Live by phony data, die by phony data

I frequently argue that income data is nearly meaningless.  Especially if one is interested in looking at the issue of economic inequality.  That’s why I rarely  get involved in blogger debates over inequality—my views are so out of the mainstream that I’d hardly know where to start.

But let’s say I’m wrong, and that income is an excellent indicator or economic well-being.  In that case I’d still argue that median family income is a nearly worthless indicator of how average families are doing.  I’ve spent a considerable time in all 5 income quintiles.  For instance in college, grad school, and then as an adjunct professor I had a pitifully low income.  At the time I was a fully independent adult, not carried on my parents’ tax return.  Yet it would make little sense to say I was “poor” in a socioeconomic sense, despite my very low consumption bundle.  At the other end of the age spectrum I know old people who pick up some extra money with part-time employment, and yet who aren’t really low income in a socioeconomic sense.  Poverty is one of those “I know it when I see it” conditions.

For years progressives have been pointing to the rather low median income data to suggest how poorly the American middle class is doing, as if the typical median income earner were a middle-aged family with two kids.  They’re not, as the income data mixes together all sorts of different types of households.

The problem with claiming Americans are doing poorly, is that you end up weakening the case for extensive welfare benefits.  For instance, what if the social benefits given to the “poor” exceeded the median income of the entire country?  Wouldn’t that create a massive disincentive to work?  I don’t know how often that happens in America, but it’s recently become a political issue in Britain:

A few days later, in the House of Lords, a coalition of Labour peers and Church of England bishops cited Charles Dickens and Victorian notions of the deserving and undeserving poor as they attacked government plans to restrict the welfare payments received by any one household to the median income of a working family. The rebels won, with the Lords voting to ease the benefits cap for families with many children. Their rebellion will be overturned: some three-quarters of voters support the cap.

There is a good argument against the cap, but it probably won’t work because it would require progressives to admit that the median income data they always cite is deeply misleading.  This is one of the many internal contradictions of progressivism.  By the way, even the Labour Party admits that Britain has a problem with welfare dependency:

Debates about capitalism dominate British politics. The Conservative prime minister, David Cameron, his Liberal Democrat deputy Nick Clegg, and the leader of the opposition Labour Party, Ed Miliband, have repeatedly spoken about building a fairer economy. Responding to voter anger, they talk of reining in bankers’ bonuses and pay packages for company bosses. All three agree that there is a need to curb welfare for the work-shy.

[Full disclosure:  I once lived in Britain.  I shared a flat with a guy in his late 20s who hadn’t worked in years, and didn’t seem to want to.]

PS.  Speaking of the internal contradictions of progressivism, Matt Yglesias points out that many progressives would like to see a significantly larger share of the workforce engaged in services, construction, agriculture and manufacturing.  That’s why Yglesias has become the most persuasive blogger on the left.  He understands that sectoral employment shares need to add up to 100%.

PPS.  Yes, the conservatives have their own internal contradictions, as when they cite data showing that many people are able to move from the bottom 20% to the top 20%.  I was one of those Horatio Algers stories.  But in my own mind I’ve been middle class my entire life.



33 Responses to “Live by phony data, die by phony data”

  1. Gravatar of Patrick R. Sullivan Patrick R. Sullivan
    28. January 2012 at 10:26

    I lived for a few months in pre-Thatcher England, which was quite an eye-opener. Especially all those Rolls Royces clogging the streets of London (something that caught the attention of Milton Friedman too).

  2. Gravatar of Benjamin Cole Benjamin Cole
    28. January 2012 at 10:30

    Interesting–I live an an Airstream trailer I dragged onto the parking lot of a factory I own. Since then I have been flush with cash. No more $1,500 a month in rent/utilities for an apartment.

    No change in “income.”

  3. Gravatar of Brett Sheckler Brett Sheckler
    28. January 2012 at 10:48


    I think you know you are stretching things here.

    Is income the only indicator of economic well-being? Not by a long shot.Is it an important indicator? Of course.

    It is one thing to say that poverty is complex, with many moving pieces. It is another thing to say poverty is one of those “I know it when I see it” conditions.

  4. Gravatar of ssumner ssumner
    28. January 2012 at 12:23

    Patrick, With MTRs at 98%, those $100,000 Rolls Royces cost $2000 as company cars.

    Ben. Good example.

    Brett, Lifetime consumption is a far better indicator of economic well-being, but even that is flawed. Income inequality data simply doesn’t measure economic inequality. It’s a garbage-in-garbage-out problem.

    Unless you think I went from being poor to being rich as I got older.

