Don’t trust historians
Let me say first that I like history. I think historians have a lot of interesting things to say and I don’t think historians should be economists. But . . .
I It depends on the definition of “great.”
Go to the following link and scroll down to the Ranking of presidents by historians in a 1996 poll. Wilson is ranked 6th whereas his successor Harding is ranked 41st. That’s 41st out of 41 presidents when the poll was conducted. Where does one even start?
1. I had known for quite some time that Wilson’s economic policies were perhaps the worst in American history. He presided over the creation of the Fed and the income tax, which went from 0% to something like 70% while he was president. In the long run the Fed may have been a good thing, but there can be no doubt that 1913 was premature, we didn’t know anywhere near enough about monetary policy to warrant a central bank meddling in the gold standard. He presided over a period of very high inflation after WWI, when we actually needed somewhat lower prices. Then we had a severe depression in his last year of office. Industrial production had fallen by 32.5% by March 1921 when Harding took office (and you think things are bad now!)
But guess what; by November 1922 we had recovered, as industrial production had already passed the previous cyclical peak. (If you factor in a 3% trend rate of growth, a new peak was reached around March 1923.) How did he do it? Recall that Hoover and FDR relied on policies of propping up wages and raising tax rates—that’s why FDR’s recovery took nearly nine years, and even 8 and 1/2 years from a July 1933 industrial production figure that was roughly as depressed as March 1921.) Instead, Harding let wages fall and cut income tax rates sharply. BTW, do you think industrial production will fully recover by January 2011, after Obama has been in office for 2 years?
2. Wilson had arguably the most destructive foreign policy in American history. When there is a delicate balance of power in Europe you don’t want to meddle unless you plan to stay there permanently. Yes, I know Wilson did intend the US to hang around, but he should have known we were an isolationist country before he brought us into WWI. All he did was assure that the strongest country in Europe lost. When we pulled out (as was inevitable), a rematch was almost preordained. WWII was the fruit of Wilson’s foreign policy. (As were more than 116,000 dead American soldiers.)
3. Wilson was one of the most repressive presidents in American history, imprisoning thousands of people whose only crime was to disagree with his political views. It was left to Harding to release people like Socialist leader Eugene Debs from prison, as well as many others.
4. And I also recently learned that Wilson was a vicious racist. Yes, I know racism was widespread at the time. But he was much worse than the Republicans who came before and after him. Check out this article from Reason magazine.
So other than being a horrible president in terms of economic policy, foreign policy, civil liberties and civil rights, he was a great president. And Harding was the worst ever. What was Harding’s great crime? A few of his aides took bribes. But isn’t that common? Didn’t Eisenhower and Johnson and Nixon and Carter and Reagan and Bush and Clinton have cabinet secretaries that got involved in scandals? I welcome any historians to write in and tell me exactly why Wilson is a great president and Harding is the worst ever. Is it just a question of Wilson being more active? Are historians using the old fashioned definition of “great,” which meant something like “powerful?”
II. Was Keynes really a savvy investor?
I got to thinking about this issue last night well reading The Lords of Finance (which by the way is a fine book so far, despite one little point I will nit pick.) See what you make of this:
“In early 1920, he [Keynes] set up a syndicate, with his brother, some of the Bloomsbury circle, and a financier friend from the City of London. By the end of April 1920, they had made a further $80,000. Then suddenly, in the space of 4 weeks, a spasm of optimism about Germany briefly drove the declining currencies back up, wiping out their entire capital. Keynes found himself on the verge of bankruptcy and had to be bailed out by his tolerant father. Nevertheless, propped up by his indulgent family and by a loan from the coolly acute financier Sir Ernest Cassel, he persevered in his speculation”
Translation, without help from his rich daddy and rich friends, this cocky, arrogant, smart-aleck would have fallen on his face, ended up digging ditches somewhere and we would never have heard of him. But he did have a rich daddy, who bailed him out.
[Alert to amateur psychologists: Yes, I was not in my high school’s Bloomsbury group, and I worked my way through college and grad school w/o financial aid.]
Don’t anyone write in and tell me that Keynes made lots of other good investments, because if you’ve got a rich backstop, none of that matters. Here’s what I’d do if Bill Gates was willing to lend me $3.57 billion dollars for a day:
I’d go to Vegas and put $5 million on numbers 1 through 34 on the roulette wheel. The odds are roughly 90% I’d win. If I did so, I’d win $180 million on a bet of $170 million. I repay the $3.57 billion and pocket my $10 million dollars and be rich for the rest of my life, clipping coupons. If numbers 35, 36, 0, or 00 came up I’d bet again, this time $100 million on each number 1 through 34. If I won, I’d receive $3.6 billion, repay Gates, and have $30 million dollars to spend for the rest of my life. The odds are nearly 99% that I’d win one of these two bets. Of course if both failed, I’d be in big trouble. But that’s not very likely is it?
What’s the point? If you have a rich backstop it’s relatively easy to come up with investment strategies that will usually (not always) make you look like a genius.) From now on I will never believe anyone who tells me that Keynes was a great investor. Does this matter? It shouldn’t, but unfortunately it does. If his investment reputation was like Fisher’s (calling stocks fairly priced in 1929) nobody would take seriously his Chapter 12 in the General Theory where he tries to shoot down the efficient market hypothesis. That is the chapter that has a lot of nonsense about musical chairs, beauty contests, and the seasonality of ice prices. But I’d like to focus on his assertion that investors only care about the short run, because I hear this argument a lot from commenters.
Suppose investors only cared about the short run, as they only intended to hold shares for 5 years. What would be the value of a biotech company that did not expect a breakthrough to occur for 15 years? To estimate it’s value, let’s suppose that the breakthrough is expected to be worth $10 billion, if the patent were sold to a big pharma company. So would the stock be worth anything today? Yes, because investors today would know that 10 years from now (if investors still had a 5 year horizon) the stock would be worth the present value of $10 billion earned 5 years later. Through backward induction we can see that the current value of the stock would be exactly the same as if investors had a long term focus. In case you don’t believe me, contrast the difference in value between a 20 and 30 year bond with equal coupon payments. All the differences occur in the out years, and yet those differences get correctly priced into the current market values of the bonds. If anything, history suggests that investors have too much of a long term focus, as they have been remarkably patient with biotech stocks lacking any earnings at all. So Keynes’ views on investment are superficially witty and sophisticated, but on closer examination are intellectually empty.
Keynes was no where near the economist Fisher was. In Chapter 12 of the GT (p. 142) he completely fails to understand how the Fisher effect can raise nominal interest rates. The AS/AD model we teach with its self-correcting economy and monetary policy being the preferred stabilization tool is pure Fisher. It is right out of his 1923 and 1925 papers where he developed a “Phillips Curve model” of the business cycle more than 30 years before Phillips. Keynes didn’t believe in a self-correcting mechanism or in the efficacy of monetary policy in a depression. His GT purports to explain changes in NGDP, but has no explanation for how NGDP is determined.
I worry that too many economists on the left have bought into his scorn for rational expectations. I vaguely recall Krugman making an offhand comment in his Lionel Robbins talks about failing to understand why anyone would pay attention to the stock market. Certainly there is a widespread disdain among economists for the message from stock investors, who Paul Einzig affectionately termed the “submen.” Why should we care what investors think? Here’s one reason:
1. If the Fed had paid attention to the 1929 stock market crash maybe they wouldn’t have let the US monetary base fall sharply between October 1929 and October 1930.
2. And maybe if the Fed had pursued an expansionary monetary policy the Depression would have been less severe in 1930.
3. And maybe if the Depression was less severe in 1930 we would never have had such serious bank panics.
4. And if we had not had serious bank panics then maybe there would never have been a Great Depression.
5. And if there had been no Great Depression it seems far less likely that the Nazi’s would have been elected in Germany. (They were a tiny party in 1929.)
Is that reason enough? Or should I also mention what happened when the Fed didn’t pay any attention to the stock crash of October 1937? Or how about last October’s crash? You may recall that at that time the Fed (and for that matter Paul Krugman) did not expect the unemployment rate to get anywhere near 9.4% this year. To be honest, neither did I. But I knew the markets were saying that Fed policy was far too contractionary. And just as in 1929, the markets were right and the Fed was wrong.
That’s why these seemingly unimportant debates over whether Keynes was a good investor actually matter.
Tags:
25. June 2009 at 05:55
Scott, great blog! Another issue with Wilson, he oversaw the institution of the Harrison Narcotics Tax Act. This paved the way for our modern drug war. How are we doing with that war?
25. June 2009 at 06:23
I like the information on the President rankings. It seems like there is always a narrative that is trying to be established. Harding, of course, is pretty bland so the scandals and failings were what defined his Presidency. Whereas all of Wilson’s heavy handedness and other failings were squashed under the biggest event of the time, WWI, which was a US victory.
It makes me wonder two things- 1) if there will be a reevaluation of the causes of the Great Depression and the efficacy of FDR’s policy if the current economy does not recover (which I don’t believe it will) in spite of the huge amount of stimulation. Or will the consensus again be that we should have done even more to stimulate out?
2) Assuming the economy does not recover and we have a lengthy period of economic problems, will history judge Obama harshly and revise W. Bush upwards as 9/11 overshadows the other failings of the Bush administration? Or will the competing narrative that Obama inherited Bush’s failings establish itself, much like with FDR/Hoover?
As far as the investor long term/short term piece, I have a disagreement. Instead of the Biotech startup, consider the following cash flow proposition that I think better demonstrates the mindset of many investors: Let’s say you have a cash cow company that is currently providing a steady income stream, but they are engaged in some problematic activity (perhaps heavy pollution or unethical behavior, or outright fraud) that will probably eventually catch up to them and jeopardize their model. Now predict how capital will be allocated, also including capital to extend regulatory forbearance via lobbying, for example. It is possible to put together a system where much of the capital moves to short term gain at the expense of long term health.
25. June 2009 at 06:31
Scott, no offense, but while you may be right that historians shouldn’t be economists, your comments on Wilson suggest economists shouldn’t be historians. Not that I’m a great fan of Wilson, but blaming him for WW II seems to be like the most superficial kind of hindsight judgement. I don’t think WW II was inevitable, although it would take a long essay to explain why so I won’t try. I would note that you yourself later acknowledge that without the GD the Nazis wouldn’t have come to power, and there’s certainly a case to be made that without Hitler WW II wouldn’t have happened. Was the GD itself inevitable? I suspect it was given the prevailing gold standard ideology and economic ideas generally (I rely here on Eichengreen). (You can argue that different policies could have been pursued but counterfactual history that assumes people will act according to the more enlightened ideas of a later time is nonsense). But even if the GD was in some sense inevitable, it was still a unique and unpredictable event. But discussions of contra factual history inevitably become metaphysical so I’d better stop here.
As to whether the U.S. should have intervened in WW I, that’s certainly a debatable issue, but I don’t think it was ex ante unreasonable. I’m reminded here of Niall Ferguson’s fine book the Pity of War. He argues that Britain shouldn’t have intervened in WW I, that it wasn’t worth the cost, that the Kaiser wasn’t Hitler, and that a German victory would have been preferable from the British point of view. I believe that ex post he is right, only ex ante it is difficult to imagine any British government acquiescing in German domination of the continent. Later when Ferguson became a frequent pundit and journalistic commentater, I was surprised to find he was a conservative. I am sure that if he had been around in 1914 he would have been a Tory and a supporter of intervention. That’s another reason why I distrust hindsight history.
And by the way, I think presidential rating polls are crap, they are inevitably infected by ideology and hindsight judgements, and in addition they degrade serious questions of public policy to the level of who was the greatest left-handed pitcher. I think less of the historians who participate in them. That said, and at the risk of contradicting myself, I highly recommend a book by Alvin Felzenberg called The Leaders We Deserved (and a Few We Didn’t): Rethinking the Presidential Rating Game, which deals with the subject in a very intelligent way. He’s more conservative than me but I agree with a lot of his judgements.
25. June 2009 at 07:13
Whenever I leave my field and get into history, I know my commenters will be more knowledgeable than I am. So far that’s true.
