Measuring Wealth Inequality

March 29, 2014

The sharp rise in income inequality in the United States is well-established. But what about wealth inequality? Income represents the flow of cash that a household earns every year, whereas wealth is the total stock of assets that a household owns, either through accumulation or inheritance.

Wealth is as important as income for thinking about overall well-being. For example, wealth may be more important than income in predicting who can send their kids to an expensive college. And wealth also represents control. Corporations are controlled by shareholders. So a higher concentration of wealth naturally implies that fewer individuals control the decisions made by firms in the economy. Similarly, non-profit organizations (including universities) and political parties pay special attention to their wealthy donors.

How has wealth inequality changed over the years? This has been a difficult question to answer in the past because wealth is highly concentrated to begin with, and we do not have good time-series data on the wealth holdings of the very rich. For example, data sets such as the Federal Reserve’s Survey of Consumer Finances (SCF) do not capture the super-rich.

Emmanuel Saez and Gabriel Zucman have preliminary work that approaches this question from a new angle. We want to emphasize that their work is preliminary research – and initial results may change as the researchers take a closer look.

Here is the bottom line from the preliminary findings: the top 0.1% of the wealth distribution has seen a dramatic rise in the fraction of total wealth held, rising from a steady level of 10% from the 1940s to the 1970s, to over 20% in 2013. Here is the key chart:




The top 0.1% have seen incredible gains over the past 30 years that have take them to the same fraction of national wealth that they enjoyed in the 1920s.

The other interesting finding in the Saez-Zucman study is that the increase in wealth is primarily about the top 0.1%. When we look at the top 1% excluding the top 0.1%, there is no gain. Here is another chart:




The gray and black line show that the top 1% to 0.1% have not seen large increases in the share of total wealth. Instead, the rise is completely driven by those in the top 0.1%. These are the very richest households in the country.

Simple measurement is often a bit boring, but it is also absolutely crucial for thinking about the overall economy. Saez and Zucman have taken an initial step toward measuring wealth inequality in the United States, and it shows a rising amount of inequality driven by the very top of the wealth distribution.

Here are some more details for those interested. The basic methodology used by Saez and Zucman exploits the fact that we can measure quite well from the IRS the flow of income to individuals generated by assets. The income flows would be dividends, interest payments, or rental income from real estate owned and rented out. We actually used a similar technique in previous research to get the net worth of a zip code.

The key difficulty then becomes capitalizing these income flows to get a value of the underlying assets generating the income. Capitalization requires accurate measure of the flows of income generated by assets, and assumptions on the rate of return we should expect the assets to generate. This is the crucial part of the Saez and Zucman study, and one likely to get the most scrutiny. A clever test the authors did was to examine how their capitalization technique works for wealthy foundations, for which they have both the income and wealth. In other words, they can test their methodology on a sample of IRS returns where they actually know wealth. It does pretty well.

Please see the slides for more details.


Bookmark and Share

29 Responses to Measuring Wealth Inequality

  1. Chris on March 30, 2014 at 12:30 ami

    Ok. But apart from being envious of those who have more wealth, why should anyone else care? What effect, negative or positive, does this changing relative distribution of wealth have on the rest of us?

    • Steve Roth on March 30, 2014 at 12:58 pmi


      Here’s one answer. It’s only one economic effect, but it appears to be extremely powerful. And it’s inexorably arithmetic:

      N.B. I’m a rich serial entrepreneur. This is not about envy. Straw man.

      • Chris on March 30, 2014 at 10:03 pmi

        @Steve: You have a very admirable bio. Very inspiring.

        I always benefit from having knowledgeable people within my field scrutinize my work. They can almost always find something wrong with it, and sometimes major issues. Since I’m not an economic modeler (I’m a quantitative portfolio manager), I’m not able to competently analyze your model. But I’d bet that if you presented it to a sharp group of expert academic economists they’d find some issues with it, and maybe even some major issues. I assume that your work hasn’t undergone a peer review and publication process. As a non-expert in this kind of thing I have little choice than to rely on experts, so there’s not much weight I can give either way to your analysis. Though there are some other, non technical things I could say about it.

