The anxiety of influence

It’s generally assumed that Milton Friedman’s Natural Rate Hypothesis led to Robert Lucas applying rational expectations to macroeconomics.  But is it possible that we’ve reversed causality?  I’m probably reading too much into this Holman Jenkins interview of Lucas, but it offers a tantalizing hint that Robert Lucas might have planted the seed that blossomed into Friedman’s greatest theoretical innovation:

He also cites Milton Friedman, with whom Mr. Lucas took a first-year graduate course.

“He was just an incredibly inspiring teacher. He really was a life-changing experience.” Friedman, he recalls, was a skeptic of the Phillips curve””the Keynesian idea that when businesses see prices rising, they assume demand for their products is rising and hire more workers””even if the real reason for higher prices is inflation.

“Milton brought this [Phillips curve] up in class and said it’s gotta be wrong. But he wasn’t clear on why he thought it was wrong.” In his paper for Friedman’s class, Mr. Lucas remembers reaching for a very rudimentary notion of expectations to try to explain why the curve could not operate as predicted.

I was a student at Chicago during the late 1970s.  At the time I read a lot of supply-side stuff in the Wall Street Journal.  I recall one class where Lucas “disproved” the Laffer Curve argument that tax cuts could lead to more revenue.  Unfortunately, Lucas evaluated the hypothesis with a standard Keynesian multiplier model (where incentives played no role.)  I imagine that he hadn’t followed supply-side econ very closely, other than hearing from his fellow economists that Laffer was a bit of a nut.  I raised my hand and suggested that Laffer’s argument was based on incentives to produce, and couldn’t be addressed with a demand-side model.

In the exam for the class Lucas asked a question on this topic, and I answered the way he taught it, not based on my class comment.  When the tests were returned I saw it was market wrong–Lucas had wanted the answer I gave in class.  I gained a lot of respect for the University of Chicago that day.

On another occasion I was talking to Lucas about growth (he was shifting to growth theory about that time).  He suggested we don’t know much about what causes growth.  I suggested that it might be free market policies.  He pointed out that communist Romania had had the fastest RGDP growth rate in recent years.  I said I didn’t believe their numbers, but fumbled around when he asked what evidence I had that they were wrong.  I learned then that there’s a big difference between knowing you are right, and being able to offer persuasive evidence that you are right.  (Or maybe there isn’t—it’s an interesting philosophical question.)

Later Lucas seems to have become something of a supply-sider himself:

For the best explanation of what happened in Europe and Japan, he points to research by fellow Nobelist Ed Prescott. In Europe, governments typically commandeer 50% of GDP. The burden to pay for all this largess falls on workers in the form of high marginal tax rates, and in particular on married women who might otherwise think of going to work as second earners in their households. “The welfare state is so expensive, it just breaks the link between work effort and what you get out of it, your living standard,” says Mr. Lucas. “And it’s really hurting them.”

I suppose you know where I’m going with this.  For a fleeting moment I wondered whether I could have planted the seed of Lucas’ interest in supply-side economics.  But then I read this:

Mr. Lucas launches into a brisk dissertation on the work of colleagues””Martin Feldstein, Michael Boskin, others””whom he credits with disabusing him and fellow economists of a youthful assumption that taxes have little effect on the overall amount of capital in society.

You mean to say he was more influenced by the empirical studies of leading economists, then by a few off the cuff remarks by an annoying first year grad student?  I’m very disappointed in Mr. Lucas.

PS.  That’s two posts within a month named after Harold Bloom books.



24 Responses to “The anxiety of influence”

  1. Gravatar of W. Peden W. Peden
    5. November 2011 at 06:09

    On Romania: I do seem to remember once, when reading some stuff from the old Soviet studies department, coming across an estimate made during the Cold War by a statistician who worked out that the (I think) Romanian agricultural statistics implied that pre-communist Romanians had a NEGATIVE overall food consumption. So there were reasons to be suspicious.

    On the philosophical question: the key word is “persuasive”. A lot depends on what you mean by this term, since you can find someone for just about anything who can’t be shaken from their belief. As for persuading people in general, if you have witnessed something fantastic that no-one else has seen, then you know something but you can’t persuade anyone else. Think of Lucy in “The Lion, the Witch and the Wardrobe”.

    The basic insight in supply-side economics (that the Laffer curve slopes downward at the extremes) is indisputable. I think even Democrats would admit that the supply-siders were right about the Kennedy tax cuts. The question is: where are we right now on the curve? And, on that matter, disputing with supply-siders can make a lot of sense.

  2. Gravatar of W. Peden W. Peden
    5. November 2011 at 06:10

    Downward at the high-taxation extreme, that is.

  3. Gravatar of Morgan Warstler Morgan Warstler
    5. November 2011 at 06:54

    That’s two posts within a month named after Harold Bloom books.

    Perhaps we can get Camille Paglia onboard.

  4. Gravatar of Becky Hargrove Becky Hargrove
    5. November 2011 at 08:14

    One problem for supply side in the present, is that what actually constitutes a service economy in monetary terms, has to change: which is why I really wish the Nobel could have gone to someone in growth theory. Even the term “job creation” was apparently just coined in 1979 according to the NYT this morning.

