Thinking Clearly about Inequality

Will Wilkinson has a fantastic new paper at Cato: Thinking Clearly about Economic Inequality.

You can see leveling in quality across the price scale in almost every kind of consumer good.19 At the turn of the 20th century, only the mega-rich had refrigerators or cars. But refrigerators are now all but universal in the United States, even while refrigerator inequality continues to grow. The Sub-Zero PRO 48, which the manufacturer calls “a monument to food preservation,” costs about $11,000, compared with a paltry $350 for the IKEA Energisk B18 W. The lived difference, however, is rather smaller than that between having fresh meat and milk and having none…

The general effect of the democratization of luxury is to increase demand among the wealthy for nonmanufacturable, inherently scarce “positional goods” whose signal of relative socioeconomic status will not be so swiftly diluted by broad mass-market diffusion.

Ezra Klein and Jim Manzi had some follow-up questions. I want to ask him a few questions, but before that a few extra theory ideas to drive the discussion.

Consumption Cascades

There are a lot of economists who argue things about consumption and positional cascades, which is a fancy way of saying “keeping up with the Joneses.” Someone buy a really nice car, so everyone else in the neighborhood has to spend more than they wanted on their car to keep up. I’m not always sure how much stock to put in this. I have a post sitting in pending I’ll probably never publish about how MTV’s My Super Sweet Sixteen is the show of 21st century inequality in the pre-crisis state, and how we need squads of critical theorists to study it scene-by-scene. (I’m obsessed with watching it.) The thing to take away from it is that a lot of the inequality in spending in the 21st century is best describe not as envy-generating but grotesque. I find watching the people watching the spectacles most interesting; they have to place themselves positionally to it without necessarily being jealous of it. All things equal, it is probably best to be the person having an five-figure Egyptian-themed 16th birthday party. They’ll certainly have health care later on. Failing that, it is almost as fun to be the person who attends but makes fun of it. But it still sets the agenda.

I often worry that economist go to this language to invoke themes of stratification, inequality of access and life choices, and power, without having to do any of the heavy lifting or probing questions that follow such an inquiry. (I have a friend who was part of a feminist economics listserv back in the day. She would start to talk about power, and then get cut off with a “no no no; we can’t say ‘power in the household’, we have to say ‘information asymmetries in the household’ instead. That’s the only way we can get taken seriously.” She did not go to economics graduate school. I hope she forgives me if I do.)

Notice that when it comes to households we suddenly have this second layer kick in; there are pooling effects when it comes to certain cities, and if controlling for urban living for college degrees can absorb large degrees of inequality then we have to worry that that is cutting off people being able to deploy their human capital in an optimal way. This is different than someone having a nice house making others jealous – this is about the actual ability to partake in the economy itself. I can go on at this at length; the key is that in addition to whatever effects “keeping up with the joneses” has, there’s another layer that goes directly to opportunities that is equally valid.

Lifetime Earnings

It’s weird to me how easily lifetime consumption smoothing can be invoked to any debt structure. Is consumption smoothing an assumption on how people behave, or a normative statement how they should behave? Here’s Milton Friedman in 1957: “The permanent income component is not to be regarded as expected lifetime earnings… It is to be interpreted as the mean income at any age regarded as permanent by the consumer unit in question, which in turn depends on its horizon and foresightedness.” It is crazy how far to the right of Friedman a lot of economic commentary has gotten – we know people don’t smooth their income; and if there is uncertainty has to how much money they’ll make they shouldn’t. I discussed this in a post that responded to Conor Clarke on consumption smoothing.

Poor Consumption

People in the bottom quintile famously consume $18,000 while only earning $11K-15K. There’s a lot of debate on how to interpret that statistical point. Some think it is measurement error, some think it is large spending fueled by debt. I also want to point out if some of that consumption is fueled by the illicit underground economy there’s a massive survivorship bias – people that are killed or end up in jail don’t get zeros noted as their consumption, though that obviously influences their payoffs for taking on that risk. I’m not sure where that literature stands.

