Beware of income inequality data

A few years back I got so exasperated reading a Journal of Economic Perspectives piece on income inequality (by Emmanuel Saez and Peter Diamond) that I did a post calling it “propaganda.”  I probably shouldn’t have used that term, but I was reminded of my frustration when reading a very good Alphaville post by Cardiff Garcia:

The issue of whether US inequality has climbed since the recession of 2008 has been relitigated this week. A short analysis by Stephen Rose claimed that income inequality had actually fallen, assigning the credit to public policy.

David Leonhardt of the New York Times discussed Rose’s findings, followed by further analyses and critiques from Ben Walsh and Nick Bunker. I’ll present the findings first before adding my own thoughts at the end.

Mainly in response to the heavily cited claim by Emmanuel Saez that 95 per cent of the income gains in this recovery have gone to the top 1 per cent of earners, Rose emphasizes a couple of broad points.

There are two problems with the 95% claim, one has already been discussed by David Henderson, while the other is often overlooked.  David pointed out that when evaluating income equality you want to remove cyclical effects, as it’s a long term problem.  It’s not unusual for the share of income going to the rich to fall during recessions (as capital gains plunge), and then rise during expansions.  It would make more sense to compare 2014 to a year with similar unemployment, say 2004.

The less often discussed problem is that talking about shares of growth can be very misleading, especially when growth is slow. I’m going to give an extreme example, just to make the point more obvious.

Suppose nominal income and the CPI rose at roughly the same rate between 2004 and 2014.  In that case real income would be roughly unchanged.  But let’s also suppose it wasn’t completely unchanged, just roughly unchanged. More specifically, assume real income rose from $15,000,000,000,000 to $15,000,000,010,000.  That is, real income rose by $10,000.

Let’s suppose that in 2004 Ray Lopez worked at a car wash in LA, making $10,000/year.  In 2014 he had two car wash jobs, and was working much harder. Assume his real income had risen to $18,000.

Now here’s my question:  Is it accurate to say that between 2004 and 2014, 80% of the entire the gain in real income for the United States of America went to Ray Lopez, car washer in LA?  You’re damn right it’s accurate!  And I’m willing to assume that the cited claim by Saez is also accurate.

But there’s another question that goes beyond accuracy; is it misleading?  To me it’s obviously misleading to say that one car washer in LA received 80% of all the real income gains in America, even if my hypothetical data were true. That’s because one could say the same thing about his cousin, if she had gone from doing one house cleaning job to two, with the same $8000 gain in real income.  Indeed I would have earned more than 100% of all real income gains, as my real income rose by more than $10,000 over that decade.  Any time an aggregate doesn’t change very much, but there are significant changes to the components within that aggregate, there are lots of ways of slicing up the data to create misleading impressions. Presenting data that way may not be propoganda, but it certainly does more to confuse than enlighten.



22 Responses to “Beware of income inequality data”

  1. Gravatar of Pemakin Pemakin
    25. February 2015 at 05:25

    Nice illustration. How does Ray get the time to write these misguided comments while working two car wash jobs?

  2. Gravatar of Benjamin Cole Benjamin Cole
    25. February 2015 at 05:36

    Anyway, tax consumption, not income.

    And remember KISS.

    How about a 20% national sales tax on all consumption above $40,000 a year per household? And that’s the only tax, at the federal level. Maybe some fat Pigou taxes (all drugs legal) and gasoline taxes.

    You get a rebate check if you spend less than $40,000 a year. That’s because I like the social fabric, and I am a softie.

    But not that soft: If Warren Buffet lives modestly, spends $39,000, then zero taxes on him!

    Anyway, stop worrying about income inequality. Worry about how to get red-hot honking on fire aggregate demand going, and Fat City.

    When there are jobs a-plenty, I will say to anyone, “Oh, stop sniveling and get a job. Marry someone with a job, and live frugally. You will do fine. It’s Fat City, and they need you working.”

  3. Gravatar of Charlie Jamieson Charlie Jamieson
    25. February 2015 at 06:21

    A tax on consumption would devastate the economy. They increased taxes on boats in Illinois one year (great idea! soak the rich!). But it wound up costing jobs for all the people who work in the boat and fishing industry and was quickly released.
    We need to get Warren Buffet to *spend* his money, or else take it from him, instead of just endlessly trading financial assets to increase his personal net wealth without making anybody else better off.

