Skip to content

The “Carl Sagan effect”

Javier Benítez writes:

I am not in academia, but I have learned a lot about science from what’s available to the public. But I also didn’t know that public outreach is looked down upon by academia. See the Carl Sagan Effect.

Susana Martinez-Conde writes:

One scientist, who agreed to participate on the condition of anonymity—an indicator of his perceived vulnerability to the Sagan Effect—left his research institute as a junior faculty member because he felt that the institute’s director—who had chided him about communicating with the press—was blocking his advancement to associate professor after there had been extensive media coverage of his work. The same researcher, who has published in the highest-impact journals, said that he has been unable to get a grant after further recent media coverage and a giving a related lecture at a TED conference. He has declined an invitation to give a second TED talk in light of the criticism, and will not do further media interviews at present. The worst for me was the grants. Since this paper [covered extensively in major international media], all my grants got rejected with terrible comments. It was suddenly completely changed. I had 25 grants rejected since the paper in [name of top tier journal].

Has Contemporary Academia Outgrown the Carl Sagan Effect? Journal of Neuroscience 17 February 2016, 36 (7) 2077-2082; DOI:

I know there are bad TED talks out there and some of it may even be pseudoscience, but how can there be an informed public about science when outreach is discouraged?

“Pseudoscience is embraced, it might be argued, in exact proportion as real science is misunderstood – except that the language breaks down here. If you’ve never heard of science (to say nothing of how it works), you can hardly be aware you’re embracing pseudoscience.” Carl Sagan – The Demon-Haunted World (1996)

My reply:

I agree that one of the duties of academic research is service, and part of this can be discharged by communication to general audiences. On the plus side, if you can communicate to the general public, then you’re reaching more people who can uncover flaws in your ideas. So one of the benefits of public exposure is that you can get some valuable critiques from the outside.

Regarding the quote at the end: 25 grant applications seems like a lot. Who applies for 25 different grants??


  1. Clyde Schechter says:

    “Who applies for 25 different grants??”

    Well, over what period of time? If this is five years of experience, it wouldn’t be surprising. Also is this only grants as principal investigator, or are applications as key personnel other than PI included? Are resubmissions of a rejected grant to the same or other funding agency included? If you count all of these things, an average of 5 grant applications per year in health care wouldn’t be unusual. And, indeed, after a couple of years of failing to bring in funding, the pace of applications might increase (and, I suspect, as a result the quality would also decline!)

  2. > Who applies for 25 different grants??

    Someone who doesn’t get the first 24?

    Also, in some areas of research it’s not uncommon to have 5 different projects you are working on. If you have an impressive 20% success rate at getting grants you apply for, that would mean 25 grant applications. (When I was at RAND, many people I knew were working on more than that, albeit usually as part of a team that collaborated on the projects.)

  3. jack pq says:

    1.) I don’t buy the ‘Carl Sagan effect’. Rather, I think there’s reverse causality. If you have trouble publishing cutting-edge work, you turn to something else: popularizing science. Moreover, spending too much time being a communicator takes time away from doing research. We can’t all be like Andrew and do both well! :) Sagan wasn’t denied tenure at Harvard because of envy, he was denied tenure because he wasn’t a top researcher.

    2.) I’m surprised that a researcher would be penalized for getting media attention and TED talks. Universities LOVE this sort of attention! But, it is true even if uni admin loves it, colleagues who review grant applications may not. I don’t think this is true in general, unless the researcher is led to make overly ambitious claims about his or her research following all the media attention.

  4. I strongly disagree that “public outreach is looked down upon by academia.” (Background: I’m a tenured academic physicist.) At most universities, outreach is considered valuable and admirable, and people who put effort into it are appreciated.

    That said, there is a strong dislike of (i) people whose “public” presence is disproportionately large compared to their research presence, and (ii) people whose public presence looks a lot like self-promotion, rather than promotion of science in general. But these aren’t criticisms of outreach and public engagement activities. And of course, it does seem like there are particular institutions and cases in which public outreach was actually penalized; I think these are, however, few.

    • Z says:

      “there is a strong dislike of (i) people whose “public” presence is disproportionately large compared to their research presence”

      Why should public presence be proportional to research presence? Public presence != recognition of achievement. Sure maybe some laypeople confuse the two, but I don’t see why that should be particularly annoying. If some kid thinks Bill Nye the Science Guy is in the pantheon of great scientists that’s no skin off my back.

      • Raghuveer Parthasarathy says:

        No one is annoyed by Bill Nye. He’s not an academic, and is not being paid by a university or by research grants to do a combination of research, teaching, and service.

        • Z says:

          I only chose Bill Nye as an example because he does literally 0 research, so I thought that would be good for illustrating the point. I didn’t realize that being paid by a university was a component of the annoyance. Why is that? Suppose Bill Nye were an adequate but not exceptional researcher working for a university in addition to taping his show on weekends. Would he then become annoying?

    • Andrew says:

      I don’t know anything about Bill Nye. But Dr. Oz annoys me, not because he does outreach but because he trades on his university affiliation to hawk pseudoscience.

      Sagan didn’t do that.

      Paul Krugman and Niall Ferguson use their academic credentials to push their political views; that doesn’t bother me so much, as they’re pretty open about it. I might disagree with particular things that are said, but I don’t object in principle to what they’re doing.

      • anon says:

        “But Dr. Oz annoys me, not because he does outreach but because he trades on his university affiliation to hawk pseudoscience.
        Paul Krugman and Niall Ferguson use their academic credentials to push their political views”

        The worst example is Steven Pinker, whose last few books are terrible.

    • Martha (Smith) says:

      “(Background: I’m a tenured academic physicist.) At most universities, outreach is considered valuable and admirable, and people who put effort into it are appreciated.”

      Indeed, my impression from the physicists I know is that research grants in physics often require plans for outreach in grant proposals. However, these are typically outreach to high school or elementary school students or teachers, rather than “self-promotion” things like TED talks.

