Should social scientists stop carping about celebrities and CEOs and start worrying more about the next financial crisis?
Tyler Cowen has an interesting article from a left libertarian slant in which he argues that “most of the worries about income inequality are bogus, but some are probably better grounded and even more serious than even many of their heralds realize.”
Before distinguishing the bogus from the serious, let’s briefly summarize some key facts about American inequality in the latter half of the 20th century. Here are some bullet points from a forthcoming paper by Sandy Jencks and coauthors (which I’m hoping to post soon) that concisely summarize the trends:
- The average increase in income inequality was relatively modest
- The increase in inequality was confined to the top half of the income distribution
- The increase in income inequality was biggest at the very top
- Inequality rose far more among households with children than among other households
I suspect that the first bullet point is the most controversial, and depends critically on what is meant by “relatively modest.” By the most commonly used measure of income inequality, the Gini coefficient, inequality rose by at most 17 percent between 1975 and 2001, and was flat between 2001 and 2008. (A technical discussion of how the Gini is calculated can be found here, but what is important is that it takes account of the entire income distribution, and not solely the difference between the very rich and everyone else.)
Skill Biased Technological Change or Demographics?
The change in the Gini is certainly not insignificant, but it is small enough that it might reflect some otherwise benign trends. Cowen mentions research by Thomas Lemieux using several decades of data from the May round of the Current Population Survey that suggests that demographics explains a large proportion of the trend in overall inequality. Cowen writes:
“There is usually greater inequality of income among both older people and the more highly educated, if only because there is more time and more room for fortunes to vary. Since America is becoming both older and more highly educated, our measured income inequality will increase pretty much by demographic fiat.”
According to the Lemieux paper, roughly two thirds of all male earnings variance can be explained by this “compositional effect”, and even more of it for women. If the compositional effect were the predominant explanation for overall inequality, then we might not worry about commonly cited factors such as the decline of labor unions, income supports such as the minimum wage, and the increasing returns to technological know-how (also known as “skills biased technological change”).
In fact, the Lemieux hypothesis does a very good job of explaining changes in income inequality across the entire distribution, but has the tendency to mask two different, and countervailing trends – increasing returns to skill at the top of the distribution, and declining returns to skill at the bottom. When David Autor and colleagues disaggregate the income data and separately evaluate the gap between percentile 90 and 50 (the rich and the middle), and 50 and 10 (the middle and the poor), they find very different stories. They argue that this polarizing tendency reflects long-term mismatches in the demands of the labor market and the skills of the bottom half of the working population on the one hand, and greater complementarities between skill and technology for the upper half of the working population. These factors, in combination with the rise in outsourcing and changes in global trade, explain much more of the rise in inequality than compositional changes, and conceptually capture two distinct phenomena.
One lesson to draw from this fairly technical economic debate is that economic inequality looks very different depending on what set of comparisons you exploit, and what you control for in your analyses. It’s not obvious that changes in education should be statistically controlled for, since they are themselves mediators of economic inequality. Another lesson is that a failure to detect measured inequality in the bottom half of the income distribution does not necessarily mean that there is nothing to worry about. Since earnings and income have stagnated for median workers, the compression of income in the middle of the income distribution might reflect the fanning out of bad working conditions to the middle of the income distribution.
Where Inequality Matters
Although I think Cowen is wrong to underemphasize the profound effect of the global economy on inequality for the bottom half of the income distribution, I share his conviction that we should all pay much more attention to the growing divergence between the very rich and the rest.
The point can be made fairly succinctly and effectively in the graph below (which is based on the Picketty and Saez income data series, and comes from this set of graphs on income inequality). In the past three decades there has been a staggering concentration of income in the top 0.1% of the income distribution. This was previously touched on in Charlotte’s discussion of the Hacker and Pierson book, which focuses on the political capture of American institutions.
