By Matt Nesvisky, NBER
Innovative thinkers are innovating later than they used to. While conventional wisdom holds that creative thinkers do their best work when they are young, a study by NBER researcher Benjamin Jones shows that over the past century the average age at which individuals produce notable inventions and ideas has increased steadily.
In Age and Great Invention , Jones considers data on Nobel Prize winners in Physics, Chemistry, Medicine, and Economics over the past 100 years, and on outstanding technological innovations over the same period. For comparative purposes, Jones also considers the ages of track and field record-setters and ball players who have received Most Valuable Player awards.
The data on the innovators reveal three initial characteristics. First, there is large variation in age: 42 percent of innovations came about when their creators were in their 30s, while 40 percent occurred when the inventors were in their 40s, and 14 percent appeared when the inventors were over 50. Second, there were no great achievements produced by innovators before the age of 19, and only 7 percent were produced by innovators at or before the age of 26 (Einstein’s age when he performed his prize winning work). Third, the age distributions for the Nobel Prize winners and the technologists are nearly identical.
The most striking finding, however, is that the age distribution shifts over time, with the mean age of great achievement rising by five or six years per century. This parallels another study showing a similar upward trend in the age of persons receiving their first patents (NBER Working Paper No. 11360). Jones finds the trends among great innovators are significant and robust, even after controlling for nationality and field of study. Indeed, these controls strengthen the age trend, causing it to rise to about eight years over the course of the twentieth century. This suggests a compositional shift in great innovation towards fields and countries that favor the young.
One possible explanation for this age shift is a decline in the productivity of younger innovators in favor of older innovators. It may well be that the younger innovators are devoting themselves to an increasing amount of education and training. Or, it may be that the productivity of older innovators is increasing in relative terms simply because innovators are living longer. If we accept that raw ability declines over the life cycle while experience increases, then the shift in the distribution may indicate the rising importance of experience over ability. Alternatively, improved health care may spell increased ability and effort at later ages.
However, Jones urges caution in interpreting these distributional shifts. The shifts, he suggests, may reflect a simple demographic effect. If the population of innovators is getting older, then the older innovators will be more likely to produce substantial innovations even if the relationship between age and innovative potential is fixed. That is to say, the greater the ratio of 50-year-old innovators to 25-year-old innovators, the more likely the Nobel Prize-winning work or the groundbreaking technological development is to come from one of the 50-year-olds. Such demographic effects may be important, because life expectancy and the average age of the population have risen substantially throughout the twentieth century.
By subjecting these various hypotheses to econometric analysis, Jones concludes that the upward trend for productive innovators does not merely reflect the aging population, but in fact is a result of a substantial decline in the innovative output of younger individuals. Meanwhile, there appears to be no relative increase in the innovation potential of those beyond middle age. Other things equal, the less time innovators spend successfully innovating, the smaller will be their lifetime output. In fact, estimates point to a 30 percent decline in life-cycle innovation potential over the twentieth century.
Jones notes that, unlike athletes, who do not require increased training demands over time, innovators appear to spend increasingly significant portions of their early years in education – a kind of human capital acquisition that might well explain the age trends in his study. Because the rules and requirements of their fields of endeavor remain fixed, athletes are not obliged to increase their human capital; accordingly, the data show no distributional shift in the ages of top athletes over the years. But thinkers must increasingly invest in acquiring intellectual capital, and the accumulation of knowledge – the rising distance to the frontier – can explain increased educational attainment.
Jones notes that economists have not focused much on the human capital investments of innovators. Because innovators customarily devote their youngest and perhaps brightest years to acquiring their education, understanding the tradeoffs at the beginning of the life-cycle may be of primary importance for understanding the ultimate output of these individuals – and for understanding why great innovation is steadily declining among younger thinkers.
De “THE EFFECTS OF WAL-MART ON LOCAL LABOR MARKETS”. David Neumark, Junfu Zhang, Stephen Ciccarella. November 2005. NBER. Working Paper 11782
Motivated in large part by local policy debates over Wal-Mart store openings, and the large size
of Wal-Mart relative to the U.S. labor market, we estimate the effects of Wal-Mart stores on employment and earnings. Critics have charged that Wal-Mart’s entry into local labor markets reduces wages and employment, and the company (and others) have countered that these claims are false.
Our analysis focuses on estimating the effects of Wal-Mart on employment and earnings,
emphasizing the importance of accounting for the endogeneity of the location and timing of Wal-Mart openings that—in our view, and as borne out by the data—is most likely to bias the evidence against finding adverse effects of Wal-Mart stores. Our strategy for addressing the endogeneity problem is based on a natural instrumental variable that arises because of the geographic and time pattern of the opening of Wal-Mart stores, which slowly spread out in a wave-like fashion from the first stores in Arkansas.
On balance, the evidence is more consistent with the claims of Wal-Mart’s critics, although
questions remain. In the retail sector, the representative Wal-Mart presence (about eight years) reduces employment by two to four percent. There is some evidence that payrolls per worker also decline, by about 3.5 percent, but this conclusion is less robust. Either way, though, retail earnings fall. Looking at total employment, some of the evidence points to employment increases, although the regional evidence for the South and perhaps the Midwest points to employment declines. At the same time, there is stronger evidence that total payrolls per worker and per person decline, by about two and five percent, respectively, implying that residents of a local labor market do indeed earn less following the opening of Wal-Mart
stores. Finally, we find clear evidence of adverse effects of Wal-Mart stores on retail employment, total employment, and total payrolls per person in the South, where Wal-Mart stores are most numerous on a total and per capita basis, and where they have been open the longest.
