28 Mar 2014

Krugman on Tax Theory

Krugman 31 Comments

[UPDATE below.]

It’s one thing for Krugman to tell us what the Republicans or people at the Heritage Foundation are “really” after, etc., but it’s quite another when he just makes stuff up about objective economics literature. In a recent post lambasting Greg Mankiw (who won’t just admit he’s a moral monster like a man), Krugman writes:

[T]he case for zero or low taxation of capital income rests on very strong, very unrealistic assumptions — basically perfectly rational intertemporally optimizing agents, with dynasties behaving as if they were infinitely lived individuals. Question those assumptions, and the whole case falls apart.

The thing here is that the standard case for, say, a flat tax on consumption–meaning zero taxes on income, including dividends and capital gains–pops out of a simple two-period model. You don’t need a perfectly rational agent who lives forever, you need an agent who can look ahead one period and who realizes “If the government taxes my interest income, that makes future consumption less attractive on the margin than sheer considerations of time preference would suggest.”

I’m not being coy here. I actually had to alter my discussion in a Flat (Income) Tax report (before it had been released–it was still in the review stage) I did for Pacific Research Institute because I realized I wasn’t handling this issue properly, after I got feedback from a mainstream tax economist. So if consideration of a simple two-period model made me “get” the standard mainstream economist’s case for a flat consumption tax, such that I actually told the graphics people I needed to tweak a few sentences, then you should get a sense of how I feel about Krugman’s claim above.

In a post the next day, Krugman goes on to claim: “[W]e don’t actually know much about how to produce rapid economic growth — conservatives may think they know (low taxes and all that), but there is no evidence to back up their certainty.”

Hmm I guess Krugman missed this empirical paper (in which the authors, to my knowledge, do NOT use a model relying on hyperrational agents with infinite time horizons), which concluded:

Our baseline specification suggests that an exogenous tax increase of one percent of GDP lowers real GDP by roughly three percent. Third, investment falls sharply in response to exogenous tax increases. Indeed, the strong response of investment helps to explain why the output consequences of tax changes are so large. Fourth, the output effects of tax changes are highly persistent.

That paragraph comes from a 2010 publication in the top-ranked journal, the American Economic Review. The authors were David and Christina Romer.

If there’s zero evidence that cutting taxes leads to economic growth, it’s amazing I was able to generate such a long essay at EconLib listing it.

Last thing: I am sure someone will defend Krugman by saying, “He said there’s no evidence to back up their certainty, not that there’s no evidence for their conclusion.” But again, if that’s the route you wish to take, then please give Krugman yet another award for breaking his own record of Most Misleading Blog Post of All Time. Krugman certainly didn’t produce quotations from conservatives saying, “I would bet my firstborn son that cutting tax rates produces faster real GDP growth,” rather he just invented a generic position and then said there is “no evidence” to back up what conservatives think they know.

UPDATE: Ken B. in the comments was pushing back on my claim that Krugman just “makes stuff up” about the optimal tax literature. It may interest some of you to see what I wrote in reply:

Ken B. I am not going to get into a discussion with you about this. If you follow Krugman’s link you will see where he is coming from, but I think those authors are being obtuse. Look, I am perfectly confident in saying, “The optimal tax literature says it would be insane to levy a sales tax on bananas at 90% but on apples of 2%.” Yet, every single model in the literature would rely on unrealistic assumptions that were violated empirically as documented by the behavioral econ guys. And I could come up with another model in which aliens said, “We will blow up Earth unless you all tax bananas at 90% but apples at 2%.” Then lo and behold, I would have completely blown apart the result, since in my new model–which doesn’t stick to all of the rigid and unrealistic assumptions in the conventional literature–the social planner does indeed set optimal tax rates that way.

If you think I’m suffering from Krugman derangement syndrome, look at the PPS in Sumner’s blog post. He goes nuts on Krugman too. So if you want to argue with somebody and defend Krugman, see if Sumner has the stamina.

31 Responses to “Krugman on Tax Theory”

  1. Ken B says:

    “T]he case for zero or low taxation of capital income rests on … assumptions — basically perfectly rational intertemporally optimizing agents, with dynasties behaving as if they were infinitely lived individuals. ”

    That’s a pretty fair summary of Chamley and Judd’s asumptions isn’t it?

    • Bob Murphy says:

      Probably. You saw the rest of the quotation right Ken? The whole case for low taxation of capital income falls apart if I doubt that people behave as infinitely lived individuals?

