Posts Tagged ‘productivity’


Eduardo Porter is right: the “long, painful slog out of the Great Recession” hasn’t been accompanied by any kind of shared prosperity.

As the chart above reveals, the share of income going to the bottom 90 percent of U.S. households has actually fallen since 2007 (from 50.3 percent to 49.5 percent)—and, in recent years, remains far below what it was (67.4 percent) in 1970.

In other words, the so-called recovery looks a lot like the unequalizing dynamic of the U.S. economy in the years and decades leading up to the Great Recession. Those who work for a living have been getting less and less, while those at the top have managed to capture and keep the growing surplus.


We’re not just talking about the white working-class. Wages “for all groups of workers (not just those without a bachelor’s degree), regardless of race, ethnicity, or gender”, have (since 1979) have lagged the growth in economy-wide productivity.

And that’s just in terms of income. As Porter explains,

by many other metrics, Americans’ well-being remains pretty low. Whether it is life expectancy or infant mortality, incarceration or educational attainment, countless statistics offer a fairly dark picture of the American experience. It is a picture of prosperity that consistently leaves large numbers of Americans behind.

The United States suffers the highest obesity rate among the 35 industrialized countries that make up the Organization for Economic Cooperation and Development. In terms of life expectancy at birth, it ranks 10th from the bottom. America’s infant mortality rate has dropped by half since 1980. Still, today Turkey and Mexico are the only countries in the O.E.C.D. to report a higher share of dead babies. Infant mortality fell faster in almost every other industrialized country.

Mainstream economists, politicians, and pundits may prefer to focus on the first part of Charles Dickens’s famous opening sentence. But that’s only true for the tiny group at the top. For everyone else, it really is—and has been for decades—”the worst of times.”

Cartoon of the day

Posted: 28 October 2016 in Uncategorized
Tags: , , ,



We’ve just learned that the corporate payouts—dividends and stock buybacks—of large U.S. firms are expected to hit another record this year. At the same time, John Fernald writes for the Federal Reserve Bank of San Francisco that the “new normal” for U.S. GDP growth has dropped to between 1½ and 1¾ percent, noticeably slower than the typical postwar pace.

What’s the connection?

Fernald, as is typical of many others who have concluded the United States has entered a period of slow growth, blames the “new normal” on exogenous events like population dynamics and education.

The slowdown stems mainly from demographics and educational attainment. As baby boomers retire, employment growth shrinks. And educational attainment of the workforce has plateaued, reducing its contribution to productivity growth through labor quality. The GDP growth forecast assumes that, apart from these effects, the modest productivity growth is relatively “normal”—in line with its pace for most of the period since 1973.

What Fernald and the others never mention is that American companies’ embrace of dividends and buybacks comes at the expense of business investment, which is an important contributor to worker productivity and long-term economic growth.

In other words, what they overlook is the possibility that the current slowdown—which, “for workers, means slow growth in average wages and living standards”—may be less a product of exogenous events and more the way the U.S. economy is currently organized.

When workers produce but do not appropriate the surplus, they are victims of a social theft. And then, when a larger and larger portion of of the surplus is distributed to shareholders (both outside investors and corporate executives)—that is, the tiny group at the top who share in the booty—workers are, once again, made to pay the cost.


Two findings stand out in a new study from the Economic Policy Institute (pdf) on black-white wage gaps in the United States:

First, since 1979, the gap between all workers’ wages—black and white, women and men—and productivity has increased dramatically. Thus, while productivity increased by over 60 percent, wages for white workers rose by only 22.2 percent and black wages by even less, 13.1 percent.

Second, wages for African American have grown more slowly (or, in the case of men, fallen by a greater amount) than those of their white counterparts. As a result, pay disparities by race and ethnicity have expanded since 1979. For example, white women’s wages increased by 30.2 percent and black women’s wages by only 12.8 percent. And while men’s wages actually declined, they fell by 3.1 percent for white men and even more, by 7.2 percent, for black men. Thus, the overall black-white wage gap increased from 18.1 percent in 1979 to 26.7 percent in 2015.

It is pretty clear from the report that overall wage stagnation (especially for the majority of workers, i.e., those below the 90th percentile), in conjunction with lax enforcement of anti-discrimination laws, led to higher wage disparities by race and ethnicity.