Questions for Will Wilkinson

1) Ezra Klein and others have already brought up debt as a critique of consumption and the lowest group of consumers. How does the analysis, that consumption has remained stable because it has been fueled by debt, hold up for the middle class? Some claim that, for the groups that consume the least, debt isn’t likely to be large enough to drive instability in the larger economy; that may be true, but it strikes me as less true for the middle-class. Is there a sense of how this question is going to be approached?

2) In so much as the previous analysis is true, that consumption is more or less flat but the riskiness of that consumption has outpaced the earnings, is this a moral or ethical problem? We know that rational agents need to be compensated for taking on more variance in earnings; in so much as the modern middle class has taken on more volatility without taking on more earnings, is there is a moral problem here?

3) I’m not a huge fan of arguments predicated on the statement “the poor are poor because they choose to be poor.” But in so much as the middle-class has become an over-leveraged risky enterprise, that in itself could influence marginal decision making against investing in additional human capital – it may be better to just sit back and buy the IKEA refrigerator instead of trying to invest that money in oneself to move into the middle-class. Does this have moral implications?

I doubt this is a big problem in the general human capital sense, but I imagine it is a huge problem in the decision to have children. Given that having a child is like dropping a volatility bomb into the household (last I read some data it is the one of the biggest predictors of defaults), and there is a story where this volatility is driven in part by inequality, particularly in that housing bundles a child’s education with it, does that have moral or ethical problems?

4) Some people have argued we should cut some of the “financial innovations” of the past 30 years in order to stop a bidding war on housing. Richard Serlin has argued (part one, part two) that we should cap interest rates to keep this bidding war on housing from going out of control.

Assuming it would work (big assumption!), what are the ethical and moral implications in such a move? In so much as it would decrease inequality by decreasing “choices”, that may lead to a wasteful bidding war, what should be our moral thought guiding the decisions in how to move here?

This entry was posted in Uncategorized. Bookmark the permalink.

12 Responses to Thinking Clearly about Inequality

  1. Dismal says:

    Please post that Sweet 16 article! I bet I would agree with it completely. I too am fascinated by that show, though recently it seems to have been replaced by 16 and Pregnant (guess the depression has hit MTV pretty hard).

  2. John McGowan says:

    The quote illustrates a very common conservative, libertarian, and business argument and tactic. The author is comparing our current standard of living to 110 years ago, specifically citing cars and refrigerators.

    Power and propulsion technologies, which play a major if not the major role in the maximum attainable standard of living, progressed rapidly from the late eighteenth century when the separate condenser steam engine was invented to the middle of the twentieth century. Depending on which technology you look at, the power and propulsion technologies either stopped progressing or the rate of progress slowed markedly some time between 1950 and 1970.

    Thus, we have refrigerators today, had refrigerators in 1970, and did not have them in 1900 (most people). Thus, we have cars today, had cars in 1970, and did not have them in 1900 (most people). Thus, we have commercial jet airplanes that fly about 400 mph today, in 1970, and not in 1900. And so on.

    Significantly, concerns about rising income or wealth inequality usually posit an increase in inequality since the 1970’s or 1980’s (actually just after the slowdown in the rate of progress in power and propulsion technology).

    What this means is that if one compares life today to 110 years ago, there are dramatic differences in the standard of living that can swamp income inequality measures. In contrast, the comparison is different if one compares today to 30 or 40 years ago.

    One may also skeptically note that the slowdown in power and propulsion technology progress actually coincides in the US with the adoption of the “free market” and “deregulation” policies that the Cato Institute favors. I would not read too much into this. It may just be a coincidence, but still it is curious. Note that these policies are often justified because they will supposedly unlock the boundless innovation of the private sector, which might — one would think — result in at least the continuation of the historical trend in power and propulsion technology, if not an acceleration.



  3. Taunter says:

    Serlin’s argument manages to be both patronizing and inaccurate.