Jinx, Thanks. I didn’t know about that. I view the war on drugs as the nation’s worst policy mistake right now, so I should have included it. On the other hand he tried to stop prohibition, so that counts in his favor.
IDB, I’d put “US victory” in quotes. What did we win? Did we gain territory, or prevent future wars? (I understand you weren’t defending the action.)
I am current working on a revisionist history of the Depression, which shows the New Deal actually lengthened the Depression by 6 or 7 years.
You ask:
“It is possible to put together a system where much of the capital moves to short term gain at the expense of long term health.”
You can certainly imagine such a model. But it doesn’t work in theory (as it implies investors are irrational) and it doesn’t work in practice (because we observe that real world markets behave exactly as they would if investors took the long view.) So I just don’t see the value of these “short term investor” hypotheses. What do they explain?
Phil P, I agree that I am a lousy historian. But let me defend my argument.
My main complaint is that 116,000 Americans lost their life for no good reason. If someone wanted to disagree, they would have to engage in exactly the same sort of wildly hypothetical speculation that you (rightly) accuse me of doing. Thus if someone says, “no, the loss of 116,000 American lives was not for nothing, it made the world safe for Democracy,” I would just point to subsequent history. So, I’d be willing to withdraw the WWII charge if the historians who defend Wilson as “great” will admit that he sacrificed 25 times as many young American men in Europe as Bush did in Iraq (a war that is widely viewed today as an American tragedy of the first order) and got nothing for that sacrifice.
Having said all that, I believe that both WWI and the Depression were necessary conditions for WWII, but not sufficient conditions. Here’s how I’d put it. Wilson behave recklessly in ignoring Washington’s advice about foreign entanglements. His action was likely to destabilize Europe after we left. I agree that the actual outcome was far worse than any reasonable person could have expected, and thus in a sense was “bad luck,” but it was nonetheless a consequence of Wilson’s policy. I do understand that there are real problems with using historical counterfactuals as “causality” arguments (i.e. what if Hitler’s two parents had never met?) and so I do understand your criticism. Indeed I partly agree. My WWII argument was mostly to head off anyone arguing against my view that 116,000 lives were lost for nothing.
Off topic, but I can’t resist the following observation:
Not just in England, but all thought Europe, almost all intellectuals supported WWI. Let me say that again, as it is an important point—almost all European intellectuals supported a useless massacre of 10 million men. (I make it sound like they all agreed with each other. They didn’t; even more stupidly they supported their own country.) If anyone ever had any doubt about why we should be skeptical of the consensus view of intellectuals, just read 20th century history and look at all the things intellectuals supported. That’s why I am not bothered in the slightest by the fact that my libertarian views are held by very few intellectuals.
That wasn’t directed at you, you were just stating a fact–that the British elite couldn’t stomach German dominance of Europe. I agree. (Of course if we could run the clock backward, we’d all take that instead of WWII.) And I agree with you about lists made by historians. Indeed that’s just part of my general distrust of “intellectual consensus.”
There is nothing more misleading than the phrase “History shows.”
25. June 2009 at 07:34
Dear Sumner,
I´m an undergraduate from Brazil and I have a doubt about this post, in this part “But I knew the markets were saying that Fed policy was far too contractionary.” , how did you know that the market thought the fed policy was too contractionary?
Keep up the good blogging work.
25. June 2009 at 08:05
Scott, I was not planning to get involved in another thread, but I just couldn’t resist. I don’t think that historian rankings of presidents are worth much, and I don’t think that historians if they are honest would disagree. It is largely a popularity contest. And historians will instinctively like an intellectual like Wilson much more than an “amiable dunce” (Clark Clifford’s putdown of Reagan, was any marginal political figure ever more highly overrated than Clark Clifford? He was a one-man self-promotion machine.) like Warren Harding. And I don’t mean popularity simply in a partisan sense. Just that they are evaluating Presidents not simply as presidents but as overall human beings and evaluating their intellects and other personal qualities not just their pure performance in office. Eisenhower was rated rather low until someone (I think it was Murray Kempton) wrote a magazine article in which he concluded that Eisenhower deliberately and strategically spoke nonsense because he didn’t want to be understood. Once that was the take on Eishenhower, he started to move way up in the rankings because he was considered to be smart and shrewd not an idiot as he had been previously.
Your take on Keynes as an investor was quite insightful. I’ll just mention that somewhere or other Hayek in one of his many ambivalent comments on Keynes noted that Keynes lost lots of money when he speculated on foreign exchange (which presumably was a field in which he had professional expertise) and made money in speculating on commodities (which was a field about which, by his own admission, he had no particular professional expertise).
I agree completely that Keynes was nowhere near the economist that Fisher was. However, for better or worse, Keynes came up with a model that made a big difference to the future course of the profession. Fisher’s contributions were huge but they were extensions of the neoclassical paradigm not paradigm shifts. So Keynes gets credit for being a revolutionary, and Fisher’s work is taken for granted because it was so easily absorbed into pre-existing patterns of thought.
That said, I would add that while I used to share your disdain for Keynes’s caricature of investing and stock market behavior, I now believe that there is more to it than you give him credit for. The point that Keynes was trying to articulate was that since asset prices reflect future cash flows that are uncertain (in the Knightian (and Keynesian!) sense) expectations themselves can influence the realizations of the cash flows that are being evaluated. This is easiest to see in what we now think of as network markets. The realized outcome is the expected outcome. So there is no (well maybe a little) fundamental underlying the realized equilibrium it purely a function of expectations. A lot of economic models simply presume that there is a unique equilibrium solution. In some cases that may be true, but in an uncomfortably large number of cases there are multiple equilibria. And the realized equilibrium depends on what is expected. If we can somehow get short-term inflation expectations up to 2 percent (or better 5 percent) the realized equilibrium will be a rapid recovery from the downturn. If we are stuck with inflation expectations of about 0 zero, we will be stuck in a high-unemployment “equilibrium.”
25. June 2009 at 09:44
You are right, I did like this post. I don’t have much to say about it except that I mostly agree.
To other posters… On WWI remember the UK was treaty-bound to fight. The US wasn’t. Whether it was wise for the UK to fight or not is a completely separate question from whether it was wise for the US to do so.
25. June 2009 at 09:45
The low rankings for Harding in various historians’ polls are really inexplicable. Not only does he rank at the bottom of every poll, but in all the polls listed on the Wiki site, he ranks *below* Herbert Hoover. So it’s not like you can blame it all on political bias.
My best guess for how to explain the rankings is this: your average historian doesn’t know much about the Harding administration, the 1920-21 downturn, or anything like that. If you ask him to think about the Harding administration, the only thing he’ll think of is Teapot Dome, and even there he’ll be kind of fuzzy on the details (Harding must have been personally involved, or it wouldn’t have been such a big deal, right?) Wilson the historian will know more about, but he won’t know much about his economic policy or his racism. If you ask him to name one thing about the Wilson administration it would probably be his support for the League of Nations, which while it didn’t work out sure sounds like a nice idea, and if it had worked might very well have prevented WWII (so the thinking goes).
25. June 2009 at 11:46
Harding’s reputation may have suffered from the fact that HL Mencken found him easy to make fun of:
“He writes the worst English that I have ever encountered. It reminds me of a string of wet sponges; it reminds me of tattered washing on the line; it reminds me of stale bean soup, of college yells, of dogs barking idiotically through endless nights. It is so bad that a sort of grandeur creeps into it. It drags itself out of the dark abysm of pish, and crawls insanely up the topmost pinnacle of posh. It is rumble and bumble. It is flap and doodle. It is balder and dash.”
However: ‘…I don’t think historians should be economists.’ Don’t get me started on Paul David.
25. June 2009 at 12:07
Scott,
If you’re going to base your case against Keynes on the fact that he got the FX market wrong in 1920 it might be an idea to check out the story. Maybe he was “on the verge of bankruptcy” and maybe he wasn’t. (That’s a rather vague notion in any case; just how close to zero net worth is that there verge?) According to his biographer, Robert Skidelsky, Keynes reckoned his remaining capital after he liquidated his speculative positions was 21,500 sterling. (I’ve no idea whether that was a realistic valuation.) The Economic Consequences of the Peace had just been published and it was causing quite a stir, so maybe it wasn’t just charity that prompted his publisher to advance him 1,500 which, together with 500 from Basil Blackett, put him back into the market. His new partners were a former pupil-turned-stockbroker and his former boss at the Treasury. Were they just “rich friends” or were they people who trusted his judgement and expected him to make them richer still? You assume the former but, as is your wont, you offer assertions in place of arguments.
AFAICT the only legally enforceable debt Keynes was faced with was the 5,000 he owed his broker (which Cassel’s loan covered) and 2,000 which he owed to his father. The Bloomsbury group were sleeping partners. That’s where your Vegas story departs wildly from the facts – if he had won his bets they would have been entitled to their share of the profits, not just principal plus interest. Having lost, legally speaking he didn’t owe them a penny. But by 1922 he had reimbursed them in any case.
BTW, what’s your problem with page 142 of the GT? You say Keynes doesn’t understand Fisher. I say you don’t understand Keynes. But since you don’t bother arguing for your assertions, I won’t bother arguing for mine.
25. June 2009 at 12:32
We may not have known how to operate a central bank, but neither did we know how to operate under the gold standard. The treasury sterilized gold inflows leading to the 1907 panic all by itself. The implication the creation of the Fed did worse is spurious.
I certainly wouldn’t expect industrial production to fully recover in two years. We are talking about a credit crisis, the worst since the depression, not just a retooling from a wartime stance. For someone that touts monetary policy, it is incredible the emphasis you place on fiscal policy. What does a president do when the central bank is screwing things up so thoroughly? FDR’s wage policy was just a crude attempt to create inflation in about the only fiscal policy permitted, however misguided it was. Yes, some of his policies were mistakes, but more of them were not less than strokes of genius. Working under the biases and misconceptions of the time, he did turn around a disaster producing growth and productivity figures we would be envious of today. I trust the voters of the time judged this one right.
Ranking are usually based on the troubles they face and how well they address them. WWI may have been a mistake but Wilson was an expert at mobilization and prosecution of it. One wishes he was better at the aftermath which was lacking, but then we probably wouldn’t have entered it.
Those who care only for the short term are not investors but speculators. Investors generally rely on income flows for valuation while speculators rely on gains. Like all investors, Keynes certainly learned from his experience. That is the most one can require of an investor.
25. June 2009 at 16:29
Scott,
Don’t forget the income tax and the Alien and Sedition Act.
Woodrow threw several dozen anti-war dissenters into prison.
He was indeed the worst U.S. president of all time.
Harding freed many of the men the Woodster jailed, including Gene Debs, who he invited to the White House.
The Fed is the worst domestic institution in American history.
Social Security and Medicare are arguably next in line, so FDR and LBJ have to rank near the Woodster. Dishonest Abe wins the body count award, and I’d rank him next to the Woodster in the worst rankings.
Let’s go:
1. Woodster
2. Abester
3. FDR
4. Truman (a real thug and mass murderer, and he tried and failed to do socialized health care, the steel company seizure, etc.)
5. LBJ
6, Hoover, the GD, RFC and other statist policies made it worse
7. Dick Milhous (lots of statist interventions, EPA, affirmative action, block grants, Section 8 vouchers, etc.)
8. TR, another “progressive,” big sticking thug
9. Jimmy the Peanut Faamah, economic disaster, synfuel boondoogle, although he did get the deregulatory ball rolling–still as the great Herb Spencer said, politicians never deserve credit for this sort of stuff
10. Andrew Jackson–a lot of libertarians like this monster because he didn’t renew the charter of BUS2, but they overlook that he didn’t do it for highminded lib. reasons, but because Nick Biddle wouldn’t toe Jackson’s line, and Andy wanted to reward his corrupt cronies in Tenn. and elsewhere by putting government deposits into their pet banks.
25. June 2009 at 23:00
Scott,
I just had quick read of Keynes’s chapter 12 (here http://ebooks.adelaide.edu.au/k/keynes/john_maynard/k44g/chapter12.html).
