There Will Be Rich Always: Finding a New Way to Think About Income Inequality

On Monday, Aaron Edlin and I published a cri de coeur op-ed in the New York Times calling for a Brandeis tax, an automatic tax that would put the brakes on income inequality. In the next few days, Aaron and I will be publishing a series of posts explaining more about our rationale and providing more details on how a Brandeis tax might be implemented.

There Will Be Rich Always
By Ian Ayres & Aaron Edlin

In one of the more memorable lyrics from the musical Jesus Christ Superstar (based on Matthew 26:11), Jesus tells his disciples “There will be poor always.”

The same is true of the rich. There will always be a top 1 percent of income earners. But what it takes to be rich can change drastically over the course of even a single generation. In 1980, you would have had to earn at least $158,000 to be a one-percenter; but by 2006 the qualifying amount had more than doubled to $332,000. (You can produce an estimate of your own household income percentile – albeit using a different definition of income that produces a much higher 1 percent cutoff –  at this site.) The rise is not due to inflation as both these numbers are expressed in inflation-adjusted, constant 2006 dollars. The rise is due to the simple fact that our richest Americans in real terms were earning much more money.

The economic changes in the past 30 years were a rising tide that did not lift all boats. Over the same period the median household income remained relatively constant, at roughly $50,000. While inequality has increased in most wealthy economies, the United States, according to the OECD, remains among the most unequal.

The vast shift in national income toward our richest 1 percent is especially vivid if their income is expressed in terms of the median household income. Indeed, an important goal of our op-ed was to suggest a new unit of measure, “medians” to help us think about what it means to be rich. In 1980, if you earned 3.8 medians, you were in the top 1 percent, but by 2006 even the poorest in the 1 percent club earned 6.9 medians. 

What we call the “Brandeis Ratio,” the average income of the richest 1 percent (which includes the billions earned by the lucky few) has grown even more disproportionate. As shown in the chart below, in 1980, one-percenters on average made 12.5 medians, but in 2006 (the latest year in which data is available) the average income of our richest 1 percent was a whopping 36 medians.

Define Rich?

In August 2008, the Reverend Rick Warren asked each of the presidential candidates Barack Obama and John McCain to:

Define “Rich.”  Give me a specific number. Where do you move from middle class to rich?

The candidates’ answers, given at what might be the high-water mark of income inequality in our nation’s history, reveals the difficulties that both parties have talking about economic class. McCain was visibly uncomfortable in naming a number:

So I think if you’re just talking about income, how about $5 million? So, no, but seriously, I don’t think you can — I don’t think, seriously, that — the point is that I’m trying to make here, seriously — and I’m sure that comment will be distorted — but the point is, the point is, the point is that we want to keep people’s taxes low and increase revenues.

In part, McCain was reluctant to name a figure which might be used to construct a new bracket to raise taxes – so he picked a large number; very large in terms of “medians.” McCain suggested that you would have to earn an income that is 100 times the median household income in order to “move from middle class to rich.” 

Obama in contrast gave a much more reasonable answer:

Here’s how I think about it, and this is reflected in my tax plan. If you are making $150,000 a year or less as a family, then you’re middle class, or you may be poor. But 150 (thousand dollars) down, you’re basically middle class. Obviously, it depends on region and where you’re living. . . .  I would argue that if you’re making more than 250,000 (dollars) then you’re in the top 3, 4 percent of this country. You’re doing well. Now, these things are all relative, and I’m not suggesting that everybody who is making over 250,000 (dollars) is living on Easy Street.

Obama should be given points for easily relating his cutoff of $250,000 to those Americans who are “in the top 3, 4 percent of this country.” But we are a little troubled at Obama’s describing family incomes of $150,000 as still being in the middle class, when these families earn 3 times what the median households earn. Survey data reveals contested definitions of middle class, and there is evidence that Americans disagree with Obama. In a 2007 study conducted by NPR, the Kaiser Foundation, and the Harvard School of Public Health, 51 percent of respondents did not consider a family of four with an $80,000 income middle class – that consensus rose to 65 percent when the family income rose to $100,000.

We as a nation seem to have trouble facing up to the fact that most Americans earn far less than what it takes to be comfortably middle class. We would do well to give more emphasis in our definition of the middle class to incomes earned by average Americans. Our aspirational middle class is making it too easy for us to forget what is happening to our actual middle class. 