      • someone says:

        @ Martha:
        No, that’s not the case. It sounds like the ones you know do “physics education” research.

        • Martha (Smith) says:

          No, the ones I have in mind are bona fide physicists who do physics research (there may be some astronomers as well).
          However, it is possible that my information may be a few years old; possibly the practice of having an “outreach component” was just a temporary phenomenon, or perhaps it was just required for certain divisions within NSF or whatever granting agency was involved.

          • Raghuveer Parthasarathy says:

            Your initial comment was correct (and “someone” is wrong) — NSF grants definitely ask for broader impacts, which typically (for all the NSF programs I’ve interacted with) take the form of outreach activities.

      • Rheophile says:

        This is kind of correct. All NSF research grants mandate that the authors discuss “Broader Impacts” of the grant, which can be arising from the work itself and from training students, bringing in high school volunteers, or outreach of the sort you mean.

        I don’t know whether more public outreach (TED talks) is a plus or a minus for NSF Broader Impacts, though!

      • someone says:

        @ Martha, Raghuveer:

        Ah, okay. Yeah, I’ve seen things asking for e.g. “contributions to society”, etc., but not specifically e.g. outreach to school kids (though that might be how applicants choose to answer it).

  5. Dale Lehman says:

    In Economics, it depends to a large extent on the politics. Paul Krugman has received a great deal of criticism for his politicized columns and the mainstream profession appears to find his public communications to be a defect relative to his Nobel-prize winning works that were more purely theoretical work. But the Freakanomics crowd get rave reviews for their making economic principles accessible to the public. What is the difference? Largely the politics of their beliefs, in my opinion.

    • jack pq says:

      I’m an economist (too?) and I disagree. Krugman is criticized by his peers because he often makes claims that are unsupported by economics (theoretical or empirical)–or if you prefer, he strongly exaggerates. In fact in some cases he literally contradicts his own claims made in papers or textbooks!

      The Freakonomics duo has been much less criticized because their claims are (usually) consistent with research, although Levitt has too been strongly criticized in some quarters, including e.g. by Nobel laureate James Heckman, and some of his papers have failed to replicate.

      • There have been several times where I read some economic opinion piece and said to myself “This person doesn’t even begin to understand basic principles of economics” and then looked at the byline and it was Krugman.

        Basic fallacies I remember from various of his columns include the broken window fallacy, the money illusion, failed to use per-capita quantities where appropriate, and argued against the existence of a poverty-trap in US welfare policy based on…. outcomes from Scandinavian welfare systems…

        Of course Krugman knows what he’s doing and understands these basic ideas from Econ 101. Therefore, my interpretation is that he’s a shill for his own political views, willing to undermine basic economic principles widely understood in economics and use his Nobel status to generate public support for things he thinks are politically important.

        • Anonymous says:

          There’s much more to economics than what one learns in Econ 101. I suspect that Paul Krugman understands economics much much better than you do.

          • Dzhaughn says:

            By all accounts he understand extremely well. But his application of it is selective.

          • Of course he understands economics well, the point is that he explicitly communicates ideas that run *counter* to basic economic fundamentals even though he must know better.

            From the days after the 9/11 attacks:

            “And there will, potentially, be two favorable effects.

            First, the driving force behind the economic slowdown has been a plunge in business investment. Now, all of a sudden, we need some new office buildings. As I’ve already indicated, the destruction isn’t big compared with the economy, but rebuilding will generate at least some increase in business spending”

            It simply isn’t true that destroying the WTC which caused “at least some increase in business spending” was in any way “favorable” on net. Krugman knows this. And yet, over and over again he’s used this broken window fallacy to argue that something destructive is “favorable” in some way. It’s just one example.

            Of course, when someone breaks something, they lose that thing, and if they replace it then they use up some resources (time and materials) which would otherwise have been available *to someone else*. In the absence of the broken window, we’d have had the window and … eventually someone else would have bought a window too using time and materials. The world would have 2 windows at time t+dt. In the presence of the broken window, now all those time and materials that would have made someone else a window goes just to replacing the one we used to already have. The second person has to use up more time and materials to get a second window. At time t+dt we have 1 window and a pile of scraps that needs to be stored in a landfill. Overall society is poorer in real terms.

            At best, the broken window moves spending from one area to another without causing too much loss. At worst it allocates a lot of resources that would have been available to other purposes making other people who would have used them poorer.

            Krugman knows this, yet he’s done this same thing repeatedly in print.

            The point is, he has a political agenda, and he turns off even basic economic common sense in order to push that agenda.

            It’s not just me saying this, it’s a commonplace thing across multiple economically trained commentators, you can google the phenomenon easily.


              “Total domestic spending + Exports – Imports = GDP

              Now suppose we have a trade war. This will cut exports, which other things equal depresses the economy. But it will also cut imports, which other things equal is expansionary. For the world as a whole, the cuts in exports and imports will by definition be equal, so as far as world demand is concerned, trade wars are a wash.”

              This is the money illusion writ large. Today you have $100 and you can buy 1 steel ingot (imported $50) and 1 bushel of peaches (produced domestically $50).

              Tomorrow there is a 35% tariff on steel imports. Peach harvesters need to spend 35% more to buy shipping containers. You still spend $100 but it buys you 1 steel ingot (imported now $67.50) and let’s say peaches had 10% shipping container costs built into their price of $50 before, so 50*(.9+.1*1.35) so new price will be $51.75 per bushel, so your remaining $32.50 buys you only 0.63 bushels of peaches.

              In the end, dollar spending may be the same, but *less real goods are procured*

              Krugman is basically saying here “don’t worry, overall across the whole world, the same number of dollars will be spent”. What he fails to mention is *it will buy less real goods and services*

              • I promise you if there’s one thing that economists will agree with me on across the board (if you put the question in sufficiently clear form), it’s that *the economy consists of the use and creation of real goods and services* and number of dollars is only a useful measurement tool which under certain kinds of circumstances can help you measure real goods and services.