The earnings declines associated with Wal-Mart do not necessarily imply that Wal-Mart stores worsen the economic fortunes of residents of the markets that these stores enter. Wal-Mart entry may also result in lower prices that increase purchasing power, and if prices are lowered not just at Wal-Mart, but elsewhere as well, the gains to consumers may be widespread. Although this paper provides what we believe to be the best evidence to date on the effects of Wal-Mart on local labor markets, we do not address questions of changes in real earnings or purchasing power, and how these changes might vary for lower- and higher-income families. In addition, while not endorsing the findings of the studies cited in the Introduction that estimate the taxpayer burden imposed by Wal-Mart (through, for example, increased
use of Medicaid and Food Stamps), even if Wal-Mart lowers earnings as well as prices, the lower
earnings will likely, in fact, increase this burden. Finally, as we noted earlier, we cannot with the data used in this paper pinpoint the mechanism leading to earnings declines, including factors such as changes in part-time work and shifts to less-skilled workers. There are, in short, numerous remaining questions of considerable interest regarding the effects of Wal-Mart on labor markets, consumption, and social program participation and expense. The identification strategy developed in this paper may prove helpful in estimating the effects of Wal-Mart stores on these other outcomes as well.
Interesante working paper publicado en NBER
How’s the Job: Well Being and Social Capital in the Workplace by John Heliwell and Haifang Huang.
Comment by Brad Delong
Most academic economics rely on concepts laid down at the beginning of the twentieth century by the British economist Alfred Marshall, who said that “nature does not make leaps.” Yet we economists find ourselves increasingly disturbed by the apparent inadequacy of the neo-Marshallian toolkit that we have built to explain our world.
The central bias of this toolkit is that we should trust the market to solve the problems we set it, and that we should not expect small (or even large) changes to have huge effects. A technological leap that raises the wages of the skilled and educated will induce others to become skilled and educated, restoring balance so that inequality does not grow too much.
So a country where labor productivity is low will become an attractive location for foreign direct investment, and the resulting increase in the capital-labor ratio will raise productivity. Wherever one looks, using Marshall’s toolkit, one sees economic equilibrium pulling things back to normal, compensating for and attenuating the effects of shocks and disturbances.
Marshall’s economics has had a marvelous run, and has helped economists make sense of the world. Yet there is a sense that progress and understanding will require something new – an economics of virtuous circles, thresholds, and butterfly effects, in which small changes have very large effects.
Perhaps this has always been so. By the standards of centuries ago, we live in a world of unbelievable wealth. Within two generations human literacy will be nearly universal.
Yet three centuries ago there was also technological progress, from the mechanical clock and the watermill to the cannon and the caravel, and on to strains of rice that can be cropped three times a year in Guangzhou and the breeding of merino sheep that can flourish in the hills of Spain. But these innovations served only to increase the human population, not raise median standards of living.
Today, if we divided up equally what we produce worldwide, would it give us a standard of living ten times that of our pre-industrial ancestors? Twenty times? A hundred times? Does the question even have meaning?
David Landes likes to tell the story of Nathan Meyer Rothschild, the richest man in the world in the first half of the nineteenth century, dead in his fifties of an infected abscess. If you gave him the choice of the life he led as the finance-prince of Europe or a life today low-down in the income distribution but with thirty extra years to see his great-grandchildren, which would he choose?
No doubt, we live today in an extraordinarily unequal world. There are families today near Xian, in what was the heartland of the Tang Dynasty Empire, with two-acre dry wheat farms and a single goat. There are other families throughout the world that could buy that wheat farm with one day’s wages.
Marshall’s economics – the equilibrium economics of comparative statics, of shifts in supply and demand curves, and of accommodating responses – is of almost no help in accounting for this. Why, worldwide, did median standards of living stagnate for so long? Why has the rate of growth undergone an acceleration that is extraordinarily rapid over so short a period? Where is the economics of invention, innovation, adaptation, and diffusion? Not in Marshall. And why is today’s world so unequal that it is hard to find any measures of global distribution that do not show divergence at least up until the 1980’s?
It has been generations since economists Robert Solow and Moses Abramovitz pointed out that Marshall’s toolkit is a poor aid for understanding modern economic growth. The real sources of growth are not to be found in supplies and demands and the allocation of scarce resources to alternative uses, but in technological and organizational change – about which economists have too little to say.
Economic historians like Ken Pomeranz rightly point out that before the Industrial Revolution, differences in median standards of living across the high civilizations of Eurasia were relatively small. A peasant in the Yangtze Valley in the late seventeenth century had a different style of life than his or her contemporary peasant in the Thames Valley, but not one that was clearly better or worse.
Two centuries later that was no longer the case: by the end of the nineteenth century, median living standards in Britain and other countries to which the Industrial Revolution had spread were, for the first time in recorded history, light-years above any neo-Malthusian benchmark of subsistence. The early industrial-era economic accomplishments occurred despite the loss of a substantial proportion of national income to support a corrupt, decadent, and profligate aristocracy. They occurred despite a tripling of the population, which put extraordinary Malthusian pressure on the economy underlying natural resource base, and despite the mobilization of an unprecedented proportion of national income for nearly a century of intensive war against France, a power with three times Britain’s population.
How, exactly, did these accomplishments occur? What were the small differences that turned out to matter so much?
Economists are now awakening to the realization that the most interesting questions they face were always beyond the reach of Marshall’s toolkit. Clearly, economics – if it is to succeed and progress – must be very different in a generation from what it is today.