        You’re analysis suggests that I would be better off financially if much of what my richer neighbor has was forcibly taken from him and given to me, not as a safety net to catch me should I fall, but simply as a way to ‘even things up’ between us. A few problems with that:

        Most fundamentally, your entire analysis assumes a utilitarian standard, as if this were not problematic. But it is. I understand that utilitarianism is a common assumption in economics, but it’s not one that – when taken to its logical limits – matches up with how most Americans think about what is fair and just. Assuming my neighbor acquired what he has through fairly dealing with others and not through some criminal or crony capitalist endeavor, then I don’t see how taking it for me could be considered fair or just, or how I could be said to deserve having it. That’s just not how we really look at things.

        Also, your suggestion that no one will be worse off with such a radical redistribution scheme doesn’t line up with the hard facts. Those countries with weaker property rights and more redistributive policies tend to have lower overall economic growth. Yes, the rich are poorer, but the poor are poorer too. There are at least two likely reasons I can think of as to why this is so.

        The first has to do with incentives. Your model doesn’t include at all the likely damage such a scheme would do to the incentive for people to work and invest, which means you would undoubtedly get less of both in the future. Maybe you’ve never been poor (I don’t know), but I have. The powerful effects of economic incentives are crystal clear in my own life and in nearly everyone else I’ve ever met in the real world. I believe that muting incentives leads to less wealth being produced.

        I also wonder how the consumption of capital would affect the growth of future increases in productivity and production. It seems to me you’re assuming that eating the seed corn will have absolutely no effect on our future ability to grow more corn. Does that make sense?

        I know very little about economics, but I do know enough to know that you’re assuming a Keynesian framework that rejects the idea that supply creates demand. And that that’s still not a settled issue.

        Overall, I struggle to see how this would really help us. Certainly the possibility of the American people accepting such a blatantly redistributive system is not likely.

        • howard on March 31, 2014 at 10:10 ami

          i’m sorry: supply does not create demand, so it’s true that you know very little about economics.

          still, let’s talk about incentives, shall we? believe it or not, people still tried to get rich in the 1950s, when marginal taxes were draconian at upper levels. Even picketty and saez, who have done the leading research work in this area, suggest that disincentives kick in at 70% tax rate, and we had 90% rates.

          now i happen to believe in incentives too, but the idea that higher taxes would in any sense destroy incentives is completely unfounded.

          as for the impact of wealth concentration? let’s start with a very first order correlation: notice when wealth concentration began to increase: the precise same moment that median household wages began to stagnate.

          no, i do not think these are unconnected.

          • marris on March 31, 2014 at 12:39 pmi

            > when wealth concentration began to increase: the precise same moment that median household wages began to stagnate

            Have median household wages stagnated in developing countries over this period. If not, then consider that the “opening up” of countries like India and China has increased the supply of labor.

            Since your analysis is framed around the idea that *they took our stuff*, maybe you should consider whether *those people* took your stuff. IMHO, good for them, since they were poorer than you to begin with.

        • Mitch on March 31, 2014 at 12:02 pmi


          Arguing against redistribution, you say this:

          “Assuming my neighbor acquired what he has through fairly dealing with others and not through some criminal or crony capitalist endeavor, then I don’t see how taking it for me could be considered fair or just, or how I could be said to deserve having it.”

          I think you are missing the fact that many of us do not believe that today’s wealth, for the most part, has been acquired via the fair dealing you suggest.

          We look, for example, at the way in which the banks were protected with taxpayer money while mortgagees were left to bleed out in the streets.

          But more basically, as we come to realize that what was once thought of as creating wealth is now revealed as extracting limited resources and making free use of limited public sinks, we find it harder and harder to see the fig leaf that you dangle.

          • Chris on March 31, 2014 at 5:37 pmi

            If the problem is cronyism as you suggest then i absolutely agree it should be eliminated. But then the problem isn’t primarily about inequality is it? It would be about rent seeking, and so the solution isn’t redistribution of income and/or wealth but constraining government’s ability to favor some businesses. It would be absurd to claim that every rich person acquired their wealth through special government-granted privileges. To punish honest hard-working people for succeeding because some rent seekers got rich would be unjust.