There were no page numbers but I didn’t see where he was misunderstanding the Fisher effect (I’m assuming you mean that nominal rates should be the sum of real rates and expected infaltion).
Based on my quick read I agree with Kevin, you must not be understanding Keynes. Perhaps you can actually explain yourself just this once.
26. June 2009 at 02:56
Adam,
The section which bothers Scott is actually Chapter 11, section III.
26. June 2009 at 04:09
I am out of my depth here commenting on US history, however, I would like to say something about Mr Woodrow Wilson.
Woodrow Wilson presided over the creation of the central bank, or more accurately the Federal Reserve. He agreed to this in exchange for campaign support. This is how he became president and thus, upon becoming president whether he thought it was a good idea or not, he saw to the introduction of the Federal Reserve.
Once installed, the Federal Reserve promised that undesired economic events will be avoided now that we have a central bank that controls the supply of money and that can artificially control the interest rate.
To get to the point, the Federal Reserve was made up of a number of bankers, who as we all know, do nothing unless it brings them financial reward. The implementation of a central bank, backed by bankers who simultaneously control the interest rate and the supply of money, seems a recipe for nothing but undesired economic events.
Basically what we have is a bank that prints money, then slaps an interest rate on it, so exactly where are the people meant to get the money from, if for every dollar, they will owe more back. This can only result in a society that collectively is in debt to the Federal Reserve. In addition, the federal reserve hoarded all gold bullion (basically stealing from the American people) and thus virtually eliminated any possibility that the people may just have enough to repay the money they owe.
For this reason and for this reason alone do I believe it sufficient to claim the Wilson, of all the Presidents, made the most profound and ill-advised decision that will forever have dire consequences for the American people. Wilson even later admitted to this in a rather regretful manner. But I guess he wanted to be President.
26. June 2009 at 04:54
Kevin, I probably haven’t looked at that passage in Keynes in at least 20 years, but it always seemed like gibberish to me. I admit that one should not lightly assume that Keynes wrote gibberish, so if you can make sense out of that passage, I am sure that a lot of visitors to this blog would be very much obliged to you.
26. June 2009 at 05:33
Victor, The best indicator is inflation expectations in the TIPS market, which had fallen far below the Fed’s 2% target. But I also look at real growth indicators such as stock prices and the consensus forecast of economists.
David, I agree that the rankings aren’t worth much, but I do think they are very revealing. I strongly believe that historians actually do think that Wilson was a much better president than Harding, and that they don’t just treat the question as merely an appraisal of IQ. But you raise a good point. Why do historians come up with such silly judgments? If could be that they think smart presidents are, ipso facto, good presidents. They may reason backward from personal qualities and simply assume that those with good personal qualities were superior presidents. I looked at the original lists on Wikipedia and noticed one pattern that would support your view. In the early polls Hoover was ahead of Coolidge (another absurdity) but now Coolidge is ahead. Perhaps the early historians remembered that Hoover had a lot of impressive personal achievements before becoming president, and just assumed that some of those had to carry over. Later historians (unless they specialize in Hoover) may have forgotten about his impressive managerial achievements in aiding Europe after WWI.
That’s interesting what Hayek said about his investments.
I’m not sure that Fisher never developed a business cycle model. I think his models were more partially equilibrium (like Friedman) but that’s also the way I think most macro models should be. I would say that Keynes came up with a new language (and hence is a great economists in that sense.) But I don’t think his GT model has actually survived (although people like Krugman are tying to bring it back.) Again, what we teach our students now seems much more Fisherian, even if the language is still Keynesian.
Maybe you can help me better understand his insights into expectations and multiple equilibria. Every time I read him I think “all you are saying is that the less ‘confidence’ there is, i.e. the more bearish the expectations, the lower will be velocity.” But that was always known, wasn’t it?
I guess I also have a problem with the fact the the shifts in expectations that Keynes saw as exogenous (such as 1929), I see as rational responses to policy errors like central bank hoarding of gold. So the system doesn’t seem so “fragile” to me. And of course this is also my argument about 2008, the financial system really isn’t that fragile, but it doesn’t do well when the Fed stops targeting 2% inflation all of a sudden.
Current, That’s a good point. BTW, I do know that given the treaty with Belgium, and the intellectual environment of the time, British involvement was inevitable. That’s too bad, as otherwise the war would have ended much more quickly.
Blackadder, I mostly agree. But I would add that although partisan political bias is unlikely, ideological political bias may explain Hoover’s ranking. Many historians do know that Hoover was more of a statist, i.e. more of a activist, whereas Coolidge and Harding were more laissez-faire in their views. So the statist bias of historians may play a role.
Patrick, Mencken is my favorite journalist. I think you are right about the role this might have played. Who is Paul David?
Kevin, This is a blog. When I read a very well reviewed history book I tend to assume the facts presented are accurate. If you consider that sloppy, fine. But I don’t go out and spend 100s of hours researching every detail of Keynes financial affairs before making a comment on a simple fact that I have no reason to question.
Obviously I don’t know if the near bankruptcy occurred, but neither do you, if are are simply relying on the vague information you present from the Skidelsky biography. Are you saying that Ahamed was wrong when he said that by the end of May 1920 “the entire capital” was wiped out? Maybe after getting a loan from his father his position quickly rose to the 20,000 pounds you cite. I don’t know. You don’t provide an exact date like Ahamed does. By the way, Ahamed apparently read the Skidelsky biography.
I had thought that readers would understand that my Vegas analogy was not meant to be exact, but rather to show that a having a rich financial backstop can make a certain investment strategy look more successful that it really was.
On the Fisher effect, I was referring to the sentence that begins “There is no escape” where Keynes assumes that if asset prices are expected to rise, they will rise immediately, and hence there will be no expected rise in asset prices that gets reflected in interest rates. This is an important error, as Allan Meltzer once noticed that the whole liquidity trap issue would disappear if the central bank targeted inflation at a sufficiently high rate. Meltzer wonder why Keynes never proposed such a monetary policy. Elsewhere Keynes argued that Fisher effects only occurred during hyperinflation. My hunch is that Keynes never thought about inflation targeting, perhaps because he thought government’s could not be trusted with fiat money. Remember that Keynes considered fiat money the worst possible monetary system.
Lord, I haven’t studied the 07 crash (that’s 1907, not 2007!) The Treasury’s role is interesting. Certainly the first 25 years after the Fed was founded saw the 3 worst depressions of the 20th century, and the Fed did play an important role in causing them. I think the Fed has done a pretty good job in recent decades, but they did a horrible job before WWII.
I think you misunderstood me somewhere, I don’t place much emphasis on fiscal policy (as a factor in business cycles.)
In the first 4 months of FDR’s administration we were in a much worse financial crisis than today, and yet industrial production rose 57%. Obviously I don’t expect such rapid growth today. But I do expect SOME growth to occur, if we had an effective stimulus policy like FDR did. But we don’t.
What should FDR have done when when monetary policy screwed up? FDR devalued the dollar. And it worked. So monetary policy wasn’t the problem holding FDR back. He got the reflation he wanted. The other New Deal programs slowed what otherwise would have been a rapid recovery. I’m not for or against FDR, I judge each of his policies on their merits.
Rankings are “based on the troubles they face.” Yes, but what if they created their own troubles? Remember the fireman that started fires, so he could be a big hero in putting them out? Is this any way to judge leaders? To give them credit for getting us into useless wars that do no good? That’s how Julius Caesar and Alexander the Great and Napoleon were judged. I thought we were beyond that.
Regarding speculation. Even people only interested in short run gains will invest in the same way they would if they cared about the long run prospects of a company. That is because short run changes in prices reflect changes in the expected long run profits of companies.
Bill Stepp, I did mention the income tax, and how Harding freed Debs. Interesting list. I have seen others put Lincoln on the top of their worst presidents lists, whereas historians often consider him the best. He is certainly a polarizing figure, but I don’t think you can just look at body counts when discussing a war that was (despite many denials) partly about slavery.
Adam, Sorry for the wild goose chase, Kevin’s right that page 142 was in chapter 11. It is hard to give in depth views in a blog. I published a long article on Keynes in Economic Inquiry in the late 1990s–in case anyone is interested in my views of Keynes. I provide many quotations there to support my arguments.
26. June 2009 at 05:43
David, I think Keynes assumed that if speculators thought asset prices would be 10% higher next year, they’d immediately rise by 10%. Interestingly, that seems to be the interpretation of some people who read that page and agree with Keynes. But that ignores the fact that with 10% expected inflation nominal interest rates (and hence the cost of speculating in commodities) will rise by 10%. For example:
http://www.marginalrevolution.com/marginalrevolution/2009/04/keyness-general-theory-chapter-eleven.html
26. June 2009 at 06:55
Scott and peanut gallery:
I’m interested in the idea that US involvement in WWI caused worse outcomes — i.e., revenge-seeking Germany — than would have occurred if we had stayed out and simply let WWI be the latest installment of the centuries-old European civil war. Some French territory would’ve changed hands, but maybe you wouldn’t have had the rise of fascism.
Can you direct me to any books or articles that discuss that in more depth?
Thanks for a fascinating debate.
26. June 2009 at 07:29
They love Wilson because of the League of Nations.
26. June 2009 at 07:45
‘Who is Paul David?’
Didn’t I say, don’t get me started?
David is a very well known economic historian (Stanford), who has an unfortunate habit of failing to check out the stories he tells. Most infamously, in a 1985 paper published in the AER, in which he fell for the intellectuals’ version of the 300 mile per gallon carburetor Detroit didn’t want us to find out about:
http://www.utdallas.edu/~liebowit/knowledge_goods/david1985aer.htm
Which began: ‘Cicero demands of historians, first, that we tell true stories.’ And, 20 some years later David still won’t admit he got snookered–nor will a lot of well known economists, including Brad DeLong and Hal Varian.
It’s an intellectual scandal of the first magnitude. Or, it would be if academia hadn’t swept the episode under the rug.
26. June 2009 at 08:12
Scott, I wasn’t trying to defend the overall Keynesian model in the GT which I continue to regard as a huge mistake from many different points of view (though amid all the nonsense there is still a lot of very good stuff to think about). However, I used to think that his discussion of playing the stock market as analogous to predicting the outcome of a beauty contest was an embarrasingly superficial, though rhetorically powerful, recitation of popular prejudices about how asset markets operate. I no longer believe that. I think that despite the exaggeration and rhetorical overkill (I can’t believe the anecdote about the stock prices of ice companies in the winter) he had a very powerful insight into the ephemeral nature of market valuations (an insight, by the way, that Austrians in particular should be able to appreciate).
26. June 2009 at 09:01
Scott: When I read a very well reviewed history book I tend to assume the facts presented are accurate. If you consider that sloppy, fine.
Well, in a blog post entitled “Don’t Trust Historians” it does seem a bit odd.
David Glasner: if you can make sense out of that passage, I am sure that a lot of visitors to this blog would be very much obliged to you.
Perhaps there is some problem I’m not seeing, but it seems to me that Keynes is making some reasonably straightforward points. Here’s my quick summary of what the third and fourth paragraphs of GT Ch 11 Section III are saying:
The higher the anticipated rate of inflation the higher the marginal efficiency of the existing stock of physical capital. Other things being equal (nominal interest rates in particular), the market value of existing capital is also positively related to expected inflation. So far as employment is concerned, it’s the profitability of producing new capital goods that matters most. The higher the market value of existing capital, the greater the level of investment. Higher nominal interest rates will choke off that source of stimulus to increased output. If they are so high that Fisher’s famous equation holds, the stimulating effect of inflation is entirely eliminated.
I don’t think there’s anything terribly controversial here, but if you see some difficulty I’m certainly interested. Obviously I’ve confined myself to the points that seem to matter; I don’t know whether Keynes’s brief mention of Pigou, for example, does the poor man justice. FWIW, Robert Mundell, whose paper on this I haven’t read, seems to think Keynes had Fisher dead to rights and, so far as the actual behaviour of real and nominal interest rates is concerned, he notes that Fisher himself had pointed out that the facts didn’t accord with his theory.
