« Apologies for the double post | Main | Portable fiscal policy question multi-pak »
General Theory, chapters one and two
I am repeatedly struck by Keynes's skill as a literary stylist. Usually this praise is denied the General Theory but I consider the book his Finnegans Wake; the most difficult passages are often the most charming but of course they are not for everyone.
I see three main themes in the book as a whole:
1. Income effects are more important than substitution effects.
2. Expectations matter.
3. The private and social returns to liquidity are very different.
#1 (as applied to macro) and #3 were most original in his time. The book as a whole circles around these themes and repeats them in varying combinations, not always coherently or consistently. You could also add the claims that 4. monetary factors render a "natural rate of interest" problematic and 5. labor markets are special. Chapter two is essentially about #1 and #5.
Keynes did go beyond the classics, even if he did often caricature them. (Keynes is brilliant as a historian of thought when praising but almost always wrong when criticizing.) Since Keynes never gives us a truly coherent model -- not even verbally -- it is easy to pick holes in the GT. Keynes had so many exciting new ideas that he never decided what his main point was or exactly under which conditions it would hold.
Much of chapter two is devoted to establishing the proposition that workers cannot in any direct way choose a lower real wage. For Keynes nominal wage flexibility doesn't solve the main problem. If nominal wages fall across the board in an economy, prices will fall and real wages will remain high. Unemployment will continue while the economy enters a downward spiral. I've already discussed that point here but to sum up my view Keynes is presenting a special case not a general case.
p.9 puts forward a version of the doctrine of money illusion.
p.15 defines involuntary unemployment, namely if the economy can be inflated into a higher level of employment. This pragmatic definition reflects that Keynes was never sure why workers minded inflation, and a cut in the real wage, less than they minded a cut in the nominal wage. But that is one of his behavioral postulates and it has survived into macro to this day.
The "neo-Keynesian" models are not so loyal to Keynes. Keynes held sticky nominal wages to be a policy prescription, but not necessarily a good description of the world.
Chapters one and two are stunning, as they announce that we are now living on a different economic terrain. But we've yet to see whether the main arguments are truly sound.
On Thursday we'll be doing chapters three and four -- be ready! And I encourage other bloggers to follow along and offer their own commentaries.
Posted by Tyler Cowen on December 8, 2008 at 06:53 AM in Books, Economics | Permalink
Comments
Testing...
Posted by: Tyler Cowen at Dec 8, 2008 6:55:00 AM
I haven't read p.15 (Internet is not so swift on my sailboat) but the reason workers prefer inflation should have been obvious to Keynes. Workers are, on the whole, debtors. Inflation favors debtholders, especially those who are in hock up to their eyeballs.
Posted by: Bob Knaus at Dec 8, 2008 7:21:46 AM
Testing...
Posted by: Tyler Cowen at Dec 8, 2008 7:47:57 AM
Further testing...
Posted by: Tyler Cowen at Dec 8, 2008 8:14:21 AM
Two points: (1) allowing that an outstanding writer can make a virtue out of obscurity and incoherence in a work of fiction, surely to invest a work of nonfiction--and one that is intended to guide the making of macroeconomic policy--with these same qualities is far from praiseworthy. To be really good, a book's style ought to be fit for it's purported purpose!
(2) "For Keynes nominal wage flexibility doesn't solve the main problem. If nominal wages fall across the board in an economy, prices will fall and real wages will remain high. Unemployment will continue while the economy enters a downward spiral." Not so. An economy faced with a shortage of money is one in which all nominal values are above equilibrium, and need to decline. Otherwise unemployment persists even if the real wage is at its equilibrium level. This is easily shown using simple supply and demand diagrams. Alas, such an exposition would probably not have much literary merit!
Posted by: George Selgin at Dec 8, 2008 8:15:59 AM
Keynes on real Vs nominal wages.
Keynes almost answers the question himself. If people are forward looking, and expectations of sustained inflation become the common perception, then workers become much more concerned about real wages.
However, initially you may not perceive the increase in prices as a general increase in prices. You must be sure that inflation is the cause of the change in prices and not a change in demand for the fruits of your labor. I.E. if you think that your prices are going up because of increased demand for your goods you would consider an expansion in production. If you think the increase in prices is just the result of increased inflation you may increase prices but keep output stable.
Also from 1920- 1935 the country saw deflation in eight of these years and only 1920 saw serious inflation. Post WW1 workers would have had little if any expectation of inflation. If inflation fears are muted, why would most workers care.
