Two Questions for Bob Murphy

Bob Murphy objects to my recent defense of Jonathan Gruber. I have two questions for Bob.

Suppose a newly elected Republican president wants to exempt all investment income from taxation. There are two ways to do this:

1) Retain the income tax, but exempt all interest, dividends, and capital gains (while also abolishing the corporate and estate taxes).

2) Scrap the income tax and replace it with a national consumption tax.

The president’s chief economic advisor, like all economists, is well aware that these two policies are essentially equivalent in the sense that, once prices, wages and interest rates adjust to the new policies, each individual taxpayer is burdened exactly as much by policy 2) as by policy 1). More precisely, at least following an initial adjustment period each individual taxpayer enjoys exactly the same lifetime stream of consumption under policy 2) as under policy 1).

Let’s suppose also that the chief economic advisor believes that policy 1) is vulnerable to scurrilous class-warfare-themed attacks and therefore cannot be sold to the American people. Policy 2), however, stands a chance of passage. He therefore goes around honestly touting what he perceives to be the clear virtues of policy 2), choosing not to mention that it’s equivalent to policy 1).

Question 1: Has that chief economic advisor behaved reprehensibly?

Question 2: Is there any essential moral difference between the strategies employed by that advisor and by Jonathan Gruber?

Note: An opponent of the ACA could, of course, argue that unlike this economic advisor, Gruber used his powers for evil. But if that’s your argument, note that we knew all along that Gruber was promoting the ACA, and therefore the recent revelations do not add to his culpability.


40 Responses to “Two Questions for Bob Murphy”

  1. 1 1 Harold

    Surmise for a moment that it was possible to disguise some earned income as dividends. Policy 1 is no longer the same in practice as policy 2.

  2. 2 2 ThomasBayes

    This might connect more with an earlier post than this one, but it does seem to connect to both . . .

    Would it be sound economics to use the following argument to justify a claim that a consumption task benefits low- and middle-income Americans at the expense of the elite?

    “Even if the costs of a consumption task are passed on in the form of higher prices, the rich are different from you and me. They spend a lot more money, and, therefore, bear a higher share of the costs.”

    How about using a similar argument to justify a claim that pollution control taxes benefit low- and middle-income Americans at the expense of the elite?

    “Even if the costs of pollution control are passed on in the form of higher prices, the rich are different from you and me. They spend a lot more money, and, therefore, bear a higher share of the costs.”

  3. 3 3 Yancey Ward

    And when it is pointed out that the two are the same, what does said economist say, Steve?

    There is a reason we have a class of lies called lies of omission.

  4. 4 4 Bob Murphy

    (1) If video surfaced of the guy laughing about how he had exploited the stupidity and ignorance of the American people, then I certainly wouldn’t be writing posts defending him. But as far as the scenario you paint above, I could imagine a particular way it would play out, where I wouldn’t say the guy acted reprehensibly.

    (Note that Steve refers to the advisor as “he.” I don’t want women to think I am dismissing the possibility that they too could throw the public under the bus while advising politicians.)

    (2) I don’t think this is a great analogy. If people reject option #1 because it taxes them more heavily than rich capitalists, then they would have to be pretty dense to think that getting rid of all those pre-existing taxes on rich capitalists and replacing them with a tax just on consumption, is somehow better in that dimension.

    In contrast, having someone say “We can either take away your tax break or impose a new tax on rich insurance companies” is not nearly as easy to interpret as the same, unless you’ve had some training in economics. *I* botched the initial draft of my post. That was an inexcusable mistake for me to make–since I taught this stuff to undergrads for three years–but it’s a perfectly reasonable mistake for a general member of the public to make.

  5. 5 5 Keshav Srinivasan

    Steve, in policy 2 do you also intend to abolish the corporate and estate taxes?

  6. 6 6 Jon Sanders

    I have paid income taxes on the money I have saved. Now you want me to pay a consumption tax when I spend my savings, effectively taxing me twice.

  7. 7 7 Bob Murphy


    Maybe this will also help clarify things: Do you think it’s “incoherent” for most Americans to have the following preferences?

    (1) I like the axiom of choice.

    (2) I dislike the claim that I can cut an orange up into pieces and assemble them back into something as big as the sun.

  8. 8 8 Bob Murphy

    And if you say, “Yes, that’s incoherent”–which is fine–would you use the term “incoherent” in a way that means, “So how can we be expected to try to talk to these people? We might as well package what we want in a way that they won’t understand”?

  9. 9 9 Roger

    I thought that the only problem with Gruber is that he committed a Kinsley gaffe. “A Kinsley gaffe occurs when a political gaffe reveals some truth that a politician did not intend to admit.”

