Will Wilkinson has gotten a lot of Internet love for his article “Thinking Clearly About Economic Inequality” (Free Exchange, Real Time Economics, Yglesias, Klein, Cowen, Rortybomb), which argues that increasing inequality is not as bad as people like Paul Krugman make it out to be. I thought it was a rhetorically clever but deeply misleading attempt to blur the obvious issue – economic inequality is increasing – by looking at it through a dizzying array of qualifying lenses.
Wilkinson marshals an impressive number of arguments to try to make the point that increasing income inequality is not the metric that we should focus on. I’ll try to take them one at a time. (Wilkinson’s arguments are summarized in the numbered paragraphs; the others are my responses.)
1. It isn’t income that matters, but consumption, since the way that income translates into utility is through consumption. “Why do we want income at all? So we can acquire things that we value.”
OK, I guess, although this ignores the desire of rich people – or even moderately well-off people – to provide for their children. Besides the psychological utility that they gain, this just means that the imbalance in consumption between rich and poor will be spread across future generations. The imbalance doesn’t go away. But this argument isn’t the big problem.
2. Actually we need to look at lifetime consumption, because people engage in consumption-smoothing. And “the run-up in consumption inequality has been considerably less dramatic than the run-up in income inequality.”
First, Wilkinson’s quotation seems to acknowledge that consumption inequality has been increasing. Second, you would expect increases in consumption inequality to lag increases in income inequality. When rich people get much, much richer very quickly – as happened in the last decade – they are not going to be able to consume that increased income as fast as they earn it. It will get deferred, or inherited, and will show up later. Poor people, by contrast, will attempt to maintain their consumption even as their incomes fall, going into more debt. This is why increases in consumption inequality lag increases in income inequality. It’s not a good thing.
Even more simply, if I make more money than you do over my entire lifetime, then over the course of my lifetime I will consume more than you (or my children will consume more than yours). Unless Wilkinson thinks that today’s successful hedge fund managers are actually going to be extremely poor in future years, this effect will not go away.
3. Furthermore, when we value consumption, we can’t look at market prices (nominal consumption); instead, we want to know the value to the consumer of the consumption. So instead we should just look at happiness. And here happiness gaps have been shrinking, not rising.
Wilkinson is honest enough to acknowledge that even the Stevenson and Wolfers study he cites for narrowing happiness gaps over the past several decades also found that “the trend toward greater equality in happiness stalled and began to reverse course in the 1990s, due in part to widening inequalities in happiness (and wages) between individuals of unequal levels of education.” And this is the period Krugman is most concerned with. But he then ignores this point in his analysis.
I’m as much a fan of positive psychology and happiness research as anyone. I am a believer, based on both personal experience and on reading summaries of the research (I’m not a research psychologist myself, of course) that above a certain level of income and wealth – which I would put somewhere squarely in the middle class – money simply does not make you happier. But this is the first time I’ve heard an economist try to justify economic inequality on the grounds that it is happiness that matters, not money. This is what you would expect from the medieval Church, not the Cato Institute: although you think you want more material things, actually you don’t, because the only thing that matters is salvation in the afterlife.
4. Wilkinson then tries to explain why increased income inequality does not translate into increased happiness inequality (although Steven and Wolfers actually say that it has translated into increased happiness inequality since the 1990s). There are two parts to this argument, but basically they collapse down to one. First, he says that the quality of budget-level products has increased faster than the quality of luxury-level products (like refrigerators), so that the differential in material comfort is decreasing for a given differential in monetary consumption. Second, he says that rich people are actually taxing themselves by spending huge amounts of money on “real estate with ocean views, or Ivy League diplomas, or goods like yachts” that do not provide value commensurate to their cost. (Along the way, he gets in a dig at the luxury goods industry, which he claims provides shoddy quality at extravagant prices, but that seems to me like an anecdote at best; after all, a Park Avenue duplex is still a lot better than a studio in Yonkers, whether or not Hermes scarves are as good as they once were.)
First off, remember that we’re talking about changes here – changes in income gaps, in consumption gaps, and in happiness gaps. Wilkinson’s argument, that increasing income gaps coexist with decreasing happiness gaps, requires more than just the observation that happiness as a function has a decreasing slope (doubling your consumption doesn’t double your happiness). It requires some evidence that the marginal happiness benefit of consumption is decreasing over time. If the shape of the happiness curve is the same in time 0 and time 1, then increasing consumption gaps will produce increasing happiness gaps, though perhaps not at a linear rate. Increasing consumption gaps can coexist with decreasing happiness gaps only if the happiness curve is getting flatter fast enough to compensate for those increasing consumption gaps.