But, and this goes beyond the report, we also need to consider the other side of that relationship—that increased racial and ethnic disparities reinforce the growing gap between productivity and the wages of all workers. Black workers are paid less than their white counterparts (of both genders), and all workers’ wages are as a result less than they otherwise would be.

In the end, then, wealthy individuals and large corporations, who capture the resulting surplus, are the only ones who benefit from racial and ethnic wage disparities.



According to the norms of both neoclassical economic theory and capitalism itself, workers’ wages should increase at roughly the same rate as their productivity.* Clearly, in recent years they have not.

The chart above, which was produced by B. Ravikumar and Lin Shao for the Federal Reserve Bank of St. Louis, shows that labor compensation has grown slowly during the recovery of the U.S. economy from the 2007-09 recession. In fact, real labor compensation per hour in the nonfarm business sector was 0.5 percent lower 20 quarters after the start of the recovery, while labor productivity had increased by 6 percent.

Clearly, the gap between worker compensation and productivity has grown during the current recovery.


But the authors go even further, showing that the gap in the United States between compensation to workers and their productivity has been growing for decades.

labor productivity has been growing at a higher rate than labor compensation for more than 40 years. As Figure 3 shows, labor productivity in 2016:Q1 is 3.8 times as high as that in 1950:Q1; labor compensation, on the other hand, is only 2.7 times as high. In other words, the gap between labor productivity and compensation has been widening for the past four decades. The slower growth in labor compensation relative to labor productivity during the recovery from the two most recent recessions is part of this long-term trend. (reference omitted)

The data in Figure 3 show that the productivity-compensation gap—defined as labor productivity divided by labor compensation—has been increasing on average by approximately 0.9 percent per year since 1970:Q1. Based on this long-term trend, the gap would have been 51 percent higher in 2016:Q1 compared with 1970:Q1; in the data, the gap is actually 47 percent higher.

The fact is, labor compensation has failed to keep up with labor productivity after the Great Recession. But, as it turns out, there’s nothing unique about this period. The gap has been growing for more than four decades in the United States.**

Clearly, the recent and long-term trends of productivity and labor compensation challenge the norms of neoclassical economics and of capitalism itself. But we are also seeing the growth of another gap—between the promises of both neoclassical theory and capitalism and the reality workers have faced for decades now.


*Neoclassical economics—in particular, the marginal productivity theory of distribution—is based on the idea that the factors of production (land, labor, capital, and so on) receive in the form of income what they contribute to production. So, for example, as labor productivity increases, real wages should also rise. Similarly, capitalism is based on the idea of “just deserts.” That idea—that everyone gets what they deserve—is essential to the very idea of fairness or justice in the way the economy is currently organized.

**The authors’ analysis is based on the gap between labor compensation and productivity. If we look at real wages (as in the chart below) instead of compensation (which includes benefits, and therefore the portion of the surplus employers distribute to pension plans, healthcare insurers, and others), the gap is even larger.


According to my calculations from Fed data, since 1979, productivity has grown by 60 percent while real wages have increased by less than 5 percent.


I’ve been writing for some years now about the emergence of new technologies, especially automation and robotics, and their potential contribution to raising already-high levels of inequality even further.

The problem is not, as I have tried to make clear, technology per se but the way it is designed and utilized within existing economic institutions. In other words, the central question is: who will own the robots?

If capital owns the robots, even if their development and use increases labor productivity, the returns mostly go to capital and the workers (those who are left, in addition to those who have been displaced) are the ones who lose out.

But you don’t have to believe me. That’s the conclusion of a recent piece published in Finance & Development, the research journal of the International Monetary Fund.

The authors, Andrew Berg, Edward F. Buffie, and Luis-Felipe Zanna, designed an economic model in which they assume robots are a particular sort of physical capital, one that is a close substitute for human workers.* They also consider three versions of the model: one in which robots are almost perfect substitutes for human labor; another in which robots and human labor are close but not perfect substitutes (i.e., “people bring a spark of creativity or a critical human touch” that cannot, at least for the foreseeable future, be replaced by robots); and a third in which they distinguish between “skilled” and “unskilled” workers.

