Equality, Value, and Merit

A common argument against absolute equality is that individuals should be paid based on merit. Should somebody who works 80 hours a week earn the same amount as somebody who sits on their couch and watches TV all week? Even the most ardent supporter of redistribution would have a hard time answering yes. One of the alleged benefits of a free market economy is that it does a pretty good job allocating resources to those who work for them. Reading Hayek, however, I find it interesting that his defense of unequal outcomes explicitly denounces the idea of meritocracy. Value, not merit, is what should determine a person’s reward.

Some clarifications are in order. “Value” and “merit” are not well defined concepts. Let’s take an example to see the distinction between these two concepts. Imagine 2 students are studying for a math exam. One student studies 8 hours per day all week for the exam, but math has never been his strength and he ends up with a hard earned B+ on the exam. For the other student math has always come easy. He takes a quick look at his notes for a couple hours the night before and breezes through with an easy A. We might say that the first student deserves a higher grade than the second. If we graded based on merit we would want to give the higher grade to the student who worked the hardest. Of course, this grading system makes no sense when we consider that a grade is meant to represent a student’s knowledge of the material. Even though he didn’t work as hard, the second student knows math better and therefore deserves a higher grade.

The same arguments can be applied to an economic context. If two entrepreneurs each develop a product, a meritocratic society might suggest paying each based on how much work they each put into its creation. However, this criteria doesn’t consider the fact that consumers might place different values on the two products. If we want to maximize the benefits to society, we don’t actually care whether a product was created by a team of people and 2 years of strenuous research and development or by a guy coming up with ideas in the shower. All we care about is the value of the two products to the consumer. In Hayek’s words, “it is neither desirable nor practicable that material rewards should be made generally to correspond to what men recognize as merit…we do not wish people to earn a maximum of merit but to achieve a maximum of usefulness at a minimum of pain and sacrifice and therefore a minimum of merit” (The Constitution of Liberty, 157, 160).

It might seem unfair that talented people tend to earn more than the less talented. The handsome actor already gets good looks and fame. How is it fair that he also gets a big paycheck? And it’s not fair. But that doesn’t mean it’s not desirable. Because without that paycheck, without that incentive, maybe he wouldn’t have become an actor at all, and the opportunity to create a product that millions would have enjoyed is gone. It’s not fair that Tom Brady gets paid so much to play a game, but the only reason he does is because so many love watching him play. The alternative might not be that he gets paid less and still plays, but that he doesn’t play at all because his incentives to work hard and become a great player are diminished.

Another problem with a meritocratic society is that merit is hard to measure. Going back to the math example, I said that one student studied more than the other. But maybe his studying was not as efficient. Maybe he was actually on Facebook half the time, or didn’t focus on the right problems. And there are other factors. Maybe the second student paid better attention in class or had worked harder in previous classes and therefore didn’t need to work as hard now. Even if we wanted to reward the students’ merit, doing so would be a challenge. Similarly, looking at two products tells us little about how much work and how much effort went into the creation. What we can see is how much people like each product (by looking at how much they pay for it).

One of the greatest benefits of a market economy is that it pushes people towards the tasks that other people actually want them to do. In Hayek’s words, “If in their pursuit of uncertain goals people are to use their own knowledge and capacities, they must be guided, not by what other people think they ought to do, but by the value others attach to the result at which they aim” (The Constitution of Liberty, 159). By rewarding value over merit we ensure that people can only earn money by offering something that others desire. Everybody acts in their own self-interest, but the market usually ensures that that interest also aligns with the interests of others. Potential earnings act as a signal that shows what society values and attempts to regulate the market will almost certainly mess with these signals.

With this perspective, it is difficult to find a reason to care about others’ wealth. Steve Jobs, Bill Gates, and Mark Zuckerberg only got rich by offering a service that other people valued. Their contribution to society is likely far greater than any monetary compensation they received. Encouraging others to continue in their footsteps, to innovate and invent, is more important to the welfare of society as a whole than any attempts to redistribute their existing wealth. In fact, attempts to accomplish the latter discourage the former. I disagree with Ayn Rand on many points, but I think the overall theme in Atlas Shrugged is about right. When society feels like it can take anything it wants from the producers, they might decide that it’s simply not worth it any more, leaving no wealth left to redistribute at all.

 

Interesting Paper on Inequality and Fairness

As a followup to my recent post on inequality, I wanted to highlight some recent research by Christina Starmans, Mark Sheskin, and Paul Bloom on fairness and inequality. Based on a survey of lab experiments and evidence from the real world, the paper argues that people don’t actually care about unequal outcomes as long as they are perceived as fair.

