking about a hypothesis I've had since the big recession hit, and I want to ask your thoughts.

The idea itself is pretty simple, although I haven't read much discussion of it in economics blogs: Globalization and outsourcing are putting severe downward pressure on U.S. wages, even for skilled work, and the "worst of this" (from the American wage-earner's perspective, notwithstanding their benefit as consumers) is yet to come.

One example popped up in the news today: The New York Times reports that some U.S. school districts are recruiting from e.g. the Philippines to fill teaching positions that pay too low to attract U.S. teachers.

[DRH note: if you follow the link in the NYT piece above, you see that average government-school salaries 2 years ago were $47,403. I was surprised at how low that was. Some teachers' claim that this doesn't allow for a middle-class life style is absurd, though. Sure it does, if both spouses work or if the teacher works a month or two over the summer.]

The examples that originally got me thinking about this idea come from the time I spent in software consulting. Most of the development work my employer did was outsourced to India. Workers there would sometimes put in 12-hour days and were paid much less than the U.S. developers, and yet by Indian standards they had good, stable, well-paying jobs. I thought to myself, Why wouldn't all software development be so outsourced? American programmers will have to start competing with Indian, Chinese, Pakistani, etc. programmers who work longer hours for less money. And these are high-tech, high-skilled jobs requiring a competitive education.

I concede that these are two mere anecdotes, but if they are representative of a wider trend in the global workforce, I would expect to see wage stagnation in the developed world, along with wage increases in the developing world.

Well, that's the idea, anyway. What do you think? Do you think there's anything to it? If you have the time and wouldn't mind, I'd love to know what you think about how this relates to slow wage growth.

Good question and nice careful use of data, being careful not to overgeneralize.

On its face, Ryan's hypothesis is plausible. Certainly it is quite plausible that an outward shift in the supply of labor in any one labor market will cause some reduction in wages or, at least, a lower increase relative to trend. Of course, this assumes all other things equal.

I can't say for sure that this hypothesis is wrong. What I do think is that it's probably wrong for the facts at issue.

Here's why: Normally, when a competing source of supply comes along, there's a decline in demand for the original supply that the new supply competes with. The decline in demand would be expected to cause some reduction in wages, all else equal, but also some reduction in employment. But recall what led Krugman, and me, to see a puzzle. It was that the labor market is sizzling hot, with, as Richard Epstein exaggerated to make a point, employers putting out signs asking criminals to apply. In short, we are seeing an increase, not a decrease, in the number of previous residents employed. (I'm assuming here that this increase in people employed represents many previous residents and not just recent immigrants.)

Comments and Sharing

COMMENTS (20 to date)
James Hanley writes:

The "criminals wanted" comment puts me in mind of a radio commercial I heard repeatedly around 2010 or so, that said, "If you have already have a job, we want you to apply with us.

I always thought how painful that must be for unemployed job seekers to hear. But also what it indicated about how the recession shaped employers' hiring abilities.

rtd writes:

Now consider the offsets as related to baby boomers along with technological/automation improvements and Mr. Long may be on to something. Labor force isn't where it once was so the claim that "In short, we are seeing an increase, not a decrease, in the number of previous residents employed." doesn't seem to completely hit the mark.

Scott Sumner writes:

David, I'd point to monetary policy. Growth in NGDP per employed person has been relatively low ever since 2008, and I think that mostly explains the slow growth in hourly wages.

David R Henderson writes:

@Scott Sumner,
David, I'd point to monetary policy. Growth in NGDP per employed person has been relatively low ever since 2008, and I think that mostly explains the slow growth in hourly wages.
Thanks, Scott. Good point. What this suggests is that what Krugman, and I, should have done is look at real increases in hourly wages and see what that looks like. I can't believe I made that elementary an oversight. But I did.

john hare writes:

I can relate to your commercial. People that I have hired from other companies average far higher quality than those that don't currently have a job or come in from a temp service.

Unemployed people and day laborers have a very high percentage of low to non productive people. Makes it real tough on the ones in those places that are able as it is tough to sort wheat from chaff in a small company.

Kevin Erdmann writes:


I would second Scott's comment, and I would go a step further. Much of the current measured core inflation is coming from rent inflation, which is more of a tax and transfer from tenants to landlords because of an endemic housing shortage. Furthermore, it is mostly not a cash-based transaction. Most rent in inflation indexes is imputed rents of owner-occupiers. This rent inflation has little to do with monetary policy or wage growth.

So, not only should we be looking at real wage growth. We should be looking at real wage growth before shelter inflation.

is a post where I revisit an old chart I have of real wages and unemployment.