                From the NBER, a footnote at the bottom of the page:

                “The NBER does not define a recession in terms of two consecutive quarters of decline in real GDP. Rather, a recession is a significant decline in economic activity spread across the economy, lasting more than a few months, normally visible in real GDP, real income, employment, industrial production, and wholesale-retail sales”

                Krugman knows this, in fact NBER even talks about REAL GDP, whereas Krugman is explicitly talking about NOMINAL GDP. Everyone who has studied economics in any way knows this. Conserving or transferring bits in a computer system (which is what dollars are) is *never meaningful by itself*

                So, anyway that’s examples of what I’m talking about that have made me decide that Krugman can be safely ignored.

              • Daniel,

                Krugman is right and you are misunderstanding his point. He is talking about “demand” in the context of macroeconomics, recessions, depressions, etc. In a recession or depression, you have unemployed labour and other resources, and the economy is below its (jargon alert) “potential output”. He is discussing the relationship between tariffs and recessions.

                Krugman writes in his blog, just before the extract you quote,

                “After all, doesn’t everyone know that protectionism causes recessions? There are reasons to be against protectionism, but that’s not one of them.”

                Then comes the extract you quote, and then he writes,

                “OK, I’m sure some people will start shouting ‘Krugman says protectionism does no harm.’ But no: protectionism should in general reduce efficiency, and hence the economy’s potential output. But that’s not at all the same as saying it causes recessions.”

                This remark about (jargon alert) “reduced efficiency” is more or less the conclusion you reach about the amount of real goods and services procured after a tariff is imposed. But he doesn’t fail to mention it, as you claim – in fact, he’s quite explicit about it. In any case, it’s not his main point in this particular blog entry, which is about the impact of tariffs when it comes to recessions/depressions/unemployment/etc.

                Put another way, he is making a point about the impact of tariffs on the gap between actual output and potential output (recession=gap gets bigger). He also notes in passing that tariffs reduce potential output. The economics is fine (he really does know his international macroeconomics).

              • Of course he knows his international economics, but his argument is disingenuous. First he says that a recession won’t occur on the basis of an equation based on nominal GDP and a definition of recession that doesn’t match the basic economic one based on REAL activity *so that he can tweak his political opponent* and then in the next sentence he uses jargon to admit that his opponent was right all along… Real economic activity WILL be suppressed below what it otherwise would be.

                It’s not that Krugman doesn’t know what he’s talking about that is my claim, it’s that he’s willing to subvert understanding so that he can win political points.

              • Oops – accidentally replied to Daniel below in a new thread – sorry about that.

              • If you impose tariffs on trade, it’s clear that prices of those items and items that depend on those items, and items that depend to second order on those items… and basically everything eventually go up. Demand for items that are more directly dependent on the tariffed item falls below what it was previously. Overall economic activity declines but especially economic activity “near” to those tariffed items.

                When the tariffed items are a small portion of the economy, the effect can be pretty minimal on broad measures. If you tariff goatskin epaulets only Cosplayers and live theater companies are likely to suffer much, and even they can probably substitute for cowhide.

                But Romney basically (paraphrased) says “widespread tariffs in a trade war will depress real economic wealth creation” and Krugman says “ha ha, little do you know because you’re nowhere near as smart as I am, that it won’t change nominal GDP of the world (maybe true but also not relevant and Krugman knows this but it wins him political points through disingenuity) and by the way of course it will “reduce efficiency” (thereby reducing real wealth creation exactly the complaint Romney made but Krugman knows that he can make a jargony distinction without a difference and again win political points).

                So, once again, it’s not that I think Krugman doesn’t know economics, it’s that he’s willing to make disingenuous or even wrong claims in print so that he can win political points. He doesn’t care about explicating economic theory to better understanding of the common person, he cares about winning political advantage.

              • Daniel,

                No, you are paraphrasing away the problem with Romney’s claim. He said that the result would be that “the economy would sink into prolong recession”. This is NOT the same thing as “reduced wealth creation”, which is what you want to say he said. The R Word means unemployed resources in aggregate. Unemployment, underutilised capacity, the kind of thing that in extremis we saw in the Great Depression (a quarter of the workforce unemployed!). Had Romney said that Trump’s tariff plans would be costly, that American consumers and businesses would be hit, etc etc, then it would be fine. That would be saying that it means efficiency loses, a reduction in potential output, etc. And Krugman made this point himself straight away in the next paragraph.

                It is very specifically the use of the R Word – the claim that tariffs would widen the gap between potential and actual output – that Krugman was complaining about. And that complaint was well founded. If you are going to criticise a bad policy, you should do it right.

                You might want to have a look at a recent Krugman blog that I think does a good job of this, in quite a bit of detail:


              • Mark, let’s continue in the other thread you started below. all in once place.

      • RJB says:

        Are there particular Krugman columns that rely on flawed economics? My impression is that his economics is usually solid, but he is more political than academic in how he attributes causality. That seems totally appropriate to his role in the NYTimes.

        As far as Freakonomics, I’ve seen loads of criticism about them, much of it on this blog, but that is more about their academic research than their popularizations.

        • Andrew says:


          I wrote a bit about Freakonomics. One theory I had was was that their success gave them a need for content, and then they started to hype various bad ideas (the purported correlation between beauty and sex ratio of births; the purported hazards of drunk walking; etc.) that fit their political preconceptions. I suspect it was a problem of trust: Dubner trusted Levitt, Levitt trusted his friends, and they both started trusting their instincts instead of tracking down the details of various iffy claims.

          • jack pq says:

            Note a big difference between the first and second Freakonomics books. IIRC, the first was based on Levitt’s published research, all of it published in top journals–although some failed to replicate. The second reported on a lot of other people’s research, and Dubner and Levitt indeed seemed sometimes credulous.

        • Terry says:

          I agree that Krugman’s economics are solid much of the time.

          He is basically an opinion machine with only one output: Republicans are bad. Since about 90% of what politicians (all politicians) say about economics is nonsense, he has lots of opportunities to use honest economics to make anti-Republican arguments.

