Research dump, 2/4/2013:

This is just a summary of some cool economics research I’ve come across recently, with some brief notes about how each study relates to broader theoretical ideas:

-The Sarbanes-Oxley Act is often credited with improving corporate governance in the U.S. in the wake of the Enron and Worldcom accounting scandals. But two professors have done some statistical analysis that suggests that accounting manipulation—and investors’ susceptibility to this manipulation—sharply decreased before Sarbanes-Oxley went into effect, but after the salient scandals. That is, improvements in corporate governance seemed to come from market participants themselves becoming more attentive to potential fraud, after the Enron and WorldCom scandals, rather than from the new regulations themselves.  So in this case at least, the market beat regulators to the punch in enforcing reform and better practice. The interesting idea here relates to how most people, when they think of ‘regulation,’ think of the laws drawn up and enforced by governments—but economists think of ‘regulation’ happening within markets, too. People in the private sector, after all, don’t like getting defrauded, so they take their own measures against it and do their own monitoring.

-Economic incentives matter. Like, a lot. As in, even in life-and-death situations. People are more inclined to crash their own planes when others can be held liable for the cost of the damage to the plane. It’s popular to accuse economics of leaving out “the human factor,” and such. But there’s plenty of interesting research out there that, like this study, shows the astonishing ways in which money influences our behavior more than we like to believe.

-A number of recent papers have argued that our measure of inflation—the Consumer Price Index—has led us to be overly-pessimistic about how lower- and middle-income Americans have fared over the past several decades. Prices for low-income consumers have inflated more slowly than prices for American goods as a whole (in part thanks to super-efficient retailers like Wal Mart), and so aggregate measures of inflation overstate the cost-of-living for low-income people. So this paper argues that while ‘income poverty’ is not declining as fast as we would like, ‘consumption poverty’ has declined encouragingly over the past five decades. In general, measuring ‘inflation’ is incredibly tricky, not just because of the difficulties of measurement, but also for broader philosophical reasons—we essentially measure ‘inflation’ by comparing the sum price of two comparable baskets of goods in different time periods, but technological changes that mean that, e.g., every consumer’s basket of goods is a little bit better in subtle ways in year 2010 than it was in 1960, and so the idea of a ‘comparable basket’ is very debatable and fraught.  And so the measurement of inflation is just one of the many areas where economics is more art than science, and in which the ‘art’ has major implications for our understanding of what’s going on in the world right now.

-I posted several weeks ago some new research and theory that suggested that, contrary to classical economic theory, increases in the minimum wage did not increase unemployment. But now, other economists have fired back with some critiques of the research design of those studies, and their own evidence that minimum wage hikes do increase unemployment. I won’t take sides in this debate, but just say that it’s a reminder that teasing out statistical trends from the chaos of real-world data is really, really hard, and, again, sometimes more art than science.

-How do financial crises (i.e., in the financial sector) become economic recessions (i.e., for the rest of us)? The answer is not as simple as, I think, an economic novice would assume. I think a lot of ordinary people look at the S&P500 figures they see on CNBC as a straight measure of how the economy is doing, but that’s not strictly correct. For example, imagine that a superior extraterrestrial species were to inform us all that the earth would be destroyed in 10 years—the stock market would tank, because the “net present discounted” value of all shares would decline since they all only have 10 years left. But as long as we all kept buying goods and living our lives like we do for the next 9 years, the real economy should do just fine: if we continued buying all the same goods, then sellers would be selling all the same goods, so the real economy wouldn’t change much. So how exactly do financial crises spread into the real economy? What connects these two universes? One answer is the “wealth effect”—i.e., if the savings you have stored in financial assets have been devastated, you’re going to want to “repair your balance sheet” by saving more, to bring your wealth back up to what you had previously thought it was. This means you’re going to spend less now, which means that the people you normally buy from are going to have less income now. That’s a major mechanism through which financial crises spread into the real economy. The great housing economist Robert Shiller and two other economists have some new data on how “wealth effects” related to the housing sector have affected consumption throughout U.S. history. Housing prices appear to affect consumer spending even more strongly than stock prices, presumably partly because more of us are heavily invested in our housing than in the stock market. So this is both (1) a big part of the explanation of the economic doldrums of the past five years and (2) a more general reminder of how stable housing markets are important to the broader economy, and policies and financial ‘innovations’ that destabilize the housing market can be a big threat to all of us…