  5. Gravatar of Tommy Dorsett Tommy Dorsett
    28. January 2012 at 12:41

    Given the inequality discussion, any thoughts on where the top of the Laffer Curve is? Obviously, it shifts over time, but the progressive left seems to have coalesced around the DS 70% rate threshold, based on dubiously low income elasticity estimates.

    Does it not even register with the progressive left that the income tax share of GDP was less with a 70-90 TMTRs than it has been with MTRs between 28-50? Do they really assume just as much pre-tax income will be earned with rates 2X today’s levels? Time to get out of the classroom and into the real world of incentives and after-tax returns, people!

    What would the top rate and threshold be if we had a Sumnerian progressive payroll tax? Would corporate income, dividends, capgains and interest be taxed? If so, at what MRT?

  6. Gravatar of Mike Sandifer Mike Sandifer
    28. January 2012 at 15:16


    I consider income inequality from the perspective of the share of real GDP that goes to each percentile. For example, the if the top 1% get a larger share of real GDP now than 30 years ago, that might be a problem to be addressed.

    Of course, when considering income meaningfully, one also needs to consider that the prices of many commonly bought goods, such as clothes, cars, etc. have fallen over the same period, but otherwise, do you see anything wrong with this perspective?

  7. Gravatar of Morgan Warstler Morgan Warstler
    28. January 2012 at 17:02

    The reason Matty has a growing audience is because he’s adopting a far more libertarian approach to life…. perhaps he’s actually / finally saying things he’s always believed… but I doubt it. Watching him genuflect is fun.

    Scott is of course right, lifetime consumption is a far more correct both in policy and philosophy.

    The winning argument, the final trump card, on income inequality is that there is a basket of goods and services that once we give them to the have-nots, they don’t deserve and won’t be able to rally public support for anything else:

    1. we are near that level now.
    2. it costs less over time to deliver that basket.

  8. Gravatar of Bob Murphy Bob Murphy
    28. January 2012 at 17:19

    A funny coincidence, last night I typed up a long post picking apart Sumner’s dismissal of “income.” If you have generally agreed with Scott’s critique of the income tax, but you felt he was throwing the baby out with the bathwater…this post is for you.

  9. Gravatar of Bob Murphy Bob Murphy
    28. January 2012 at 17:21

    Johnny Cochrane alert! Here’s Sumner:

    Yet it would make little sense to say I was “poor” in a socioeconomic sense, despite my very low consumption bundle.

    And this is Scott’s illustration of why “income” is a bad measure, and should be replaced in our minds by “consumption” to show who is rich and who is poor.

  10. Gravatar of dtoh dtoh
    28. January 2012 at 18:14

    For the umpteenth time, the reason not to tax capital is because you have asymmetric returns (a lot of losers/few winners) for many types of investment (particularly those that are important to new business formation, employment and productivity increases). As a result, even low rates of tax on capital gains will quickly turn expected positive pre-tax returns into negative expected after-tax returns, thus choking off investment.

    If you don’t understand this fundamental concept, you should not be discussing tax rates or income inequality.

  11. Gravatar of Morgan Warstler Morgan Warstler
    28. January 2012 at 22:10


    1. you make the same dumb mistake as Scott, you look at the world through “investor” eyes. investors are betas, they are our second concern. the main concern is what is best for entrepreneurs.

    2. more losers = less investment = better return on investment

    fundamentally, BUY when everyone else is selling. START when everyone else is thumb-sucking.

    In my experience the worst damn time in business cycle is when there’s TOO MUCH money, you face bidding wars on talent, you have to raise more money, give up more equity, and generally deal with a bunch of shitheads copying your work.

    The best time is when it takes skill to raise cash, when you can carve out something generating revenue quickly, and entrenched corporations (customers) are just impressed you are alive…. rather than overwhelmed with calls from 1000’s.

    In another example, take housing / real estate – what we want is CHEAPER rents for everyone, who cares if values fall, and idiots who thought their hoe was heir retirement eat it. What we want is people spending less on the dumb bricks and mortar shit, and more buying shit on the Internet.

    We treat 1031 as zero capital gains. This is dumb. Instead we should treat gains made by sweat equity serial entrepreneurs as the as the best kind of capital, they are the real producers.

    one more time for you and Scott: moving money around is a second class sport.

  12. Gravatar of ThatGuy ThatGuy
    28. January 2012 at 22:22

    All bow to dtoh.

  13. Gravatar of ThatGuy ThatGuy
    28. January 2012 at 22:26

    And to Morgan Warstler, who apparently has had too much to drink.