In all three cases, output per person rises—but so does inequality. As the authors explain for the first version:

If we assume that robots are almost perfect substitutes for human labor, the good news is that output per person rises. The bad news is that inequality worsens, for several reasons. First, robots increase the supply of total effective (workers plus robots) labor, which drives down wages in a market-driven economy. Second, because it is now profitable to invest in robots, there is a shift away from investment in traditional capital, such as buildings and conventional machinery. This further lowers the demand for those who work with that traditional capital.

But this is just the beginning. Both the good and bad news intensify over time. As the stock of robots increases, so does the return on traditional capital (warehouses are more useful with robot shelf stockers). Eventually, therefore, traditional investment picks up too. This in turn keeps robots productive, even as the stock of robots continues to grow. Over time, the two types of capital grow together until they increasingly dominate the entire economy. All this traditional and robot capital, with diminishing help from labor, produces more and more output. And robots are not expected to consume, just produce (though the science fiction literature is ambiguous about this!). So there is more and more output to be shared among actual people.

However, wages fall, not just in relative terms but absolutely, even as output grows.

This may sound odd, or even paradoxical. Some economists talk about the fallacy of technology fearmongers’ failure to realize that markets will clear: demand will rise to meet the higher supply of goods produced by the better technology, and workers will find new jobs. There is no such fallacy here: in our simple model economy, we assume away unemployment and other complications: wages adjust to clear the labor market.

So how can we explain the fall in wages coinciding with the growing output? To put it another way, who buys all the higher output? The owners of capital do. In the short run, higher investment more than counterbalances any temporary decline in consumption. In the long run, the share of capital owners in the growing pie—and their consumption spending—is itself growing. With falling wages and rising capital stocks, (human) labor become a smaller and smaller part of the economy. (In the limiting case of perfect substitutability, the wage share goes to zero.) Thomas Piketty has reminded us that the capital share is a basic determinant of income distribution. Capital is already much more unevenly distributed than income in all countries. The introduction of robots would drive up the capital share indefinitely, so the income distribution would tend to grow ever more uneven.

The only difference in the second case (in which robots and human labor are close but not perfect substitutes) is that wages eventually rise (after, say, 20 years, when the productivity effect outweighs the substitution effect)—but by then it’s too late (as capital continues to have a higher share of income, although not as much as in the first case). And, in the third case, the growing gap between labor and capital (as in the other models) is exacerbated by growing inequality between skilled and unskilled workers.

In all three versions of the model, then, most of the income goes to owners of capital (and, in the third version, to skilled workers who cannot easily be replaced by robots). The rest get low wages and a shrinking share of the economic pie.

And the authors’ conclusion?

We have implicitly assumed so far that income from capital remains highly unequally distributed. But the increase in overall output per person implies that everyone could be better off if income from capital is redistributed. The advantages of a basic income financed by capital taxation become obvious. Of course, globalization and technological innovation have made it, if anything, easier for capital to flee taxation in recent decades. Our analysis thus adds urgency to the question “Who will own the robots?”

The assumption about the unequal distribution of capital income is, in fact, the appropriate one for the existing set of economic institutions. As the authors understand, the only way to change their dystopian prognosis is to fundamentally change the distribution of capital income.

And, if we’re going to be honest, the only way to do that is to eliminate the private ownership of the robots and the rest of capital.


*The model is also based on the presumption that all markets, including the labor market, clear, that is, they assume away unemployment and other such “complications,” which turns out to give those who deny the negative effects of robots and automation their strongest possible case.


From the very beginning, one of the central claims on behalf of capitalism has been that it leads to increases in productivity—and, as a result, an increase in the wealth of nations. The idea is that, the more national wealth increases (the more commodities are produced per person hour worked), the higher living standards of ordinary people will be (i.e., real wages will increase).* We find that story in pretty much every text of mainstream economics, from Adam Smith to Deirdre McCloskey.

That’s why Karl Marx spent so much time (hundreds of pages, in fact) discussing productivity (along with machinery, mechanization, technical change, and so forth) in his critique of political economy. So, John Cassidy gets it wrong.