          His flaw is that he is selective in his vitriol. He only lambastes Republican idiocy and almost never Democrat idiocy, so his output is highly biased.

          Outside economics, his reliability is lower. He is so blinded by his hatred that he says all sorts of things that are wrong or questionable or empty expressions of raw hatred.

  6. steven t johnson says:

    jack pq doesn’t buy the Sagan effect, on the grounds that showboats are losers who can’t do good science, and that’s why the good scientists look down on them. (Not a literal quote but, accurate.) Then jack pq actually cites Freakonomics as superior scientists. Freakonomics is the people who were convinced that abortion lowered crime rates, no? Well, you know, if you go by a Popperian disproving conjectures is science, and statistical analysis hasn’t refuted this, then maybe it isn’t a deal breaker. But for me, after I read how freakonomics is about the effects of incentives, I remember it. Then I read an elaborate analysis of geoengineering projects that can save humanity from anthropogenics global warming. ( The speculations are aimed largely I think at refuting panic mongers who would advocate interference with the market, caps on this and that, prohibitions of that and this, and sundry nasty unhinged schemes.) At that point I expect the proper analysis of whose incentives for geoengineering will save us. And when I don’t read it, despite the arch claim incentives are the game, then, yeah, that’s a deal breaker.

    jack pq’s notion that popularization is a consequence of failure successful scientific research is remarkable. From Say and Bastiat and Malthus, to von Mises and von Hayek and Buchanan, it seems to me you could argue that successful outreach to the right people (the Right people?) has been a well-trodden road to academic fame. I suppose you could argue that Friedman was an outstanding researcher. I’m not so sure that even if the permanent consumption function remains as unrefuted as double predestination’s biblical warrant, that it is nonetheless not good science, not even by economics’ standards.

    But switching back to the natural sciences, isn’t the all-time great example of a showboat scientist reaching out to the public Albert Einstein? All the physicists are certain General Relativity is wrong and QM is right. Einstein may have had a sinecure but he was despised. Is this the Sagan effect?

    • jack pq says:

      1. I don’t buy the Sagan effect as *the only* explanation, nor as the reason Sagan specifically was denied tenure. I suggested alternative explanations. Is it possible that in some cases someone is penalized for popularizing science? Sure.

      2. I never said Levitt was a superior scientist (and, superior to whom?). I said his pop books were usually consistent with research (the econ parts). I also said some of his papers were criticized and/or failed to replicate. The abortion paper is one of them.

      3. Not sure I get your point about Say, Bastiat etc. My point was just comparative advantage: some people realize they are better at writing books *about* science than doing science. The authors you mention weren’t aiming for outreach, they wrote academic books that became classics over time and, since they didn’t use math, laypersons could read them too.

      4. Re: Einstein you’d have to ask a physicist.

  7. I’m sorry, Daniel, but Krugman is being pretty mainstream here, and is very clear if a bit jargony. He is saying imposing tariffs doesn’t cause recessions. A recession is an increase in the gap between actual and potential output. This is why his use of the word “demand” is important – in a recession, demand falls short of potential output and you get unemployed resources (unemployment goes up, capacity utilisation goes down). He is saying that the effect on demand is a wash, so gap between actual output (determined by demand) and potential output doesn’t change. He’s simplifying a lot but this is just a blog entry. This is fairly mainstream, textbook stuff (I haven’t checked but Krugman’s international macro text is one of the more popular ones and I wouldn’t be surprised if it’s covered there in more detail).

    You are banging on about the impact of tariffs on potential output (or “full-employment output”) but that is not what he is writing about. And he is even explicit that tariffs reduce potential output – he says so immediately after making his main point. I don’t see anything to complain about w.r.t. the economics. At most, you could complain about the jargon, but even then, this is a blog entry, not an op-ed.

    • Oops, sorry, was meant to be a reply to Daniel above.

    • Carlos Ungil says:

      Do oil shocks cause recessions?

      • Thanks Carlos. Here’s the price of West Texas Intermediate crude (indexed to 100 at the end of the 1975 recession end, and GDP, with gray bars for the recession periods. Every time the price of oil goes up rapidly, we have a US recession. (we have some other recessions too, but if you see a spike in oil prices, immediately after a grey bar for a recession occurs)

      • Conventional view is “yes”. But quantifying the causal effect is tricky. There is a big literature on this. Main names that come to mind are Jim Hamilton and Lutz Lilian. Googling them + “oil price shock” will generate lots of hits.

        But if your point is that an oil price shock is like a tariff … well, besides oil having fewer substitution possibilities (at least until recently – domestic US tight oil producers would presumably increase their output fairly quickly in response to a tariff on imported oil), you want to go through the full thought experiment of putting a large tariff. This includes spending the tariff revenue (or cutting taxes to keep revenue constant) as well as the diversion of demand to domestic oil producers.

    • If this is standard, then perhaps mainstream economics has a serious problem with jargon confusing itself.

      1) Only *real* demand matters. I hope you agree, it just doesn’t matter if the number of dollars spent stays the same.

      2) If you impose tariffs you raise prices on goods artificially above their market clearing price.

      3) Increased prices reduces demand, this is basic economics.

      4) The only place where Krugman uses the word “demand” is “For the world as a whole, the cuts in exports and imports will by definition be equal, so as far as world demand is concerned, trade wars are a wash.”

      But this is explicitly using an equation that ONLY holds for *Nominal* GDP. What he’s saying is that even if trade completely crashes that Sum_i(GDP[i]) stays constant because every dollar not spent in imports is a dollar not received in exports. In this equation he completely ignores the *real* consumption. In my example, the same number of dollars are spent, but fewer peaches are bought.

      The NBER defines a recession (quoted above) as “The NBER does not define a recession in terms of two consecutive quarters of decline in real GDP. Rather, a recession is a significant decline in economic activity spread across the economy, lasting more than a few months, normally visible in real GDP, real income, employment, industrial production, and wholesale-retail sales”

      Whether it’s two consecutive quarters of *real* GDP decline, or just some broad decline in *real* activity, NBER is very clear that it’s *real* activity that matters. I’ve never heard any Economist who when asked whether dollars or real quantities matter says “it’s the dollars”.

