Believe those who are seeking the truth. Doubt those who find it. Andre Gide

Wednesday, December 15, 2010

Okun's Law Rules the FOMC

Zzzzz...oh...what's that? The FOMC say's what?
Information received since the Federal Open Market Committee met in November confirms that the economic recovery is continuing, though at a rate that has been insufficient to bring down unemployment.
(Full statement available here.)

Stop the presses! New headline: Central bank warns recovery too slow to curb unemployment (Financial Times, Dec. 14, 2010).

Oh no! Well, we'd better get that growth going then. What's everyone waiting for? Increase G! Increase M! Increase...increase.....Zzzzzz.

Sorry, but I stayed out way past my regular bedtime last night. And for some strange reason, I find myself mulling over this curious phrase: "...though at a rate that has been insufficient to bring down unemployment." Why did the FOMC include it in their statement? What does it mean? And where does it come from?

Let me start with the last question first. It comes from the late Arthur Okun, who discovered something called Okun's law back in the 1960s. Evidently, Okun's law was widely taught back in the day. These would have been the days when most of the now senior members of the FOMC were impressionable youngsters; see -------->

Alright, so now we know where it comes from. But what does it mean? Obviously, it refers to some sort of ironclad law of economic nature, right?

Uh,, not exactly. When the law breaks down, proponents like to refer to it as a "rule of thumb," instead. To be honest, it's really just a statistical correlation. When economic growth goes up, the unemployment rate goes down. At least, on average this is what happens at business cycle frequencies.

Of course, it's also true that when the unemployment rate goes up, economic growth slows down. To put things another way, output tends to go up when more people are working. I know it's quite the shocker, but many economists believe this to be true. This is why they pay us the big bucks.

But why am I confusing you in this manner? Let's see what that great purveyor of purloined propositions (PPP) has to say about the subject: Growth and Unemployment.
What you see is that unemployment tends to fall when growth is high, rise when it’s low or negative. You also see that growth has to be fairly fast — more than 2 percent — just to keep the unemployment rate from rising. Why? Well, productivity is rising, so that you can produce any given level of output with fewer workers; so output has to rise to keep employment from falling.
(That darned productivity growth--scourge of the labor market.)

Alright, so what is PPP trying to say here, especially in that last sentence? As far as arithmetic goes, it's obviously correct (the language is sloppy, but let's cut him some slack). But arithmetic is not theory. He is trying to explain the empirical correlation of Okun's law. And explanation requires theory. Where is the theory?

To see what I mean here, let me alter the last sentence in the quote like this:

Well, productivity is rising, so that you can produce any given level of output with fewer workers; so employment has to rise if output is to rise at a more rapid pace. 
As far as arithmetic goes, this statement is also correct. But the sentence appears to convey a different message, doesn't it?

Evidently, there's more to Okun's law--the way it is commonly expressed--than a simple correlation. It seems to be sort of a "theory" too (I use the term loosely here). The theory is to be found in the assumption that the direction of causality runs one way only, and that it runs from output to employment, rather than the other way around. Output growth causes--nay, output growth is needed--to bring unemployment down. I guess the idea that more employment might cause more output--reversing the causality--seems less obvious. 

The way Okun's law is commonly expressed naturally leads to more sympathy for "demand side" policies, like increasing G or M, to stimulate employment. I am not saying that this shouldn't be done. What I'm saying is that the arithmetic supplied by PPP is not what supports such a policy. My use of the arithmetic, for example, might instead be used to support "supply side" policies, like a cut in the payroll tax. It would be equally wrong to abuse arithmetic in this manner.

Anyway, I think it is time to bring my morning rant to a close. But I still haven't answered the question of why that silly phrase was included in the FOMC statement. Would anything of any substance have been lost if the statement had not included it? It's not as if QE2 can only be justified if one believes in Okun's law (PPP version). I guess it's difficult to rid oneself of some youthful impressions.


  1. David,

    Would you think that the "output leads employment" version of Okun's "law" runs counter to Say's Law? Not the Keynesian version of Say, which goes something like "supply creates its own demand" and is obviously nonsense. I mean this one:

    "t is worthwhile to remark that a product is no sooner created than it, from that instant, affords a market for other products to the full extent of its own value. When the producer has put the finishing hand to his product, he is most anxious to sell it immediately, lest its value should diminish in his hands. Nor is he less anxious to dispose of the money he may get for it; for the value of money is also perishable. But the only way of getting rid of money is in the purchase of some product or other. Thus the mere circumstance of creation of one product immediately opens a vent for other products. (J. B. Say, 1803: pp.138–9)[3]"

  2. The Fed legally is suppose to satisfy their dual mandate, part of that is to key employment near its maximum legal consistent with minimal inflation. So, my guess is that they included it because they are legally suppose to care about employment. Plus, everyone seems to care care about unemployment the most so the Fed does have to play up to those demands.

    The problem I have with Okun's Law is that it seems to suffer from a coordination problem. OK, output is low - why would I, an atomistic agent who has no effect on the economy individually, hire someone when output and employment are low - who is going to buy my stuff? Only if we could coordinate then people could hire and expand together, but decentralized producers and investors can't coordinate like this. So, how do we escape this coordination problem with Okun's Law?

    That's why I have my own strange interpretion of Okun's Law. I personally think what really goes on is that people receive news about the future, whatever that news may be, that makes them revise their expectations of future growth upwards. To cash in on this expected future growth, they invest and expand their labor force. But this process leads to a self-fufilling prophecy since the investment and expansion of their labor force leads to higher growth in the aggregate immediately. Under my interpretation, the government just hiring people wouldn't seem to lead to these higher expectations of future growth. Why would temporarily hiring people lead them to believe that growth would be higher in the future such that they should expand themselves? In my interpretation, the best policy would be austerity measures coupled with reduced corporate and capital taxes. Of course, I realize my interpretation of Okun's Law is hardily standard so take it with a grain of salt.