  10. 10 10 Steve Landsburg

    Jon Sanders: Yes, the two taxes are inequivalent for the generation that is alive during the transition. They are equivalent going forward for future generations. Given that, if you went with the consumption tax, you’d have to do one of three things: Either a) say that this is unfair to a single generation, but one generation is not enough to care about; b) say that this is unfair to a single generation, but we’re not concerned with fairness anyway, or c) make a one-time transfer to people with existing savings to counteract the unfairness. My preferred choice is c), but I think I could live with any of them.

    Of course for purposes of the Gruber analogy, it’s important that the two policies be fully equivalent, so we should assume that c) has been chosen.

  11. 11 11 Keshav Srinivasan

    Bob, what do you mean by liking and disliking those claims? Do you mean believing them or not believing them?

  12. 12 12 Steve Landsburg

    Bob Murphy: Okay, I concede that people can hold contradictory beliefs without realizing that they’re contradictory, and that it does not follow that we have to throw up our hands and give up on communicating with them. You’re right about this.

    But I stand by my example of the economic advisor who chooses to tout a consumption tax he believes can pass, as opposed to an equivalent tax he believes cannot pass. I continue to be pretty sure I’d have no problem at all with that behavior, and I continue to believe it is not significantly different from what Gruber did.

  13. 13 13 Max

    The acid test is, could a fully informed (and disinterested) person accept the policy? If so, then it’s a legitimate “product” to be sold using whatever sales pitch works. Selling to the less-informed is not the same as scamming, although to an academic like Gruber they may both be unwelcome reminders of an imperfect world.

  14. 14 14 Bob Murphy

    OK Steve let me think about it for a few days and I’ll post something at my blog. (I’ll paste the link here too for the 4 people who care.) What I still don’t like about your consumption tax example is that nobody could be confused as to what they were getting with it. Nobody could say, “Oh wait a second, when you got me to support scrapping the old income tax code and replacing it with a consumption tax, I had no idea that meant rich people wouldn’t be taxed on their dividends.”

    In contrast, if you applaud John Kerry sticking it to the big rich insurance companies, and then the market reacts in a way that is equivalent to losing your own tax break, *that* seems like you were taken advantage of.

  15. 15 15 Harry

    Steve Landsburg: So the trained economist advising the president of option C, which explains away the problem of screwing the people who have saved their money through another Rube Goldberg system to adjust for unfairness? I know this question has little to do with Jonathan Gruber, but since you raised the subject, one of the controversies about replacing the income tax with the “Fair Tax” is exactly how one goes about it fairly. Moreover, there are good arguments for eliminating taxes on investment income, some of which might have popular appeal to young and old alike, so it is not obvious that option 1 be rejected in favor of option 2 as politically palatable. A national sales tax a la the European VAT in the neighborhood of 20% in my opinion is a political and economic turkey. Maybe we can make a better analogy to defend the Machiavellian Dr. Gruber, maybe even one that plausibly shows that a Republican president would be as corruptible as a Democrat president.

    Dr. Gruber was one of the hired guns. He happened to be the architect of Romneycare, so he was among the right people to do the job, but so was Ezekiel Emmanuel. Now that the s… is hitting the fan, it is no wonder that people are pointing fingers. Does Nancy Pelosi remember Zeke?

  16. 16 16 Jimbino

    I don’t see how a consumption tax could be equivalent to an income tax for the person who can take earnings in a no-income-tax state and do most spending in a no-consumption-tax state.

  17. 17 17 nivedita

    A couple of questions that are only tangentially related to the topic.

    1) What is the general argument for why the legal incidence of the tax doesn’t matter? I’ve only seen the argument in terms of supply/demand curves and relative elasticities determining tax incidence, but it seems to me that the result should be more general: it should hold even for monopolistic suppliers, where there is no well-defined supply curve.

    2) What happens when the tax is not uniform? i.e. Legal incidence doesn’t matter when the amount of tax per unit depends only on the price. Does it also not matter when different consumers pay tax at different rates? If the “Cadillac” tax were implemented by making the portion of the total insurance premium above some fixed number count as taxable income to the employee, different employees pay different rates depending on their marginal income tax rate. Is there a tax that can be imposed instead on the insurer (or the employer), which would have exactly the same economic consequences?

  18. 18 18 Steve Landsburg

    Jimbino: We are talking about national tax policies.

  19. 19 19 Steve Landsburg


    For your question 1, just write down the maximization problem. The seller maxes (P(Q)-S)Q-C(Q)-TQ where P(Q) is the price at which quantity Q is demanded, C(Q) is the cost of providing quantity Q, S is the per-unit tax on demanders and T is the per-unit tax on suppliers. If you increase S and decrease T by the same amount, the problem is unchanged.

    Your question 2 is considerably more difficult in principle but I suspect not in practice, since pretty much all the people getting Cadillac plans are likely to be in the same (highest) bracket.