      If you raise prices through tariffs on imports in the US, the quantity of imported goods demanded will decline in the US. The quantity of exported goods from trade partners will decline in foreign countries. Foreign countries will either produce less due to declining demand, or will consume the goods, but they had a surplus of them (hence why they were exporting them). There is no reason to believe anything but that net *real* consumption of traded goods will decline due to declining demand.

      If your country has widespread international trade and you impose widespread tariffs on those traded goods, you should expect “a significant decline in economic activity spread across the economy, lasting more than a few months” due to the concomitant increase in the price of imported goods or goods produced domestically using imported materials, hence “industrial production, and wholesale-retail sales” declines. These declines in retail sales and industrial production should affect employment in sectors “close to” those industries where the tariffs are imposed… the whole thing snowballs so that overall *less real economic production occurs*.

      That you, a trained economist, think otherwise is kind of shocking to me.

      Do you seriously believe that cutting off trade between two countries would cause the amount of *real* production and consumption to increase? Why would anyone trade if they could make more real wealth by not trading?

      • You say that Krugman’s main point is something about demand in a standard textbook argument, but I disagree, here is Krugman’s main point, he puts it right there at the bottom line:

        “But, you know, protectionism was the only reason he gave for believing that Trump would cause a recession, which I think is kind of telling: the GOP’s supposedly well-informed, responsible adult, trying to save the party, can’t get basic economics right at the one place where economics is central to his argument. “

        The purpose of the blog post is to get a political dig at Mitt Romney to claim that Romney doesn’t know a thing about economics and people should just listen to Krugman who’s the expert. Yet, Romney’s argument, broadly stated is my argument: artificial increases in price cause reduction in demand and that causes reduced *real* economic activity here in the US (and by the way, also everywhere else but Romney didn’t care about that, Krugman is the one who mentioned other countries).

        If you think my argument is wrong, please point out specifically where I’m wrong in the above. I really want to know how imposing government price increases causes total real demand to increase.

    • LInking from marks comment above

      Mark, you claim that “recession” is a “jargon term” which specifically means a “The R Word means unemployed resources in aggregate”

      Of course, *that isn’t the definition that NBER uses* their definition is a broad decline in real economic activity across the economy, I’ve quoted it elsewhere. So for example, if a natural disaster wiped out huge quantities of production capacity and overall production declined, they’d call a recession by their definition. So it’s totally OK in my opinion for Romney to use a decline in real production as a definition of recession, since experts in measuring recessions do so.

      Next, as a popularizer of economic ideas, if Krugman wants to bring jargon into the mix and use a different definition, it’s his *obligation* as the expert to explain exactly what he means. His attempt at that is to use a totally *invalid* measure of recession by any means: total change in nominal GDP across the world.

      If you impose tariffs on oil so that imported Brent Crude goes to a trillion dollars a barrel. Zero barrels will be sold (to the US). Yet huge quantities of resources are already committed to *potential production of Brent Crude* across the globe to meet US and other demand. Actual production will be well below potential production. There are unemployed resources in aggregate, according to your definition. So either way. I don’t see how I’m wrong.

      • Hopefully this one will go in the right place…

        The NBER definition of “recession” is specifically empirical/historical and is about trying to date recessions. Measuring output gaps is very tricky to say the least because it depends on unobservables like “potential output”. And there is a big debate in macro where people argued that there are “real business cycles” caused by fluctuations in Y* rather than Y. Believe me, we don’t want to go there. (And not only because some of the economics profession comes out of it looking pretty silly; it’s a detour.)

        Anyway, for an economist the Krugman blog entry you are unhappy with is perfectly intelligible and makes a fairly standard point. What he means is clear from the context and terminology. Romney should not have used the R Word. You as a non-economist might not find the blog entry clear, but it’s just a blog entry, not an op-ed designed for public consumption. I expect lots of non statisticians find the discussion here opaque to say the least!

        • The imposition of a tariff doesn’t have the problems with measuring “potential” output. The day after a tariff is imposed potential output is equal to at least the actual output the day before the tariff… The only physical change in the universe is that some paper has a signature on it.

          Imposing a tariff on Brent Crude doesn’t break the oil pumps in Norway. It does however *decrease the demand in the US for Brent Crude due to artificially high prices* Now why you keep talking about how I’m saying potential production is decreased I don’t know. I’m talking about tariffs reducing demand and causing perfectly good production capacity that people would like to utilize on both sides of the trade to go to waste. Furthermore there’s a ripple effect. If Brent Crude used to be traded, at higher tariffed prices now less of it is, it’ll be substituted for something higher price from somewhere else, or it won’t be consumed at all. Gas prices will go up. Transportation prices of goods will go up. Retail prices go up. Demand for retail goods that are transported goes down at their new higher prices. Retail stores will lay off workers or close. Demand for trucking labor goes down. People who used to drive trucks get laid off or furloughed. All of it *reduces demand in aggregate* and causes *reduced actual output* relative to the perfectly unchanged potential output that was there the day before the tariff goes into place. Of course, it’s a time-dependent gradual thing. Eventually through time of course, people compensate by rearranging their assets and labor… so that potential production of each good does change eventually. But it takes time and in the mean time perfectly good economic production capacity is wasted. Furthermore we should assume the new mixture of goods is *less desired* by society because society was free to consume that mix of goods back before the Brent Crude tariff and they didn’t. Real consumption declines.

          When we have oil shocks such as OPEC’s cartel price fixing in the late 70’s, as Carlos pointed out, almost every time the NBER will call a recession. OPEC could have pumped a lot more fuel, but they didn’t, their capacity didn’t decline it was their prices that increased.

          In the end, if we prefer where we are after the tariff, it’s only by accident. Because if everyone was aware that it was better to not trade Brent Crude and make all those changes in everything… they’d have done it on their own without the tariffs.

          How can that be at all controversial?