    It is patronizing to suggest that people should have government-mandated caps on the amount of leverage they can take for individual items (or, presumably, for all manner of consumer goods). In effect, that puts the government in charge of household budgeting, with whatever the social convention might be dictating the allocation to housing, transportation, etc. If you don’t have the freedom to allocate your money how you wish, you don’t have much control over it, do you?

    Beyond the morality of his solution, his diagnosis is terribly flawed. Suppose all of the circumstances he imagines were true: the growth of two income families, the bidding wars, etc. It does not stand to reason that house prices would grow in real terms. Car prices have lagged inflation, and they are just as easily positional goods. The cost of CONSTRUCTING a house has largely tracked inflation over the past couple of decades and lagged inflation before that (the labor component tracks inflation and productivity improvements in pre-assembled components/joists reduce time to build). It is entitled land – and, more precisely, entitled land where Serlin wants to live – that has appreciated wildly.

    With limited exceptions for places such as Phoenix and Las Vegas that grew so rapidly by immigration that the physical grid was maxed out, the prices of entitled land have skyrocketed because of zoning – because all of a sudden in the 1970s there was an alliance between entrenched landholders looking to maximize value, urban reformers (eg Jane Jacobs) horrified by the construction of the 1950s/1960s, and environmentalists (eg California Coastal Commission) looking to minimize development. Although these groups did not like each other, they each found a powerful incentive in using zoning to hold back the growth in housing stock.

    Without touching anyone’s right to spend his money however he chooses, the government could immediately and permanently reduce the price of housing by reducing zoning restrictions. Except people who vote believe high housing prices = good thing.

    • Taunter,

      You write, “If you don’t have the freedom to allocate your money how you wish, you don’t have much control over it, do you?”

      So you oppose any usury laws? Do you also oppose restrictions on buying Meth? That means “you don’t have the freedom to allocate your money how you wish”, at least not completely as with usury laws, so it means, “you don’t have much control over it”?

      Do you oppose laws that do not permit anyone to buy high explosives because, “If you don’t have the freedom to allocate your money how you wish, you don’t have much control over it”?

      Do you oppose laws that do not permit anyone to buy nuclear material because, “If you don’t have the freedom to allocate your money how you wish, you don’t have much control over it”?

      If you’re an extreme enough Libertarian you would oppose all of these laws. If you’re an extreme enough Libertarian you would be willing to allow enormous suffering and lost utility, lost quality of life, to avoid giving up even the tiniest bits of economic freedom that would barely be noticed, if at all.

      I’m obviously not an extreme Libertarian.

      Getting rid of low interest rate restrictions has added fuel to the bidding wars for homes, but that effect is on the demand side. That doesn’t mean there aren’t also effects on the supply side such as the zoning laws you mention. And that doesn’t mean you can’t still lower prices by lowering demand.

      It doesn’t matter how cities are zoned, if interest rates are restricted to moderate enough levels that mortgage providers will not take much risk in lending, and will therefore not lend more than say 2.5 times family income, then that’s the most money a family’s going to have to pay for a home, at least from borrowing. And that’s therefore the highest price that will clear. Increases like we’ve seen recently would simply not have been possible without the great relaxation of interest rate limits; individuals and families just would not have been able to borrow enough money to pay prices that high.

      Harvard bankruptcy and financial distress expert Elizabeth Warren describes this well in her 2005 book “All Your Worth”:

