Swedish inequality datapoint of the day

By Felix Salmon
March 25, 2011
NYT, the wealth-inequality survey by Michael Norton and Dan Ariely is back in the news.

" data-share-img="" data-share="twitter,facebook,linkedin,reddit,google" data-share-count="true">

Thanks largely to the NYT, the wealth-inequality survey by Michael Norton and Dan Ariely is back in the news. You might remember it from back in September. Here’s how it was reported in HufPo:

The respondents were presented with unlabeled pie charts representing the wealth distributions of the U.S., where the richest 20 percent controlled about 84 percent of wealth, and Sweden, where the top 20 percent only controlled 36 percent of wealth. Without knowing which country they were picking, 92 percent of respondents said they’d rather live in a country with Sweden’s wealth distribution.

Similarly, Tim Noah, in Slate, said the survey showed respondents favoring “a wealth distribution resembling that in Sweden”. And Chrystia Freeland has the same idea: “Americans actually live in Russia, although they think they live in Sweden”, she writes.

The Norton and Ariely paper is easy to misread in this way. Americans Prefer Sweden is one heading; the text does little to dispel that idea.

As can be seen in Figure 1, the (unlabeled) United States distribution was far less desirable than both the (unlabeled) Sweden distribution and the equal distribution, with some 92% of Americans preferring the Sweden distribution to the United States. In addition, this overwhelming preference for the Sweden distribution over the United States distribution was robust across gender, preferred candidate in the 2004 election and income.

If you look at the referenced Figure 1, it labels three different charts as “Sweden (upper left), an equal distribution (upper right), and the United States (bottom)”. It also comes with a note:

Pie charts depict the percentage of wealth possessed by each quintile; for instance, in the United States, the top wealth quintile owns 84% of the total wealth, the second highest 11%, and so on.

The clear implication is that in Sweden, the top wealth quintile owns 36% of the total wealth, as demonstrated in the “Sweden” pie chart. But that’s not true. Go back to footnote 2 (yes, a footnote), and you find this:

We used Sweden’s income rather than wealth distribution because it provided a clearer contrast to the equal and United States wealth distributions; while more equal than the United States’ wealth distribution, Sweden’s wealth distribution is still extremely top heavy.

This is an important point, which nearly all the discussion of the paper has missed. Mark Gimein has put together the charts showing what the truth of the matter is. The first two charts are reality, while the third is the fictional “Sweden” of the Norton-Ariely paper:




The point here is that wealth inequality is ever and always enormous. The US and Sweden are very far apart, when it comes to inequality, but if you look at wealth inequality rather than income inequality — which is the subject of the Norton and Ariely paper — then countries tend to look more alike than different. A huge part of the population of just about every country is going to have zero wealth — if you live paycheck to paycheck, for instance, or if you’re young and haven’t been earning money for long, or if you just spend a lot. That doesn’t mean you’re poor.

In countries like Sweden, indeed, the social safety net is strong enough that you don’t need to build wealth in the same way you do if you’re Chinese, say. Wealth is a form of insurance, and when insurance is nationalized, you need less wealth. As a result, people can enjoy the fruits of their money, instead of saving it up for emergencies or for retirement — and only a small percentage of the population really spends a lot of effort in a successful attempt at accumulating more.

Indeed, Sweden and the US are even closer together, in terms of wealth inequality, than the charts above suggest: as Gimein notes, the Swedish data exclude money held offshore, the value of family owned firms, and the considerable wealth of super-rich Swedes like Ikea founder Ingvar Kamprad, who left the country to avoid taxes.

Ariely told Gimein that “we created a more equal society than the most equal society in the world,” while calling it “Sweden”. Which might be interesting as an academic exercise, but the message was lost on most of the people who read the paper, and who thought that there really was a society where the lowest quintile owns 11% of the wealth.

Wealth inequality is a problem — but it’s one of those things, like homeownership rates, where public policy only makes a very small difference to some very large numbers. Norton told Gimein that he and his colleagues are now exploring “whether educating Americans about the current level of wealth inequality (by showing them charts and pictures) might increase their support for policies that reduce this inequality.” Well, it might. But it’s important not to mislead people about what’s possible.


We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see http://blogs.reuters.com/fulldisclosure/2010/09/27/toward-a-more-thoughtful-conversation-on-stories/

That is a publishable paper?!? Not a journal I’ve ever heard of, but still… What has happened to academic standards?