    The car wash illustration is silly. Ray’s income has fallen. He has doubled his jobs but not doubled his income.

    Wealth inequality is a growing problem. It’s a problem of inefficiency. For every dollar created we are generating less and less economic growth. This is because more of the money creation is going to people who don’t spend or produce.

  4. Gravatar of Major.Freedom Major.Freedom
    25. February 2015 at 06:26

    The Ray Lopez example is not useful, because for one thing, it is an example of “most gains going to the 99%”, which is exactly opposite to the very claim being made about income inequality. It is not that the content of the example is “misleading”, it is that it is merely a negation of the argument being addressed. Tantamount to a response of “Nope”.

    Tweak the example. Suppose that there are 100,000 Ray Lopez’s, each earning over $200k a year. Now take your scenario of GDP rising by a historically consistent long term growth rate, say 2%, e.g. from $10 trillion to $10.2 trillion first year, then about $10.41 trillion the year after, etc.

    In two years real growth was about $410 billion. Now would it be “misleading” to refer to the fact that the top 1% became say $400 billion more wealthy while the 99% became $10 billion more wealthy?

    Fact are not “misleading”! What is misleading is how such facts are presented, in what context. You did not even address the context. You proposed an example of a tiny growth accruing to a single poor person. That is not misleading, it is just empirically vapid.

    Inflation increases income inequality, and inflation is a political creation

  5. Gravatar of Ray Lopez Ray Lopez
    25. February 2015 at 06:58

    Fair point by Sumner, akin to those stats that say: “If an investor missed just 40 of the biggest up days in the market over the last 20 years (1987-2007), their return would have totaled 3.98% versus remaining fully invested and achieving an average annualized return of 11.82%.” The point being that in a composition of data there are individual components that can mask or mislead about the composition.

    I wish however Sumner would discuss why he feels monetarism works, and why he feels money is not neutral. Please professor, cite the data that shows the economy or market responds to the Fed long term. Just today, the economist who ‘discovered’ Sumner, Tyler Cowen, said, as an example of a truism, “During the upward phase of the recovery, monetary policy just doesn’t matter that much. ” – See more at:

    If the learned professor Cowen disagrees with the learned professor Sumner, one of them must be wrong. Who is it?

    As for my personal finances, as I said I’m comfortably in the 1% due to my parents net worth, so, absent being disinherited, don’t cry for me (Argentina). I’ve made, from my own business (online) and working for others, as little as $0 (in 1999, when everybody and their grandma was employed, but I was starting my own business) to as much as $150k, and last year I made about $12k, which is three times the average a worker makes full time in the Philippines, and I’m about to buy a condo here in PH, for the princely sum of… 200k (pesos, or about $5k). I have lost a lot of business when I relocated overseas, due to prejudice from US clients who thought I was ‘off-shoring’ and compromising on quality, but I love being my own boss. I’ve not worn a suit and tie and gone into the office in over a decade… priceless.

  6. Gravatar of Patrick R. Sullivan Patrick R. Sullivan
    25. February 2015 at 07:40

    The newest JEP gives Thomas Piketty another drubbing. As Timothy Taylor summarizes it;

    ‘In case you didn’t catch all that, Piketty is noting that r>g is not useful for discussing income inequality, and does not necessarily lead to wealth inequality, and that the future of wealth inequality is highly uncertain.’

    And in another interview, Piketty let’s it all hang out;

    ‘I think we also need to have new forms of governance and capital ownership. For instance, in the book I mention the difference between the private and social value of capital in corporations …. Therefore developing new forms of ownership, new forms of sharing of power between those who own capital and those who own their labor, is extremely important to me. …. We need progressive taxation of private capital, and at the same time, a new thinking of what capital ownership means and how we organize its owners.’

  7. Gravatar of ssumner ssumner
    25. February 2015 at 09:09

    Ray, You said:

    “I wish however Sumner would discuss why he feels monetarism works, and why he feels money is not neutral. Please professor, cite the data that shows the economy or market responds to the Fed long term.”