A larger goal of our op-ed is to spark a different kind of debate about income inequality. Rev. Warren was crafty in that he only asked the candidates the descriptive question: what is rich. But we want people of goodwill to ask a normative analog. In our op-ed, we claim that it “would be bad for our democracy if one-percenters started making 40 or 50 times as much as the median American.” 

Is there any Brandeis Ratio that would give concern to this year’s Republican candidates? We suspect they would be unwilling to express unease with any level of inequality produced by market forces. They would argue that any government attempts to lean against inequality by taxing job-creators would invariably produce distortions (including sending one-percenters abroad, beyond the reach of our tax laws) that would end up making things worse off for middle class Americans. This is a debate worth having, but we cannot begin unless we find a way to engage in normative debate about income and wealth inequality. Asking whether the government should be concerned about the rapid rise of the average one-percent income from 12.5 to 36 median incomes is our way to provoke a more explicit debate.

Seeing Medians

Framing income inequality in terms of “medians” is also part of a larger goal of making the median household incomes more salient. As we said in our op-ed, “Part of our goal is to change the way politicians speak about income equality. Framing the income of the wealthy in relation to the median income will help us all keep in mind the relative success of the middle class.”

It might even be useful to describe other things in terms of medians. A new Cadillac Escalade will run you 1.4 medians. A  year’s tuition at Yale Law School is about .88 medians. We might even restructure government salaries so that they automatically adjust with the median. Paying a congressman 3.5 medians (instead of the current $174,000), might make it easier for our representatives to remember and even pursue the interests of their typical constituents.

To raise the prominence of the median measure, government could standardize a “mi” symbol. A stylized icon figure representing a median earner might even be more effective in letting us remember that to earn 36 median incomes is to earn as much as 36 actual households:


Leave A Comment

Comments are moderated and generally will be posted if they are on-topic and not abusive.



View All Comments »
  1. MrAtoZ says:

    Oh, and I forgot to include another critique of this proposal: if we are to be concerned about “medians” (which is the apparent purpose of this exercise), why wouldn’t you take the median income of the top 1% to determine the Brandeis ratio, rather than the average of the top 1%?
    I am going out on a short limb here in assuming that the average income of those in the top1% is skewed upward by the billionaires in that cohort, while using the median will reflect that there are more earning somewhere closer to $332k, as opposed to those few making billions at the top end of the bracket.

    The reason median vs. average is a relevant critique is that themedian income of the top 1% will result in a Brandeis ratio far lower than the 36 to 1 expressed above.

    Well-loved. Like or Dislike: Thumb up 8 Thumb down 1
  2. Tim says:

    I would love to see a longer-term graph of Brandeis ratio, because this is really a utilitarian, macroeconomic question: what ratio, over a given economic window, has returned the greatest growth in GDP or GNP? Answering that would allow us to reverse engineer tax codes to that effect. Over-focusing on the sociological concept of “fairness” (whatever that means- “the fair is in September”, as my father used to say), causes us to lose sight of the fact that this is a fundamentally economic problem, with secondary social baggage. Income inequality isn’t bad because it’s unfair, in a capitalist system: it’s only bad if it causes the economy to function poorly!

    Thumb up 4 Thumb down 3
  3. James says:

    I can only wonder at a standard that defines “middle class” as $150K/year, then states that most aren’t earning that much. Isn’t middle class, by definition, what the middle earns? Since I manage to live quite comfortably by spending only a fraction of that, I can only suppose that those $150K-earning “middle class” folks are actually spending most of that money on consumer debt, gambling, or something similar.

    One point I think needs to be addressed. Many of the previous posters bewail the fact that – in their opinion, anyway – most of the top 1% are corporate CEOs and the like. So how about some actual data? What fraction of the wealth goes to the executives, what to financial services people, what to tech types, what to entrepreneurs, and so forth.

    Thumb up 2 Thumb down 1
    • Jim... says:

      $150K can be rich or poor, depending largely on (1) where you live and (2) the size and composition of the household. A single person making that in Spokane might be well off, but a family of 5 in Manhattan would not.