  3. Prof J: I'm afraid that I cannot say (ha ha) one way or the other. I've seen that quote from JBS before, but I confess to finding parts of it confusing (e.g., why is the value of money necessarily perishable?)

    Ted: I can appreciate the logic underlying your multiple equilibrum story. Everything you say makes sense to me until the very last seems to me that the government might, in principle, be able to kickstart the economy (move to the high level equilibrium) via a large scale government employment program. At least, I know that I can write down a model where this is true.

  4. David,

    Since Say would have been writing in France at the end of the 18th century, I should think he was referring to the currency manipulations of the French crown and the damage done to the purchasing power of money. The Mississippi bubble comes immediately to mind.

  5. I'm pretty sick today so that sentence was a bit sloppy. Let me explain my position on this a bit better.

    Now, I agree, that it's possible that a large scale government spending program could kick start the economy. I just simply don't think the government expenditure multiplier is as large as these models claim. I don't think for plausible levels of government spending that it can do enough to really jump start the private sector again. I think to really get the economy going it would require an implausible large spending program. And, indeed, at that size I begin to wonder how this interacts with expectations of future distortionary taxation and the endogeneity of the monetary regime. For example, if producers and investors believe that government spending will be financed by an increase in corporate and capital gains taxes - how stimulative will this spending really be at jump starting the private sector? I mean, if I take some of these lump-sum tax NK models where the monetary regime (passive versus active) is imposed exogenously seriously I could conclude we should spend $25 trillion dollars since apparently this would induce a $50 trillion increase in output according to these models with minimal consequences for inflation. Call me skeptical.

    Also, I have to ask, what type of model do you have in mind - perhaps there is one I haven't considered. I'm not an economist by any means so perhaps I'm unfamiliar with the literature you are referring to, but my understanding is that the only models where government spending has a large multiplier are New Keynesian models with a binding zero nominal interest rate and sticky wages and prices. Perhaps a neoclassical model with sufficiently productive government spending would yield large multipliers as well, but I'm going to assume that most of this spending isn't particularly productive. The models I seen seem so unrealistic and every time something more realistic is added to the model the muiltiplier gets smaller and smaller, which makes me think an "accurate" model would yield very small multipliers.

  6. Prof J: Right. So what he says makes sense in an inflationary environment. But what about in a deflationary environment, where the real rate of return on money is quite high? The Great Depression comes to mind, obviously.

  7. Ted:

    Well, let's just say that I share some of your skepticism. But from a theoretical perspective, it's quite possible. I actually present an example of this in my undergrad macro notes, available for free on my SFU webpage. The example I use assumes increasing returns in the production function; and this leads to a multiplicity of equilibria. The economy can get stuck, in the way Keynes described, in the low-level equilibrium. A program to increase employment leads the private sector to revise upward their forecast on the returns to their own hiring decisions (there is a strategic complementarity). The fiscal stimulus can, in principle, jump the economy to the high-level equilibrium. (In fact, even the credible threat of such a program can do the trick in the the government would not even have to spend a dime). Anyway, there you have it. Details available in my online textbook.

  8. Alright, I see what you are getting at. I agree that's theoretically possible. I could think of some ways you could flip that result around, but fundamentally I see what you mean.

    By the way, I just skimmed your textbook and read just a few sections. It's a really great textbook. You seem to take the idea that perhaps your students aren't mathematically illiterate morons seriously, which is refreshing. A lot of the very popular undergraduate textbooks I've seen are comically simplistic. My friend's are in my high school's AP macro class and their textbook looks like it was written for a 7th grader, not an undergraduate class. Random suggestion through, since it said on your web page you were writing the 3rd edition, it might be interesting to add a few lines about second-generation speculative attacks to your speculative attacks section. The logic is pretty simple and a small discussion might be interesting (well, at least I would find it interesting).

  9. David,

    That's a good point on the perishability of money. I don't see how it would debilitate Say's Law, though. Most of the stuff I buy are oriented towards my periodic needs, so holding off indefinitely isn't really an option. Thus, like the producer in Say's Law, my grocer and gasoline provider will lay off their goods on me tout de suite. And if the producer couldn't lay it off, my man would have a sale (my clothier back home does this frequently).

    I have to read Say in the entirety, though, to get a strong sense of what he's saying. I shouldn't attempt to represent without so doing.

  10. Ted,

    By any chance, are you The Teenage Economist?

  11. Prof J: If he's a teenage economist, I'm very impressed!

    Ted: Glad you enjoyed my notes. I will take your suggestion to heart, once I get back to revising the text.

  12. David,

    Me too. There's a fellow that rights a blog for a newspaper (forget which one) who goes by "The Teenage Economist." Wondered if he is the same.

    P.S. I'm also heartily looking forward to the 3rd edition.

  13. David,

    While we're on the subject, I'm also a fan of your textbook. Did you ever publish solutions to the problems anywhere?

  14. vimothy: I'm sorry, I did not. And I'm sorry that the new edition has been delayed. Dang financial crisis and my new job at the Fed has distracted my thoughts! But hopefully it will all translate into a much better and complete text. Thanks.

  15. Oh well, in any case, we very much like the microfounded thing--it beats Mankiw and ISLM any day of the week!

  16. Free sex videos free porn aways way. Home you teen sex pornhub hd free videos. Smart porn tv free sex watch nowe !