26. June 2009 at 09:19
Kevin, For Keynes and Mundell to get that result, aren’t they assuming that the future cash flows which increase because of inflation are being discounted by a discount factor which does not adjust to reflect the inflationary expectations. That seems like an inconsistent assumption to me. The inflation is either anticipated or it is not. What am I missing?
I think that the reason that the data didn’t support the Fisher hypothesis for the Gibson paradox is that under the gold standard, inflationary upticks were often random or self-reversing, so inflation now did not necessarily imply inflation in the future, indeed, inflation now might well be create deflationary expectations. Under a fiat standard in which the inflation rate is chosen by the monetary authority, the public is likely to respond to inflation now by assuming that inflation will persist in the future. So empirical support for the Fisher hypothesis depends on the monetary regime from which observations are drawn.
26. June 2009 at 09:24
Many of the points Scott is making here about Chapter 11 are in “Failure of the New Economics” by Henry Hazlitt on p.160-165.
http://www.mises.org/books/failureofneweconomics.pdf
With characteristic grumpiness he writes…
“If the marginal efficiency of capital embodies expectations, so do interest rates. To assume otherwise is to assume that entrepreneurs are influenced by their expectations but that lenders are not. Or it is to assume that entrepreneurs as a body can be expecting prices to rise while lenders as a body do not expect prices to rise. Or it is to assume that lenders are too stupid to know what borrowers know.”
Of course it is also true, as I pointed out here recently, that knowing the change in the general price level doesn’t mean you can know the change in the level that applies to you or your business.
In the Appendix to that book Hazlitt puts in some charts showing the stock price of Ice companies in the 30s. As you may expect they don’t do what Keynes claims they do.
Another interesting point Hazlitt makes is that this is Keynes who said “In the long run we’re all dead” and that economists should only be concerned about the short run. Then in Chapter 12 he attacks speculators for only looking at the short run.
I agree that there is some good stuff here about emphemeral market valuations as David puts it. It would be interesting to unpick this chapter sometime and separate the wheat from the chaff.
26. June 2009 at 09:32
Continuing your logical chain, but in true Internet fashion…
6. And so if an economist relies on Keynes, he is relying on policies that enabled Hitler and the Nazis to take power.
7. Therefore, all liberal economists are Nazis and no better than Hitler.
QED
26. June 2009 at 09:49
Kevin Donoghue: “Other things being equal (nominal interest rates in particular)”
Kevin Donoghue: “I don’t think there’s anything terribly controversial here”
There is something controversial. How can the interest rate be held equal?
For investments in capital we are talking about the short term rate and long term rate, which floats. It is affected by expectations too as we have been discussing in the “Was Krugman right in 2002” thread.
Think about it. In the paragraph you write below:
“The higher the anticipated rate of inflation the higher the marginal efficiency of the existing stock of physical capital. Other things being equal (nominal interest rates in particular), the market value of existing capital is also positively related to expected inflation. So far as employment is concerned, it’s the profitability of producing new capital goods that matters most. The higher the market value of existing capital, the greater the level of investment. Higher nominal interest rates will choke off that source of stimulus to increased output. If they are so high that Fisher’s famous equation holds, the stimulating effect of inflation is entirely eliminated.”
You could hold M.E.C. steady and allow the interest rate to change. What would that show?
Keynes insists that M.E.C. is “dynamic” and the interest rate is “current”. But there is no real difference read the following excerpt and see if you agree.
Keynes: “The mistake of regarding the marginal efficiency of capital primarily in terms of the current yield of capital equipment, which would be correct only in the static state where there is no changing future to influence the present, has had the result of breaking the theoretical link between to-day and to-morrow. Even the rate of interest is, virtually, a current phenomenon; and if we reduce the marginal efficiency of capital to the same status, we cut ourselves off from taking any direct account of the influence of the future in our analysis of the existing equilibrium.”
As I’ve been explaining else-thread. Even if you subscribe to the pure time preference theory of interest the interest rate is affected by expectations.
26. June 2009 at 10:57
I’m stunned every time I witness the party line leftist mentality of the History Department professors …
26. June 2009 at 11:03
To an academic “great” = advanced socialism and the Democrat Party.
I’d suggest Wilson did more the FDR to advance these.
The surprise is that Wilson isn’t #1 on their list.
26. June 2009 at 11:14
Scott,
This is a really good post. Thanks.
I am, however, surprised that you, of all people, think that Harding’s handling of the 1921 depression was good. According to your Monetarist rule-book, shouldn’t Harding have expanded the money supply to fight the depression? Yet he didn’t. And there was a fast recovery.
Doesn’t that make you doubt the veracity of the monetarist view? Doesn’t that support more the Austrian deflationist liquidationist view?
Yes, it’s true that the Fed didn’t expand the money supply in 1929 to fight the depression. But I would argue that that wasn’t the cause of the depression. The cause was the protectionism and the price control that prevented the adjustment from taking place. Had the Fed let prices adjust, recovery would’ve been swift. It would have been more painful than 1921, probably, but only because more monetary expansion was carried out before 1929 than before 1920.
26. June 2009 at 12:25
I guess that the easiest way to have a great impact is to destroy rather than create. Awful bias.
26. June 2009 at 14:06
[…] Wilson, Harding, Keynes, by Bryan Caplan In his latest one-two punch, Scott Sumner…1. …explains why Woodrow Wilson was the worst president in U.S. […]
26. June 2009 at 22:06
Scott,
Hello. First time commenting here.
I’ve wrote my interpretation of Keynes’s view of the Fisher Effect in my comment of WCI blog “In praise of MV=PY” post (himaginary@June 25, 2009 at 10:04 AM) as response to your comment.
http://worthwhile.typepad.com/worthwhile_canadian_initi/2009/06/in-praise-of-mvpy.html
Allow me to reproduce what I wrote there:
“P.142 seems prelude to chapter 17. The uniqueness of Keynes’s idea, as I understand it, is in that he redefined the Fisher effect as arbitrage between goods and money. From his perspective, real rate of interest is interest on real goods, and money rate of interest is interest on money (of course). If the former is smaller than the latter, money becomes more valuable than real goods, hence the price of goods goes down. It goes down to the point where it bounces back, the point where expected change of the price of goods becomes positive. Then, money rate of interest equals real rate of interest plus expected rate of inflation, i.e. the Fisher equation holds. And this mechanism is exactly identical to what Krugman described as “What happens is that the economy deflates now in order to provide inflation later” when he explained his model .”
As for Keynes as an investor, does one’s reputation as an investor matter so much to one’s reputation as an economist? I’m not so sure. For example, LTCM fiasco didn’t make people to back off from using theory by Scholes and Merton. Rather, current crisis made them so.
FYI: As for Wilson and racism, we Japanese experienced a symbolic episode at that time.
http://en.wikipedia.org/wiki/Paris_Peace_Conference,_1919#The_racial_equality_proposal
27. June 2009 at 01:00
Without entering into the economics and discussions of the efficient markets, which has been done in more detail elsewhere…the entire proof that Keynes was a bad investor is that he got wiped out?
I have no idea about currency markets in the 1920s, but here is one pretty obvious hypothesis which you need to knock down.
The information available when Keynes & Co. made their investment wasn’t sufficient to predict ‘a spasm of optimism’, which is admitted to be short-lived. Keynes could still have wonderful insight and understanding, but information shortage would account for his error. No matter how good you are, what you don’t know can easily kill you. This hypothesis is theoretically neutral – efficient markets or some kind of ‘investor sentiment’ model, take your pick.
More to the point, I’m struggling to think of any ‘great investor’ who hasn’t been taken to the cleaners at some point. Buffet and Soros, the first to come to mind, both have suffered significant reverses, just at points in their career when they had already made it and could afford the losses.
27. June 2009 at 06:17
Frank, I’m not quite so suspicious of the Fed, but I agree they caused great harm before things got a bit more stable after 1982 (until 2008.)
Dave, Sorry but I don’t know much WWI history. I read one book by John Keegen, but based on reviews the book by Niall Ferguson (The Pity the War?) might be more along the lines of what I am suggesting. Keegen would disagree with my opposition to US and UK involvement.
The book from Dawn to Decadence by Barzun discusses how almost all European intellectuals supported the war.
Silvermine, Yes, I guess Wilson was idealistic (or an idealistic racist.)
Patrick, I hadn’t heard about that scandal. (Perhaps because it was covered up.) But I have heard many academics claim the auto companies held back mysterious technologies to enrich . . . the oil companies? I never understood that argument, and I don’t believe in conspiracies. Thanks for sharing that info.
David, You may be right, but I’ll continue to think that it reflects the incredible unpredictability of the world we inhabit. Does Keynes’s beauty contest theory have any practical implications? That’s the issue I raised in my anti-anti-EMH post from a few months ago. The anti-EMH theories have no practical value.
Kevin, Very clever, but there is an obvious distinction between trusting a historian’s recitation of facts, and trusting a historian’s interpretation of those facts.
Fisher’s view on the asymmetry of the Fisher effect seems implausible, but Fisher had a better understanding of abstract macro theory than did Keynes. Fisher simply didn’t push rational expectations far enough.
The problem with Keynes’ statement is that he is claiming that asset prices should respond immediately to expected inflation, and thus that there would be no expected appreciation of asset prices to influence market interest rates. If you think I misinterpret Keynes, then explain away this “smoking gun” in a letter from Keynes to Robertson:
“Now, as against this, my contention is that the expectation of a change in price has no effect on the rate of interest, since it leaves unchanged the relative attractions of cash and loans. Are you denying this? Do you think that an expectation of higher prices in [the] future causes the rate of interest to rise?” (Feb. 1935, Collected Writings, Vol. XIII, p. 518.)
David, Very well put. I have some sympathy for Keynes overlooking the Fisher effect, as the price level was close to a random walk during Keynes lifetime, as you said. I am not as anti-Keynes as my grumpy post suggests. Sometimes I just like to puncture his overblown reputation. But he was a fine economist and a great writer. I think you and I have very similar views of Keynes in that regard.
Current, Good points. In fairness to Kevin, this is an asymmetry in Fisher’s approach as well. But I think Keynes’ problem’s are much deeper.
I’ll finish replying an another comment
27. June 2009 at 06:56
TomB, I assume you are joking in your assertion that Fed policy in 1930 relied on Keynes.
Current, Thanks. And again if there is any doubt about what Keynes meant, check out the quotation I provided above in the response to Robertson. Keynes just doesn’t “get” expectations.
Greg, Yes, I think historians have a statist bias. It wouldn’t even surprise me if this is true of some on the right (say neoconservative historians.)
Saifedean, Check out my newest post, which is a long rambling reply to your question. Here I will add that the 1929-33 deflation lasted almost four years, vs. one year in 1920-21. So the Great Depression would have been much worse than 1921, even w/o Hoover’s statist policies in other areas.
jean, There are plenty of economists out there trying to discredit monetary stimulus. I am trying to be constructive in this blog, to show how monetary policy could solve many of our problems that currently seem insolvable. I don’t know how much you know about intellectual debate, but to advance your own constructive ideas it is sometimes necessary to show how existing ideas are counterproductive. Keynes would not object to my criticism, as he was 100 times more sarcastic than I am. He relished the give and take of intellectual debate.
Himaginary, Thanks for commenting here. And thanks for the history of Japan after WWI. I am missing something in your argument, but it may be my fault. You describe a situation where the interest rate on money exceeds the real interest rate on real goods. But isn’t this what happens during hyperinflation? And yet obviously money doesn’t become more valuable than goods during hyperinflation. Perhaps it would help if you indicated whether you agree with my assertion that Keynes denied that a rise in expected inflation would raise nominal interest rates (as I indicated in another quote provided earlier in a response to Kevin.)
I agree that one’s reputation as an investor shouldn’t matter, but I think it does. Especially if you claim that your theories can allow you to beat the market. If it was discovered that Shiller’s investments had all done poorly, I think it would hurt his reputation as an economist.