      If your dad wanted to buy a car he couldn’t really afford, he couldn’t get the car loan. If your mom wanted to rent an apartment that was out of balance with your family’s income, the landlord wouldn’t rent it to her. If your parents tried to take out a loan – say, for an addition on the house or just to make ends meet for a while – they had to meet with a banker, face-to-face, to explain why they wanted the money. The banker would have asked for pay stubs, tax returns, and all kinds of financial records, so he could evaluate the prospects for repayment. And if things looked out of balance, the banker would have rejected the loan.
      As a result, in those days it was really really rare to spend too much on the basic monthly bills. Why? It’s not because your parent’s generation was smarter or thriftier or “more in touch with what matters” No, things were different in your parent’s day because the rules were different. Your parents lived at a time when the government strictly regulated the banking industry…As a result, in your parents’ generation there were no zero down mortgages. Almost no one was “house poor”, spending too much on a home or apartment. There were no offers on TV to “cash-out” your home equity. No one had a car payment the size of Texas, and car leases hadn’t even been invented…
      In today’s world, you can get a mortgage that is too big for you – and the banks will help you do it. You can get a car lease that chews up half your income. You can wind up with a student loan bigger than some mortgages. And as sure as the sky is blue, you can rack up credit card debt without blinking an eye, even if you don’t have 50 cents to make the payments. (page 17)

      Families could over time just save more money to add to the mortgage money in their bids, but that would take many years, and in the meantime many sellers wouldn’t be able to wait. In addition, over those years pressure would build to do something about the zoning laws.

      Plus, a substantial factor in home price appreciation is increase in size (especially relative to shrinking family size) and amenities. This is very positional and not affected by zoning laws, although I would also like to see zoning laws for the purpose of driving up housing prices eliminated, except for cases where they serve some other purpose which makes them worth the costs.

      You do note, “Car prices have lagged inflation, and they are just as easily positional goods.”

      It’s an interesting point, but positional/prestige goods externalities can be very large without it meaning that these goods always grow much faster than inflation (or incomes). They simply couldn’t always do this because eventually they would exceed 100% of income. In economics terms, the stock could be extremely inefficiently large, even if the flow rate is average.

      You might ask, why hasn’t vehicle spending increased like housing over the years of relaxed interest rate limits and lending, as they’re a highly positional good too? Well actually vehicle spending has increased faster than it looks – per household – as households now own more vehicles.

      And there are other, different factors going in the other direction. Working against vehicle price increases has been technology improvements bringing down costs of production (whereas there is no cost of producing land for a home). This is a case where a factor on the supply side worked to decrease price, as opposed to zoning regulations in housing, a supply factor that worked to increase price.

      Nonetheless, even with financial distress epidemic, typical families spend for power windows, seats, and many other do-dads they didn’t spend on a generation ago, as well as about 100 more horsepower over the last 30 years.

      In any case, with such increases in home prices, it’s just very hard for families to increase their per vehicle spending that much, especially with substantially increased vehicles per family. And, the supply curve for vehicles is much flatter than the supply curve for homes in good suburbs with other well educated, law abiding families. Thus, the price will change much less due to shifts in the demand curve.

      It’s important to note that not all positional externalities are prestige ones, especially with homes. Positional externalities essentially can occur any time when one persons consumption affects others utility, for example when one person being in a top 50% school district means another person can’t be. Cornell economist, and New York Times Economic View columnist, Robert Frank explains this well in a July 11th column, “The Invisible Hand, Trumped by Darwin?” (at:

      In Darwin’s framework, then, Adam Smith’s invisible hand survives as an interesting special case. Competition, to be sure, sometimes guides individual behavior in ways that benefit society as a whole. But not always.

      Individual and group interests are almost always in conflict when rewards to individuals depend on relative performance, as in the antlers arms race [Individual male elk evolve ever larger antlers to win mating battles, but it makes the species as a whole much less able to elude predators. If there was a regulation that could cut all antlers sizes by half, all would maintain their relative position and mating ability, and all would be far safer and better off]. In the marketplace, such reward structures are the rule, not the exception. The income of investment managers, for example, depends mainly on the amount of money they manage, which in turn depends largely on their funds’ relative performance.

      Relative performance affects many other rewards in contemporary life. For example, it determines which parents can send their children to good public schools. Because such schools are typically in more expensive neighborhoods, parents who want to send their children to them must outbid others for houses in those neighborhoods…to earn extra money for houses in better school districts, parents often work longer hours or accept jobs entailing greater safety risks. Such steps may seem compelling to an individual family, but when all families take them, they serve only to bid up housing prices. As before, only half of all children will attend top-half schools.