      As to the CEO vs. entrepreneurs etc.:

      “Wall Street and Main Street: What Contributes to the Rise in the Highest Incomes?” by Steven N. Kaplan and Joshua Rauh from the Review of Financial Studies

      Thumb up 0 Thumb down 0
      • EB Hansen says:

        At some level though, people choose to live in those more expensive places – higher cost of living often coincides with a more desirable place to live.

        150k gets you a much smaller living space in New York than Scranton, but you get your family access to one of the most in demand locations in the country.

        Household size is also – generally – based around the preferences of the heads of household. Some couples may decide that having a nice car is a good way to spend their income, some may prefer children.

        Thumb up 1 Thumb down 0
      • James says:

        I would, of course, argue that anyone condemned to live in Manhattan is poor, even if they have a billion or two in the bank.

        But to your general argument, certainly there are some things, notably housing, that differ considerably in price from one location to another. Most things, however, do not: Amazon does not (at least AFAIK) charge different prices depending on where you live.

        Income can also be independent of location. Salaries may run higher in Silicon Valley than in Salt Lake City, but these days many of us can telecommute, live where we choose, and be paid the market rate for our work.

        Thumb up 0 Thumb down 1
    • Enter your name... says:

      That $150K is supposed to be the top end of middle (or the bottom threshold of upper), not the middle of it. You could just as accurately say that “middle class is $30K”, even though most households earn more than that, because $30K puts the typical (one person) household just above what most people think of as lower class.

      Thumb up 1 Thumb down 0
  4. Jim says:

    A careful analysis of this issue is beyond the scope of a newspaper column or a blog post.

    Apart from playing fast and loose with terms (salary is not the same as income, which is not the same as wealth) and cherry picking statistics (2007 was the peak for inequality), there are important confounding econometric issues.

    Household size and composition are extremely important — consider four couples where each spouse makes $100K. The maximum household income ($200K) equals the median which also equals the minimum. Two of the couples get divorced The maximum remains $200K, but the median is now $100K.

    Demographics are also important, as retirees often have very little income, but are spending down their earlier excess spending.

    Tax policy is important as well — for business owners, the relative level of corporate versus individual tax rates has significant bearing on how taxes are paid. What may look like dramatic increases in individual income may only represent similar earnings, but a change in tax policy.

    And finally as others have pointed out this also ignores income mobility.

    There may be a problem, but this is op-ed masquerading as economics.

    Thumb up 2 Thumb down 2
    • rehajm says:

      …Passive income like dividends and capital gains are included in the income numbers while passive income like government subsides for food, housing, medical care, etc., and receipt of private charity, are not….

      Thumb up 1 Thumb down 2
    • Travis says:

      Income mobility is a myth. For someone who is harping about how this is not economics, it doesn’t behoove you to bring up a concept of something that has essentially been dead in the united states for the past 2 decades.

      see, e.g.,, page 5

      If you’re going to criticize something for not being economic enough, perhaps you shouldn’t be ending with platitudes that hold little to no water in todays world.

      While there are a lot of things in your commentary, it is no better than Mr. Ayers’ proposal in that you simply do more hand-waving about what the median household income is or should be when the central core of the proposal, whether it would be beneficial to tax the top 1% at a rate relative to the median household income (or rather, tax inequality), is completely ignored.

      Thumb up 0 Thumb down 2
      • Enter your name... says: says you’re wrong: nearly half of people in a lower income quintile 15 years ago saw their income quintile rise during the decade that followed.

        If you want a longer-term perspective, when I look at my own family, 6 out of 8 cousins on my mother’s side of the family currently earn in an income quintile higher than their own parents. And the converse is also true: all of the cousins on my father’s side of the family are in an income quintile distinctly below their parents’.

        Thumb up 1 Thumb down 0
  5. AaronS says:

    It is not “medians” that are important, because it can be so easily distorted. For instance, if you have two people, one with nothing and one with a billion dollars, the median is $500 million dollars, which would give us a quite distorted view of the actual situation.

    What we must FIRST ESTABLISH is what a person SHOULD have. I mean, if we can say that a person is “too poor,” then we can also judge what is required to make them NOT “too poor.” Before I was unemployed for nearly three years, I might have answered this question very differently; but now, knowing what I do, I think we are often wrong about what poor people really need.