John Fembup writes:

Rising medical care costs drive up premiums for companies that provide group insurance and drive up direct costs for companies that self-fund their employee medical plans. Everywhere I’ve worked, medical benefits are considered a part of compensation budgets. So as medical costs rise, they absorb funds that could otherwise be allocated to wages. I’m not suggesting this explains flat wages. I am suggesting it’s a plausible, meaningful factor

Don Boudreaux writes:

All: I'm going to offer here a more general point, one that does nothing to undercut (or to support) any of the specific points made by David in his post, or by him or anyone else in the comments. My point is this: it is highly implausible that a general increase in access to goods and services - which is what, in effect, Ryan Long described - will lead to overall real-wage stagnation in rich countries.

A larger effective supply of labor on global markets - a larger supply made possible, say, because of falling costs of conducting global commerce - means more work effort to produce for global markets. Why would anyone suppose that an increase in the global supply of the resource "labor" (or "low-skilled labor") is more likely to lead to a stagnation or a fall in overall living standards in rich countries than is, say, an increase in the global supply of fossil fuels or of some other resource that is widely used in production?

An increase in the global supply of fossil fuels will destroy some jobs in rich countries (or put downward pressure on some people's real incomes), but why should it lead to overall, widespread stagnation in rich countries? There's no reason that I can think of; indeed, we expect quite the opposite.

So why reason differently about labor? One response to my question is that an increase in the supply of labor in poor countries will indeed raise average real incomes in rich countries, but that it will do so by increasing the incomes of owners of capital while doing nothing for, or even decreasing, the incomes of labor. Yet this response is inadequate for at least two reasons.

First, an increase in the supply of human labor from poor countries is not unique in destroying the demand for particular forms of labor in rich countries. Increases in the supply of all resources - and, of course, improvements in technology - can have the same effect, yet many fewer economists worry that increases in supplies of non-human resources and improvements in technology are general sources of secular wage stagnation in rich countries.

Second and more fundamentally: if increases in the global supply of labor increase the returns to capital in rich countries, these higher returns should spark increased capital investment in rich countries - increased capital investments that increase the demand for labor in rich countries. The story here is a standard one that economists tell about competition and economic growth. I believe that the logic of this story remains valid.

If, however, this story now fails to explain reality, the reason for the failure of this story to explain reality must be found in some effect beyond 'increased supply of labor on global markets.' Something must be obstructing new capital investment. Perhaps the culprit is poor monetary policy. Perhaps the culprit is unwise government regulations or poor tax policy. Perhaps the culprit is animal spirits. Perhaps the culprit is [fill-in-the-blank]. Whatever the culprit might turn out to be, the economist must first look for such a culprit before granting any validity to the popular claim that a larger supply of one particular resource on global markets - labor - is a source of a widespread, secular decline in the real wages (real material living standards) of workers in rich countries.

Alan Goldhammer writes:

John Fembub's point about medical price inflation is spot on. I've seen a greater than 10% increase in costs/co-pays over the last three years for three employers that I am familiar with. I don't think most employees understand what a drain this ends up being on prospective salary increases.

It was interesting to watch Berkshire Hathaway's vice chair Charlie Munger speak out on this over the weekend (Berkshire had their annual meeting). He said the current system is riddled with inefficiencies (though he did single out Kaiser Permanente as a good model that should be emulated more widely) and noted that if the Democrats take over both the legislative and executive branch that we will see some form of single payer system implemented.

T Boyle writes:

I know of outstanding teachers, who US schools would love to hire, but can't, because they can't get good visas for them. Outstanding teachers can command above-average wages, and they won't accept temporary positions.

Schools need visas that support permanent employment, like H-1Bs, but the government won't issue those for teachers: the government makes no distinction between outstanding teachers, who could command high wages, and all holders of teaching credentials.

In any case, the IT outsourcing and tech companies vacuum up all the H-1B visas in less than a day each year. Although H-1Bs are intended for employers offering long-term positions here, a high percentage of those H-1Bs are actually being used to bring people into the country for a few months of training, after which they are sent back to their home countries - but the visa has been used and isn't available to attract someone to our country on a permanent basis.

According to the NYT story linked here, schools are now bringing teachers in on J-1 visas. The J-1 is an exchange visa, explicitly intended for temporary workers, and actually prohibits any transition to permanent employment. And, of course, such visas are largely useless for attracting teachers who can command above-average wages in permanent positions elsewhere.

In other words, visas for permanent employees are being abused by tech companies to bring in temporary workers, resulting in temporary visas being abused by long-term employers who actually want to hire permanent workers, but can't get visas for them.

And, we continue to perpetuate the idea that immigrants are bad for wages and working conditions - because we only admit immigrants who are willing to settle for that.

What a mess...

Roger Sweeny writes:


If there is an increase in the supply of fossil fuels, I figure that the price of fossil fuels will go down and people who sell fossil fuels will get less money.