      • Chris on March 30, 2014 at 10:16 pmi

        BTW, I think you mean ‘red herring’ and not ‘straw man’, which is an argument in which one argues against a misrepresented version of the opposing position. I neither represented a position nor argued against one. I merely asked a reasonable question.

        • JJ on March 31, 2014 at 8:40 ami

          I was going to say “red herring” as well

    • Burzghash on March 30, 2014 at 1:14 pmi

      You should care, because these moneyed interests continue to buy off our entire political process. They use their vast wealth advantage to bend everything from laws, to judges, to tax code, to corporate personhood to their advantage. At this point, you and your economic class are no longer represented by your leaders, who only have an interest in raising enough money from rich donors to get reelected.

      To paint it as being a matter of “Hurpa durp you’re just jealous!” is the assessment of the mentally vapid and intellectually lazy, and demonstrates just how good moneyed interests are at getting you to regurgitate exactly the message they want you to. “There’s no problem here, you’re just jealous!”

      How embarrassing we have become intellectually as a country.

      • Number Six on March 30, 2014 at 5:12 pmi

        Spot on!

      • Chris on March 30, 2014 at 10:06 pmi

        My “economic class?” I grew up for much of my life as the child of a poor single mother and started my adult life as a high school dropout with so few economic prospects that I had to join the army. I eventually got an MBA from the school at which Amir currently teaches. So what’s my “economic class”?

        The notion that I should think of myself as a member of some static “economic class” is childish and insulting; it denigrates who I am as an individual. It’s ironic that same the people who harp on about economic classes are the same people who never tire of working to remove the very incentives and economic freedom that poor kids like me relied on to move out of poverty. It’s as if they have time travelled straight out of the middle ages, with its serfs, lords, and guilds. If you actually want to help people move out poverty, then give them a very good reason to do so and get as much of the beaurocratic crap out of their way so they can more easily do it.

        • Richard on April 1, 2014 at 1:59 pmi

          By traditional sociological standards (Veblen, among others) your social class is the one you grew up in. Hence the snide remarks to the nuveau rich, the middle-class insecurities of the upwardly mobile, the hauter of those raised wealthy who are now poor.

          Social sciences like psychology and sociology feels like slumming these days, largely because econ “feels” more quantitative (even if disciplines like psychology have *real* experiments and sociology has r-squareds which econ sometimes doesn’t approach). But spend some time in the slums: it will do you good.

      • Chris on March 30, 2014 at 11:02 pmi

        BTW, I asked a perfectly reasonable question. Personal insults are not a response someone uses who has facts and logic on their side. Just more evidence that those who advocate collectivistic, redistributive policies care nothing at all about actually promoting civility in our society.

        • ottnott on March 31, 2014 at 12:01 pmi

          The broad answer to your question is that this enormous concentration of wealth is leading to an enormous concentration of political power that can destabilize a democratic country.

          The narrower answer to your question is to point out that “taking” and “redistribution” comes in more forms than you have acknowledged here.

          When industries successfully block regulations so that they can continue to make our air/food/water less healthy for us and products more dangerous, who is doing the taking?

          When productivity gains become uncoupled from average wages, so that labor is awarded a diminishing share of the value of its output, who is doing the taking?

          When dirty, toxic facilities are easily pushed out of the backyards of the wealthy and into the neighborhoods of the poor, what is being redistributed?

          When the rich receive “probation with counseling” for horrendous crimes (google DuPont heir raped 3-year old for today’s example) while the poor end up in jail for minor crimes or even just for being poor (google Colorado cities jail poor who can’t pay fines for minor offenses for one example), what is being redistributed or taken?

          What does this have to do with the wealth of the 1% or 0.01%? That group controls the major media outlets and is exerting large and growing control of the political process that determines who will be writing the laws and regulations for us.