    Yes, if a poor person spends his welfare money PERFECTLY, he may have plenty. But how likely is that a person in need of welfare has the ability to spend his money perfectly? He likely wouldn’t be on welfare if he could do it just right, right?

    I think that we need to build the floor a lot higher than we have. People in America, if they work (or are willing to), should have the wherewithal for a safe and decent place to live (not just the inner-city, high-crime areas, etc.), reliable transportation (it makes life a whole lot easier), nutritious food, cellphones (really, can we reasonably exist without them anymore?), internet connectivity, education opportunities, and a little extra beside…for the kids, gas, Christmas, birthdays, and the other little sweetnesses of life. If we have the ability to do more than is absolutely needed, why shouldn’t we? I mean, must we make minimum wage as MINIMUM as we can? Is there something wrong with poor folks having a little something more than bare bones?

    OK, so once you have established this floor of necessary income, THEN you can measure how many (let’s call them) LIVES a person has. If a person needs 50,000/family to live a decent life (as we define it), then someone making 100 times that much, has 100 LIVES (so to speak) of wealth.

    Maybe we could say that anything above 100 lives should either be plowed directly into making jobs and so forth…or be taxed heavily? A person with a billion dollars has enough to spend $100,000 PER DAY FOR THE NEXT TWENTY-SEVEN YEARS.

    If we can discern that some people don’t have enough, are we unable to discern when some people have ENOUGH (or too much)? Let a billion dollars be the ceiling. Anything over that amount should be taxed mightily…OR allowed to be plowed back into the AMERICAN economy in specific and meaningful ways that lifts all boats higher.

    Just seems to me that I’d be able to do just fine with $100,000/day.

    Thumb up 2 Thumb down 2
    • Enter your name... says:

      You might like to read about what the Canadians have done with a “basic needs” concept.

      Some of their choices interested me: they included long-term costs (like the occasional need to replace a piece of furniture—mattresses do wear out eventually) and renter’s insurance, which they determined was ultimately slightly cheaper than the risk of disaster-induced bankruptcy.

      Thumb up 1 Thumb down 0
  6. jeffreytg says:

    Income is the wrong measure: a 70 year old retired couple with no mortgage and no children at home could live on $30,000 per year income better than a 40 year old couple with 2 kids and either renting or making mortgage payments while taking in $60,000 per yer in income.

    Thumb up 2 Thumb down 0
  7. Noni Mausa says:

    Another element that effects the meaning of the “median” measure within a country has to do with stability versus precariousness of the average person’s life situation. This is not the same as mobility, up or down, but definitely effects it.

    Compare two countries with very similar income medians, and rather similar tax structures for the middle class, Canada and the US. The simple addition of universal health care increases the value of the Canadian median income by thousands of dollars a year, and removes one of the factors that can tip a family into a downward spiral into near-permanent poverty.

    To avoid such risk, a wealthy person in a country with poor social supports needs to be far wealthier than his counterpart in a better-supported country. It could be argued that the undercutting of universal social supports is less a result than a cause of income inequality. In an environment of poor social supports, people with a talent for acquisition are strongly motivated to accumulate far more leverage (money, property, power etc) than otherwise, and fight more aggressively to retain it.


    Thumb up 2 Thumb down 0
  8. David Miller says:

    If the proposal merely increases the tax rate on taxable income, then it will have only a negligible effect on the truly wealthy. The proposal needs to incorporate a tax on economic income.

    Take Warren Buffett. He paid $6,923,494 of tax last year on taxable income of $39,814,784 (17.4%). But his Berkshire Hathaway stock increased by $3 billion, and he will never pay any income tax on this appreciation. So his tax rate on economic income (including unrealized appreciation) was more like 0.2%.

    Even if he were taxed at a 100% rate (but only on taxable income), he’d pay tax of $39,814,784, but this would represent only a little more than 1% of his economic income.

    I think the only cure is a mark-to-market tax on publicly-traded securities (and derivatives that reference publicly-traded securities). I’ve proposed such a tax for the 1/10 of 1% wealthiest and highest income individuals.

    I testified before Congress this month about such a tax.

    A summary of my proposal is here.

    The full version is here.

    Thumb up 1 Thumb down 0