  20. 20 20 J Paulson

    I think your argument about the two system being equivalent may be true in a theoretical sense, but in the real world there could be a huge benefit to a consumption tax.

    Various estimates of the underground economy float around the $2 trillion mark. If the consumption tax were paid at the point of consumption, these folks would be forced to pay their share. As a result, those of us law abiding folks would see a decrease in our tax burdens for a given level of revenue to the government.

  21. 21 21 Marco

    J Paulson: If the consumption tax rate gets large enough, then even consumption will go underground. Hence that VAT, with many, small increments may reduce cheating even though it seems to introduce more compliance points.

    Steve Landsburg / Jimbino: For the 99%, restricting this to national taxation solves the problem of earning in a low income tax jurisdiction and consuming in a low consumption tax jurisdiction, but there will be highly visible avoidance. For example, I might go golfing in Bermuda instead of Pebble Beach, sailing in the Greek Islands instead of Catalina Island, send my children to Swiss boarding schools instead of Exeter (assuming schooling is consumption and not investment). I do not claim this will a huge problem in absolute collections (remember J Paulson is correct on the underground economy, and he omits the legal gaming of the complexities of the current system), but I expect it will create bad ‘optics’.

  22. 22 22 Steve Landsburg

    Marco: Whether or not consumption can easily go underground depends not on what you’re taxing but on how you collect it.

    Instead of collecting consumption taxes at the point of sale, I can just as easily have you fill out a tax form once a year that has you document your income (just as we do now), document your savings, and pay tax on the difference. This makes your golfing in Bermuda no easier to hide than your trip to the local supermarket.

  23. 23 23 Marco

    Thank you, Steve Landsburg. But then the consumption tax loses some of its attractiveness: It is not only not simpler than the existing income tax, but it is more complex because it adds to the problems of our current scheme (defining income and collecting it from millions on individuals) and introduces the complexity of defining investment (my example of education, which is surely both investment and consumption).

  24. 24 24 Steve Landsburg

    Marco: But if you’re going to have a consumption tax, you’re going to have to decide whether (or how much) to count education expenses at consumption. That’s exactly the same problem whether you tax education at the point of sale or whether you tax [Income-Saving] once a year, and have to decide whether a year of school counts as a form of saving. So I don’t see where the once-a-year tax collection makes things any less simple.

  25. 25 25 Marco

    Instead of having hundreds of millions of taxpayers reporting income and savings a point of sale consumption tax scheme has maybe one million businesses reporting consumption. And the old Pareto’s Rule of Thumb means that you probably get close to 100% compliance out of a small number of big hitters (Wal-Mart, Amazon, etc.).

  26. 26 26 Ross Levatter

    It seems to me the real problem with what Gruber did is (rather explicitly, per some of his comments) work to hide the actual costs of Obamacare because, in his view, it was worth having. To say something is worth having independent of cost is thoroughly anti-economic thinking. To say, “*I* think it is worth the cost so I’m hiding the cost from anyone who might disagree” is a thoroughly anti-democratic way of thinking.

  27. 27 27 Jimbino


    You may be talking of national tax policy, but I’m talking about working in in the USSA and spending for you vacations and yachts in Taiwan, and so on, which companies and rich people can do.

  28. 28 28 Johnson85


    It seems like your analogy is poor. To make your analogy better, the tax rate in policy one and two need to be set such that it is effectively a tax increase for the average person. Then you need the politician to sell policy one by claiming that unlike policy two, it is not a tax increase.

    Had Gruber simply advocated for Obamacare and relied on the ignorance of the voters, that would have been politics as usual. But he went out of his way to mislead the voters (including misleading them about his role as a paid consultant). Hopefully he will pay a personal and political price for what he did. We’re obviously gonig to have laying politicians because that’s what is rewarded, but I’m not sure if our gov’t can even function if politicians and their paid lackeys can boldly lie to get major expansions of gov’t passed without repercussions.

    , and representative democracy when you have a gov’t behemoth like ours is pretty meaningless if politicians and consultants can mislead the public without repercussion.

  29. 29 29 Al V.

    To Marco’s points in #21 and #23, the effect of a national consumption tax would be that many things that are today identified as purchases would be redefined as investments. For example, rather than purchasing a car, I could purchase an ownership share in a corporatation that provides me with a free car as one of the benefits of ownership. I would bet that people would get quite creative in avoiding the consumption tax.

  30. 30 30 Steve Landsburg


    You may be talking of national tax policy, but I’m talking about working in in the USSA and spending for you vacations and yachts in Taiwan, and so on, which companies and rich people can do.

    Why does it matter whether I vacation in Taiwan or in Omaha, as long as I’m paying the same tax rate on purchases in both places?