I suppose transaction costs and other issues make rewriting labor contracts more troublesome then the effort is worth. I imagine the 20's were more marked by rapid changes in the type of products and services being delivered so fear of remaining competitive was a greater fear then inflation.
An interesting special case would be agriculture during this period
Posted by: DanC at Dec 8, 2008 8:23:15 AM
This is a great idea Tyler, and I hope more bloggers and commenters participate.
Perhaps something we could do with other classics.
Posted by: Jason at Dec 8, 2008 8:45:43 AM
I had kind of hoped that you'd say something about the various prefaces to the GT (or have you done this previously, and I've missed it?).
After all, there are some really important passages in the English and other (German, Japanese, French, other?) prefaces. For example, Keynes on writing the GT as being a "struggle of escape" from old ideas - gripping stuff.
I also think it is important that in the preface to the German edition Keynes wrote: "The Manchester School and Marxism both derive ultimately from Ricardo,—a conclusion which is only superficially surprising." He has already made clear that he is out to destroy the Ricardian assumptions - and this means that he is out to destroy the Ricardian premises of Marxism as much as of the laissez-faire branch of "classical economics".
Both Skidelsky and Markwell in their studies of Keynes stress that Keynes was strongly ANTI-Marxist, and I think that this is implicitly evident even as early as the German (if no other) preface. Keynes's anti-Marxism is very important - important to our udnerstanding that he was out to save capitalism by reforming and managing it, not out to undermine it (as extreme Austrians are sometimes tempted to claim).
No doubt we will get around to this?
Posted by: Tanya White at Dec 8, 2008 9:29:45 AM
What version (publisher, edition, etc.) are you using. The pages in my copy don't seem to be matching up with yours. I'm finding the relevant passages, but it is bothersome. Could you find some other way of referencing? Chapter #, Section #, paragraph?
Beyond that, this is terrific. It's been a while and I'm enjoying getting back into Keyne's work.
Posted by: Tim at Dec 8, 2008 9:43:58 AM
Dr. Selgin,
I'd like to comment on...
'(2) "For Keynes nominal wage flexibility doesn't solve the main problem. If nominal wages fall across the board in such an economy, prices will fall and real wages will remain high. Unemployment will continue while the economy enters a downward spiral." Not so. An economy faced with a shortage of money is one in which all nominal values are above equilibrium, and need to decline. Otherwise unemployment persists even if the real wage is at its equilibrium level. This is easily shown using simple supply and demand diagrams. Alas, such an exposition would probably not have much literary merit!'
______________
Good point about the need for nominal values to fall when there is a shortage of money. However, I don't think I agree that a shortage of money is all that important in the type of recession where the GT would be most germane. If the problem were a shortage of money, shouldn't we expect interest rates to be high? And yet, interest rates are typically low in a 'low animal spirits' type depression like the one in 29-32... In these depressions, prices fall while interest rates fall, indicating that individuals are, in aggregate, trying to save more than they want to invest, so there is an excess demand for bonds and an excess supply of goods for current consumption or investment. In this situation, a low interest rate addresses the situation more directly than a decline in money prices, as that directly addresses the imbalance between desired saving and desired investment. Of course, Keynes' didn't feel that would be effective due to inelastic expectations viz the long term interest rate. So he favors fiscal policy.
I don't think I agree with Keynes' views on wages and the labor market--Tyler has nudged me toward the view that lower nominal wages can do some good. However, I think his broader point is right on--depressions are mainly a problem related to the loanable funds market/goods market/aggregate demand (3 ways of saying the same thing), not a problem related to the labor market.
Steve
P.S. A quick suck up: I read your Theory of Free Banking a few years back and I really enjoyed it, especially the point about how banks' reserve ratios would tend to adjust in a way that stabilizes MV. I was really impressed with that insight.
Posted by: steve at Dec 8, 2008 10:06:44 AM
Easily the highest IQ book group I've ever participated in, but I don't think I'll ask any of you out on a date, which is the real reason I usually participate...
Posted by: Andrew Meyer at Dec 8, 2008 10:08:12 AM
An excellent point about liquidity that is often forgotten. In fact, given the current situation, it is hard to understand why the different returns to liquidity have not received greater attention in the recent finance literature which has been struggling with defining and measuring liquidity.
I will have to re-read this part, I smell a paper....
Posted by: Rgregory at Dec 8, 2008 10:24:32 AM
Here's the relevant part of the post on my blog referencing the book club.
