Income Inequality and the Management of Organizations

This post is the first part of a blog forum about inequality and organizational theory. Bruce Kogut and Jerry Kim of Columbia University wrote the post, and Shamus Khan, also of Columbia, wrote the commentary.

Bruce Kogut and Jerry Kim

Why should we care about income inequality?  For many, an egalitarian society is a just society.  The argument for this belief ranges from  “I just think that” to sophisticated reasoning about declining marginal utilities and the distribution of natural abilities and the tradeoffs between efficiency and equity.  And for those troubled by such tradeoffs, this latter reasoning boils to the hoped-for compromise that ‘we can get a lot more equity at very little sacrifice in efficiency’.  Arguing about that claim, economics has spilled a lot of the proverbial ink.

But why should we in management and organizational theory care about income inequality in the context of what we study and teach?   The short answer is that the primary source for the massive growth in inequality in the US is due to the greater share going to those who are the top managers of public and private firms.  The share of total income in the US earned by the top 1 percent of income earners has gone from 9 percent in 1970 to 23.5 percent in 2007.   While an increase in inequality is to be found in many rich countries, the US distribution is remarkably more skewed.

If we take a closer look at this 1% (which was approximately for incomes greater than $400,000 in 2007), a substantial number of high-income earners are managers. In fact, according to a recent analysis of individual tax return data, close to half of the top 0.1%—those that make upwards of $2 million—can be categorized as non-finance executives, supervisors or managers.   What is it about business organizations today that are distributing such wealth to their top managers?

An answer to this story must be consistent with an important stylized fact: the share of remuneration of top managers has increased much faster than the market value of firms. One common argument is that technological change and globalization have led to increasing returns to general managerial talent. We call this ‘the best athlete’ argument.   However, the historical evidence suggests that such change predates this recent turbo-charging of pay.

Moreover, if an increased demand for scarce managerial talent drives salaries, then we should see a persisting inequality of pay among top executives.  After all, given all the pay for performance, shouldn’t executive pay be more unequal than the pay in the general population? Salaries in Major League Baseball (MLB)—a context that is perhaps closest to a perfect market for talent—exhibit a level of inequality that is slightly higher than what we find in the overall population, but more importantly, consistent over time.  The inequality for executive compensation, on the other hand, fluctuates a great deal, peaking at more than 15% in 2000 above of the inequality found in
baseball and then falling by 2009 about 15% below.

The alternative explanation that we offer in this working paper is that social norms have shifted in the past two decades. The internet boom was an exogenous shock: high stock prices and the adoption of pay for performance knocked some salaries into the stratosphere.  Social comparisons by benchmarking kicked in and inequality fell, BUT now the proportional share of executive compensation to firm market value went dramatically up.

Executives are surely often very hard working and markets tend to reward that.  However, there is also ambiguity that permits gaming of the system. A strike out or a single is pretty unambiguous and the market for players seems to sort into a fairly stable pay distribution.  This is not the case for executives whose performance is harder to observe.  You would think that boards would do their best to get rid of all the noise they can, such as stock market fluctuations.  But a good deal of research says this is not the case.

That directors of boards do not try harder to purge noise suggests that the rent-seeking theory of pay can explain this explosion.  However, managerial rent seeking to rise up the income rank seem to have been around much earlier than the start of this historical trend.  And similarly, the argument that the growth in pay is due to technical progress that drives up the value of executive talent also does not explain the pattern, for technical progress has been around for some time in any event.  But the real problem with this argument is that it just not particularly clear why technical progress should cause executive pay to skyrocket in absolute terms relative to skilled labor in general.

At all levels of society, social comparisons are bound to be made and to provide footholds into gauging self worth.  It’s not compelling that these comparisons need to require such historically astounding degrees of inequality as justified by stories of sudden technical change driving the demand for executive talent –for which there is little evidence.   And, to depart from data, one suspects that such stunning inequities have unpleasant cultural and democratic consequences that are bound to be reflected in the attitudes and behaviors of the business leaders of our age.

Organization theorists and economic sociologists are well positioned to join other social sciences in the study of the role that social norms and peer benchmarking as well as gender and racial bias play in generating inequality.  This is a global issue that nevertheless begins in the decisions made by business organizations.   That the topic of economic inequality is also entwined in the great political and normative debates of social justice makes our collective contribution all the more desirable.

Discussion by Shamus Khan

Bruce and Jerry’s piece is a wonderful primer on elites and inequality. My response is simply a chronicle of my thoughts upon reading it (and their paper, which I would strongly recommend to all).