          Romney was saying that higher prices would reduce demand for imported goods and that would ripple through the economy reducing demand for things that depend on that imported supply, and causing people to lose jobs and produce less vs what could have been done in the absence of the tariffs. That’s obvious to anyone.

          • Daniel,

            I’m trying to keep it simple and concise, and it’s hard without using bits of jargon and defining terms. You want to use terms like “potential output” and “demand” in ways that don’t map onto standard economics usage when analysing macro policy. It’s understandable … but it’s not a good recipe for productive debate. Let me try one more time, but focusing on the concepts and trying to avoid jargon. I think I can do it with just two examples.

            One is the Great Depression. Lots of underutilised resources. Unemployed people, closed factories, terrible human hardship. Huge costs borne by society. Call this kind of cost “underutilisation”.

            The second is something I know about, the economy of the USSR. Huge inefficiencies, low standard of living. Causes are institutional – central planning is an inefficient way to run an economy. Call this kind of cost “inefficiency”. Note that there is full employment of labour and capital.

            Both extreme examples, chosen only to illustrate the concepts.

            The kind of costs that result from tariffs come mostly under the “inefficiency” heading, not the “underutilisation” heading.

            Romney’s economist speechwriters – I assume he didn’t write it all on his own, and that he had the help of some economists on his team – implied the big costs were under the “underutilisation” heading. That’s wrong, and they should have known better, and it is what Krugman was complaining about. It might not look that way to a general reader, but when a leading politician makes a speech touching on economic policy, s/he should anticipate scrutiny by economists. If Romney wrote it on his own, then perhaps we shouldn’t be so harsh (but even then, he should have had someone check it).

            • Hmm… maybe “underutilisation” isn’t very good and “unemployment” is better. Avoiding jargon is hard… Anyway, “Great Depression” is the key.

            • Hmm… I guess this helps me understand where you’re coming from, but I just don’t buy it as a useful distinction nor do I believe that Krugman in this case is successfully making that distinction. Nor do I think his GDP based argument has anything to do with this distinction. Nor do I think it really argues that Romney doesn’t have a grasp on economics.

              In the USSR we had “full employment of labor and capital” because if you didn’t do your job as Stalin wanted you wound up in Siberian gulags. The history of the USSR is a history of starvation and misery vastly outpacing the Great Depression (I’ve been told the stories by Russian immigrants with first-hand experience). To say that this wasn’t “inefficiency” because the “potential output” was equal to the actual output… because there was no way possible to reallocate the capital to where it needed to be…. well I recognize if you want to compare planned economies and market economies this could be something of a useful distinction… but if you want to compare a market economy before a tariff to a market economy after a tariff… the usefulness of the distinction is far less.

              In a moderately market based economy such as the US, you won’t find everyone working their heart out digging ditches to nowhere inefficiently simply because they’re afraid that Stalin will ship them to Siberia if they don’t. They’ll be out of work. So the symptoms are different, but the cause is the same: inefficient allocation of productive resources so that actual output is below the potential output.

              When broad tariffs are imposed in the US, people lose jobs and factories close etc because what was otherwise a perfectly good economic activity is now not viable at the newly higher prices for the raw materials etc. In Stalin’s USSR you kept doing your job anyway because the Party told you to. Either way *actual output* is much lower than *potential output under efficient utilization of the resources*.

              If Krugman’s argument is “don’t worry, tariffs don’t cause recessions they just make the US more like the USSR destroying our potential” I don’t see that as a helpful economic argument against Romney knowing what he’s talking about. That’s pure “playing gotcha”. It’s like accountants telling you you aren’t bankrupt because if you stop paying your employees and shut down you can pay off all your creditors and walk away with $10. Ok technically if you go into the petty cash drawer and tear up a $20 bill you’ll be bankrupt… but … you obviously don’t know anything about accounting because the proper definition of bankrupt is that there is no action you could take to pay your creditors, and you could actually just shut everything down and still have a
              Hamilton to buy lunch.

              Nothing about that kind of thing changes my mind about Krugman’s use of his status as revered Nobel Economist to push his political agenda at the expense of basic economic understanding and communicating economic ideas to the public. Krugman doesn’t write his blog posts for an audience of Economists who study the differences between the US and USSR, he writes them at the NYT for an audience of HR people and retail sales people and middle managers at canning factories, and fishermen and truck drivers and farm operators and things.

              If you think somehow that Romney meant “Tariffs cause actual output to decline” and “tariffs also cause the potential output to fall because they’re part of the institutional restrictions that define potential output” and “yet still, the actual output will fall by even more than the institutional restriction caused reduction in potential output” and “so we will wind up in a technically defined pure economic recession according to Mark’s version of the definitions, not the NBER or IMF’s definitions” then … you are wrong.

              If you give Krugman a pass because that’s a perfectly reasonable thing for an expert in Economics to think… then you are wrong. It’s not a perfectly reasonable interpretation of what Romney said. Only a tortured soul who’d eaten the whole MacroEcon enchilada could get that out of Romney’s speech.

              It’s absolutely the case that an expert in a public forum intended for non-experts needs to give the non-expert every benefit of the doubt, and correct or at least mention every usage of jargon that might confuse. This is known as the Principle of Charitable Interpretation:

              Krugman consistently takes the opposite tack for political purposes. It’s not that he doesn’t know what he’s talking about, it’s that he stubbornly refuses to interpret non-experts on the opposite side of his personal politics as if they could have any idea what they are talking about, and tears them down by consistently doing the opposite of the Principle of Charity.

            • OK, getting there…

              On the USSR – you are generalising from the Stalin period in a way that isn’t very helpful. Yes, you were supposed to work, but most people wanted to anyway. Fear of the Gulag was not why most people worked (post Stalin, at any rate). And there was always work to be had, even if you didn’t have lots of choice in what (or where – there were internal controls on movement) it was. Firms were always scrambling around looking for workers to hire. It was very much a full employment economy.

              But that’s a digression and I think the main point is clear enough. The USSR was a full-employment low-efficiency economy.