    I have 100s of posts showing the short run non-neutrality, including my newest. As for the long term, well then money IS neutral.

    Patrick, Thanks for the tip.

  8. Gravatar of Doug M Doug M
    25. February 2015 at 09:39

    Money is not neutral, even in the long run!

    In the classic thought experiment, suppose everybody woke up one morning and everyone had twice as much money in their pocket. What would happen… If money is neutral, all prices and wages double, and life goes on as if nothing had changed accept the numbers…

    Buy my debts don’t revalue when money is created…debts are priced in nominal dollars. If I had a lot of debt, I am relatively wealthier.

    And If I have a lot assets, I now have a looming capital gains tax liability, and I am relatively poorer.

    Regarding Ray Lopez, I don’t suggest you don’t give a character in your hypothetical the same name as one of your regular commentators.

    Income inequality is BS. Wealth inequality is slightly more relevant. But, the dynamics of inequality are far more relevant. How many second-generation billionaires are there? A couple of Waltons, a Getty, any others?

  9. Gravatar of collin collin
    25. February 2015 at 10:04

    Wow! Inequality going down! Avoiding measurement issues, I am not surprised it might be controlled the next decade: Why:

    1) Labor participation has dropped and does not look like it will recover very quickly. The skilled labor market is tight and it started to impact younger workers and lower wage positions.

    2) The main reason US inequality shot up in the Bush years was the explosive increase in wage and working with India and China. As their wage are closer to lower US wages, we will witness on shoring. Although it has not happened with manufacturing, but offices are bringing back some positions.

    3) There is a growing shortage of a lot of blue collar working class positions. (A lot Diesel mechanics are retiring right now.)

  10. Gravatar of Charlie Jamieson Charlie Jamieson
    25. February 2015 at 10:19

    Wealth and income inequality matter in two ways:
    – social and political: if the labor and productivity rewards of the 95 percent flow to the 5 percent, there will be unrest.
    – rising inequality leads to stagnant growth. We are also seeing this today. This is because owners of large piles of financial assets are not putting their money to work and circulating it in productive ways.

  11. Gravatar of Ryan S Ryan S
    25. February 2015 at 13:18

    For my research last semester, many journalist and economist were theorizing without facts or good data. There were several issues but to name a couple:

    1. Economist have looked at data and incorrectly when assuming that Between-country income inequality has been rising over the last 20 years (rich nations are pulling way ahead of poorer ones). There is evidence that a large number of “poor” nations are falling behind. However, a majority of those “poor” nations contain about 10% of the world’s population. However many nations in Asia have caught up significantly to the richer countries, and they contain 40% of the world’s population. The data should be weighted to reflect that global inequality has fallen steadily for the overall world population.

    2. Gini data is very inconsistent and misleading:
    This measurment is a “catch all” number for inequality so there are naturally going to be imperfections. More importantly, different countries have different metrics and methods for gathering gini data. This makes it difficult to gauge how “unequal” America actually is compared to the world. Most importantly, this data includes people who work 40 hours a week and those who work 1 hour a week. Gini is typically an overestimation of the nation’s income inequality. Looking at people in the labor force that are full year, full time workers (at least 30 hours a week year round) we see much less inequality.

  12. Gravatar of ssumner ssumner
    25. February 2015 at 15:02

    Doug, You said:

    “Regarding Ray Lopez, I don’t suggest you don’t give a character in your hypothetical the same name as one of your regular commentators.”

    Don’t not confuse me with double negatives.

    Ryan, Yes, income inequality data is highly unreliable.

  13. Gravatar of Doug M Doug M
    25. February 2015 at 15:18

    “Owners of large piles of financial assets are not putting their money to work and circulating it in productive ways.”

    They certainly do. Those financial assets provide the capital for businesses to build offices, equip factories, expand, grow, produced, hire, etc.

    They also provide the capital for our our Government to function.

    They also give middle class people the ability to buy their own homes.

  14. Gravatar of Mark Pash Mark Pash
    25. February 2015 at 15:51

    There are many reasons why inequality exists in an economy. Before I give my laundry list, I will discuss some general macroeconomic reasons. One, we are NOT created equal! We have many different talents, abilities, appearances along with unique emotional and physical makeups.