      No it doesn’t.

      • Ken B says:

        But Krugman doesn’t actually say individuals act that way. He says the assumptions are
        1. Intertemporal rationalization by indivduals, and
        2. Dynasties acting like they were long or infinite lived individuals ie maximizing something for our descendants.
        Which is not quite what you say.

        Now your argument if I get it is 1 is enough. Is that correct?
        Intertemporal thinking ahead. And it pops out of a model when you do just two periods. Is that correct?

        So then you are saying 2 is just smoke Krugman is blowing. Correct?

        I have a follow up, but let’s get this clear first.

        • Bob Murphy says:

          Ken B. I am not going to get into a discussion with you about this. If you follow Krugman’s link you will see where he is coming from, but I think those authors are being obtuse. Look, I am perfectly confident in saying, “The optimal tax literature says it would be insane to levy a sales tax on bananas at 90% but on apples of 2%.” Yet, every single model in the literature would rely on unrealistic assumptions that were violated empirically as documented by the behavioral econ guys. And I could come up with another model in which aliens said, “We will blow up Earth unless you all tax bananas at 90% but apples at 2%.” Then lo and behold, I would have completely blown apart the result, since in my new model–which doesn’t stick to all of the rigid and unrealistic assumptions in the conventional literature–the social planner does indeed set optimal tax rates that way.

          If you think I’m suffering from Krugman derangement syndrome, look at the PPS in Sumner’s blog post. He goes nuts on Krugman too. So if you want to argue with somebody and defend Krugman, see if Sumner has the stamina.

          • Bob Murphy says:

            Oh another thing Ken B.: Even on narrow technical grounds, yes Krugman is indeed just making stuff up. Go read the very paper he cites. There are TWO benchmark results that argue for low or zero capital income taxation. One result uses infinitely lived agents (or overlapping generations blah blah blah), but the other standard result relies on a two-period model. So the authors do try to knock out both results, but the point is Krugman’s summary is wrong. He makes it sound like it’s one particular set of rigid assumptions, when no, at best he has to say, “There are two completely different modeling approaches to get at this, one assuming infinitely lived agents and the other using a two period analysis. But my clever friends who have built a career on justifying soak-the-rich tax policies have come up with arguments to challenge both results.”

    • Bob Murphy says:

      Let me put it this way Ken B: You read Landsburg a lot right? At least once a year Lansburg goes off on how stupid it is to tax dividends and interest, because they’re just effectively ways of using today’s wage income to either buy an “apple in 2014” versus an “apple in 2016.” Have you ever seen Landsburg preface one of those blog posts with, “Now this is important: Assume the guy in my thought experiment has a constant utility function of U(t) = C(t), where t runs from 1 to infinity.” ?

      • Ken B says:

        “You read Landsburg a lot right?”

        Nope. I read him when he blogs.
        This reminds me of an old joke.
        “Sex is better in your 60s. Especially the one in the spring.”

  2. Tel says:

    Taxing tobacco is a way to discourage smoking, taxing alcohol is to discourage drinking, charging people a fine for speeding (to all intents and purposes a tax also) discourages speeding.

    … very unrealistic assumptions — basically perfectly rational intertemporally optimizing agents …

    Translation: regular people are too dumb to figure out that we are taxing their savings.

    Uhhh, actually, we aren’t that dumb, but quite a lot of people just stopped caring about having savings thanks to Keynesian oppression.

  3. Tel says:

    [W]e don’t actually know much about how to produce rapid economic growth…

    But we sure know what prevents rapid economic growth: North Korea, Venezuela, Zimbabwe, etc.

    • guest says:

      Krugman should be asked, “Rapid economic growth” for which SPECIFIC individuals?

      He’s framing the issue as if the “collective” is what economically grows. The premise is wrong.

      It doesn’t matter AT ALL what are the aggregate numbers. Only individuals economically grow.

      • Major_Freedom says:

        The only way to sacrifice individuals on the top of the temple, for thugs like Krugman, is to pretend it’s only the totals that really matter. It also helps when the rich are sacrificed more, because something something fairness.

        • guest says:

          Whenever you say that thing at the end, it reminds me of the Underwear Gnomes.
          😀

      • Tel says:

        He’s framing the issue as if the “collective” is what economically grows. The premise is wrong.

        It doesn’t matter AT ALL what are the aggregate numbers. Only individuals economically grow.

        That happens to be your view, there’s no reason Krugman would see things that way. The aggregates are all important to him.