              In terms of the impact of imposing a tariff, of course there will be local effects. (By local I mean industry as well as region specific.) Some industries are hurt. But others benefit because of demand diverted to domestic producers. And there’s also the tariff revenue, which can be spent domestically (or taxes cut by an equivalent amount). Quantifying the overall impact is for trade specialists (see e.g. the recent Krugman blog entry I pointed to). The standard view for a large economy like the US is that the overall cost is not negligible but not huge either. But in any case this is still not about “recessions” in the business cycle sense. Think “frictional unemployment” vs “unemployment in the Great Depression”. They are fundamentally different.

              (NB: It also gets very interesting when you come to industrial strategy and economic development. There’s a long running debate on whether/how tariffs can/should be used to protect domestic industries during industrialisation, but I digress again.)

              As for how Krugman should pitch his blog entries, it is perfectly reasonable for him as an academic economist to pitch some blog entries at people with some economics background if he wants to when he feels like it. Why not? As I keep saying, you are complaining about a blog entry, not a NYT op ed. Had it been an op ed, you’d have a point … but it wasn’t.

              And it’s also perfectly reasonable to complain about sloppy economic analysis in such a blog entry. Again, why not? He thinks a major economic policy speech by a leading politician should withstand some scrutiny by experts as well as being accessible to the general public. Why is that unreasonable? Especially since fixing it would be easy and would not detract from the main message (=a trade war is a dumb idea, which Romney/Krugman/you/I all think is the case). And would lend credibility to the argument in case a trade war does start up, which sadly seems to be happening. Romney predicted the consequence would be a “prolonged recession”, Krugman says that’s wrong (I think K is right), and if such a recession doesn’t come to pass it discredits the anti-trade-war argument. Why not just say instead that it will lead to various kinds of other costs that are supported by evidence and a consensus among trade economists, and that are pretty likely to happen?

    • Also: “You are banging on about the impact of tariffs on potential output (or “full-employment output”) but that is not what he is writing about. And he is even explicit that tariffs reduce potential output”

      Tariffs *don’t* reduce potential output. That’s not my claim. Potential output is reduced when you blow up oil fields or burn peach orchards or the government shuts down steel mills because they fail to meet some regulation. Tariffs reduce *actual demand* by artificially increasing prices above the market clearing price, leaving potential output the same (tomorrow if you turned off the tariff the steel mill would run full blast and produce the steel you want at the market price).

      • Sorry, but tariffs reduce “potential output” as conventionally defined. The Wikipedia entry for potential output is not bad. “Potential output in macroeconomics corresponds to one point on the production–possibility curve for a society as a whole, reflecting its natural, technological, and institutional constraints.” Tariffs (and tax structure in general) come under “institutional constraints”.

        • So you’re saying “tariffs and bombs in the oilfields are the same thing” they both affect “potential output” in the same way. Because of this, actual output and potential output are equal after imposing tariffs *by definition* and since *recession* is *by definition* a gap between actual output and potential output, then *by definition* after a tariff there isn’t a recession.

          Now I *know* that economics is off the rails…

          except that NBER doesn’t use that definition of recession, and writers on output gap at the international monetary fund don’t use that definition of potential output:

          “Potential output is the maximum amount of goods and services an economy can turn out when it is most efficient—that is, at full capacity. Often, potential output is referred to as the production capacity of the economy.­”

          Since imposing the tariff “reduces efficiency” according to Krugman, it has to be the case that it is not reducing *potential output* but only *actual output*.

          So, Economics is saved by not making these mistakes in defining away a real world phenomenon using formal jargon definitions.

          • Note, I’m actually OK with your definition of recession as “reduced production relative to potential”. An economist friend pointed out that you could have tariffs reducing actual output but not actually causing the growth rate to go negative… GDP need not decline, it should only be below what it otherwise would have if resources were employed efficiently. A gap between Actual and Potential works as a definition, and NBER’s definition is just a way to *detect when that occurs* ok.

            It’s that you somehow magically define away that gap by saying that tariffs reduce demand and potential by exactly the same amount by definition… that simply can’t be a useful or correct definition of potential productivity. It’s not the one you use either. Elsewhere you say:

            “The people are there and want to work, the factories are there but not operating, so the explanation is not a fall in Y*.”

            A few days after a massive tariff on Brent Crude, the oil workers are laid off, they’re sitting around wanting to work, the pumps are stagnant but they’re still there, no-one is shipping… truck drivers and tanker operators are standing by not getting paid. It can’t be an explanation that Y* just declined. We could easily pump all that sweet light crude that people want…. They just don’t want it anymore at the artificially high price.

          • I’m trying to be helpful, and patient, and clear, and not go into the minutiae of variations in jargon usage, as well not get wound up by unfounded blanket statements about an entire academic discipline, but it’s hard…

            Anyway, the answer to your question about tariffs and bombs is “yes, sort of”. Using the very simple distinction I pointed to above, these are more akin to the USSR (“inefficiency”) case than to the Great Depression (“underutilisation”) case. Blowing up your own factory means you get less output from given resources; so does imposing tariffs. But the resources you do have are still fully employed.

            The Great Depression is different – a horrifying but compelling demonstration of a very specific kind of market economy failure. Millions of people who wanted to work but there were no jobs; thousands of factories closed because there was no demand for their output. Understanding how that works, and what to do about it, is what the Keynesian revolution in macro policy was about. Romney’s use of the R Word alluded to this kind of cost in connection with tariffs, but shouldn’t have.

            • Here’s Romney’s speech transcribed:

              He uses the word “recession” twice. The second time is just to reiterate the first. Here’s the first:

              “If Donald Trump’s plans were ever implemented, the country would sink into prolonged recession. A few examples. His proposed 35 percent tariff-like penalties would instigate a trade war and that would raise prices for consumers, kill our export jobs and lead entrepreneurs and businesses of all stripes to flee America.”