    Second, the institution of capitalism-competitive markets by its sheer nature of operation has a natural flow of money to the wealthier. The main reason for this flow is that it is easier to make money if you already have some money. This goes for businesses as well as individuals. This naturally causes a shift towards business consolidation creating monopolies or oligopolies.

    For those doubters, I have never seen a statistical, economic study proving otherwise. If you do a little macroeconomic historical reading of the last 150 years, you will determine this yourself.

  15. Gravatar of Liberal Roman Liberal Roman
    25. February 2015 at 17:17

    I see these composition vs aggregate errors so many times. Like when they say “Well, if you take out Apple, S&P earnings were only blank” And it’s always some low number.

    Or when they say “If you take out Texas, US job growth is only blank”

    If Apple is #1 in total earnings growth, I am sure if I take out #2 and run the same “analysis” I’ll find equally “shocking” results.

    Lies with statistics is what I call it.

  16. Gravatar of Steve Steve
    25. February 2015 at 19:42

    If you want to see extreme inequality, look at the share of real income gains over the last decade that went to people who were teenagers in 2004. AGEISM!!!

    And don’t get me started on the share of the national debt increase that was caused by people born in the 1940s.

  17. Gravatar of A A
    25. February 2015 at 20:34

    Charlie Jamieson,

    Warren Buffett already spent the money. That’s how he got the shares. If financial transactions are trivial, then what is the benefit of claiming his shares? There is a weird anti-buyback argument along these lines, where it seems like the authors want to lower real required returns, but they don’t know it.

  18. Gravatar of Charlie Jamieson Charlie Jamieson
    26. February 2015 at 04:30

    Financial wealth is not trivial, as the gains to the wealthy are mainly in the form of financial assets. Those asset prices are actively supported by the central bank.
    Remember when the values of real estate fell, the Fed allowed the owners of real assets — houses — to be liquidated, but bought the financial assets leveraged on those real assets at their par value.
    And as financial assets are a claim on other people’s labor and productivity, there is something very unstable about today’s system.
    For example, a lot of the profits created at Coca-Cola accrue to Warren Buffett, even though he does not make, manage, market or sell the product.
    He’s made a living in the secondary markets … hasn’t created many businesses himself…
    To Doug M’s point, buying shares in the secondary market is saving, not investing. It drives share prices up but it’s not new equity. The buying and selling of existing shares doesn’t create new economic activity except for those in the financial industry.
    Capitalism is a great system, but as Mark points out, money flows to the wealthy unless something is done to break it up every now and then. Otherwise you get the French Revolution and/or the Great Depression.

  19. Gravatar of A A
    26. February 2015 at 07:27

    Charlie Jamieson,

    If the movement of existing shares is economically neutral, then the following statement doesn’t make sense: “We need to get Warren Buffet to *spend* his money, or else take it from him, instead of just endlessly trading financial assets to increase his personal net wealth without making anybody else better off.”

    If you claim his shares for yourself, your money increase is a money drain from the liquidity providers. There is no prima facie welfare benefit from “spending” the shares. Maybe you are really arguing for different priorities amongst the owners of capital. But try to unpack what you think you mean by “spending” the wealth. Are you assuming that existing resources are misallocated, so your hypothetical control over a liquidity pool is welfare enhancing?

  20. Gravatar of TallDave TallDave
    26. February 2015 at 10:49

    Yes, exactly, there is not a giant fixed pile of income that people show up and take a slice of, plus the top 1% are a different group of people every year.

    Scott starts to peel back the onion, but if you go a couple layers deeper you almost immediately realize that without the people in the top 1%, there would have been no income gains at all.

  21. Gravatar of Charlie Jamieson Charlie Jamieson
    26. February 2015 at 12:46

    ‘but if you go a couple layers deeper you almost immediately realize that without the people in the top 1%, there would have been no income gains at all.’

    Sorry, but there was X amount of income gains generated, and the top 1 percent took it.

  22. Gravatar of Check your assumptions: Inequality Data | For My Benefit Check your assumptions: Inequality Data | For My Benefit
    27. February 2015 at 10:46

    […] From Scott Sumner: […]

Leave a Reply