Marx (not to mention other nineteenth-century critics of capitalism) never denied that there was a connection between increases in productivity and a rise in workers’ wages. That would be silly, both theoretically and empirically. All he ever did was deny there’s an automatic or necessary relationship between them—and, perhaps more important, that increases in real wages didn’t mean workers weren’t being increasingly exploited.



The first point (one even Cassidy, in a way, concedes) is easy to show: for a time (until 1973 or so, in the United States), workers’ real wages increased at roughly the same rate as productivity. Then (from 1973 onward), productivity continued to grow but workers’ wages stagnated.

One key question, for the pre-1973 period, is why productivity and workers’ real wages increased in tandem. Cassidy assumes that wages grew because of the increases in productivity. Nothing could be further from the truth. The explanation for the increases in productivity (having to do with the growth of manufacturing, capitalist competition, the role of U.S. corporations in the world economy, and so on) is separate from the change in wages (based on fast economic growth, unionization, a shortage of labor power, and so on). There’s simply no automatic relationship between increases in productivity and increases in real wages, which has been confirmed by their divergence after 1973.

The second point is, in my view, even more important. It’s possible for workers’ wages to increase (at or even above the rate of growth of productivity) and for capitalist exploitation to also be rising.

Let me explain. In Marxian theory, the rate of exploitation (s/v) is the ratio of surplus-value (s) to the value of labor power (v). The value of labor power is, in turn, equal to the exchange-value per unit use-value (e), or price, of the commodities in the wage bundle (q), or the real wage. So, we have v = e*q and, in terms of rates of change, Δv/v = Δe/e + Δq/q. Mathematically, exploitation can increase (Marx referred to it as relative surplus-value) if the value of labor power is decreasing (Δv/v is negative) even if real wages are going up (Δq/q is positive) as along as the change in the price of wage commodities is negative (Δe/e) and its absolute value is greater than the change in real wages (|Δe/e| > |Δq/q|).

For example, in terms of numbers: if real wages increase by 10 percent (workers are buying more things) but the prices of the items in the wage bundle (food, clothing, shelter) decrease by 20 percent, then the value of labor power (what capitalists have to pay to get access to the commodity labor power) will decrease by 10 percent. Voilà! Higher real wages can be (and, throughout much of the history of U.S. capitalism, have been) accompanied by rising exploitation.

And that’s precisely one of the effects of increasing productivity: it lowers the exchange-value per unit use-value of wage commodities.** Less labor is embodied in each unit of bread, shirts, and housing. The fact that workers are able to purchase more of those commodities (say, at the same rate as productivity is increasing) doesn’t mean exploitation is not also increasing.

That’s even more the case when real wages are stagnant (Δq/q is equal to zero). Then, the decline in the price of wage goods (again,  decreasing by 20 percent) is translated directly into an increase in exploitation (via a 20 percent decrease in v).

But what if rate of growth of domestic productivity begins to decline (as it did after 1973, and even more so in recent years)? Then, the domestic contribution to the decline in the price of wage commodities would fall (say, to 10 percent). But, at the same time, if jobs are offshored and cheaper wage goods are imported from abroad (think Walmart), that also leads to a decline in the price of wage goods (say, by another 10 percent). So, even with declining domestic productivity growth, the combination of domestic production and imported goods can lead to a decline in the price of wage goods (for a total, as before, of 20 percent) that, with constant real wages, decreases the value of labor power (by 20 percent) and increases the rate of exploitation.

So, while Cassidy and many others are worried about a slowdown in productivity growth (linking it to workers’ wages and living standards), workers know that increases in productivity don’t automatically lead to increases in real wages. And even if real wages do rise, it’s still likely they’ll be more exploited than before.

Their employers, not they, will be ones to benefit.


*It is merely presumed that the standard of living of those at the top, the capitalists and other members of the 1 percent, would be just fine.

**Another, separate issue is why productivity itself might increase. The Marxian argument is that, during the course of competing over “super-profits,” that is distributions of the surplus-value among and between capitalist enterprises, capitalists will engage in technical change, which in turn leads to increases in productivity and a decline in the value of the commodities they produce. An interesting question, then, is why productivity in the United States and other advanced countries has slowed down. One reason may be a decline in competitive pressures among enterprises that produce goods and services.