  5. Gravatar of flow5 flow5
    5. November 2011 at 08:38

    The goal of supply-side economics is to increase the production of higher quality goods and services which can be marketed at competitive prices. It reduces monopolistic elements in the price structure; increases labor productivity; reduces unit labor costs; reduces transfer payments to the non-productive sectors; eliminates excess regulatory burdens, excessive rates of taxation on producers and savers, etc. Even more intractable are constraints imposed by resource and technological factors.

    The demand for capital goods is a derived demand, derived from primary consumer demands. That even in a capitalistic system the end and objective of all production is human consumption. The demand for inventory or plant and equipment, however far removed from the ultimate consumer, is derived from final consumer outlays in the marketplace.

    Demand is always paramount in successful business planning and commitment decisions. If sufficient demand is not expected to exist, it matters not what the expected costs will be. “Sufficient” demand, of course, covers all costs plus and expected after tax profit margin. An economy such as ours which is geared to mass production requires concomitant mass consumption. Payrolls must be sufficient to buy the goods and services produced – at the asked prices.

    Only in the frictionless world created by the mathematical model builders are the asked prices in equilibrium with consumer spendable income. In the real world, there is always a purchasing power deficiency gap of varying proportions. This is just another way of saying that to have high levels of production and employment, we need not only a vastly more competitive price structure, we also need a steady but slightly inflationary monetary policy (prices increase c. 2-3 percent annually), and a tax policy that contains some elements of compulsory income redistribution – downward.

  6. Gravatar of flow5 flow5
    5. November 2011 at 08:44

    “over 90% of the growth in nominal GDP has been from inflation, with only 9.3% from “real” GDP growth over the last 2.5 years”

    The inflation indices do not accurately reflect all prices, particular prices, nor price levels. The implicit price deflator “is Keynes’ measure of the level of prices for all new, domestically produced, FINAL goods and services in an economy”.

    The consumer price index peaked on 7/1/2008 @ 219.133 but didn’t exceed that figure again until 28 months later on 11/1/2010 @ 219.240. The proper appellation for this phenomenon coming out of a recession/depression is REFLATION, not INFLATION

  7. Gravatar of TheMoneyIllusion » The anxiety of influence | Social Fobi – Det Du Behöver Veta TheMoneyIllusion » The anxiety of influence | Social Fobi - Det Du Behöver Veta
    5. November 2011 at 09:04

    […] here: TheMoneyIllusion » The anxiety of influence This entry was posted in Uncategorized and tagged applying-rational, generally-assumed, […]

  8. Gravatar of Patrick R. Sullivan Patrick R. Sullivan
    5. November 2011 at 09:18

    ‘Romanian agricultural statistics implied that pre-communist Romanians had a NEGATIVE overall food consumption. So there were reasons to be suspicious.’

    I remember reading a piece in a journal, in the early 80s, demolishing the demographic statistics coming out of Cuba. The math wasn’t even complicated. The only people who believed Communist statistics were left-wing ideologues.

  9. Gravatar of ssumner ssumner
    5. November 2011 at 09:25

    W., Peden, I hope you are right, but I see lots of progressives praise the Eisenhower tax code.

    Morgan, I’m certain Camille would be a market monetarist. She’s equally contemptuous of reactionaries, and progressives who live in a dream world.

    Becky, How does the economy need to change?

    flow5, I’m not sure who you are quoting, but it’s nonsensical, as what matters is RGDP relative to trend, not in absolute terms.

    We might want to redistribute, but not for macro reasons.

  10. Gravatar of ssumner ssumner
    5. November 2011 at 09:28

    Patrick, You said;

    The only people who believed Communist statistics were left-wing ideologues.

    Yes, and the CIA!

  11. Gravatar of Patrick R. Sullivan Patrick R. Sullivan
    5. November 2011 at 11:20

    Yes, Scott, but the CIA had been penetrated by Stalin since its inception as the OSS. People such as Julia Child’s friend Jane Foster.

  12. Gravatar of Jason Jason
    5. November 2011 at 11:58

    Funny comment about he CIA. I believe that the CIA knew they were exaggerated, but that the USSR was smart and didn’t implausibly exaggerate. Their definition of “implausibly” might have differed from the economists’, and I bet there was an anchoring effect from the communists’ numbers.

    I went and looked at what data** I could find on Romanian GDP to see what the difference turned out to be*** for my own personal edification. In doing so, and remembering the anxiety everyone had in the 1970s that communism actually worked (or that is, as I learned in my cold war history class from the late 90s), I came up with a new theory of the fall of communism based on NGDP targeting.

    In a sense, central planning may be like NDGP targeting on steroids. When you are desperately far below trend output, communism is an excellent catch-up growth strategy that basically fails when you achieve your 19th century growth trend. Or, in the case of Romania, your pre-WWI growth trend. Romania achieves this around 1980 and the former Soviet Union in the 1970s. Being beaten into utilizing excess capacity apparently works when excess capacity is large (things are terrible). When it is smaller (things are bad, but not terrible), more subtle methods like monetary policy may be required.