          Just this weekend, two leading GOP contenders for the GOP Presidential nomination traveled to a single billionaire in Las Vegas to court his tens (in 2012) and possibly hundreds of millions of dollars in support for the 2016 election. This man got wealthy from building and controlling large gambling enterprises — not the sort of business known for providing society with life-enhancing innovations.

          Christie and Walker are not courting this man because he is recognized as having particularly valuable insight into domestic or foreign affairs. They are courting him, because he is willing to apply massive amounts of money to influence our country’s political leadership. In 2012 he gave the money to Gingrich-affiliated organizations. He has made it clear that he plans to keep the money under his own control going forward.

          In terms of political voice, that is quite a redistribution.

      • JJ on March 31, 2014 at 8:41 ami

        Intellectually lazy argument at least , morally corrupt argument at most

  2. Number Six on March 30, 2014 at 8:50 ami

    For there to be large, vibrant middle class, income (or wealth) distribution needs to more closely resemble a bell curve (few poor/many in the middle/few rich) than the present downward slope (many poor/some in the middle/few rich). If a healthy middle class matters for the sake of a civil and just society, then redistribution is the only alternative to torches, pitchforks, and guillotines.

    • Chris on March 30, 2014 at 10:07 pmi

      I’m all for a strong middle class. But I’ve noticed that the most uncivil and unjust societies are those in which the citizens have very weak or non existent property rights, where others are viewed more as potential drains on their efforts and ambitions than as a means to advance them. It seems to me that what matters most fundamentally for a civil and just society is the recognition that people have a right to pursue what they perceive to be good for them through voluntary interactions with others and to keep the results of such activity.

      • ottnott on March 31, 2014 at 5:01 pmi

        In contrast, I find that the most uncivil and unjust societies are those in which the citizens have very weak or nonexistent rights.

        Their rights to their property are a secondary matter, because they lack so many of the personal rights needed to enjoy life itself, much less to acquire and enjoy any property.

  3. dwb on March 30, 2014 at 9:35 ami

    Top 0.1%: Well, duh, lots of billionaires who got their billions in new technology industries with monopoly like characteristics (Microsoft’s Gates and Ballmer, Zuckerberg, Ellison, Jobs, etc).

    • KR on March 30, 2014 at 8:36 pmi

      Which of course brings the question – would you rather have “equality” or PCs, iPhones, and Facebook?

      • invhand on March 31, 2014 at 12:02 pmi

        Really, that is not the question. There are at least 3 legitimate questions:
        1) Would Steve and Woz or Mark Z really have been less likely to make their contribution if they could only expect half the money? This is a hard question to answer because we never get to see the counterfactual.

        2) How much extra efficiency have we got in our capital markets because investment bankers (construing the term very loosely) make a lot more money than they used to? It is not just the entrepreneurs who make more than they used to!

        3) How much drag on the economy comes from the increased power of vested interests to defend those interests?

  4. TMS on March 30, 2014 at 2:09 pmi

    If wealth shares are measured using capital income flows from IRS data does that not neglect capital income that is not federally taxed (e.g. assets held in pensions, IRA’s, Roth accounts)? If the share of nontaxable accounts grow over time, isn’t the share of wealth owned by lower groups misrepresented in any method that uses IRS tax data to measure income flows from wealth?

    Also, using taxable income flows to estimate wealth shares avoids the fact that taxable income has an elasticity that changes in response to changes in marginal tax rates. For instance, the 2003 tax change that reduced the top rate on taxable dividends from 39 percent to 15 percent led to surge in personal dividend income in the NIPA data and by extension IRS data. Measuring this surge in dividend income and then applying it to a corresponding increase in wealth share is obviously incorrect.

    It seems to me that if you construct wealth share data from IRS tax data that there is an awful lot of phantom gains and losses in wealth share that will arise due to changes in income tax law. Those who face the highest marginal tax rates on capital income clearly alter their realization of capital income in response to changes in marginal tax rates.

  5. Jim on March 30, 2014 at 2:49 pmi

    On income inequality, Saez and Piketty’s results are not as cut-and-dry as many believe.