      If you still doubt that positional/context/prestige externalities are extremely large, costly, and inefficient, I suggest reading Frank’s book “Luxury Fever”.

  4. Mike says:

    John – Yes, agreed. And I think that some of the goods that have had inequality grow in the past 30 years are things that can’t necessarily be hand-waved by saying the least-good is nearly as good as the most-good; insurance, education, access to places to deploy human capital.

    Taunter – I think Serlin’s argument is to cap the interest rate on a mortgage (usury laws) – you could still take out as much principal as you wanted, and thus adjust accordingly leverage up according. Actually wait, it is to cap as % of income. Ok, so I’m really just thinking outloud about usury laws, and using Serlin as a jumping point.

    “It does not stand to reason that house prices would grow in real terms.”

    Agree with you completely about zoning, but I think schools play at least a part of the role in this; there are handfuls of studies where identical houses within blocks sell for 10-20% more as a result of being in one district over another. Given that education is bundled with housing, and education is widely considered to be one of the more important inputs for human capital formation, it doesn’t surprise me a bidding war breaks out on that axis independent of zoning laws. Returns to higher education are higher; returns to marginal level of education are marginally higher the higher up you go (last I remember the research); college competition (AP Classes, etc.), and with general anxiety about future generational placement it would surprise me if this didn’t drive the result a little bit.

  5. Taunter says:

    Is the problem with education really that we OVERvalue it as a society? Difficult to believe.

    First of all, there is no inherent reason for municipally-administered or municipally-funded education. If the core issue is access to education, that could most effectively be solved by centralizing the education provision function. Other countries without our administrative system – France, for example – have virtually no difference between their schools (it’s an old joke that the French education minister can tell a visitor exactly which page of Voltaire is being discussed at that moment in every fifth grade classroom).

    Secondly, to the extent that there is local variation in schools, one would imagine that reflects the willingness of the local population to fund its schools. So it is rather circular to state that “people with aspirations need to move to X district.” If the SF-LA train were developed and a prosperous commuter village developed in Fresno, Fresno would have better schools – because the residents of Fresno would want better schools.

    If we are talking about a nationwide phenomenon – as opposed to a David Brooks phenomenon that affects attorneys, doctors and journalists in the suburbs of NFL cities who have seen their income position relative to finance decline over the past thirty years – we would expect this pressure to have good schools to result in an arms race among municipalities to provide better schooling; if you can increase your house price by 20%, investment in local education is at least as well spent as investment in a swimming pool.

    Unfortunately, I suspect that most people in the US underinvest in education. Either that or we are naturally a very dumb population compared with other First World countries.

  6. Duff Samoa says:

    Would also be interested in seeing the sweet 16 article as well..

    Although I haven’t watched it for a while (had to look up to remember Hart as his episode was my favorite), it always amazed me how superficial it all was and how clueless everyone on the show actually was.

  7. Clint says:

    I think the biggest problem with Wilkinson’s piece — and with income inequality in general — is that it’s not produced through fair mechanisms.

    Inequality by itself isn’t immoral — so long as everyone begins from an equal starting point. But that’s not the case in the United States. Education, government, Media and other powerful institutions have exclusive access, usually through financial barriers. To the extent that such institutions influence wealth (which is a lot), there is unjust inequality.

    Further, the risks of inequality are enormous. They include starvation, homelessness and death. In a society of great wealth, it’s immoral to not guarantee protections against such existential threats to human life.

    I wrote a longer piece about inequality, partially in response to the Cato Institute piece, available here.

  8. Pingback: links for 2009-07-21 « Embololalia

  9. Pingback: Obfuscating Inequality « The Baseline Scenario

  10. Pingback: Hindsight Is 80/20: Information and the Zero-Sum Society « Generation Bubble

  11. Pingback: Spending Spree «  Modeled Behavior

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