        If you look at Detroit, it would be difficult to say what happened was good for the community… probably if you search hard enough you might find a few individuals (generally corrupt individuals) who came out of it OK. The high brass in North Korea also do OK, while most of the people go hungry. You always find some individual does well out of wars, strife, etc.

        I agree that it is difficult to measure an objective improvement to a community as a whole. These sort of measurements are strongly subjective. Even accepting the subjectivity of it, who is jumping up and down eager to live in North Korea?

        Very few people, I’d guess.

        • guest says:

          Only individuals act.

          In both Detroit and North Korea, government prevents the individual from growing economically.

          By treating people as a collective, government is disincentivising specific individuals from engaging in production, which leads to shortages.

          If you’re rich, but there are shortages, then there’s no reason to stay.

          But if you take present-day Detroit, and just take the government out of the way …:

          Amid Detroit’s Bankruptcy, A New Vibrancy Emerges
          https://www.lewrockwell.com/2013/12/karen-de-coster/detroit-is-increasingly-vibrant/


          In spite of a coast-to-coast recession and ensuing uncertainty regarding this country’s economic prospects, the core of the city center, as well as many of the city’s historic neighborhoods, have been experiencing an economic revitalization driven by private investment, philanthropy, and perseverance on the part of communities that are reshaping their existence through hard work, hope, and an ample supply of faith and effort.

          The New York Times referred to this phenomenon as “a private boom amid Detroit’s public blight,” and CNN Money called Detroit “the nation’s hottest downtown.” Even as city services disappear further into the abyss, urban gentrification is full steam ahead and occupancy rates have skyrocketed to as high as 99 percent in Midtown, a place that was once designated as one of the seediest districts in the city.

          • Tel says:

            If you believe that only individuals act, why bother complaining about the actions taken by government?

            • guest says:


              If you believe that only individuals act, why bother complaining about the actions taken by government?

              Yes. Very astute.
              😀

              I concede that this is imprecise. “Government” doesn’t do anything – individuals which comprise the government do.

              But I am using the word “government” metonymically, to refer to the use of force that is typical among all of those individuals which comprise the government.

              I was wondering if I would ever have to break that down.

              • Tel says:

                Sure, and there would also be a force that is typical amongst all the North Korean elite who look after their own affairs, and also call themselves “government”.

                They are good at looking after their own affairs, they just don’t do a good job of looking after the country as a whole. Then again, since they act as individuals, why would they care about looking after their country?

              • guest says:


                Then again, since they act as individuals, why would they care about looking after their country?

                Because they believe in the Labor Theory of Value, and therefore that profits made by hiring laborers to operate privately owned capital consist of a partial theft of the value that a laborer imparts to the finished good.

                If you’re a Marxist, you believe that materials and capital have their own use-value, which is all well and good; The Austrians would agree with this.

                But if in order to make a finished product, you employ labor, well then any value that the finished good sells for over the price of the materials that went into the good, as well as the maintenance of the capital, MUST be a result of the labor that went into the process.

                Therefore, according to Marxists, all profits made by capitalists paying wages to laborers, MUST be stealing a part of the laborers earnings.

                So, in order to keep this theft from happening, the Marxist central planners want to prevent capitalists from making too much, so as to reduce the amount of “theft” occurring.

                This is why they like regulations.

                But since value comes from the consumer, and all value (and costs) are subjective to the individual, the Labor Theory of Value is wrong, and the Marxist central planners end up destroying price information that people need in order to plan their own lives for themselves.

                This is why central planning is a fatal conceit.

                I offer this video again, for anyone trying to understand the Austrian’s theory of value:

                The Birth of the Austrian School | Joseph T. Salerno
                http://www.youtube.com/watch?v=dZRZKX5zAD4

          • Tel says:

            Good luck to the people of Detroit, by the way… if it really is making a resurgence.

            • Ken B says:

              Count me skeptical. I live in a western suburb.

  4. Major_Freedom says:

    One can use his warped logic and conclude “The case for taxing capital gains rests on very strong, very unrealistic assumptions — basically perfectly rational intertemporally optimizing government agents, with dynasties behaving as if they were miniscule lived individuals. Question those assumptions, and the whole case falls apart.”

    You see, in Krugmanistan, governments are smarter spenders of money than everyone else, period.

  5. Ken B says:

    Point of order. I have not pushed back, at least not yet. I asked you to clarify exactly which parts you think Krugman got wrong. If you read my comment carefully you will see I am trying to isolate parts of Krugman’s claim and deal with them separately.