-A big debate in the econ-policy world right is, essentially, “Are the Luddites, this time around, actually right?” That is, since the industrial revolution, in every generation people have been worried that new technologies would replace labor, leading  to widespread unemployment and poverty. But so far, in every generation, the Luddites have been wrong, and we’ve found ways in which people can ‘complement’ those technologies, instead of being replaced by them, such that technological advances since the industrial revolution have led to a higher standard of living for everyone. But could this change in the future? Could this time really be different? One fear is that once Artificial Intelligence becomes sufficiently advanced that it can match an adult human brain, then, by definition, there will be nothing a robot can’t do to substitute for human labor, and so the Luddites will finally be right. In a new paper, Jeff Sachs and Laurence Kotlikoff present a scary stylized model, in which current, established generations of skilled employees use future artificial technologies to substitute for young, unskilled labor, which prevents young people from accumulating the skills that could allow them to complement intelligent robots. So, in their model, technological progress can lead to ‘long-term misery,’ each generation worse off than its prior. Scary stuff.

-There’s a growing body of evidence that, in elite professional services in particular, hiring decisions are made not on the basis of objective competence and skills, but on a vague sense of cultural ‘fit.’ Why is this? Well, in the narrow instrumental sense of the word ‘rational,’ hiring on the basis of ‘fit’ can be a ‘rational’ decision. If you’re a law firm of Democrats, and you have to choose between a Democrat or a suspected-Republican job candidate, both of about equal resume-excellence, it’d be instrumentally rational to choose the Democrat, because most of the important skills will be learned on the job anyways, and many of the lawyers at your firm might feel awkward interacting with the Republican, which would detract from your organizational efficiency. More, you, as a hirer who is politically like-minded with the rest of the firm, might think you have good reasons for believing Republicans generally don’t get along well with people—might (s)he make offensive remarks or behave selfishly toward the rest of the firm team? So you can see how discrimination on the basis of ‘fit’ is ‘instrumentally rational,’ and, hence, why it gets done. But it’s a big problem for those of us who hold to the moral ideal that people should succeed and fail in our society according to their individual merits, and not according to our culture’s taxonomies of people—and also for those of us who think that eccentrics are good for us, and cultural, affiliational, and political diversity should be rewarded and not punished. (My readers, to really feel this idea, should replace the word ‘Republican’ above with the name of a group they strongly identify with.)

-Speaking of that idea: Ethics compels us to try to make more space and opportunity in our society for people with disabilities. How can we do this with policy? Austria has experimented with giving firms tax incentives to hire more disabled workers—specifically, one disabled worker per every 25 total workers. This appears to have worked decently well on increasing employment for the disabled. It’s also, predictably, caused firms to “manipulate employment of nondisabled workers,” (i.e., hire, say, exactly 99 workers and not 100, so that they are only ‘liable’ to hire three disabled workers and not four). That’s not, by any means, proof that those costs are not worth the benefits of this policy. It’s just another reminder that public policy is really, really hard and always involves tough trade-offs and dealing with the fact that people in the private sector are, legitimately, nervous about anything that can hurt their bottom lines.

-What do people do with their extra time when unemployed? Mostly, sleep and watch TV. This is not a jab. The reason this is interesting is that there’s a big debate in economics about the possibility that people will use productivity gains with future technologies to simply work fewer hours instead of making more money—an idea that hearkens back to Keynes. But what would people actually do with all this extra time? Well, we already have some evidence, thanks to the natural experiment provided by the great recession…

-This paper finds that people who, early in their careers, engaged in entrepreneurial enterprises subsequently, throughout the rest of their careers, earn higher incomes than those who were in other respects similar to them. The authors propose as an explanation that entrepreneurship increases your “human capital” (teaching you new skills, habits, confidence, self-actualization, etc.). This strikes me as plausible. But it’s of course impossible to do a randomized study here, and so it’s also likely that the kinds of people who become entrepreneurs had a lot of unobserved good qualities from the very start, and so the “control group of comparable workers” doesn’t really do it.