If there is an increase in the supply of labor in various skilled jobs, I figure that the price of that labor will go down and people who sell that labor will get less money.

Fred Foldvary writes:

The average product of labor is rising, but labor is paid for its marginal product, which is not rising, due to globalization as well as labor-replacing technology, and driven by high taxes on labor relative to taxes on land and capital goods. In a competitive industry, firms make only normal accounting profits, and so the surplus, average product minus marginal product, is captured mostly by higher land rent, as locations become more productive. The rent is partly implicit and partly disguised in interest, dividends, capital gains, and retained earnings.

Jon Murphy writes:

@Roger Sweeny

If there is an increase in the supply of fossil fuels, I figure that the price of fossil fuels will go down and people who sell fossil fuels will get less money.

It's not obvious such an outcome will occur. If we assume everything else equal, then a shift out in the supply curve will mean producers get lower prices, yes, but it also increases their volume. What they lose in price they can make up in volume.

We also need to consider why the supply curve is shifting. If it is due to a decrease in costs, then that'd mean rising profits, even at a lower price point.

The same line of thinking goes for labor.

Roger Sweeny writes:

Jon Murphy,

I agree. However, it doesn't seem terribly relevant for worker's compensation.

Your point one says that an increase in labor supply (ceterus parabus, of course) will cause wage rates to go down, but hours may go up more than enough to make for a larger paycheck. That doesn't seem terribly likely to me, and for many workers extra hours are a cost (and not a small one).

Part of my ceterus parabus was that domestic costs aren't changing. The increase in supply of fossil fuels or labor is because of changes abroad, new labor or new supplies of fossil fuels.

Don Boudreaux writes:

Roger Sweeney,

Thanks. But over the past couple of centuries there has been an enormous increase in the supply of labor and, simultaneously, an enormous increase in the real wages of labor. The analysis cannot possibly be so simple as to note that an increased supply of labor means lower wages.

Of course, everything turns on what is and what is not appropriately packed into the ceteris paribus conditions. Modern history tells us pretty clearly, I believe, that when the general supply of labor increases, neither the demand for labor nor the kinds of labor that are contracted for remain fixed. Among other changes, new products come to be produced and the division of labor deepens.

But let me make a more general point. Suppose that, rather than rich countries trading more openly with poor, high-labor-supply countries, rich countries instead enjoy a strange meteorological blessing - namely, goods of the sort that would be supplied by trading with poor countries instead rain down from the heavens upon the denizens of rich countries. Would this blessed meteorological event make the denizens of rich countries poorer? It's difficult to to see how. Such a blessed meteorological event would surely generally enrich the denizens of poor countries, even if it might lower the incomes of some people.

Yet if this blessed meteorological event makes the denizens of rich countries richer rather than poorer, why would the very same happy consequence not arise if the increased supply of goods comes from a blessed opening of trade with poor countries rather than from a blessed meteorological event?

Don Boudreaux writes:

Correction to my last comment: in the third paragraph I meant to write the following:

Would this blessed meteorological event make the denizens of rich countries poorer? It's difficult to to see how. Such a blessed meteorological event would surely generally further enrich the denizens of rich countries, even if it might lower the incomes of some people.
Robert Simmons writes:

$47,403 x 2 is $94,806. Outside of a few places like Manhattan that will provide a very nice lifestyle for a household. Even in Manhattan you'd do ok.

Floccina writes:

One thing that I alway think of when this comes up, is that is does not mean that workers wages are not rising, because as new, probably less experienced and capable, works get jobs they will start at lower wages bring down the average. So the picture is a little better that the average indicates.

Timo writes:

The question seems to be: If the economy is expanding so much, aren’t people spending more?
My experience was this: I was technically skilled but ended up unemployed for years (off and on) during terrible Obama economy. Now employed in a growing economy, I am starting out as mostly “inexperienced” due to skill loss (changing technology) and now (at 50 years old) only command a slightly higher salary than when I started my career. The second factor, in my experience, is the extensive debt that I took on during the Obama years is being paid off now and doesn’t allow me to consume as I normally would.
A terrible economy during ones prime earning years often does extensive damage which is often not reversible or destroys the productive value (hence spending power) of a whole generation of individuals. Now I am paying off debt and scrambling for retirement money and can’t spend.

Anisha Meka writes:

I think that one of the most important reasons for stagnant wage growth is: rising costs of healthcare in the US. A large chunk of the paycheck goes toward paying for healthcare. Another reason is the issue of student debt. When one cannot get a job at the level that there degree permits, they will resort to lower level jobs in order to fulfill loan payment obligations. I strongly feel that these domestic issues, rather than the issue of outsourcing is the cause for stagnant wage growth.


Return to top