Barney, You make a good point, and I agree. I’m afraid that I didn’t express my ideas well. Let’s start with the fact that Keynes does have a good reputation as an investor. My point is that it is difficult to really know how good someone was if that had a wealthy backstop. Recall the old joke about “the markets can stay irrational longer than you can stay solvent.” In statistics there is a famous example of a coin flip bet, that is endlessly repeated. You keep betting heads, and then go double or nothing if you lose. Eventually you will almost certainly come out ahead, but you can only play this game if you have a wealthy backstop to cover losses on the way toward to eventual “heads.” It is not fair to compare Keynes to an ordinary investor that doesn’t have someone to bail them out if they go bankrupt. I’m not saying Keynes might not have been a fine investor, but the quote I provided suggests that those who want to make that argument have a much higher bar to overcome than they might have thought.
In my own case I put a lot of money into East Asia in the 1990s. After the 1997 crash I was convinced it was a great time to buy (and it was) but I didn’t have anyone rich to advance me a lot of money. I have still done fine in the long run, but not nearly as well as if I’d had a rich guy to allow me to go double or nothing after my first bet lost.
27. June 2009 at 08:39
“Recall the old joke about “the markets can stay irrational longer than you can stay solvent.””
Perhaps you will also recall the Keynes was the originator of this joke, that it was occasioned by precisely the failed investment that you refer to (in which he bet strongly, brilliantly, but disastrously prematurely against the German mark), and that this incident, which you seem to think you are the first to bring to anyone’s attention, is the most famous incident of Keynes’ entire career as an investor. Only a hack like Mankiw could find this over-exposed nugget “thought-provoking.”
27. June 2009 at 08:52
David,
Your comment at 09:19 seems to be concerned with the empirical question, whether Mundell, Keynes and Fisher were actually right in thinking that Fisher’s theory is rejected by the data. All I was dealing with in my 09:01 comment was your remark that the passage in question seemed like gibberish to you.
As to the inconsistency you see, I don’t read Keynes as ruling out the possibility that a rise in expected inflation, which stimulates investment, will cause a rise in interest rates just sufficient to fully offset that stimulus. He treats that as a special case: “Professor Fisher’s theory could be best re-written in terms of a ‘real rate of interest’ defined as being the rate of interest which would have to rule, consequently on a change in the state of expectation as to the future value of money, in order that this change should have no effect on current output.” At full employment there’s not much scope for an increase in output, so I think we can take it that Keynes would happily concede that Fisher’s equation, like other features of the ‘classical’ model, is valid when comparing one full-employment equilibrium with another. But what happens when there is mass unemployment and wages are sticky is another matter.
Current,
When I say that, “other things being equal (nominal interest rates in particular), the market value of existing capital is positively related to expected inflation” all I’m getting at is this: in any sensible model in which nominal interest rates and expected inflation are variables, the NPV function which gives you the market price of existing capital is increasing in expected inflation (and decreasing in interest rates of course). I’m certainly not saying that it would be sensible to regard nominal interest rates and expected inflation as independent variables.
Scott,
I see that Greg Mankiw has taken up your theme, so your line of Keynes criticism obviously resonates with Republicans. You could attack his taste in paintings while you’re at it – I seem to recall one of his Bloomsbury friends saying he bought the worst painting Cezanne ever produced.
Let’s get back to the economics:
The problem with Keynes’ statement is that he is claiming that asset prices should respond immediately to expected inflation, and thus that there would be no expected appreciation of asset prices to influence market interest rates.
Well he’s right about asset prices surely? If asset prices don’t respond there’s a free lunch there for the taking. Keynes didn’t subscribe to the strong-form EMH, but he certainly didn’t think market players were stupid. The line of causation he describes is: asset prices jump, production of capital goods is thereby stimulated, consumption is increased via the multiplier (assuming there is spare capacity) and, finally, increased transactions demand for money raises interest rates. So inflation generally does boost interest rates, but it does so through a process which increases income. Naturally the slope of the yield curve anticipates the path of short-term rates.
Without knowing the context of his letter to Robertson my best guess is that he is saying something like that to him also – an expected price increase can’t be the proximate cause of an increase in short-term interest rates. But of course (although he doesn’t mention this in the bit you quote) rising prices will shift the MEC etc. as above. You can quibble with his view of the transmission mechanism, but you’re a long way from establishing your claim that he didn’t understand Fisher. If anything you are highlighting something Keynes’s admirers have always known: that he gave a great deal of thought to these issues and discussed them at length with his peers. Your portrait of a spoiled brat isn’t even a good caricature.
27. June 2009 at 08:56
maynardGKeynes, I didn’t claim to have brought this nugget to people’s attention, I said I read it in a famous book.
But I still think people don’t fully understand the implication of this information (which was new to me.) Sorry but I am not impressed by someone who starts out investing, fails, gets even more money from a rich dad and then succeeds the second time. I think I could do pretty well under those circumstances as well. Would you consider my investments in East Asia in the mid-1990s “brilliant”, because the East Asian stock markets exploded upward between 1998-2007. After all, I was just early—they are great growth stories. If only I had had a rich dad to bail me out in 1997, after I had blown my first investment.
If this is what the anti-EMH Keynesians consider evidence that smart guys can beat the market, more power to them. And good luck on Wall Street.
27. June 2009 at 10:06
First of all, thanks for taking the time to reply to my comment, which I appreciate. Here’s where I respectfully disagree with you. I can’t see how Keynes’ apparent ineptitude as an investor undercuts his greatness as an economist — they are fundamentally different skills. FWIW, he did have a good career as the “David Swensen” of Oxford for many years, which also proves nothing. The case of Irving Fisher is actually quite the same as that of Keynes. His reputation as an economist is, if anything, even higher today than it was in 1929, when he made his famous gaffe about the stock market. His theory of debt deflation and interest rates is still the best there is. As far as EMH, I think Keynes was actually attacking Say’s Law, which said that markets were self stabilizing. This is doesn’t have much to do with EMH, which is more about information arbitrage.
BTW, I’m sure Keynes obnoxiously thought he was smarter than everyone else too, but unlike Greg Mankiw, he was probably right.
27. June 2009 at 13:24
Scott,
I first saw the “was Keynes a savy investor” piece on Greg Mankiw’s website. It initially reminded me of stories that I heard in college in the early 1970’s about how Keynes made his stock picks in the morning before getting out of bed. So I read on ….
Wow! Then I remembered my first experience with the passion that Chicago PhD’s hold for Keynes, and this experience came in a corporate finance class taught by a card carrying Chicago man – in the mid 1980’s.
You really have to admit, the stuff above is a shade bit nasty. The thing that really caught may attention was the statement that Keynes argued against the efficient market hypothesis in chapter 12 of his General Theory. Now my acedemic “chops” do not in any way place me in this league, but my guess is that the efficient market theory was not being debated in Keynes day, at least under that name.
Jim
27. June 2009 at 13:57
A good book on the US entry into WW1 and what it meant is “The Illusion of Victory” by Thomas Fleming.
27. June 2009 at 18:41
Scott,
Thank you for responding to my comments both here and at WCI. I’m really impressed by you responding to each comment diligently.
“Perhaps it would help if you indicated whether you agree with my assertion that Keynes denied that a rise in expected inflation would raise nominal interest rates (as I indicated in another quote provided earlier in a response to Kevin.)”
In Keynes’ idea, as I understand it, real rate of interest and money rate of interest are determined independently. That’s what articulated exactly in the passage in his letter to Robertson you quoted. So, in his idea, the relative price of goods and money must change to adjust the disparity between them. And, only then, expected inflation rate kicks in. In other words, expected inflation rate only serves as adjustment factor between pre-determined two rates, and not included in either of them.
“But isn’t this what happens during hyperinflation? And yet obviously money doesn’t become more valuable than goods during hyperinflation.”
That’s tricky part of Keynes’ (and Krugman’s) logic, because it’s two-stage logic. Hyperinflation is the process of adjustment toward parity, not the result of it. If real rate is higher that money rate, expected inflation rate must become negative. But the only tool the economy has got is current price of goods. So, in order to attain negative expected inflation rate, the price of goods must be high enough to the point that people think it would fall thereafter. I think that’s what hyperinflation means in Keynes’ logic.
28. June 2009 at 04:42
Kevin, First of all, my personal comments about Keynes were meant as a joke. I thought that when I poked fun at myself right afterward people would see that. I am not blaming you as my sense of humor isn’t very good, but just mentioning that. Of course sarcasm and personal insult is exactly what Keynes was famous for (Economic Consequences of the Peace), so I thought those who like Keynes would enjoy my sarcasm. I don’t do this with other economists, even those I disagree with. And I have said that Keynes is a great economist in past posts.
No, asset prices should not respond immediately to an increase in expected inflation. Higher interest rates should prevent that from occurring. That is exactly Fisher’s point Suppose the expected rate of inflation rises from 0% to 10%. In that case the expected rise in asset prices would also rise from 0% to 10%. There would be no discontinuous jump in asset prices at the point of inflection. Now if you want to be picky I agree that a few asset prices might jump, if they are regarded as inflation hedges. Gold would be an example. But Keynes clearly meant something more. He clearly thought the fact that asset prices would jump was some sort of insight that proved Fisher wrong. Well if it were true Fisher would be wrong, and we’d have to tear up the modern textbooks. Now there are lots of good arguments for modifying Fisher’s conclusion. If prices are sticky a monetary shock may have real effects, and then asset prices might jump. But that is a completely different issue. Go back and study the context for the letter to Robertson, I don’t think you will find any good excuses for what he said, which is absurd.
Keynes was a product of the gold standard, where the expected rate of inflation was near zero. In that environment changes in actual inflation were unexpected, and one didn’t observe much of a Fisher effect. But unlike Fisher, Keynes lacked the imagination to understand what would happen if there were once and for all changes in the trend rate of inflation. Indeed if he had that imagination, everything we know about Keynes suggests that we would have favor mild positive inflation (say 2 or 3% a year) not stable prices (which he actually favored.) This is because in his model real output is higher with mild inflation, and there is less chance of liquidity traps.
BTW, the Fisher effect does not depend on an income effect, if by “income effect” you mean real income.
maynardGkeynes, I agree that Keynes was a great economist, and have said so in other posts. His greatness comes from his inventing a new language, which others have adopted. But I still think he is overrated. He is viewed as greater than he really is. The GT is a deeply confused book. The Tract, on the other hand, is great.
I have met Mankiw and he seems like a very nice guy. If you think he comes across as arrogant in his blog, what must you think of that even more famous blogger, and follower of Keynes? The guy who (unlike Mankiw) is always calling other famous economists stupid? I actually think Mankiw’s blog is pretty polite, relative to some other bloggers.
In Chapter 12 he clearly uses anti-EMH arguments. Obviously the EMH didn’t exist at that time, but these are exactly the same arguments that people today use against the EMH. They are not arguments against Say’s Law, which as far as I know was not applied to anomalies in financial markets.
Jim, You misunderstood the argument. Check out what I said to the two previous commenters, it answers your questions. (I wasn’t “nasty” and it is an anti-EMH argument.)
Thanks sourcreamus.
Himaginary, If that is what Keynes meant, then he is wrong. You do not need to have hyperinflation, or even high inflation, for prices to be expected to fall. The fall in prices in the US in the early 1930s was not preceded by hyperinflation, and neither did Japan experience hyperinflation before the current (year 2009) deflation set in. And I am quite sure that Krugman would not make that argument. Can you provide a quote from Krugman?
28. June 2009 at 08:43
You’re not the only non-Keynesian who finds much to like in early Keynes, even among Chicagoites.
Mario Rizzo has a “Hayekian appreciation” of Keynes here.
maynardGKeynes, I think Scott already said this but it might have been in an older post. He doesn’t expect great economists to be great stock-pickers either. The EMH says we shouldn’t expect anybody to be able to regularly beat the market (other than through the law of large numbers producing a few improbable lotto winners). If Keynes was in fact able to beat the market that could serve as a bit of evidence counter to the EMH (which Scott says Keynes himself opposed). By giving a revisionist account of Keynes’ stock-picking Scott is showing that it isn’t evidence against the EMH.
