Ed Dolan's Econ Blog

Globalization and Inequality: Is there a Superstar Effect, and if so, What does it Mean?

If two things happen at the same time, it is natural to think that one may cause the other. A case in point is the much-discussed linkage between the growth of global trade and the increased inequality of income distribution. The following charts show one way of representing these two trends for the United States. The same general pattern appears for many other countries, and seems to hold up for many different measures of income inequality and globalization. (For example, the charts in this VoxEU post by Paolo Epifeni and Gino Gancia show much the same thing in an entirely different way.)

The odd thing is that conventional theories of international trade do not account well for the relationship. Long-established models, for example, the Hecksher-Ohlin theory, do allow for the possibility that expansion of trade can change the relative earnings of different factors of production, including both labor vs. capital and skilled vs. unskilled labor. However, when economists run the data, they find that standard versions of the models show trade accounting for only a small part of the recent growth of inequality. (For example, see this 2010 paper from the Peterson Institute for International Economics.)

Economists used to take comfort from such results. Consider Paul Krugman for example, a long-time friend of free trade and foe of inequality. In an often-cited 1995 paper, Krugman concluded that expansion of trade accounted for no more than 10 percent of the growing gap between incomes of skilled and unskilled workers. Today, he and others are not so sure. In a 2007 VoxEU post, Krugman wrote, “It’s no longer safe to assert, as we could a dozen years ago, that the effects of trade on income distribution in wealthy countries are fairly minor. There’s now a good case that they are quite big, and getting bigger.”

Despite economists’ increasing conviction that there must be some linkage between globalization and inequality, it has proved surprisingly difficult to find a plausible causal mechanism that fits the data. Many think that part of the answer to the globalization-inequality puzzle must lie in the way the growth of information technology simultaneously reduces the cost of trade and raises the need for skilled workers. However, the impact of skills on earnings does not always show up in the ways one might expect. For example, in the United States, the income gap between workers with a bachelor’s degree and those with only a high school diploma widened during the 1990s but has narrowed slightly since then. Not all skills seem to be getting their expected rewards.

One hypothesis is that technology is a complement to skilled, nonroutine tasks but a substitute for skilled, routine tasks. Think of an investment bank. Those responsible for nonroutine tasks like working out new trading strategies or new forms of securitization are able to command higher rewards for their talents when they sit in front of a bank of computer screens than they formerly could when they sat at their desks with a telephone in each hand. Meanwhile, in the departments that deliver established products and carry out established strategies, the provision of services may not be entirely automated, but it is computer-assisted to a degree that reduces the bargaining power of the workers who carry out skilled, but routine tasks. The end result is that even if both the nonroutine and routine tasks require an MBA, not all MBAs are rewarded equally in the marketplace.

The routine/nonroutine distinction is consistent with a thought-provoking new interpretation of the evidence on globalization and inequality by Jonathan Haskel, Robert Z. Lawrence, Edward E. Leamer, and Matthew J. Slaughter that appears in the most recent issue of Journal of Economic Perspectives. These authors point out that a large part of the increase in inequality is attributable to rising incomes of a top 1 percent of highly skilled, highly compensated workers that they call superstars. Superstars accounted for less than 8 percent of all U. S household income, excluding capital gains, in the 1970s. In 2007 their share of income peaked at over 18 percent, and has fallen only slightly during the Great Recession.

Why, though, should globalization increase the share of income going to superstars, as opposed to merely skilled workers? Haskel et al. conjecture that reduced costs of trade in services, whether entertainment, financial services, or technology, increases the relative demand for output of just those firms that make intensive use of the nonroutine skills of superstars. As an example, they suggest that in a world where information technology makes it easier to deliver the same books to readers throughout the world, superstars like J. K. Rowling earn even bigger rewards than they would in a world where readership was more fragmented.

I anticipate that apologists for the 1 percent will take heart from this new line of research. On the face of it, the superstar hypothesis suggests that globalization allows more people to benefit from the exceptional talents of the few, to the mutual benefit of all concerned. In such a world, any barriers to the free exercise of superstar talents, such as trade restrictions or high income taxes, benefit only the envious. The fact that superstars are richer than ever before only means that they deserve their rewards more than ever before.

Maybe, but I am skeptical. I can think of several hypotheses that are consistent with the superstar-globalization-inequality nexus without necessarily implying that the growing wealth of the 1 percent is fully deserved. Instead, it may only mean that globalization is increasing the ability of superstars to extract rents rather than increasing their actual contribution to world economic welfare.

One indication is the resistance of large, global corporations to scrutiny of their compensation policies by their own shareholders. It seems that the most superstar-intensive companies are precisely the ones  that most fiercely resist say-on-pay rules, even weak ones that allow only nonbinding shareholder oversight of compensation policies. The fight against transparency in compensation suggests that someone has something to hide.