Having wrestled with this question a bit myself, I’m still uncertain how to best measure “wealth distribution”. I believe the typical net worth calculation includes DC accounts but not the value of DB plans. It doesn’t count Social Security promises. Doesn’t make any attempt to estimate future earnings potential. The scion of a wealthy family may have nothing in his name during college (the trust fund that provides him with income is a separate entity) but is nonetheless stinking wealthy! And so on…

For many purposes, income is more meaningful than net worth, just as earnings and cash flow are more meaningful than asset prices. And income is more evenly distributed (especially post-tax income).

Posted by TFF | Report as abusive

The top quartile in the United States is a group within which there is much mobility, with folks continually entering and departing. I lack data for both the USA and Sweden on this, but typically Europe has far more old money than does the new world. One can be aspirational in the United States, but not so much in societies that are redistributional in structure and over-regulated in practice.

Posted by OregonJon | Report as abusive

I should add that people typically don’t think things through carefully. Show them a pretty chart and they’ll tell you, “Sure, that sounds like a great idea!”

That doesn’t mean they support the policies that would be necessary to achieve that wealth distribution.

Posted by TFF | Report as abusive

If I understand what you are saying, the research is intellectually dishonest, and the reporting on it journalistically dishonest. It’s one thing to show people graphics of either income or wealth distribution, and ask which society they want to live in (that’s probably reasonable research), and another entirely to ascribe them to countries, then use different measures for those countries designed to maximize the difference (highly dishonest).

Posted by Curmudgeon | Report as abusive

I don’t see what’s dishonest about the paper. They fully disclosed what they were doing. It’s not their fault that most journalists are economic ignorami.

In any case, the take-away from the study is that:

(a) Americans think they already live in a society that is AMAZINGLY equal; more equal than the most equal modern societies on Earth;

and (b) they believe that this kind of equality would actually be desirable.

So the overall thrust of the journalistic commentaries — that the only reason Americans fail to support more progressive taxation and efforts to reverse the concentration of wealth, is that they don’t understand what kind of society they live in — is entirely accurate, even if the journalists are getting a key detail wrong.

Posted by Auros | Report as abusive

Has there EVER been a society with that kind of wealth equality? I suspect that we’ve seen a general growth in equality over many centuries, at least since the invention of the “trades and merchants” class. Possibly reversing a bit over the past twenty years, but not nearly to the extent of the feudal days.

People believe they live in a society of broad equality because they look around themselves and realize that the majority of people are pretty well off. The difference between my own condition and that of Bill Gates is less material than the difference between a lord who had enough to eat all winter and the peasant who starved for 3-4 months of the year.

Posted by TFF | Report as abusive

I was logging in to write almost exactly the first comment TFF did.

If I’m 62 and I get 1500/month from SS and 1500/month from my union pension than I make $36000 a year with zero assets. That’s more income than a million dollars generates in the current rate enviroment.

We need a national savings program. Keep social security payouts just as they are… bust the earnings to pay the benefits. Next step increase payroll taxation to 12% and throw the new 5% into an UNTOUCHABLE retirement account in the name of every worker. Any approved financial company can manage the accounts. Some low cost provider with a strong reuptation like Vanguard can be the default option using active management for 50 basis points.

This way the next time we publish a study about income inequality not just the hard working responsible people will have all the money.

Posted by y2kurtus | Report as abusive

Felix, you’re doing Norton and Ariely a disservice by misrepresenting how they disclose their use Sweden’s income rather than wealth distribution. It is clearly disclosed in the BODY of their article; only the explanation as to why is relegated to the footnote. Here’s the text:

“Americans Prefer Sweden
For the first task, we created three unlabeled pie charts of wealth distributions, one of which depicted a perfectly equal distribution of wealth. Unbeknownst to respondents, a second distribution reflected the wealth distribution in the United States; in order to create a distribution with a level of inequality that clearly fell in between these two charts, we constructed a third pie chart from the income distribution of Sweden (Fig. 1).2″ [The "2" at the end is the reference to the footnote you quote.]

While I think that they should have used scare quotes around ‘Sweden’ to more clearly indicate what they had done, their reasons are well justified as a matter of research methodology (though there is a problem with the methodology of this task that Noah catches).