I had forgotten much of the first two chapters. I was actually embarrassed at how much. But there were two things I took out of these chapters. One was Keynes’s discussion of real and nominal wages. I think he describes the problem of the "money illusion" quite well in this early chapter. The idea of workers being reluctant to reduce money wages when real wages are rising is important. It could have been even more difficult when most people were not aware of the connection of the true impact of changes in price level. And they were probably also less aware of price behavior outside their local community. Certainly those in larger cities may have better understood the extent of price changes, but those in less populated areas could have seen it as a less pervasive phenomenon. I expect that would affect their expectations.
For me, the second thing was the sense that this was more a general theory about special circumstances.
Posted by: Tim at Dec 8, 2008 10:36:30 AM
"For me, the second thing was the sense that this was more a general theory about special circumstances."
I disagree. Classical equilibrium is a special circumstance. Disquilibrium is a special circumstance, certain to be temporary. Non-optimal equilibriums, sometimes enduring, are the usual case. The GT is not "Say's Law is right, with exceptions". Say's Law is wrong.
Posted by: bob mcmanus at Dec 8, 2008 10:52:00 AM
Why does Keynes think that the assumptions of classical economics fail to apply under conditions of less-than-full employment when they apply under conditions of full employment? I don't understand the argument.
I think the thought is this: (1) The economy gets into a state of uncertainty (not risk), (2) this causes people with a high propensity to save to hoard/save rather than consume or spend, (3) this causes the level of interest necessary to attract investment high, (4) the higher level of interest requires that capital make more profits, (5) but under the conditions they have to cut back production to make returns to a smaller number of investors and this involves laying workers off, (6) which in turn creates involuntary unemployment (sticky nominal wages, efficiency wage theory, etc. all help to create the conditions in step 6, right?).
So it's not the following: IF more than full unemployment, THEN the assumptions of classical economics don't apply - as if this were a CAUSAL explanation. It's just a correlation. The assumptions of classical economics stop applying because conditions of UNCERTAINTY somehow suspend them. Is this the idea? That classical economic assumptions only apply when people can take rational risks, i.e. they have some idea of the probabilities of certain outcomes?
Posted by: Selfreferencing at Dec 8, 2008 11:50:32 AM
How does the Data on this look? If workers are unwilling to take a drop in the money wage, aren't they just fired and replaced with someone who does take that money wage? Does that just fired worker walk down the road and get another job that pays less?
Posted by: libfree at Dec 8, 2008 11:52:13 AM
What exactly is the difference between frictional unemployment and involuntary unemployment? It seems that there's no difference extensionally, only in terms of what causes it. Frictional unemployment occurs when people want to work but can't because the economy is restructing and they're looking for new jobs. But involuntary unemployment occurs when there's insufficient aggregate demand.
But isn't the reason that people don't have jobs in both cases is not that they won't work at a certain wage (as in voluntary unemployment) but that they can't find one or haven't found one yet? It seems to me then that the only difference between frictional and involuntary unemployment is the REASON that workers can't find jobs when they want them - normal restructuring, or business cycle conditions. Do people agree?
Posted by: TheGarbageMan at Dec 8, 2008 11:54:54 AM
Keynes is a clever writer. I enjoyed this:
The classical theorists resemble Euclidean geometers in a non-Euclidean world who, discovering that in experience straight lines apparently parallel often meet, rebuke the lines for not keeping straight—as the only remedy for the unfortunate collisions which are occurring.
Posted by: Bernard Yomtov at Dec 8, 2008 12:15:21 PM
I found Keynes' discussion of workers inability to accept a lower real wage tedious to read. Perhaps this is due to my subscription to sticky nominal wage theory as an accurate representation. I hope that goes on to extrapolate it as a "policy prescription" as you call it.
Posted by: Dann Ryan at Dec 8, 2008 12:54:33 PM
Keynes was the best stylist who ever wrote serious economics in the English language. Any one who doubts that should consider his account of the general price level in the General Theory. The elegance with which he avoids giving any such account a lesson to us all - a lesson in the perils of style as well as of its use.
Reviewing the treatment of wages and of the nature of different types of unemployment in the General Theory, I come back to my feeling of the 1970s that he was struggling for lack of the concepts of human capital and of lifetime earnings. To take two examples:
- a large part of 'frictional unemployment' occurs because labour incorporating human capital is very far from homogeneous. It is therefore worth both workers and employers investing time in achieving a good match between the vacancy and the worker.
- the reluctance of workers to accept lower wages is, at least in part, because a lower wage may imply lower than expected lifetime earnings.