-One of the more uncomfortable ideas in modern descriptive social science has been that, while most of us applaud the intrinsic value of diversity on strictly moral grounds, some statistical research has found ethnic diversity to correlate with some undesirable consequences. E.g., the most politically functional and least-corrupt states in the world are also among the most ethnically homogeneous; the most corrupt states in the world are disproportionately among those states with the highest levels of linguistic diversity. Thus, many modern liberal political scientists have even actively advocated political ethnic balkanization in states as in, e.g., between Sudan and now South Sudan. The idea is that, even in the modern world, we humans are deeply biased and tribal, and so leaders in ethnically diverse states feel driven to legislate in the interests of their own groups, rather than for the common wealth. But an interesting new paper has tried to statistically disentangle the effects of ethnic diversity versus the effects of economic inequality, and suggests that it is not ethnic diversity per se, but diversity coupled with economic inequality that leads to political dysfunction. I.e., it is in states with high levels of inequality between ethnic groups that ethnicity becomes salient, so that people start to think, “It’s those people, them, who control all the wealth and political power.” Thus these states have more adversarial politics and less support for free markets, both of which compound their problems. But the authors argue that this effect disappears as inter-ethnic inequality decreases—which might provide us some hope that, in the future, Sudan-style balkanization could be dodged  and problems solved with growing wealth and equality.

-One of the most interesting ongoing debates in the econ-policy world centers on Intellectual Property. Basically, most of us think we need some I.P., which gives entrepreneurs incentives to develop new innovations, by allowing them to profit from a monopoly on those innovations for a time. But we also all see the abuses—isn’t it crazy that in 2013, we can only do 1-click shopping on (since Amazon owns the patent on this ‘innovation’ which is some 14 years old)? So what’s the ideal I.P. policy? I dunno, but this guy has a model that suggests the answer is “strong, but narrowly defined,” I.P. rights in general. That is, we want to protect your monopoly on your product, just as you made it, without forbidding others to riff off of it in interesting new ways.

-Moral psychology has found that we do not condemn indirect harm as strongly as would be morally rational. I.e., rationally, it would seem equally immoral to spread dirty rumors about someone on the internet, as to hire somebody else to do the same—but moral psychologists have found that we condemn the latter less strongly, and thus let ourselves “get away with” indirect immorality more easily. These professors have made an interesting connection between this idea and our understanding of group negotiations. Specifically, we are more likely to press for goals that we would not individually feel comfortable pressing for when we are negotiating with and on behalf of a group. I.e., working with a group allows us to “diffuse responsibility” for morally bad deeds in ways that can make us misbehave.

-Companies with “well-connected” boards of directors earn “excess returns” (i.e., above the market rate). Does this seem obvious and uninteresting to you? If so, you may be misunderstanding. Keep in mind the basic financial theory here: because investors “price in” their expectations for the future of companies into stock prices, an “excess return” doesn’t just mean that the company “does well and gets bigger” in the future. Rather, it means it does even better and gets even bigger than the best financial analysts were predicting. So anything that predicts “excess returns” is almost by definition interesting and surprising. So it can’t be the case that these excess returns are explained by the fact that “well-connected people choose to get on the boards of successful firms.” Rather, there must be something causal going on. More significantly, there’s something causal going on that the investment community as a whole doesn’t yet understand and, thus, isn’t properly pricing in. It could be some surprising network effects; it could be something more corrupt.


2 thoughts on “Research dump, 2/4/2013:

  1. I love your social analyses and your questioning of the validity of data that society (and markets) take for granted based. There are always intangibles and equal and opposite reactions to every action.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s