28. June 2009 at 09:32
I too am puzzled at the animus one poster exhibits toward Mankiw. I’ve never met Mankiw in person, but have exchanged a few e-mails with him since he began blogging.
He’s always been polite to me. Disagreements were dealt with cordially, and when I offered him examples that bolstered some point he’d made, he was gracious in crediting me.
Since, I mentioned Hal Varian above as someone still denying that Paul David’s theory of market failure published in the AER in 1985 was based on a crude hoax, I’ll also mention that I’ve had extensive dealings with Varian (once debating him over a span of a couple of weeks) and though I (and David Friedman) could not convince him he was in error over a point in his and Carl Shapiro’s ‘Information Rules’, the debate was scrupulously polite between the three of us. He still will respond to my e-mails.
Which brings me to Paul Krugman’s role in that fiasco. In some ways Krugman’s sin was even worse than Paul David’s (and not neglect his colleague’s role, Brian Arthur too). Krugman took the ball from David and Arthur and ran with it, publishing a chapter in his 1994 ‘Peddling Prosperity’, titled ‘The Economics of QWERTY’, gleefully touting it as a commonplace example of market failure that dopey ‘conservative economists’ won’t recognize.
However that was four years AFTER Steve Margolis and Stan Liebowitz published their first demolition of the QWERTY/Path Dependence myth; ‘The Fable of the Keys’ in, iirc, the Journal of Law and Economics. Shortly followed by other journal publications in similarly non-obscure places such as the JEP.
That is, by 1994 there was no excuse for Krugman not to have been aware of the facts of the matter. Then, when Stan and Steve included Krugman in their criticism in this Cato ‘Regulation’ piece:
http://www.cato.org/pubs/regulation/regv18n3/v18n3-4.pdf
Krugman responded in a letter, by attempting to change the subject (scroll down to ‘The Prevalence of Path Dependence’):
http://www.cato.org/pubs/regulation/regv18n4/reg18n4-letters.html
Which was immediately, and very effectively, rebutted by Liebowitz and Margolis (‘Academic Scribblers Leave Paper Trails’). To which Krugman never responded. The closest he ever came to admitting error was in telling a WSJ reporter in 1998: http://www.utdallas.edu/~liebowit/
‘[Krugman] concedes that evidence of being locked into a bad technology “turns out to be fairly weak.”‘
No kidding.
28. June 2009 at 09:42
How did Keynes’ father get rich?
Not by writing on economic “methodology”.
Maybe the father was the genius, and we should pay more attention to _his_ economics.
28. June 2009 at 09:51
Scott, David,
Here’s my attempt to make sense of Keynes in chapter 11 of the General Theory:
http://canucksanonymous.blogspot.com/2009/06/keynes-on-fisher-effect.html
Basically my interpretation is that Keynes is claiming that an increase in expected inflation reduces the real value of existing debt on firm balance sheets and thus causes credit spreads to tighten. The lower credit spreads partially offset the rise in the risk free interest rate, thus lowering the real rate a firm faces and stimulating further real investment.
28. June 2009 at 09:55
I hope your history is international in scope.
I was looking at some output numbers, and noted that the depression was a very different thing in Britain that it was in America — which makes a difference when you attempt to make sense, for example, of Hayek’s public pronouncements at the time. Hayek was buried in theoretical work, and what he did know empirically was a popular empirical stuff about Britain — which didn’t show anything like the same empirical profile as the U.S.A. in the 1929-1932 range.
Scott:
“I am current working on a revisionist history of the Depression, which shows the New Deal actually lengthened the Depression by 6 or 7 years.”
28. June 2009 at 16:04
I like your roulette strategy, but you’d be better off finding a single-zero wheel. You’d decrease your chance of losing from 1 in 90 to 1 in 152.
28. June 2009 at 20:01
Re Wilson:
“My main complaint is that 116,000 Americans lost their life for no good reason … he sacrificed 25 times as many young American men in Europe as Bush did in Iraq (a war that is widely viewed today as an American tragedy of the first order)”
And one might note that the population of the US at the time was about 100 million, so the comparable number of deaths today would be closer to 350,000 … or 75 times as many as in Iraq.
Considering Wilson’s policies on war, press suppression, civil rights, racism, his gross screw-ups at the close of WW I and after (compounded by his self-righeousness over it all) and his surreptitious handing of the keys to the govt to his wife after his stroke (Hillary could only dream!), it’s hard to imagine how generally left-leaning history faculties think so highly of him.
Or maybe not.
After all, FDR used military tribunals to shoot people without trial (including at least one US citizen) sent loyal American citizens to concentration camps because of their race, and “Roosevelt’s greatest contribution came in his using all his devious sneakiness as a ruthless political intriguer to put the United States in harm’s way in World War II. Great man. Great President”, wrote Brad DeLong on his blog. Indeed, FDR is the patron saint of the liberal Democratic side of the world. So maybe it’s not exactly what one does that makes one a good or bad president, but whether the cause one does it in is noble enough … and we get to decide that for ourselves after the fact. “History is written by the winners”, and all.
Slipping sideways and off the thread topic a bit, it is interesting to see how much human life apparently has gained in value over the last 100 years. In the 1970s, 50,000 deaths in Vietnam over most of a decade were a politically unsustainable national tragedy; now circa 4,500 in Iraq are. But 95 years ago, the world’s most developed nations took over a million casualties in single battles such as Verdun and the Somme, and just kept on going year after year. And in World War I, what were they all fighting over, anyhow?
A lot of people have observed that envronmentalism is a luxury good. Morality too. The value of animal life. (After that jet landed in the Hudson river the govt started clearing the geese off the NYC airports — now protesters are marching through Manhattan to “Save the Geese”.) And the value of human life too — the EPA now values a life at $6.9 million for decision making purposes. (Try fighting WWI on those terms!)
Maybe the single best benefit from economic development if the rise of these “luxury goods” (damned filthy geese apart) … but I digress. Back to arguing about Keynes.
29. June 2009 at 07:03
Now if you want to be picky I agree that a few asset prices might jump, if they are regarded as inflation hedges. Gold would be an example.
Other (and in some cases better) examples would be equities, land, buildings, oil-wells, machinery, ships, aircraft, financial assets denominated in hard currencies, some livestock, paintings, jewellery and antiques. This list is by no means exhaustive. Mrs Thatcher recommended well-stocked freezers when she was leader of the opposition to the Labour government.
If prices are sticky a monetary shock may have real effects, and then asset prices might jump. But that is a completely different issue.
Keynes doesn’t assume sticky prices for either goods or capital assets but he certainly does assume sticky wages. I really don’t know how you manage to convince yourself that the real effects of monetary shocks are a different issue. He was writing about the impact of expectations on investment and employment. That is the issue. The Fisher equation just came in for a brief mention.
Go back and study the context for the letter to Robertson, I don’t think you will find any good excuses for what he said, which is absurd.
Here’s what he said, so you tell me: “my contention is that the expectation of a change in price has no effect on the rate of interest, since it leaves unchanged the relative attractions of cash and loans.” Far from being absurd that’s elementary optimization, as far as it goes. (It doesn’t go far enough but then it is just one sentence.) Suppose I hold $200 in cash earning no interest and $1,000 in a savings account earning 3%. Based on news that Ben Bernanke has ordered a fleet of helicopters, I revise my expectation of inflation from 1% to 5%. Is there some reason why I should now wish to hold more in cash and less in my savings account? Or vice-versa? Common sense suggests I should reduce both my $200 cash and my $1,000 interest-earning deposit and switch my wealth into real assets. But as Keynes points out, if my concerns about inflation are shared by other agents, the prices of real assets will already have risen.
Incidentally, you really don’t need to go digging out the Keynes-Robertson correspondence to show that Keynes was in earnest about his liquidity-preference theory of the interest rate. In GT Ch 13 Section II he states baldly that the rate of interest is simply “the ‘price’ which equilibrates the desire to hold wealth in the form of cash with the available quantity of cash….” If you think that’s absurd you’re not the first by any means. The great liquidity-preference versus loanable-funds debate dragged on interminably. But your claim that Keynes didn’t understand Fisher is another matter. You’ve a bit of work to do when it comes to establishing that. Are you serious, or is it just another of your jokes? I think he understood Fisher pretty well and, up to a point, he agreed with him – just as he agreed with Marshall, up to a point.
29. June 2009 at 07:22
Thanks TGGP.
Patrick, Thanks for the info on path dependence. The letter from Krugman is very revealing. I agree with you about Mankiw.
Greg, I don’t know anything about the dad’s views on economics, but my experience is that there is zero correlation between investment success and economic knowledge. Indeed I believe that there is probably zero correlation between investment success and any kind of measurable knowledge. Otherwise mutual funds run by those with a certain measurable advantage (say higher IQ) would outperform the funds run by stupider people.
Adam P, I don’t understand why someone like Keynes who is such a great writer is such a poor communicator. It is never hard to figure out what Fisher and Hawtrey “really meant.” If Keynes really meant all the things his defenders claim, why didn’t he just say what he meant? The GT is almost unreadable. And it’s not just me, I remember Patimkin saying the same thing.
My second point would be that even if you are right, it doesn’t really explain his dismissal of the Fisher effect. He should have said “the Fisher effect is true, but there are also other effects that work in the opposite direction.”
And by the way, I have also seen him dismiss the Fisher effect for government bonds, which could not be explained away using your risk spread explanation.
Thanks for the attached paper, I’m behind in my work right now but will definitely get to it.
Greg, My history has a US focus, but uses international factors where needed. Thanks for mentioning that UK difference. I don’t have anything critical of Hayek in the book right now, but your observation does partly (not completely) explain his views. (The UK depression was milder, but still significant.)
Tom, Thanks, I’ll keep my eye out for wheels with 37 numbers.
Jim, The evolutionary psychologist Stephen Pinker has recently been talking a lot about the growing value of human life. He point out that the world has been getting steadily less violent (despite headlines) since the STONE AGE. You are right that we would no longer tolerate such huge casualties, and it is a good thing.
29. June 2009 at 07:57
Scott,
I agree with what you’re saying, I was just offering one possibility that might make a bit of sense.
Of course, it could be simply that Keynes had in mind a liquidity trap. If the economy is liquidity trapped with the real rate stuck above its equilibrium level then the Fisher effect wouldn’t hold, an increase in expected inflation would just decrease the real rate until it hit it’s equilibrium level. Only then would it raise nominal rates.
29. June 2009 at 08:16
Scott,
“Can you provide a quote from Krugman?”
That’s the sentence I keep quoting in my comments here and WCI blog:
“What happens is that the economy deflates now in order to provide inflation later.”
http://web.mit.edu/krugman/www/japtrap.html
Please note that this sentence refers to the situation like US in the early 1930s and Japan in 1990s, where the economy needed expected inflation. The economy experiencing hyperinflation is in just opposite situation, i.e., the economy needs expected deflation so that valuation of money rises against that of goods. In Krugman’s words, in that situation, the economy inflates now in order to provide deflation later.
I agree that price does not necessarily need to fall to cause expected inflation, or need to rise to cause expected deflation. If monetary policy (and/or fiscal policy) can generate proper expectation on inflation rate, that would be the end of the story.
The problem occurs when monetary policy is null and void (perhaps because of political problem), and above-mentioned price-adjustment process cannot bring about parity between attraction of money and that of goods. Let me quote a paragraph from GT chapter15:
“The most striking examples of a complete breakdown of stability in the rate of interest, due to the liquidity function flattening out in one direction or the other, have occurred in very abnormal circumstances. In Russia and Central Europe after the war a currency crisis or flight from the currency was experienced, when no one could be induced to retain holdings either of money or of debts on any terms whatever, and even a high and rising rate of interest was unable to keep pace with the marginal efficiency of capital (especially of stocks of liquid goods) under the influence of the expectation of an ever greater fall in the value of money; whilst in the United States at certain dates in 1932 there was a crisis of the opposite kind — a financial crisis or crisis of liquidation, when scarcely anyone could be induced to part with holdings of money on any reasonable terms.”