Cowen’s explanation for this divergence focuses less explicitly on the role of special interest politics, and more on the growing scope and sophistication of financial markets. Much of the income growth at the top has gone to people working on Wall Street – that is, bankers and lawyers that have gained sophistication and know-how in complex financial transactions. The growth in wealth in the finance sector has greatly outpaced other forms of executive compensation:
“For 2004, nonfinancial executives of publicly traded companies accounted for less than 6 percent of the top 0.01 percent income bracket. In that same year, the top 25 hedge fund managers combined appear to have earned more than all of the CEOs from the entire S&P 500. The number of Wall Street investors earning more than $100 million a year was nine times higher than the public company executives earning that amount.”
None of this would be problematic, Cowen implies, if the finance industry was creating a product of genuine societal value. In fact, the high returns to finance are due to a perverse incentive to “go short on volatility”, i.e. to bet against big, unexpected declines in prices (such as the declines in housing prices that drove the housing crisis in the United States and Europe). This is normally a practice that goes unnoticed since large declines are outliers, but when it does go badly, the losses are huge, and are often pushed on to taxpayers and the unemployed.
The potential moral hazard in the financial market is one reason to be especially concerned about income inequality between people in the financial sector and the rest of society. The costs of financial crises are generally not borne by those who perpetuate the bad deals – at worst, they end up getting sacked from their corporate jobs only to find work with some other financial firm. The real losers are those who end up having to pay directly through a bailout, or indirectly through the ripple effects in capital investments and employment:
“In normal years the financial sector is flush with cash and high earnings. In implosion years a lot of the losses are borne by other sectors of society. In other words, financial crisis begets income inequality. Despite being conceptually distinct phenomena, the political economy of income inequality is, in part, the political economy of finance.”
Because the large banks have become more concentrated and powerful in recent years, we can expect that the “too big to fail” problem will become larger and riskier for the market as a whole in the future. The upshot of Cowen’s argument is that changes in the financial sector have the potential to generate another future financial crisis, which would then meaningfully affect the ability of most people to get credit, get a job, and meet their material needs.
Fights We Shouldn’t Be Fighting, and Fights We Could Better Fight
Have scholars of economic inequality been picking the wrong targets? If our ultimate objective is to influence policy that protects the economic opportunities of working and middle class families, then there may not be much to gained – either politically or practically – by going after celebrities or CEOs. To use Cowen’s reasoning, perhaps we think that these people should pay more in taxes, but do we really begrudge them the returns to hard work that benefits the rest of society? Put slightly differently, it makes the most sense to target groups that make money in ways that are deliberately irresponsible, or allow them to gamble with other people’s money in ways that actually undermine the economic prospects of average households.
This is especially true where political capital is limited. In a slightly different context, James Surowiecki argued in the New Yorker earlier this year that the debate about tax reform in the United States has ended up threatening many groups that would be less adversely affected, but feel more economically insecure. The fight would be more productively waged by winning over the merely rich (the “lower upper class” including doctors, lawyers, accountants, etc.) and by focusing on reining in the excess wealth of the 0.1 percent of income earners:
“The lower upper class exerts a cultural influence out of proportion to its size, and so its anger toward the upper upper class—toward outrageous executive salaries and Wall Street shenanigans—could be a powerful force for reforming the way we deal with inequality.”
The suggestion, therefore, is to stop quibbling with the merely rich, and focus on either devising better regulations to curb the practices of the ultra rich in finance, or changing the tax system on billionaires in order to recoup some of the money that has been funded through government bailouts. The second solution is not ideal, especially because the effect on future risk-taking behavior is ambiguous. Even more pessimistically, Cowen argues that there may be little scope for regulation of the financial sector because of the complexity of financial deals and the ability of people in the financial industry to sidestep the regulations.
Let me conclude with two thoughts.
First, Cowen’s article very clearly points to the need to develop a new research agenda for studying inequality across the entire income distribution. The types of policies that inequality scholars are used to thinking about including changes to the tax code, the welfare system, and education are necessary but not sufficient to protect ordinary families from future financial risks. There is a lot of territory to chart in the area of financial regulation.
Second, the problems of inequality have not changed in American society – there remain large imbalances in the resources commanded by the less educated and structurally disadvantaged, but the old discourse of inequality may not be sufficient in the post financial crisis world. The lesson is that we should not jettison the old study of inequality, but rather seek new and better ways to describe the new trickle down economics – the rippling down of financial risks from large banks, to family bank accounts, jobs, and home loans.