    Neither of your comments have anything to do with the article. But you both seem like smart guys. Really.

  14. Gravatar of Liberal Roman Liberal Roman
    29. January 2012 at 00:07

    Is Matt Yglesias still a “liberal”? I wouldn’t know it from reading his latest posts. Basically, he is the complete opposite of Krugman. He spends most of his posts decrying stupid regulations, bringing liberals back to reality and yes he does give some lip service to redistribution of wealth. But man has he changed lately.

    I think you converted him Scott.

  15. Gravatar of dtoh dtoh
    29. January 2012 at 02:45


    I agree with you about investing at the bottom of the business cycle, but you’re wrong on everything else in your post.

    Entrepreneurs are usually the investors who are worst hurt by the asymmetric nature of returns. Professional investors (e.g. VCs) can average out their losses against their gains and also offset losses against gains for tax purposes.

    Less investment from others is good for the entrepreneur but not the economy and it has nothing to do with the tax problem. It just means you will have 19 bad investments and 1 good investment instead of 38 and 2.

    We don’t just want cheaper rents. We also want better housing. Doesn’t happen without investment.

    I’ve done both the entrepreneur gig and the investment banking gig. TBTF causes an over allocation of the brightest talent to the finance sector. But even without that, the sector would still attract the best talent because a) efficient capital allocation has a huge impact on output, b) it’s a highly efficient sector, and c) at the top the difference between the very best and the very good is huge. Thus the high compensation.

  16. Gravatar of Morgan Warstler Morgan Warstler
    29. January 2012 at 07:07

    TBTF causes an over allocation of the brightest talent to the finance sector. But even without that, the sector would still attract the best talent because a) efficient capital allocation has a huge impact on output, b) it’s a highly efficient sector, and c) at the top the difference between the very best and the very good is huge. Thus the high compensation.


    I could make two changes to banking regulations and overnight chase smart guys out of finance.

    I could make one change to tax code and entrepreneurs would rule the roost.

    Entrepreneurs WANT the money, but they also want the world to be the way they see it.

    Saying “asymmetric returns” doesn’t work in real business, because real entrepreneurs are SERIAL entrepreneurs. I have more friends who’d say their most profitable newco wasn’t their favorite newco – than anything.

    I liken it to college football.

    the top 2% of highschool players (small business) make it into the college level – the top 2% of SMBs that earn 50% of SMB revenue.

    You are concerning yourself with which of them become NFL superstars, and noting how few this happens too.

    I’m unconcerned with that piece, I know it happens, I claim the job growth and super cheap consumer advantages of it happening (modern superstars = facebook, skype, etc), but I con’t really CARE about that piece.

    What I’m concerned with are the maybe 1% of the population that have the chops for serial entrepreneurship, who will start newco after newco, each time inventing / crafting something new and each time hoping into that 2%.

    I know that right now these guys fight with an arm tied behind their back (the tax code favors financial investors), and if we just even that piece up,it won’t be even close to a fair fight.

  17. Gravatar of ssumner ssumner
    29. January 2012 at 12:07

    Tommy, There are different Laffer curves for different taxes. If you compare the US and Europe, they raise similar amounts of revenue with very different taxes as a share of GDP. That suggests to me that Europe is near the peak, maybe past the peak. But these are extremely complex questions, and the honest answer is no one knows. It also matters how the money is spent.

    I favor a progressive consumption tax, plus a land tax and a pollution tax.

    Mike, I think it’s a huge mistake to pay any attention to income. Consumption inequality is much more meaningful.

    Morgan, Matt’s not libertarian, he’s rational.

    Bob, Someday we’ll sell our house. It will merely convert one asset into another. Our wealth won’t change. But the US government will say we are “rich” that year. Does that make sense to you?

    dtoh, Yes, that’s ONE reason not to tax capital. But even if there weren’t asymmetric returns, we still shouldn’t tax capital.

    Liberal Roman, He still seems liberal to me, but I suppose it’s a matter of perspective. He still favors fiscal stimulus and a single payer health system, if I’m not mistaken.

  18. Gravatar of LeeN LeeN
    29. January 2012 at 12:57


    Your ideas on consumption inequality are not completely out of the mainstream. Robert Frank has been pushing for progressive consumption tax for a while, though from a different angle. He’s been arguing that “arm races” for luxury goods have been a major source of waste and a consumption tax would curb the waste. He comes at the issue from a progressive angle, but wants the same end goal. Also, pre-tax deductible contributions to retirement savings plan have similar features as a consumption tax. I think your views aren’t out of the mainstream only your reasoning and justifications for specific policies.