They highlight several studies that show that in laboratory settings people (even children) are likely to distribute resources equally. However, in many of these settings, equality and fairness are indistinguishable. Since none of the participants did anything to deserve a larger portion, participants could simply be attempting to create a fair distribution rather than an equal one. And experiments that explicitly distinguish between fairness and equality do find that people care more about the former. For example, people were not unhappy with allocations that were determined randomly even if the outcome ended up being unequal as long as everybody began with an equal opportunity. Children who were asked to allocate erasers as a reward for cleaning their rooms were more likely to give the erasers to those who did a good job.

In reality people also seem to prefer an unequal distribution of income as long as it is perceived to be fair. In surveys, while people’s perception of the true income distribution is often highly skewed, their ideal distribution is not one of perfect equality. Of course, looking at these surveys does not necessarily tell us much about what the “best” income distribution would be, but rather the one people (think they) prefer. As I argued in my last post, I think too much weight has been placed on income or wealth inequality when really all that matters are differences in people’s happiness or utility. The evidence presented here does not go that far, but it does suggest that people realize that different behavior should lead to different rewards in some cases.

One reason that I think the debate has focused mostly on income or wealth inequality rather than on fairness or another measure of inequity is due to issues with measurement. Everybody has different ideas about what is fair so it’s easier to frame the question in terms of something that can be easily reported numerically. We may want to reconsider our acceptance of those statistics as a meaningful representation of a social problem. The whole paper is well worth reading and it opens up some interesting questions about human behavior. I will have at least one more post related to inequality coming in the next week or so.

What Kind of Inequality Matters?

Thomas Piketty’s book, Capital in the Twenty-First Century, a thorough analysis of the causes and effects of inequality, recently became an international best-seller. It’s not often that thousand page economic treatises attract popular attention, so clearly there’s something important to discuss here. Looking at some of the data on inequality, it’s not hard to see why many people are concerned. Here’s a chart showing the share of income held by the top 10% in the United States since 1910:

Notice where the two peaks occur – 1929 and 2009. I seem to recall something important happening in each of those years. Whether inequality was a symptom or a cause of the broader problems that led to the Great Depression and the Great Recession is an interesting question and definitely deserves scrutiny. For the purposes of this post, however, I want to address a simpler topic. Should we care about inequality on its own? And, more specifically, what kind of inequality should we care about?

For the first question, let’s do a simple thought experiment. You can choose to live in one of two societies. In Society A, everybody makes $50,000 per year no matter what their profession is. LeBron James and a janitor get paid the same amount. In Society B, average income is the same $50,000 per year, but it is now dispersed, so that some people earn less than average and some earn far more. Now assume that you are guaranteed to begin at average income (to avoid questions of risk aversion). Which society would you rather live in?

The answer to the hypothetical depends in part on whether you care about absolute or relative income. Does it matter if you are rich, or does it only matter if you are richer than others? In Society A everybody is on the same level, which might seem to be an appealing feature.

Except as soon as we start to think a bit harder, we realize that people in Society A aren’t equal at all. At least to some extent, differences in income do come from differences in effort. Some people work harder than others. Should people get paid the same regardless of effort?

Of course, this reasoning attacks a bit of a straw man. Hardly anybody would argue for full equality of income. But that doesn’t mean that there aren’t some situations where reducing income inequality could be helpful. I don’t believe in free will, which means that I think that where you are today is determined by circumstances you had no control over. But even with free will, it’s impossible to deny that some people are luckier than others. Some people are born into families with higher incomes or better connections. Some people are just smarter, or more talented. Two people can put in the same amount of effort and come out with wildly different outcomes. Isn’t there some justification for correcting these kinds of inequalities?

Now we need to bring in the second question: what kind of inequality matters? To this point, I have focused entirely on income inequality, but money is only as good as what you can buy with it. Somebody who earns $1,000,000 per year but saves $950,000 is no better off than someone who earns $50,000 per year (until they start spending those savings of course). We also need to consider a dynamic component to inequality. The chart above shows only a snapshot of inequality at one point in time, but there is large variation in earnings over a person’s lifetime. So a better measure of the kind of inequality that actually matters would be total lifetime consumption inequality (due to measurement difficulties, the question of whether consumption and income inequality move together is still under debate – see a nice survey here).