First, I think we need to establish why we should care about inequality, particularly if it is driven by wage growth of elites rather than a lowering of the floor of wages for the poorest. Christopher Jencks has done some work arguing that greater inequality is either neutral or bad for societies (but never good), and I’m a little more convinced by it than I am recent work by Wilkinson and Pickett, which makes the same basic argument. But the Jencks and Wilkinson and Pickett arguments are hardly definitive, and it strikes me that some of the best reasons for caring about inequality are because of the relationships between inequality and mobility. Alan Krueger has argued that there is a “Gatsby Curve” (slides here), drawing upon the work of Miles Corak. This has come under some criticism, but I think Corak’s counter-arguments are quite convincing here: there is a tie between rising inequality and declining mobility.

While the rhetoric of the “1%” has recently led to an important social awakening of the findings of a 2003 study by Piketty and Saez, Kogut and Kim somewhat quietly point to an important corrective: there’s a lot of variation within this category, and perhaps we should be looking at the top 10% within it. From that Piketty and Saez and paper I would particularly point readers to Figure IV, which gets a little less attention than I think it should. One of the things it shows is that in earlier eras (1929), the very wealthy were fundamentally different from others within the rich, insofar as their income stream was from a different source (capital income rather than wage income). A quick way to think of this is that this is a Marxian condition – the super-rich were owners of capital like factories. Today it’s still the case that the very rich are more likely to have capital income than the somewhat rich. But the story is not nearly as dramatic. (Importantly, these data don’t include capital gains). And so today’s rich are more “Weberian” – their position is not as tied to capital ownership but instead tied to wage wealth (money). I point to this because Kogut and Kim argue about the importance of shifting norms, and this might help explain why such shifts have happened. That is, wages have embedded within them the idea of pay-for-performance. So if your wages are high you must be really smart and really hard working and provide lots of value, otherwise you could simply be replaced at a lower wage rate. Such arguments work better with wage income than with capital income.

The working paper by Kim, Kogut, and Yang provides a welcome evaluation for DiPrete, Eirich, and Pittinsky’s “leapfrog” argument (particularly, why the leapfrogging happens when it does and not in other times – why pay is similar among CEOs in the 2000s rather than an arms race). This argument differs from Olivier Godechot’s work, which makes an argument for managers “holding up” their firms insofar as they have control over assets and can thereby punish firms that do not comply with their salary and bonus demands. For Godechot, this is particularly unique to finance, where such assets are easily transferable to other firms.

For Kogut and Kim, an exogenous shock of the internet boom created inappropriately “stratospheric” salaries for some, which others started to benchmark themselves against this. This brings up some interesting implications. The first is that the financialization of the economy that has been the focus of so many might be a red herring. While Godechot’s argument is conditional upon finance (or a market sector with easily transferable assets), the mechanism that Kogut and Kim offer does not require the rise of finance. What do you need? Interconnected networks of CEOs wherein a subgroup suddenly is considerably overpaid – high pay diffuses through social networks. What matters is the network structure, not the structure of the economy.

Yet most readers will wonder: such exogenous shocks are not uncommon within markets, and so why did the diffusion happen now and not before (and not globally – the US was not the only nation to experience a dot-com boom)? There is a hint of an answer; pointing to McCall and Percheski and Western and Rosenfeld (unions), they suggest that organizations that might have constrained this process or sanctioned those engaging in it have been weakened or disappeared. This makes me wonder if we’re really talking about new norms (which I’m defining as expectations for action). Instead, what if the norms are somewhat constant and the constraints are what vary? This is important because if the norms vary we can argue about returning to the era of a “socially responsible CEO” but if they are constant, this kind of discussion doesn’t make sense and instead we ask ourselves about how to construct constraining institutions.

Finally, I’d point to the importance of inter-elite contention: something often ignored when thinking about constraints upon elites. And this returns us to the finance point. When constraining elites we often think in Marxian terms (even if we’re not Marxists). That means we think of elites as having fairly consistent (or shared) interests, and that to constrain elites, masses must act. But it may well be the case that inter-elite contention is far more effective than any other mechanism for constraining elites – by creating collective action problems and competing interests. Here, then, shifts in markets where certain sectors become increasingly dominant will have large negative effects. And as such, the massive transfer of wealth to the finance sector – perhaps around the order of more than 6 trillion dollars since 1980 – has created conditions wherein within-elite constraints have declined.