    [paper], [good talk with Wolfers and Frank]

  19. Gravatar of LeeN LeeN
    29. January 2012 at 13:26

    links didn’t seem to work:,

  20. Gravatar of Donald Pretari Donald Pretari
    29. January 2012 at 13:50

    I think that there’s too much subjectivity about how one feels one is doing for data to be decisive. If the Govt uses a particular set of data, it’s usually because it helps the Govt do what it wants to do. Ease & Usefulness are the desired traits, not Truth.

  21. Gravatar of dtoh dtoh
    29. January 2012 at 15:48

    But the impact caused by the distortion from asymmetric returns overshadows everything else. Agree on the progressive consumption tax. I also agree on the land tax, but would be curious to know your logic.

    You’ve been reading too much Ayn Rand.

    Sure there are “change the world” typeS, but there are also a lot of “make a buck” types and “don’t want to take orders from anyone” types as well. Talent helps, but virtually no one is smart enough or omniscient enough to ensure success. It’s all about tenacity trial and error. Even the geniuses rarely succeed. I think we’re talking about the same people. The tax code kills this group but not because it favors financial investors.

    Sure you could change the tax code and drive the smart people out of banking, but then you would be under-allocating resources to an important sector of the economy.

  22. Gravatar of ssumner ssumner
    30. January 2012 at 05:45

    LeeN, Within the economics profession my ideas on inequality are 100% standard, everyone uses consumption as the metric. But out in the blogosphere everyone talks about income.

    Donald, Good point.

    dtoh, The land tax doesn’t tend to discourage the production of new land, except in Holland.

    I don’t know enough to comment on the importance of asymmetric returns.

  23. Gravatar of Morgan Warstler Morgan Warstler
    30. January 2012 at 07:20

    dtoh, no I’ve just spent my whole life in venture funded start ups. And I grew up surrounded by small business owners.

    There are serial entrepreneurs, much like professional pokers players, who spin up venture after venture and have far more success than failure. They have any and all of the motives you describe, and sub-dividing them doesn’t change my point.

    The data shows that:

    These guys make up the 2% of SMBs that earn 50% of SMB income, and it is from the pool of their creations where the super nova newcos (that generate the new job growth).

    Favoring this class over all other classes of investors is the most positive tax policy we can employ.

    Again, I’m not saying other classes of investors are worthless, I’m saying they are not as important. Money is easy to find, idea/talent/execution is harder. And the correct tax policy can increase the quality of ideas, talent, and execution.

  24. Gravatar of dtoh dtoh
    30. January 2012 at 14:17

    There are serial fund managers who have success year after year too. Just like the serial entrepreneurs, it’s easy to identify them after the fact. Also, success gives you the experience needed to improve performance, but it also makes it easier to attract money and people on subsequent ventures so it’s not at all clear to me that serial entrepreneurship is much more than the same mix of luck, timing and talent that you see in any venture.

    Also a lot of super-novas come out of the blue; I think you would be hard-pressed to demonstrate that they predominately come from the pool of serial start-ups.

    I may be wrong on this, but I’ve worked with a couple hundred venture backed start ups so my sample size is pretty good.

    Be curious what you would do on taxes. Can’t think how you would make it simpler than no tax on capital, but I guess you could have negative tax rates.

  25. Gravatar of Cthorm Cthorm
    30. January 2012 at 15:53

    >Favoring this class over all other classes of investors is the most positive tax policy we can employ.

    Sorry Morgan, but I don’t see this as a self-evident point. I have an obvious conflict-of-interest here, but I think you do too. A tax code that makes no distinction between types of investors (or really any tax payer) is always better, even if your stated aim is to help favor these serial entrepreneurs. The skills of “serial fund managers”, as dtoh calls them, are complimentary to those of serial entrepreneurs. Why should venture capital be any different from other asset classes, once the tax code is simplified? Venture funds specialize in identifying good ideas and entrepreneurs, while “serial fund managers” specialize in efficient (i.e. profitable) allocation of capital. More liquidity in venture funds would only increase the volume of start-ups that could be funded, certainly more so than an implicit subsidy in the form of a favorable tax code.

  26. Gravatar of dtoh dtoh
    30. January 2012 at 16:18

    Morgan’s logic is right. It’s just that his premise is wrong. He thinks the economy is entrepreneur constrained and not capital constrained.

  27. Gravatar of Cthorm Cthorm
    30. January 2012 at 16:35


    There is plenty of capital available. The problem is there are some ‘industries’ where regulations, tax rules, etc. make it impossible to give liquid access to capital markets. These barriers are what is making capital constrained.