              As I say, the Principle of Charity requires us to believe that Romney has a meaningful thing to say here, that he uses the word “Recession” in a way typical for someone *like him*. For someone *like him* the meaning of recession is the NBER one since that’s the one used to define when recessions occurred in govt publications he’d be familiar with: reduced real output relative to previous levels. Then he gives examples of the kinds of real output that would be reduced: higher prices for consumers resulting in lower demand for goods and services, reduction in employment in industries that depend on exports to China (since China would impose their own tariffs thereby increasing prices and reducing demand for those products), and reallocation of investment resources to other areas where the tariffs don’t apply and efficiency is higher.

              None of those things is against very basic principles of economics. They all make perfectly fine sense. They all come down to basic concepts of supply and demand. Every single person listening to his speech who wasn’t a card carrying Economist knew exactly what he was saying. To say that he doesn’t have the slightest clue what he’s talking about here is pure politics.

              • You’ve just about exhausted me (I don’t mean my patience, I mean I am genuinely tired – it’s late here in the UK). But providing the context of the Romney speech is helpful, and I should respond.

                The Krugman point is twofold. First, a “prolonged recession” is not a likely consequence. (I think he’s probably right.) Economic costs, yes, and for good reasons (which K acknowledges). But not a “prolonged recession” that would show up using the NBER methodology etc. Second, major policy speeches by senior politicians get scrutinised. Why not take the trouble to get it right? Especially if it’s easy to do, doesn’t change the message, and is more likely to stand up in the face of evidence should an actual trade war start?

                That’s it for me for today. Apologies to all and sundry for taking up so much bandwidth and not even getting the threading right. :(

              • I think we’ve both made our points, reasonably well. I stand by mine, and I certainly don’t take this one blog post as the reason for my dislike of Krugman, it was just a particular example of a general phenomenon that I believe is problematic. Good night, and take care. Thanks for making your point in a way that I could at least understand what it was.

  8. I’ll repeat from above, with tweaks.

    You are paraphrasing away the problem with Romney’s claim. Romney reportedly said that the result would be that “the economy would sink into prolong recession”. This is NOT the same thing as “reduced wealth creation”, which is what you want to say he said. The R Word (“recession”) means unemployed resources in aggregate. Unemployment, underutilised capacity, the kind of thing that in extremis we saw in the Great Depression (a quarter of the workforce unemployed!).

    Had Romney said that Trump’s tariff plans would be costly, that American consumers and businesses would be hit, etc etc, then it would be fine. That would be saying that it means efficiency loses, a reduction in potential output, etc. And Krugman made this point himself straight away in the next paragraph.

    It is very specifically the use of the R Word – the claim that tariffs would widen the gap between potential and actual output – that Krugman was complaining about. And that complaint was well founded. If you are going to criticise a bad policy, you should do it right.

    Another way to put it is to use the “output gap” = potential output Y* – actual output Y. (Standard term, widely used.)

    Standard story: when there’s a recession, Y-Y* gets negative and big, Unemployment grows, capacity utilisation falls, etc. Usually explained in terms of a fall in aggregate demand (what Krugman meant by “demand” – standard shorthand in macro). Great Depression an extreme case. Huge fall in Y. Mass unemployment and bankruptcies consistent with a big fall in Y rather than Y*. The people are there and want to work, the factories are there but not operating, so the explanation is not a fall in Y*.

    Tariffs cause Y* to fall. That’s what you are banging on about, and Krugman is very quick to acknowledge this as well. But a “recession” is about a negative output gap opening up, i.e. a fall in (Y-Y*). Krugman’s point is that the main argument against tariffs is the hit to Y*. Not a hit to (Y-Y*).

    Classic example: it used to be thought that the Smoot-Hawley tariff played a big role in causing the Great Depression, but that’s no longer the standard view. It caused a decline in trade (so Y* fell) but isn’t the main reason for the opening up of a huge negative output gap (Y-Y*), mass unemployment, etc.

    • Argh! The wrong place again! Sigh.

    • So you’re saying that if you raise a trillion dollar tariff on imports of Brent Crude it will magically break all the oil pumps in Norway so that the oilfields fail to have as much production capacity as before (reduce potential output), but fortunately the demand for oil will plummet too so that magically there is no gap between what could have been produced in the absence of the tariffs and what is actually produced now that no-one can afford to buy the stuff… Got it.. Peace out.

  9. Argh! The wrong place again! Sigh.

  10. Dale Lehman says:

    I do hesitate to continue this thread, but I have to say that I stand by my original comment. David and Mark’s attempts to define what Economics 101 (or 201 or 601) actually says about recessions is of some interest (although it also reinforces why I am happy to not be teaching economics any more). But mostly my takeaway is that the difference in reactions of economists between Krugman and Levitt/Dubner is largely politics. Whose “science” is better? Debatable. What is not debatable is that Krugman is partisan, and in ways that many economists find distasteful – for two main reasons. First, this type of overt politics has always made economists uncomfortable since it makes economics appear less “scientific.” Second, most economists are trained to dislike many liberal (read as Democrat) policies.

    The criticisms of Levitt/Dubner have been milder – more about whether their results were valid or could be replicated. The critique has not bee nearly as personal and emotional, however. This is because their work is much more mainstream – finding unintended consequences and showing how much smarter economists are than everyone else (which provokes the response of some to try to show how much smarter they are than Levitt/Dubner).

    So, I’ll stand by my original comment.

    • I share your view in general, though I think there’s more to the discomfort with overt politics than just not wanting to look “unscientific”. (And I suspect you think the same.) For one thing, some like the appearance of “objectivity” and “scientific” results because it supports conclusions that they hold for partisan reasons, consciously or not.

      Which, come to think of it, is not a million miles away from incentives for p-hacking and related themes that come up here a lot. Economists are like everybody else.

      That said, though, this particular example (economic effects of trade wars) is one where there is a lot of consensus. Krugman’s trade chops are widely respected in the profession, even by those who disagree vehemently with his politics, and his analysis and conclusions in the blog entries we were discussing are pretty conventional.

Leave a Reply to Daniel Lakeland