Gimein says: “The effect of this and other economic research isn’t to prompt debate. It’s to shock people with numbers, even if they’re made up.” But here he betrays his own failure to carefully read their paper. The results of the second task, shown in figures 2 and 3, are based on subjects responses as to ideal distribution, not on Sweden’s income distribution. There subjects preferred an even more equal distribution than the “Sweden” distribution from the first task. Those numbers are “made up” in the slightest, nor do they involve “mislead[ing] people about what’s possible”. What those results show is that the public is already deeply confused as to the realistic range of wealth distribution.

You’re also wrong in your characterization of Freeland’s reporting. When she says “Americans actually live in Russia, although they think they live in Sweden”, she’s clearly referring, not to the preferred distribution results from task 1, but to the actual vs estimated vs ideal results from task 2. As she, correctly, goes on to say, “they would like to live on a kibbutz”, i.e. their ideal is essentially utopian. When she says “they think they live in Sweden” she’s understating their estimation of US wealth inequality, as the estimation is actually more equal than the real Swedish distribution.

Noah does get it wrong, though he has a good criticism of Norton and Ariely’s task 1 methodology. But he misses entirely their ideal distribution results from task 2 that are even more equal than the “Sweden” preference from task 1.

Posted by ginsbu | Report as abusive

Large swathes of people do not, and will not, build wealth, period.

A lot people will locate whatever there is available to consume in the present, they will consume all of it. These people, and there are a lot of them, will never have any meaningful wealth.

Sadly that is the present attitude of the leaders of our State.

Posted by DanHess | Report as abusive

y2kurtus, one of the dangers of that approach is you make it increasingly difficult for RESPONSIBLE individuals to manage their earnings.

For starters, the Republican proposal (and any that is likely to pass) would restrict the investment options to mutual funds — likely low-cost index funds. While this is superior to high-cost index funds (which is essentially how most actively managed funds behave), index funds are a highly risky way to invest. They enforce whole-hearted participation in every bubble, while making it impossible to time holdings to match assets to obligations. If hedge funds were restricted to trading indexes and index futures, they would not make any money.

Second, there are three basic financial truths — college, your house, and retirement. (Because of the time frame involved, their PV isn’t as different as you might think.) There are times when it makes sense to be saving heavily for retirement, but other times when it makes sense to be aggressively saving enough for a 20% downpayment and another 10% as reserves. (Could easily be a full year’s earnings.) The more restrictions you place on the accounts, the harder it becomes to plan for the future.

Roth IRAs are the most flexible account structure I know. I would be interested in seeing statistics — are people as successful in accumulating money in Roth accounts as they are with other structures? (Obviously a strong demographic dependence that would need to be factored out.) If not, then perhaps your proposal is necessary.

Still, it would make yet ANOTHER bite out of the take-home pay. Even if it is money I would save anyways.

Posted by TFF | Report as abusive


I admit that any forced savings program would be a hindrence to any responsible individual who would prefer to wisely manage their own money. My experience as a financial advisor is that perhaps 20% of Americans have saved prudently enough to achive the retirement they envision.

The other 80% would have been better off had the goverment deprived them of a starbux/day for their 40 working years and allowed them a roast beef sandwich/day for their 22 year retirement.

While I’m pesimistic enought to think that any sensable solution is unlikely I don’t see why a national 401k style system could not mandate the savings as a % of earned income and deny all access to the money except for the most extreem circumstances and leave the money management up to the individual.

At scottrade there are restrictions to what I can invest my ROTH IRA in. I can’t use margin (probably a good thing.) I can’t buy options (probably a good thing.)

Most people would probably keep their money parked in whatever the default option was. Yet if you use a balanced fund as a default then the manager can swing at least some of your money between stocks and bonds if they feel a bubble brewing.

A well run national savings plan is probably wishful thinking but like getting HIV through unprotected sex, having a heart attack from being 100 pounds overweight, or getting lung cancer from smoking, the solution to the problem exists… just a matter of early education and willpower.

Posted by y2kurtus | Report as abusive

Any word what Swedes would prefer?

The way to compare income and wealth is to measure what wealth is capable of safely generating, 4 or 5%.

Posted by MyLord | Report as abusive

This is only the start of the problems with this survey. O blogged about it. http://lennartregebro.wordpress.com/2011  /04/15/does-americans-really-want-swede ns-wealth-distribution/

Posted by LennartRegebro | Report as abusive