  31. 31 31 Advo

    Are we sure that increasing investment capital at the expense of consumption is really going to produce higher economic growth?
    As an extreme thought experiment, consider what would happen if we all saved as much as possible and restricted our consumption to the bare minimum (i.e. bread, basic clothing, cheap transportation). If we did that, there’d be something like a permanent depression and little in the way of technological progress. We’d all be impoverished forever. This extreme example makes it clear that somewhere there is an optimal balance between saving/capital investment on the one hand and consumption on the other. How do we know on which side of that optimal point we are? Given that long-term interest rates are extremely low DESPITE the end of QE, and given the various problems caused by disinflation/deflation (see: Japan) perhaps we should give some thought to increasing consumption instead of trying to drive it down further?

  32. 32 32 Steve Landsburg


    As an extreme thought experiment, consider what would happen if we all saved as much as possible and restricted our consumption to the bare minimum

    Usually, when we do thought experiments in economics, we retain the standard assumption that people are behaving rationally. Therefore, in your thought experiment, it must be that people want to reduce their consumption drastically (probably because they’ve decided that they really hate working). Therefore the outcome you describe is a good one.

  33. 33 33 Harold

    #31. Interesting idea – it is basically Keynes is it not? Reduced aggregate demand results in recession.

    If the reason were that everyone permanently adopted something akin to Bhuddism, and decided that desire leads to unhappiness. What would everyone do? There would be precious little reason to increase savings, because that is consumption deferred. Everyone would scale back on un-desired effort (work) until they just met their needs, and would spend the rest of the time on leisure. For some, work would be leisure. It would be necessary to do enough work to feed and clothe everyone, and therefore we would need some infrastructure, transport etc.

    Assuming people desired to heal the sick and have children, then society would devote considerable effort in these areas. The resources we have would go further, and the population would grow. the economy would therefore grow with it, but it would be based on basic food clothes essential infrastructure and medicine. Eventually the size of the economy would grow larger than the current one, and would continue to sustain a larger, growing population.

    Thus eventually, this change in preference from consumer goods to other things would lead to a growing economy.

  34. 34 34 Advo

    The outcome may be caused by government influence – perhaps the government massively taxes consumption while subsidizing savings or perhaps it has imposed puritanical consumption laws. Alternatively, the individual decision may be rational but the aggregate outcome suboptimal. (see: Paradox of Thrift)

    In any event, you avoided the main point of the thought experiment. From a certain point on, increasing the savings rate will lead to lower economic growth as well as slower innovation and productivity growth.

  35. 35 35 Advo


    I am not talking about a population that reduces consumption in favor of leisure time or whatever.
    I’m talking about a population that works hard and minimizes consumption in order to maximize savings.
    WHY they are doing that (sumptuary laws, religious ideology, whatever) is beside the point for the question of what the economic consequences will be. As is the speculation about population growth.

  36. 36 36 Harold

    #35. If people reduce consumption in favor of saving, then to be consistent they would be anticipating greater consumption in the future. It sounds like the paradox of thrift, and it is certainly possible that the economy would shrink and overall savings would go down. The counter argument seems to depend on using the savings to increase efficiency of production. Plausible, but not certain.

  37. 37 37 David Cushman

    Suppose we use the Solow growth model with population and technology growth given, and we start from long-run equilibrium. Let the saving rate rise, but labor works as much as before, as Advo specifies. In the short run, consumption falls. In the long run, capital and output per person rise, but consumption per person may rise or fall. The latter depends on where the country is initially with respect to the golden rule point, or whether the changes are large enough to overshoot it. Indeed, there would be some sufficiently high saving rate that would push the country to lower consumption per person. Much of the higher output would then be going to offset depreciation of the very large capital stock.

  38. 38 38 Harold

    #37 Thanks for the clarification. Sort of on the one hand, or on the other hand thing. If output per person rises in the long run, how can consumption per person fall? Who is consuming the extra output? Is this due to population growth or are we not in a closed economy?

  39. 39 39 David Cushman

    “If output per person rises in the long run, how can consumption per person fall?”

    When saving rises, capital accumulation rises because saving = investment, and investment either adds to existing capital or replaces worn out capital. Over time, capital per person rises to a new, long-run, high, steady-state value. With all this capital, there is also a lot of output per person. This could lead to more consumption. But, if the amount of capital is sufficiently large, given a depreciation rate the amount of output required to replace depreciated capital will also be large, and therefore the amount available for consumption (per person) could be lower than before the saving rate rose.

  40. 40 40 Advo

    You also have the phenomenon of diminishing returns and capital misallocation. The marginal additional capital will be put to less and less productive uses.

  1. 1 Quicktake: Landsburg on Gruber | RWCG
Comments are currently closed.