    • Bob Murphy says:

      Fair enough Ken B. I hope my responses give you what you need. Click on Mankiw’s link too; that paper by Kehoe et al. specifically explores the relaxation of the assumptions in the classic papers.

      • Ken B says:

        Already had browsed it.
        My take is that Krugman’s “with” is conjunction, so what he says is wrong. People thinking ahead, direct consequences in other words, incentives can do the job without ricardin equivalence or what ever infinite life stuff you choose.
        He does address intertemporallity seprately but not convincingly. This is clearly a case where the agents don’t need to be perfect for a qualitative argument to be persuasive.

        • Major_Freedom says:

          Persuasive to who?

          • Ken B says:

            To Bob apparently, and he hopes to his readers. If you think Bob’s argument unpersuasive talk to Bob.

            • Major_Freedom says:

              Thought you meant something else, because that was too obvious.

  6. Gamble says:

    When tax rates are steep/excessive, it does not matter what system you have in place. Tax rates multiply one and other, soon enough you reach 100% or more…

    Sit down someday, add up all the taxes, at every point, and hopefully realize the sum total is far greater than the individual points. Then smoke a fatty and realize confiscation rates may currently exceed 100%. It is only, subsidy, exemption and other tricks that place some of the confiscation back into play. Political economy.

    I know I am talking way past current thinking, I may be delusional, but there are a few of us who think like this. Mainly those of who have went to great lengths to expand our minds. Our altered DNA allows us to see truth…

    If you want 1 consumption tax, the rate still needs to be less than 10%…

    Rates and frequency are more important than the where and how.

    In summary, they confiscate all of it; waste some of it, utilize some of it, give back some of it…

  7. Innocent says:

    Okay perhaps I am being silly here but don’t people place sin taxes on things in order to discourage consumption. Just from a purely simplistic approach, why would we expect increases in taxes to increase consumption or even keep consumption at the same levels? I know the Keynesian model says you are simply taking from Peter who is not utilizing the recourse and giving to Paul that is but that would only be true if you were not already borrowing money for said consumption at a Federal level anyway.

    Increased taxes, just like sin taxes, are not felt immediately, well people who honestly cannot afford to participate in the vice drop out immediately, so people who could afford to invest a little would drop out fairly quickly ( I would suggest this is where a 1% increase in GDP taxes lowers GDP growth by 3% ) but even this issue does not follow the tax long enough to see what ultimately happens as decades would have to pass for the effect to be truly felt, this is of course conjecture on my part as no real world case beyond ‘sin taxes’ can really be used to model this. It can take decades for this ‘sin tax’ to truly change the behavior of a society,

    Why would we expect there to be any less time taken in changing the way we have historically invested?

    Additionally this leads to mal-investment over time in an effort to return what is perceived to be an adequate return in comparison to what you had before. Investors chase ‘the dragon’ rather than accept the status quo.

    Man is not a rational animal he is a rationalizing animal. I swear Mr. Krugman thinks that people are fools that cannot see their own self interest moving forward. Perhaps people are slow to change but people over time figure out what ‘works’ for themselves. It typically is this false sense of security that causes mal-investment. I remember back in 2003 when I was house shopping and housing was going up 10%+ year over year I asked a real estate agent what the heck was going on, that this could not continue and there would be a fall. Her answer to me was, ‘Oh that can’t happen with real estate.’… I was shocked and refused to purchase a home until the market came down…

    Now admittedly I thought the crash would happen in 2005 and not be as severe as it turned out to be. But hey who can’t underestimate the herd mentality. Oh well…

  8. Gamble says:

    One benefit of flat consumption tax AND NO OTHER TAXATION is that it is easier to mentally comprehend market effects and total severity/burden.

    Currently, a single dollar is taxed so many times and places, we don’t know if we are coming or going. Think about it, a dollar is created in the center and tippy top of pyramid and this dollar is taxed relentlessly, numerous times. We never get to create more dollars as we produce and consume. The creation is central and out of sync. Talk of multiplier effect and all the rest is somewhat of an illusion or smokescreen if you will.

    In todays system, nobody and I mean nobody, understands the real taxation amount and burden. This is a political system, not an economy. Listen to me, this is not an economy based on scarcity, supply demand, consumer preference. This is a political system based on power, control and favoritism…

Leave a Reply to Ken B

Cancel Reply