    This doesn’t work as well for China and it is hard to say what is going on there with this model.

    ** I looked at some post-fall of communism data on the growth in Romania and then in the former Soviet Union and China (although it derives from one person’s thesis who aggregated several pieces of data from OECD and other sources for estimates before the 1960s and goes until 2003; I supplemented it with modern OECD data and adjusted the 1990 $ PPP).

    *** The rate does not seem to be that far off from that (in)famous graph from that 1970s economics textbook showing the Soviets were going to beat us. Basically, they hit the wall of trend NGDP output right after that, and growth slowed way down.

  13. Gravatar of Morgan Warstler Morgan Warstler
    5. November 2011 at 13:13

    When I was 20, I begged Camille to take me on as an unpaid personal jester. (I love lesbians.)

    She deserves a far larger audience.

    like you.

    She should own up to to her personal conservatism, instead of trying to straddle.

    like you.

  14. Gravatar of ssumner ssumner
    5. November 2011 at 14:08

    Patrick, Good point.

    Jason, Aren’t you confusing NGDP and RGDP?

    Morgan, She’s not a conservative, she’s a liberaltarian.

  15. Gravatar of Jason Jason
    5. November 2011 at 15:13

    Duh, you’re totally correct Scott.

    The only way to salvage the argument would be to assume a relatively constant rate difference (on average) between NGDP and RGDP.

  16. Gravatar of Becky Hargrove Becky Hargrove
    5. November 2011 at 16:24

    This answer is a bit rough but I’ll give it a shot: In the coming years, the portion of the service economy that is supported by government will continue to be trimmed, and while that would seem to be the cue for production/manufacturing to pick up the slack, many of those disappearing jobs involved consumption at relatively high levels. That’s why I was wishing for growth theory that would be able to show – for instance – something approximating production codes that would show how much nonsense – or lack thereof – actually existed between the product and the ability to get the product to the average consumer (Not the production measured at the level of the single organization, but the organizations involved in the process itself) Plus, any median indexing in the aggegate can be misleading because above a certain base of need, prices start to become affected by status signalling. The only reason I brought up signalling in price is because it create numerous multi-tiered economic effects beyond a certain point. To sum this up, how much might the consumption patterns of those leaving (mostly) government service occupations actually change?

  17. Gravatar of Becky Hargrove Becky Hargrove
    5. November 2011 at 16:28

    Sorry, that should have read “production efficiencies”

  18. Gravatar of Lorenzo from Oz Lorenzo from Oz
    5. November 2011 at 17:32

    On CIA reading of Soviet stats: they had a bureaucratic incentive to read threat estimates high.

    Soviet stats were known internally as “ceiling statistics” (look at the ceiling and think of a number).

    Morgan: Perhaps we can get Camille Paglia onboard. That made me laugh out loud, thanks.

  19. Gravatar of Morgan Warstler Morgan Warstler
    5. November 2011 at 20:57

    Sumner, I don’t think social conservatism is real.

    Like cannibalism, it existed only when communities could be “small.”

    It is blown away as soon as society is large. Said another way, the Internet delivered gay marriage in 15 years, what do you think we’re going to deliver in the next 15?

    Look, Camille happened before the Internet, like you, and bless her heart, she’s taken to it like a champ, but she still has a pre-1993 view of social vs. economic conservatism.

    Christians will retreat to state’s rights, which will leave only economic issues for national politics.

    In that view of things, Camille, like you is a flat out Republican.

  20. Gravatar of dtoh dtoh
    7. November 2011 at 04:10

    Just for the record, if you assume asymmetric rates of return on capital (i.e. a lot of losers and a few winners), you don’t need anything other than simple arithmetic to understand why the Laffer Curve works.

    It amazes me that economists go through contortions on the issue of taxes on capital when it all it requires is 3rd grade math to understand this.

  21. Gravatar of Becky Hargrove Becky Hargrove
    7. November 2011 at 05:09

    “Annoying first year grad student”
    1) First to harangue the teacher
    2) first to dissent in some way
    If so, teachers knew I sometimes fought to restrain myself. In an english essay class the other classmates hated me until they realized I really liked the teacher.

  22. Gravatar of Cthorm Cthorm
    7. November 2011 at 07:57


    Tell that to my relatives and the rest of the Christian Right. The potency of social conservatism may be much diminished but I think it’s a stretch to assume it’s national importance is gone for good. This is my view: social conservatism will continue to be a political force for people who see Christian Values as one of the defining parts of their identity. I really wish the Family Research Council and the rest of the movement would just keep their values to themselves and their families. They are clearly the force that keeps the GOP from becoming a big tent party.

  23. Gravatar of Morgan Warstler Morgan Warstler
    7. November 2011 at 08:14

    Cthorm, they don’t have to keep it to themselves, they just have to keep it to their state.

  24. Gravatar of ssumner ssumner
    10. November 2011 at 20:11

    Becky, I’m not sure how to answer your question–I guess that’s why we have markets.

    Morgan, Or she’ll become a Republican when it becomes the party of PJ O’Rourke.

Leave a Reply