Posted by: David Heigham at Dec 8, 2008 1:34:33 PM
"p.15 defines involuntary unemployment, namely if the economy can be inflated into a higher level of employment. This pragmatic definition reflects that Keynes was never sure why workers minded inflation, and a cut in the real wage, less than they minded a cut in the nominal wage. But that is one of his behavioral postulates and it has survived into macro to this day."
Say what?! Keynes in the passage in italics is giving in words a standard mathematical definition of the conditions required for the real wage to be greater than the marginal disutility of labor at the current equilibrium. The argument has nothing whatsoever to do with inflation.
He is essentially expressing himself in pure Classical economic vocabulary here: "involuntary" unemployment implies identically that there are more workers available at the current wage than are currently working and (given certain assumptions on strictly increasing functions) therefore w > u'(leisure).
Posted by: ccm at Dec 8, 2008 3:38:16 PM
"Keynes never gives us a truly coherent model -- not even verbally -- it is easy to pick holes
in the GT."
"Keynes had so many exciting new ideas that he never decided what his main point was or
exactly under which conditions it would hold."
Exactly then, why are there so many adherents that foam at the mouth when somebody
questions Keynes' theories.
Posted by: Superheater at Dec 8, 2008 4:25:36 PM
I'm trying to wrap my head around Keynes' Chapter 2 discussion using some recognizable textbook model. Chapter 5.4 of David Romer's "Advanced Macroeconomics" ("Alternative Assumptions about Price and Wage Rigidity. Case 1: Keynes's Model") might be useful for simplified nuts and bolts. For whatever reason, the money wage is fixed. When the price level is 'low' the real wage is high and the economy is on the upper left part of the labor demand curve, with involuntary unemployment (and, equivalently, where the real wage is higher than the marginal disutility of work). The economy can only move towards full employment if the price level rises relative to the fixed money wage (e.g. due to a rise in aggregate demand, which increases prices in the goods market). Thus the Aggregate Supply curve is upward sloping - output and employment rise with the price level because this pushes down the real wage, allowing firms to hire more workers.
On the empirics Romer observes that "This view of aggregate supply therefore implies a countercyclical real wage in response to aggregate demand shocks. This prediction has been subject to extensive testing beginning shortly after the publication of the General Theory. It has found little support: most studies have found that the real wage is approximately acyclical, or moderately procyclical."
Posted by: mb at Dec 8, 2008 5:10:55 PM
Steve writes:
If the problem were a shortage of money, shouldn't we expect interest rates to be high? And yet, interest rates are typically low in a 'low animal spirits' type depression like the one in 29-32... In these depressions, prices fall while interest rates fall, indicating that individuals are, in aggregate, trying to save more than they want to invest, so there is an excess demand for bonds and an excess supply of goods for current consumption or investment. In this situation, a low interest rate addresses the situation more directly than a decline in money prices, as that directly addresses the imbalance between desired saving and desired investment. Of course, Keynes' didn't feel that would be effective due to inelastic expectations viz the long term interest rate. So he favors fiscal policy.
Steve,
You have confused nominal with real interest rates. In fact, real rates were at close to all-time high levels during the debt-deflationary early 1930s; it is real rates that matter, not nominal rates. (For a good graphic of just how high there were, see Kenneth Fisher's book, The Wall Street Waltz.) High real rates were one reason why investment fell during this period. Other reasons for this included tax increases by both Hoover and FDR, and overall "regime uncertainty" (the effects on investment and the business climate of the increase in uncertainty caused by the attack on property rights and the shifting regulatory apparatus, including the increase in required bank reserves).
Posted by: Bill Stepp at Dec 8, 2008 5:57:25 PM
Seems like the core of Keynes's theory of wage setting is the stated in the following:
"In other words, the struggle about money-wages primarily affects the distribution of the aggregate real wage between different labour-groups, and not its average amount per unit of employment, which depends, as we shall see, on a different set of forces. The effect of combination on the part of a group of workers is to protect their relative real wage. The general level of real wages depends on the other forces of the economic system."
Thus, he argues, workers are not really trading off consumption and leisure. Rather, they are trading off leisure and relative social status or something of the sort, as expressed by their relative wage. Price movements don't change thiese relative standings and hence have no effect on the labor supply decision.
Whether this is realistic or not, I don't know. But it is radically different from the two favorite contemporary stories: money illusion (Lucas) or nominal rigidity (Taylor).
Posted by: Felipe at Dec 8, 2008 7:34:46 PM