But we’re still not quite there. Why do we consume anything? Presumably because it makes us happy, or, in the words of an economist, because it gives us utility. Simply giving people more stuff might not actually help them at all unless it’s stuff they actually want. So shouldn’t we actually care about total lifetime utility? And as soon as we jump into the world of utility, the problem gets much more difficult.

Consider an extremely wealthy person. Incredibly talented and smart, he excelled in school, founded a business, and became one of the most successful CEOs in the world. He has a beautiful house, ten expensive cars, flat screen TVs, season tickets to the Patriots. He can buy anything you could ever want. Except he works all the time, hates his job, and has no time for his family or friends. Despite his money, despite his consumption, he is miserable.

Another individual earns far less. She isn’t poor, but she earns right around median income. She doesn’t have a luxurious life, but she can afford the basics. More importantly, she’s happy. She has a loving family, great friends, a job she likes. Would she be happier with more income? Probably. But she doesn’t need luxuries to live a good life.

How do we make this society more equal? Simply looking at income would suggest a transfer from the wealthy man to the average income woman. This transfer would of course reduce income inequality, but it would increase utility inequality. The woman is already pretty happy and the man is not. Taking money from him and giving it to her would only increase the happiness gap. Is this outcome desirable? I don’t think so.

Then maybe we should try to minimize utility inequality. But how? Taking money from the woman would probably reduce the woman’s utility and eventually it would be as low as the man’s, but giving it to the man would probably do little to increase the man’s utility unless he takes comfort in the fact that others are as miserable as he is. The woman’s happiness comes from pieces of her life that can’t be transferred to others. Despite being born with all the skills necessary to succeed, the man would likely view the woman as the more fortunate one.

In general, trying to equalize utility gives some strange implications. Let give a few more examples.

Two people work in the exact same job and get paid the same wage. Seems perfectly fair. But what if one of them enjoys working and the other hates it? In dollars per hour, they are equal. In utility per hour, one receives more than the other. Reducing utility inequality would require that people who enjoy their jobs be paid less for the same work.

Some people prefer living in cities while others would prefer to live in smaller towns. Houses in cities are usually much more expensive, which means to achieve the same utility, a city lover will have to pay far more. In this case, income equality greatly benefits people who hate cities. Utility equality would suggest transfers from people who love rural areas to those who love cities.

Consumption equality could also generate large utility inequality. If one person places a lot of meaning on material goods while another values other aspects of their life, they would need different levels of consumption in order to achieve the same utility. Should we give more to the materialist than the ascetic simply because giving to the latter wouldn’t help them anyway?

And even these examples ignore the largest problem with trying to achieve equality in utility – it’s difficult to measure and impossible to compare across individuals. I have trouble defining my own preferences and determining what makes me the happiest, I certainly don’t trust others to do that for me.

So utility equality is probably not an option even if it were desirable. But income equality almost certainly worsens the problem of utility inequality. The people who make a lot of money are much more likely to also be people who place a high value on money. Those who earn less are more likely to enjoy a simpler life. In fact, there is little evidence that the rich are any happier than the rest of us. Taking their money makes them even worse off while helping those who are already pretty happy despite their relatively low income. The happy get happier while the miserable get more miserable.

Notice that I have deliberately avoided using examples with truly poor people. I can certainly see an argument for redistributing income to the poorest. Nobody should have to live at subsistence levels if they are willing to work. But being concerned about poverty and being concerned about inequality are not the same. It is possible for a society to have zero poor people and still be incredibly unequal and also possible to be almost perfectly equal with everybody poor (as it was for most of the history of human existence).

Have we gotten any closer to answering the original question? What kind of inequality should we care about? If you’ve made it this far, it should be clear that there isn’t an easy answer. We often use the term “less fortunate” as a euphemism for poor people and that almost exclusively refers to poverty in a monetary context. We view income as if it came from a lottery and then aim to use redistribution to correct for discrepancies. Why is that? Aren’t people that can be happy despite low income really the most fortunate? Isn’t money just one of many factors that matter for a person’s happiness? And aren’t many of these other factors difficult to measure and even more difficult to redistribute?

If we answer yes to the above questions, reducing the kind of inequality we care about becomes a much harder task. Can we really correct the deeper inequalities that arise due to people’s preferences and talents – some of which will lead to higher incomes and some not? Or should we accept that inequality is an essential part of society, accept that treating everyone equally necessarily produces inequality in outcomes, that differences in wealth don’t necessarily lead to differences in happiness, and that correcting differences in happiness is almost impossible?