29. June 2009 at 08:24
Kevin Donaghue,
I see you understand that the interest rate takes some accounts of expectations, and think that Keynes thought likewise. Fair enough. What then is your interpretation of the end of Chapter 11 in GT? I don’t understand your previous posts. Why do you say “The higher the anticipated rate of inflation the higher the marginal efficiency of the existing stock of physical capital.”?
Scott,
What book accurately describes what Fisher wrote? I haven’t got any of Fisher’s books. Is there a description online or somewhere accessible that is accurate?
I think you are a little too kind to Keynes. You and I interpret Keynes as working from the premise of expecting a stable price level. He should though have been aware of the fact that this need not be the case.
The second edition of “The Theory of Money and Credit” was published in 1924 in it Mises wrote: “Lenders and borrowers are not in the habit of allowing for possible future fluctuations in the objective exchange value of money.”
Which backs up the view that economists of the time were not accustomed to thinking in this way. He then grumbles for a few pages about how the public don’t understand this. However, he then writes:
“So far as variations in the objective exchange value of money are foreseen, they influence the terms of credit transactions. If a future fall in the purchasing power of the monetary unit has to be reckoned with, lenders must be prepared for the fact that the sum of money which a debtor repays at the conclusion of the transaction will have a smaller purchasing power than the sum originally lent. Lenders, in fact, would do better not to lend at all, but to buy other goods with their money. The contrary is true for debtors. If they buy commodities with the money they have borrowed and sell them again after a time, they will retain a surplus over and above the sum that they have to pay back. The credit transaction results in a gain for them. Consequently it is not difficult to understand that, so long as continued depreciation is to be reckoned with, those who lend money demand higher rates of interest and those who borrow money are willing to pay the higher rates. If, on the other hand, it is expected that the value of money will increase, then the rate of interest will be lower than it would otherwise have been.”
So, Mises at least understood that in a “continued depreciation” interest rates would rise. I expect other economists of the time did likewise. So Keynes seems behind the curve here.
29. June 2009 at 10:35
Kevin, I think you’re right. Any time the market real interest rate exceeds the natural real rate, meaning desired saving exceeds investment demand or we’re in a recession, then an increase in expected inflation causes the real rate to fall until it reaches its natural level. Once the real interest rate reaches its natural level then nominal rates start to rise.
The Fisher equation still holds but it is the real rate that does the initial adjusting whenever we are at less than full employment.
29. June 2009 at 11:22
Kevin, and more to the point, I agree that’s what Keynes was saying too.
29. June 2009 at 20:07
‘The letter from Krugman is very revealing.’
I thought you’d find it so. And, I suspect you thought I provided it for that very reason.
29. June 2009 at 23:24
I was joking about Keynes’ father – an economist best known today for his book on “methodology” of economics.
30. June 2009 at 01:50
Scott, David,
I’ve been re-reading chapter 11 of GT and I think it’s quite clear that what Keynes is saying is that the Fisher equation still holds but sometimes it is the real rate that adjusts to the change in expected inflation.
Keynes writes: “Although he does not call it the ‘marginal efficiency of capital’, Professor Irving Fisher has given in his Theory of Interest (1930) a definition of what he calls ‘the rate of return over cost’ which is identical with my definition.”
He clearly means that marginal efficiency of capital = real return to investment = real rate prevailing in the market.
He then writes “The mistake lies in supposing that it is the rate of interest on which prospective changes in the value of money will directly react, instead of the marginal efficiency of a given stock of capital. ”
Thus, he is saying that it is the real rate that adjusts to make the Fisher equation hold whenever the market real rate is above its natural level, that is whenever desired saving exceeds investment demand.
30. June 2009 at 04:00
Kevin,
You said:
“Here’s what he said, so you tell me: “my contention is that the expectation of a change in price has no effect on the rate of interest, since it leaves unchanged the relative attractions of cash and loans.” Far from being absurd that’s elementary optimization, as far as it goes. (It doesn’t go far enough but then it is just one sentence.) Suppose I hold $200 in cash earning no interest and $1,000 in a savings account earning 3%. Based on news that Ben Bernanke has ordered a fleet of helicopters, I revise my expectation of inflation from 1% to 5%. Is there some reason why I should now wish to hold more in cash and less in my savings account? Or vice-versa? Common sense suggests I should reduce both my $200 cash and my $1,000 interest-earning deposit and switch my wealth into real assets. But as Keynes points out, if my concerns about inflation are shared by other agents, the prices of real assets will already have risen.”
Change the second line to “expectation of a 1,000,000% change in prices in Weimar Germany” and I think you will see how absurd his statement is. And everything that happened in Germany happens to a lesser extent when expected inflation goes from 2% to 10%. Nominal interest rates rise. That is one of the most well-established propositions in monetary economics. And the statement “the price of real assets would already have risen” totally misses the point. As you point out, many real assets, not just gold, may hold some attraction as inflation hedges. So if you go from 2% to 10% expected inflation, the RELATIVE price of inflation hedges like land might rise relative to non-hedges. And that might occur immediately, as you say. But that is merely a relative price change, and even after it occurs the nominal price of land will be expected to rise at another 10% a year, just like non-inflation hedges. Were this not to occur the original jump in the relative price of inflation hedges would go away. So it doesn’t change the logic of the Fisher effect at all.
You said:
“Incidentally, you really don’t need to go digging out the Keynes-Robertson correspondence to show that Keynes was in earnest about his liquidity-preference theory of the interest rate. In GT Ch 13 Section II he states baldly that the rate of interest is simply “the ‘price’ which equilibrates the desire to hold wealth in the form of cash with the available quantity of cash….” If you think that’s absurd you’re not the first by any means.”
That is precisely the problem that I have been harping on in my blog for months. Too many people simple assume that expansionary monetary policies reduce interest rates–as they simply focus on liquidity preference. But monetary policy they raises inflation expectations, this will raise nominal interest rates. Under high inflation the nominal amount of money rises, but the real amount falls–that’s how the liquidity preference equilibrium is maintained. Keynes actually denied that the rise in T-bond yields in November 1933 could possibly be due to FDR’s avowedly inflationary gold policy, on the grounds that it was an expansionary monetary policy, and we know that sort of policy lowers rates! Here is the quotation:
“If you are held back [i.e. reluctant to buy bonds], I cannot but suspect that this may be partly due to the thought of so many people in New York being influenced, as it seems to me, by sheer intellectual error. The opinion seems to prevail that inflation is in its essential nature injurious to fixed income securities. If this means an extreme inflation such as is not at all likely is more advantageous to equities than to fixed charges, that is of course true. But people seem to me to overlook the fundamental point that attempts to bring about recovery through monetary or quasi-monetary methods operate solely or almost solely through the rate of interest and they do the trick, if they do it at all, by bringing the rate of interest down.” (J.M. Keynes, Vol. 21, pp. 319-20, March 1, 1934.)
Modern new Keynesians would say it should be the “real” interest rate in the last sentence. I don’t even agree with that. But virtually all modern economists agree that Keynes left out the Fisher effect here. (Here Keynes does concede that hyperinflation would hurt bond holders, but he doesn’t indicate whether it would raise rates, or just reduce the real value of bonds.)
Keynes’ error does real harm. Even today many people assume Fed policy has been “easy” just because rates are low.
30. June 2009 at 04:27
Kevin and Adam, I read Adam’s post and I’d like to amend the remarks made to Kevin. I don’t think Keynes was stupid, he was certainly capable of understanding the Fisher effect. Rather, I think Keynes’ thinking was too much a prisoner of his times. Under the gold standard the price level was almost a random walk, so there really wasn’t that much of a Fisher effect in Keynes’ lifetime, except in unusual cases like Germany 1923. My problem with Keynes as an economist is that he wasn’t very good at abstract macroeconomics, he liked to think in terms or real world examples, and that made him miss some things. A couple examples:
1. There may well have been expansionary monetary policies that increased inflation and yet didn’t budge nominal interest rates, for some of the reason’s Adam said. Thus because of the gold standard, an actual 5% inflation might only raise inflation expectations one percent. If this caused real growth (which it obviously might), then nominal rates might not rise at all–as risk spreads would fall on many loans. I am pretty sure that Keynes observed things like that in his lifetime. But he also said monetary policy only works by reducing interest rates. So he might look at that picture and say to himself “see we are in a liquidity trap, and monetary policy is mostly ineffective.” I keep claiming that monetary policy was highly effective in 1933, but he would probably disagree because rates didn’t change very much. So his error led to bad judgment regarding policy counterfactuals.
2. Recall that in the GT he worried about economies stagnating for long periods as investment opportunities dried up, and rates fell to low levels where monetary policy was no longer stimulative. Thus he proposed the government take a large and long term role in investment projects. If he had more of an imagination for theoretical concepts, he might have proposed a fiat money system that targets a positive rate of inflation. That prevents this secular stagnation for occurring, even if the Wicksellian real rate falls to low levels. Meltzer pointed this out. In all of Keynes’s writing I have never found even a single word contemplating the central bank targeting a long term secular rate of inflation that is positive (but not very high–say 3%). If anyone knows of one, I’d be very interested. BTW, he did of course favor temporary reflation in the Great Depression.
In contrast Fisher did come up with a foolproof way out of liquidity traps—his compensated dollar plan. I think Fisher had a more powerful imagination for abstract concepts. Keynes was of course much wiser about the real world than Fisher.
I am running late today, and will catch up on the other comments later in the day.
30. June 2009 at 06:47
It is ironic then that Keynes, in his 1920 book The Economic Consequences of the Peace,essentially described Wilson’s presence in Paris as that of a moron. All of the intellectual opinion about Wilson was exploded at the conference, but Keynes of course only cut off Wislon from his ‘A’ list, and never examined why his own culture was so gulllible in the first place.
Keynes was not promarily an economist, which is not a bad state for an economist to be in. That would describe Adam Smith. But his economics were indeed bad. His oratorical skills were good.
30. June 2009 at 09:50
Adam,
I think we’re broadly in agreement but I’m not sure. I can’t really make sense of Keynes using a one-good model, which is what I take you to be using. It’s worth noting that Hicks and Meade, who were among the first to present simplified models of the GT, both went for a two-sector setup. I follow their lead (see below). It was only when Hansen entered the fray that IS-LM became a one-good model.
Current,
Sorry for the late response. I’m not sure what you’re asking me. Maybe Keynes doesn’t make sense to you, so let’s make sure we’re on the same wavelength as regards his MEC concept. Apologies in advance if you already know this stuff. Keynes defines the marginal efficiency of an asset as an internal rate of return: put the replacement cost on one side of an equation and the NPV of the prospective yields on the other and solve the equation for the the IRR. I quite like this way of thinking about capital assets but it is certainly open to criticism.
Let’s put a simple model together. Suppose that the economy consists of two sectors: a consumption-goods sector, which doesn’t interest us at present, and a capital-goods sector, where the action is. The capital-goods sector produces indestructible machines which will earn rents (or quasi-rents as Keynes would have called them) forever. Suppose this “series of annuities” referred to by Keynes at the start of Chapter 11 is Q next period, (1+g)Q in the period after that and they grow at a constant rate g after that in perpetuity, because g is the rate of inflation – the machines churn out the same product in every period but the price goes up. Suppose the nominal interest is r, where r>g since otherwise the value of a machine turns out to be infinite. Then the price of a machine is given by the Gordon model equation:
P = Q/(r-g)
To calculate that price we have to know r. Suppose we don’t what r is. (Maybe we’re the central bank and it’s up to us to set the interest rate.) If we do know the replacement cost, C, of a machine, then we can solve for r in terms of C, which will force the market price P into line with the replacement cost:
r = g+Q/C
r is the MEC. It’s an interest rate of sorts, but not a market interest rate. It’s given by the capital-goods supply curve (which tells us what C will be for any given level of investment) and the expected cash flows Q, Q(1+g), etc.
Like I say, pretend you’re the Governor of the Fed. Inflationary expectations jump (g goes up). What do you do to r? Clearly the answer is: it depends. If, like Paul Volcker in 1980, your aim is to squeeze those inflationary expectations right back down, then you must raise r enough to hurt the poor bastards who have bought too many machines. The increase in r must be greater than the increase in g. But if, like Keynes, you want to end a slump you try to keep the interest rate right where it is and let the price of machines shoot up.