    See this paper by Armour, Burkhauser, and Larrimore (or their more recent paper):$FILE/A01_Larrimore.pdf

    After accounting for tax/transfers makes income inequality trends look much different. Many people make policy conclusions using pre-tax/transfer income definitions, which doesn’t make sense.

  6. Jay on March 30, 2014 at 2:52 pmi

    You write this post as if the households in a given sample in 1960 are the same households in a given sample in 2000.

    What is P(in top 0.1% | in top 0.1% T years ago)? T = 5, 10, 15, etc?

  7. Gus Halberg on April 1, 2014 at 6:08 ami


    You sound like an intelligent person, so I am boggled when you ask ‘why should anyone care’, which is more or less ‘what difference does it make?”

    Is that a serious question?

    Look, if money were only money, you would be correct. But it’s naïveté of a staggering level not to realize that money is power. Or, more correctly, Power. Wealth differentials create enormous power differentials, especially within democratic societies. Go back and read ‘Wealth of Nations” and you’ll see that Adam Smith understood this very clearly. The sort of government interference in the markets that he disparaged came about via the capture of government by monied interests: what we call ‘crony capitalism’.

    So wealth disparities should concern everyone. As money is concentrated, so is power, and those with money are not accountable to anyone.

    Second, your bit about property rights is also a bit of a red herring. Or at least a bit of verbal ‘sleight of hand’ (mixed metaphors, anyone?) By ‘property rights’, what you mean is ‘the right not to be taxed by the government’. Sorry. There is no such right. You want a civil society, with roads and schools and, well, property rights, you gotta pay for it. And those who benefit the most should pay the most. The biggest beneficiaries of civil society are corporations who need the roads to ship their goods, the schools to produce an employable workforce, and, biggest of all, a military.

    This latter is what allows the free flow of goods around the world, It allows companies to increase profits by shipping jobs to low-wage places and then having the confidence that those goods will reach the US unmolested by pirates or piratical states. Remember the shores of Tripoli? So, corporations have an obligation to pay for that military to a level above and beyond what the average guy should pay. After all, the corp is benefiting, while the average guy has probably lost his job because of that military security.

    So, as for property rights, in the “ancien regime” of France, the wealthy were exempt (mostly) from taxes. Is that the sort of society you want to recreate? No, that would be a straw man. But this state of low taxes and small government was the norm for most of this country’s existence. It didn’t work. Would you prefer a society like the US in the period 1950-1970, or like the late 1800s? No, the choice is not binary, but you get the point.

    • Chris on April 1, 2014 at 4:32 pmi

      Assuming for the sake of argument that some people have gotten rich from cronyism, it makes no sense to me to punish all members of the class of rich people for the sins of some. Tell me, are you also the kind of person who believes all racial minorities are criminals because some of them are?

      If cronyism is the problem (which I believe it is), then the logical position would be to advocate for less government involvement in economic matters. But I’m guessing you’re not a libertarian, so your use of cronyism is a red herring.

      Transportation and infrastructure is a small portion of the US Federal budget, between 3-12% depending on what you include in that category. Your use of the need for infrastructure spending as a justification to raise taxes on the rich is a red herring.

  8. Beth Leath on April 1, 2014 at 12:21 pmi

    I have enjoyed both the article & the back and forth of the comments.

  9. Ajit on April 2, 2014 at 3:05 pmi

    Reading the comments, I tend to agree with Chris who is making a subtle point but being attacked by people who are over reading.

    I find it disturbing that so many of the comments, in response to Chris, act like the answer is obvious. That high wealth disparity must be the result of rent seeking, monopolies, crony capitalism and other forms of covert theft. What’s more, it’s further implied that the middle class stagnation has been the result of their activities and the economy is surely hurt by this kind of disparity.

    of course, it is certainly possible, but this much is true: There is no compelling evidence as of now that actually proves any of these statements being true. We don’t know for sure that this all the result of rent seeking anymore than are we sure that having a huge wealth disparity means we’re necessarily worse off as an economy.

    I wish people would wait for the evidence before being completely convinced.