And here Wilkinson undercuts his own argument. He points to the relatively small practical difference between a $300 refrigerator and a $10,000 refrigerator, which is far bigger than the difference between no refrigerator and a refrigerator – which was the relevant difference maybe fifty years ago. Good point. But he also talks about how vanilla and pepper suffered the same fate – only much longer before. The lesson is that different products and services that people want change over time, and at different times, from being rare luxuries to being relative commodities. Just because one former luxury good is now a commodity good doesn’t mean there aren’t other valuable goods that many people cannot afford.
Take organic fruits and vegetables, for example, which many parents would like to buy for their children, but are simply too expensive for tens of millions of households. Or private school in places where the public schools are not very good. Or being able to move out of an area that is plagued by air pollution, or by crime. These are not frivolous luxuries like Sub-Zero refrigerators. Or even take air travel, which has gotten much cheaper over the past three decades and that we commonly think of as having been “democratized.” Flying a family of four even just from the Northeast to Orlando can easily cost $2,000 in air tickets alone – something that is far beyond the reach of many families who would love to take their children to Disney World just once in their childhoods. In short, for all the product categories where you can say the rich are wasting their money on Sub-Zero refrigerators, there are other product categories where having more money makes a big, big difference in your material quality of life.
And let’s not mention health care, which is literally a matter of life and death.
Finally, there’s something strangely patronizing about this argument. Wilkinson cites research showing that the rate of inflation for the basic necessities that poor people buy – food, shelter, clothing – has been lower than the rate of inflation for other goods, like “home cleaning, lawn care, psychotherapy, and yoga classes.” That is an important finding. The implication is that, relatively speaking, the buying power of the poor (and hence their consumption) has grown faster than their income, while the buying power of the rich has grown slower than their income.
If the falling relative price of basic necessities (other than health care, of course) has reduced the proportion of people who go without basic necessities, then that is a great thing. But that is not the same thing as a decrease in inequality. Whether or not the poor have what social scientists think they need – food, shelter, and clothing (but not health care!) – they may still want home cleaning, lawn care, psychotherapy, and yoga classes. In this model, more leisure time, better psychological balance, and less back pain are all valuable things that rich people have much more of than poor people. And no matter what you do with the numbers, if nominal consumption inequality (inequality in the amount spent on consumption) is going up, you cannot make inequality in consumption of these goods go down. In a simple model, rich people benefit the same amount in nominal terms as poor people from a fall in the price of necessities, and therefore if the nominal consumption gap is increasing, the gap in the amount of money left for yoga classes is necessarily also increasing; the rate of inflation of yoga classes cannot change that. Given an increasing nominal consumption gap, rich people may have a lower percentage rate of growth of abstract consumption units than poor people, but their level of consumption will always grow faster than that of poor people.
Ultimately, the “rich people are fooling themselves” argument relies on a theory of false consciousness: rich people don’t know what is good for them and are wasting their money; poor people are better off not taking yoga classes. But this, I fear, moves us out of the realm of economics altogether. I know that behavioral economists for decades have been showing that people make irrational choices. Fine; let’s try to help people make more rational choices. But remember, it’s still their money. And they want more of it. How people spend their money reveals how they value things, and if they pay $50 per hour for yoga classes, then maybe yoga classes are worth $50 per hour to them, at least in their conscious brains.
From a moral standpoint, if there is a problem in one person having ten times as much money as someone else, that problem does not go away because he blows it on cocaine. From a policy standpoint, whether or not people use their money in ways that increase their happiness, money is the thing that they care about, and trying to base policy decisions on happiness is both paternalistic and impractical. If it turns out that rich people are no happier than poor people, should we simply stop worrying about poverty?
Wilkinson brings up a lot of interesting ideas and cites some interesting research, but does little to challenge Krugman’s core point: people like money, people use money to buy stuff, the money gap between rich and poor is increasing, and the stuff gap is increasing as well. I’ll accept that nominal consumption inequality may be growing slower than income inequality (although I suspect the difference is just being deferred into later decades), and inequality in stuff actually consumed may be growing slower than nominal consumption inequality (due to different inflation rates), but I don’t see a solid argument for why they aren’t growing. And that’s still a problem.
That was just the first nine pages – hopefully I’ll be back to talk about the rest.
By James Kwak