Another thing that raises my suspicions is the frenetic competition for “talent” at the highest corporate levels. It seems to me that a simple feedback mechanism is at work here. The most highly compensated individuals, whether best-selling authors, financial executives or high-tech entrepreneurs, work in fields where personal success inherently depends on a mix of talent and luck. It is a law of human nature that those who succeed in such a context exaggerate the degree to which success was the result of their own talent while giving luck less than its full due. They, of all people, are least well positioned to make a balanced judgment of the value of attracting more “talent” to the corporate team. Instead, they are likely to systematically overpay for perceived talent that is in part only a reflection of past luck, and to systematically underestimate the likelihood of regression toward the mean in the performance of their apparently talented peers. The result is a sort of “winners curse” in which elite professionals end up working for the organizations that most egregiously overestimate the degree to which their past success was truly due to talent.

Finally, we must take into account the possibility that the high earnings of superstars are, to some extent, the product of their ability to gamble with other peoples’ money. Bonus-based compensation schemes with inadequate clawback for losses, combined with the moral hazard that exists when losses can be shifted to shareholders, unsecured creditors, or taxpayers, are likely to produce compensation packages that have an expected value higher than the true value of services rendered.

These considerations suggest a darker interpretation of the new research into globalization and inequality. Yes, the rise of inequality may be due to the fact that globalization has favored the growth of the most superstar-intensive firms. However, what if that only means that it favors the growth of firms with the least transparent compensation policies? The ones most susceptible to the winner’s curse in the competition for talent? Those best positioned to gamble with other peoples’ money? If that is the case, it is more difficult to interpret growing inequality as a benign by-product of mutual gains from expanding global trade.

8 Responses to “Globalization and Inequality: Is there a Superstar Effect, and if so, What does it Mean?”

princess1960June 11th, 2012 at 11:32 am

ok economia is not just ''mechanisme'' ..i think the'' strategy '' needed changed with much better efect..and this is very important because some time gv trying to find some were is not so neccessry (is )but not to efective and i am agree the tax for '' rich street '' needed more attention and i am sure will have efect ..thank you

DA SimarmataJune 12th, 2012 at 1:03 am

There is a great role of transparency in hiding the way to set the compensations related to "talent'. But another factor might be due to theoretical nature especially in the macroeconomic theory. The well known principle is the constant return to scale, while in fact it was increasing return to scale. There is one part of the product originated in the increase of the scale of the system. This part is not taken into account and is attributed to capital and other classified talented human factor. All these are enabled by the non-transparency of the compensation scheme.

Bryan WillmanJune 12th, 2012 at 9:35 pm

I think part of the issue is overly vague terms and crude structure of analysis mixed with debates that seem to focus on proving or disproving single relatively simple causes.

For example – the 1% in the US is more than 3 million people. They can't all be manipulative corporate executives or manipulative bankers. On the reverse, there are millions who are unemployed – it's lunacy to claim that all of them, or none of them, are lazy or incompetent. Lunacy to claim that there are no skills mismatches, and lunacy to claim that skills mismatches explain all of unemployment.

Bryan WillmanJune 12th, 2012 at 9:39 pm

Part II – Likewise, it's really foolish to argue that globalization causes, or reduces unemployment or inequality as though that can be a general rule that applies in all places at all times.

The question is never "does globalization help or hurt …" but always "how does globalization interacting with { long list of coinfluences } affect …."

O. SwarthoutJune 13th, 2012 at 1:37 pm

In my opinion, Ed Dolan's statement " we must take into account the possibility that the high earnings of superstars are, to some extent, the product of their ability to gamble with other peoples’ money" is an excellent observation of Wall Street shenanigans these days. For the ordinary investor holding 401k and IRA's for retirement, earnings have been flat or dismal. Meanwhile, Wall Street superstars are pulling down outlandish bonuses and commissions based on fast and loose trading rules and collection of endless fees .
The 06/10/12 New York Times posted a related article:

TechAugust 6th, 2017 at 8:38 am

There is an awesome part of straightforwardness secluded from everything the best approach to set the remunerations identified with "ability'

Max994October 8th, 2017 at 10:42 am

Follow the link of the address fireboy and watergirl and play the most entertaining brain twister video game for free and online.You love to play this cool math game more than any other online game.Thank you so much for your attention.

Max994October 8th, 2017 at 10:42 am

Follow the link of the address fireboy and watergirl and play the most entertaining brain twister video game for free and online.You love to play this cool math game more than any other online game.Thank you so much for your attention.

Most Read | Featured | Popular