  28. Gravatar of Morgan Warstler Morgan Warstler
    30. January 2012 at 22:37

    1. to my thinking, a system that rewards entrepreneurs first with taxes on capital will quickly favor them on regulations too – they will become exponentially more powerful amd turn other laws to their interests- they already are the source of new jobs and lower cost of goods and services (productivity gains).

    2. my plan is simple and a small ask. let SMB owners declare a salary (basically their annual consumption) and pay taxes on it. and let them take all other profits / income tax free like a 1031 right now) from any venture and move them to any other venture.

    then, a bit more out there, we can then get really aggressive about not letting ANY business have ANY kind of luxury business expense – first class air, hotels more than $200 a night, dinners over $50 – treat all that as INCOME and tax the individual who eats, travels, and sleeps in the lap of luxury accordingly.

    If you spend time looking at the stats on millionaires, you find a ton of SMB scrimpers and savers in small towns who start and own different newcos over their lifetime.

    And on the flip side you have a bunch of non-millionaires living in Blue State big cities living high on the expense acct. hog.

    rewarding the first and screwing the second is basically all you need to do – overnight we’ll get that 5M+ monthly job quits / hires / separations that JOLTS data confirms will be a job growth month.

    We’ll be codifying creative destruction into our system – architecting our policies to get more of it.

  29. Gravatar of Cthorm Cthorm
    31. January 2012 at 08:12


    Have you ever read Animal Farm? Does it ever occur to you that by crafting a system to favor one group over another you’ll only strip the very traits you praise from the More Equal Pigs?

    I don’t want to trade entrepreneurs for crony capitalists, because I fear we’ll just make the former like the latter. Why does the system need to be tilted at all? The market will reward the entrepreneurs plenty well if we just even the playing field. Even more so if we ensure it stays level. Crafting the tax code to favor SMBs is not so different from gerrymandering political districts.

  30. Gravatar of dtoh dtoh
    31. January 2012 at 10:32

    You’re being silly on the expense account issue. You can’t get a room for $200/night in many major cities. There are a lot of people who spend hundreds of hours a year in the air. They couldn’t function if they had to ride in the back of the plane all the time. Shareholders and managers are perfectly capable of setting appropriate expense account guidelines. You don’t need the tax code to do this.

    What we need to do is get rid of all taxes on capital. That would be a huge boon for the economy. You could replace it with higher taxes on wages or a progressive consumption tax.

  31. Gravatar of Morgan Warstler Morgan Warstler
    31. January 2012 at 10:54


    Why does the system need to be tilted at all?

    1. Penance. jk.
    2. The paradox of the rent seeking capitalist.
    3. Because old people are not as important as young people.

    Big Business = Big Government. Large corporations will always seek favor with government, they will enjoy regulations that weigh more heavily on smaller players. You know all the arguments.

    Since this is true, or at minimum probable, we HAVE TO form policy that actually gives the small guys the extra juice, it is a hack to preserve liberty.

    Look I’m familiar with all the efficiency arguments about large corps, but in my experience those are junk, the arguments about scale really are about throwing weight around.

    Don’t think of it as affirmative action. Think of it as tax credits being given to small companies so that they do not bear the cost of regulation – the EFFECT is that large companies stop seeing advantages in regulations that they alone have to pay. In this case it is simpler and more about creating a political power behind distributed SMB economics.

    Finally, tax policy that favors SMBs over others leads to more creative destruction, which leads to faster technological change, which puts more power being placed in the hands of young at the expense of the old.

  32. Gravatar of Morgan Warstler Morgan Warstler
    31. January 2012 at 10:59

    dtoh, you aren’t actually answering my argument just restating the thing I’m sayng is wrong:

    “What we need to do is get rid of all taxes on capital.”

    I refuse to have this discussion and will support taxes on capital UNTIL the profits made YOY by SMB entrepreneurs are considered capital.

    Right now, the 2% of SMBs that make 50% of SMB revenue – the fat majority of them are passing profits though as income – and this is WRONG.

    The HIGHEST form of capital gain is profits made by entrepreneurs investing in themselves.

    We should all codify and admit that is true.

  33. Gravatar of dtoh dtoh
    31. January 2012 at 16:40

    Your solution is not wrong, it’s just inadequate. Why tax business profits at all or why require that they be re-invested. Why not just totally eliminate all taxes on capital.

    For successful small businesses, every dollar paid in taxes reduces investment by a dollar. But for start-ups, the incentive of getting to take the money out is important. Why take that away?

Leave a Reply