If you fool around with this model for a while I think you’ll see what Keynes is getting at. Bear in mind however that Keynes was assuming sticky wages in 1936, which made sense at the time, while Volcker (equally sensibly) was not assuming any such thing in 1980.
30. June 2009 at 10:20
Adam P, Possibly, but Keynes typically asociated liquidity traps with deflationary environments. I understand the theory behind what you say, but I don’t think that would account for all the various quotations I have provided in this comment section. I still think he was a bit confused about how to integrate expected inflation into macro.
Himaginary, 2 points:
1. I don’t know if Krugman would agree with you about hyperinflation. I thought that Krugman argued the two cases are not symmetrical, as there is no upper bound on nominal rates. Doesn’t the zero bound on nominal rates drive the overshooting argumetn you provided? This isn’t my area, so i may be wrong here.
2. The quote you provide comes right before Keynes talks about the failed 1932 Fed OMPs. These failed because of the outflow of gold. Keynes himself may not have understood this point, but it is exactly what you say, not a failur eof monetary policy in a technical sense, buut a political failure, a failure to discard the constraints of gold. WEhen FDR did discard those constraints in 1933, monetary policy instantly became more effective. My dispute with Krugman is not so much technical as political. I think it should be possible right now for the Fed to commit to at least 2% inflation of 5% NGDP growth over the next year. He doesn’t think so.
Current, I may be too kind to Keynes, but in fairness he didn’t assume a stable price level, he assumed a near-zero expected rate of inflation, which is very reasonable under a gold standard, or even under currencies losslely linked to gold, as was pretty much the case all the way up to 1968 (excluding wars.) I don’t think the GT is a fixed price model, or a depression model, I think it is a zero expected inflation model.
I’ve read The Purchasing Power of Money, which is his classic on monetary economics, and also The Money Illusion, and the History of Price Stablization movements. Some key articles ar ehis two “Phillips Curve” articles from the 1920s, one was reprinted in the Journal of Political Economy a few decades ago. The reprint was something like “I Discovered the Phillips curve.”
You might be surprised given what I said about Keynes, but Mises is right. Lenders and borrowers weren’t in the habit of allowing for price level changes, as the price level in 1914 was similar to 1879. Yes, there were ups and downs in price levels, but they were pretty unpredictable. The Fisher effect only became a big deal when we went off gold in 1968, although it also occurred to a lesser extent in other periods, obviously including hyperinflation.
And I agree that Mises got the abstract idea better than Keynes. But surely Keynes understood it at some level, I really don’t understand why he said all the silly things he did, but I’m not letting him off the hook.
Adam P#2, I disagree. In a recession monetary expansion raises the expected real growth rate. That faster expected growth shifts the IS curve. It’s not clear that the Wicksellian equilibrium real rate falls at all (although it is possible.) I agree with you that Keynes may have assumed that, as he certainly didn’t deal in M-policy shifting the IS curve.
Patrick, Yes, I’ll leave it at that. I just read a whole bunch of Brad DeLong today, and I try to stay away form the name calling (although that may sound crazy after my sarcasm about Keynes–but he’s dead and I was half-joking.) I know a lot of bloggers do that to be more readable, and that’s their choice. Again, I’m not saying I’m perfect, but I do try to avoid insults. Usually I say arguments are moronic, not people are moronic. But it can be a thin line.
Greg, Yes I remember that now. At first your joke went right over my head.
BTW, Slighhtly off topic, but I think it is possible that Keynes was an above average investor. My general view of life however is that people believe too many things that aren’t so, and I try to puncture those beliefs.
Adam P. I am confused. Isn’t the Fisher EQUATION a tautology, or a definition? I thought all the debate was over the Fisher EFFECT? The idea that a rise in expected inflation will cause nominal rates to rise. Did I misunderstand what you are driving at? (I’m doing these replies fast, so I may have.)
James, That’s a very astute observation. I am embarrassed to admit that I haven’t even read TECOTP yet. (But of course I have read many, many, excerpts, so I know how well written it is.
30. June 2009 at 11:55
Scott,
I was about to reply to some of your comments about Keynes but then I saw you had more-or-less withdrawn the ones I most objected to. I did respond to Adam and Current but my comment is stuck in your moderation queue.
Meanwhile here is what I love about blogs. John Quiggin criticises Lee Ohanian in a blog post and the man himself shows up to defend himself. It’s a great shame we can’t get Keynes and Pigou in on the act.
1. July 2009 at 03:00
Scott,
I’ll clarify. I think Keynes is saying that the Fisher effect does not happen in recessions and thus the real rate does the adjusting to make the Fisher equation hold.
The argument would be that in a recession we have an excess of desired savings over desired investment, thus a decrease in the real rate does not initially cause nominal rates to be bid up because there is an excess supply of loanable funds. The lower real rate causes saving to fall and investement to rise but only once they are equated does there begin to be a constraint on loanable funds that would bid up nominal rates.
Now, can this situation of excess supply of lonable funds happen? Yes, in a liquidity trap. If the nominal rate is greater than zero then the Fisher effect should always hold. But Keynes might have been thinking of a liquidity trapped economy.
Keep in mind that I’m trying to figure out what Keynes thought. However, this and my suggestion of credit spreads are not exclusive. Explining a liquidity trap as a negative natural real rate combined with a real risk premium on investment is sensible (the risk premium explains how investemt demand is depressed even though capital should always have a marginal product greater than zero). This does seem to me to be what Keynes was saying and it is correct, he just should have more clearly explained that it only applies to liquidity trapped economies.
1. July 2009 at 05:03
Kevin#1, Why don’t you think Volcker assumed sticky wages? Remember that Volcker actually wanted a much more gradual slowdown in inflation than what he got. I think he did believe in sticky wages, how else can we explain unemployment rising to almost 11% when inflation fell?
Kevin#2, Thanks for the tip about Ohanian. Did you know I am currently involved in a 3 or 4 round debate with him over at CBSMoneyWatch.com. When completed, I may discuss the debate in a post.
My posts reflect the different moods I am in when I write them. So one day I might write something really negative about Keynes, on other days something positive. Unlike you I don’t find the theoretical apparatus in the GT useful, but I can see his excellence as an economist in other works he did, such as the Tract. And he obviously communicates very well, except in parts of the GT, where the meaning is sometimes not clearly spelled out.
Adam and Kevin,
I think Adam may be on the right track, but I doubt Keynes would agree that his views on the Fisher effect were limited to liquidity traps (which he thought were rare.)
Let’s see if all three of us can agree on the following propositions (which might be worth a journal article.) I’ll start with my contribution, and then add a more favorable take that somewhat reflects your views (as i interpret them.)
1. Keynes either did not understand or did not consider changes in the secular rate of inflation. His views on the Fisher effect are not accurate, if applied to scenarios where central banks targeted inflation at 2% for a few decades, but then 5% for the next few decades. Nominal rates would rise about 3% in that case, just as Fisher indicated.
2. We can give Keynes a bit of a pass, because during his lifetime the expected rate of inflation over any significant period of time was usually near zero. Exceptions like German 1920-23 were viewed as pathological cases and were sometimes exempted from his generalizations. So he didn’t observe many real world examples of actual Fisher effects, and thus it is not surprising he was skeptical.
3. Contrary to what his critics sometimes assert, he did not assume fixed prices. He knew that prices were somewhat procyclical. But the secular trend was near zero. Suppose we were in a recession and the central bank expanded the money supply a bit to boost the economy. Also suppose it was expected to lead to modertately higher prices for a year, after which things would level off. In that case asset prices might rise right away. Indeed all prices would rise right away if not for price stickiness. And here is my key point, price stickiness is not enough to generate a Fisher effect. So if asset prices rise right away, and sticky prices (of goods not easily arbitraged) rise gradually over a year, then in terms of broad indices like the CPI there will be a bit of expected inflation, but rates won’t rise, and indeed may fall due to the liquidity effect. That seems to me to be the best way of defending Keynes.
To summarize, a true Fisher effect requires asset prices to be expected to rise at a certain rate over time. (in my previous example asset prices would first rise at 2%, then 5%.) If we are on the gold standard, and long term inflation expectations are near zero, then any demand shock will be immediately and fully reflected in asset prices. There will still be a bit of expected inflation, but only because sticky prices have yet to adjust.
This distinction roughly corresponds to the distinction between income and profit inflation in the Treatise. But not exactly. In the Treatise income inflation was a lagged reaction to profit inflation (if my memory is right.) I am talking about both occurring together in my fiat secular inflation example. In my second example (gold standard) profit inflation occurs first, causing assets to rise, and then income inflation catches up, but then income inflation ends when prices have adjusted.
1. July 2009 at 08:34
Scott,
OK, when you put it like this I agree to some extent with what you’re saying about Keynes. I see what you mean about the difference between expectations of a stable price level and actuality. I agree with you too that Mises was correct at the time that debtors and creditors did not consider inflation. (My own family suffered and benefited from this when the changeover came.)
However, Keynes wrote what he claimed to be a _general_ theory. He was aware of what happened in Germany in the 20s, he wrote about it. A truly general theory would have dealt with this. (The same criticism applies to Mises later book Human Action which claims to be very general but is quite specific to the time in places).
Kevin,
Thanks for explaining what Keynes means by M.E.C. I’m much more clear about that now.
But, I’ve read through this thread and through GT Chap 11 part III, and I don’t think that you are really right.
Firstly, which situation are we talking about here, one where the Fed set the rate or it is set in the market? Which do you think Keynes is talking about? (I think Keynes is talking about the market situation).
Since I subscribe to the Austrian view I think that a lot of the assumptions used to get to Chap.11 are not really reasonable. But the concern in this thread is doing mainstream economic theory right.
That Keynes subscribed to his liquidity preference theory of interest doesn’t help him here. The influence of inflation on the interest rate is just another reason why that theory is wrong.
The Capital equipment price P in your example above may adjust to expectations in it’s future profitability (though that’s unlikely to happen in a very short space of time). That doesn’t mean that the interest rate doesn’t also change.
So, I don’t really see what you’re getting at.
1. July 2009 at 23:10
Kevin,
I’m not using a one good model, but in any event I think we’re saying close to the same thing. We’re just using a slightly different language.
The fact is that Keynes quite clearly thinks the Fisher effect doesn’t always happen. The question then is, why not?
I’ve suggested two things that are coherent and might reasonably be what he thought. Recall that the challenge from Scott was to explain it without assuming that Keynes was just mistaken.
A central bank targeting the nominal rate is also a way that this could happen, the increased investment demand is accomadated by an increase in the money supply so the nominal rate needn’t change. In this case the Fisher effect is happening though, it’s just being offset by the increased money supply. I just didn’t get the impression from reading the chapter that this is what Keynes actually thought.
2. July 2009 at 05:00
This is one of those cases where I think the IS-LM approach represents Keynes faithfully enough, so let’s go that route, starting from the equation I wrote upthread for the price of capital goods, P, where r is the nominal interest rate and g is the expected rate of inflation for the quasi-rents:
P = Q/(r-g)
Suppose the supply curve for capital is just a straight line through the origin so that:
Investment = kP = kQ/(r-g)
I’m thinking of this as being in money terms, but I suppose Keynes might have preferred to divide by the wage rate and think in wage-units. It doesn’t much matter for now. Let’s be equally simplistic about the consumption function:
Consumption = a + cY.
Then we can derive an IS curve:
(1-c)Y = a + kQ/(r-g)
If the IS curve is drawn in (r,Y) space a rise in g shifts the curve vertically. The new equilibrium depends on the LM curve, which Hicks drew with a flat portion on the left (his liquidity trap region) getting steeper as Y increases and vertical at full employment. Clearly, on the vertical section the Fisher effect is king. Since Y cannot change the only way to keep the equation true is to raise r by precisely the same amount as g. The same holds true at lower levels of employment if the central bank is very conservative and refuses to allow an increase in Y. That too gives you a vertical LM curve.
But where there is unemployment and a humane central bank, the increase in demand resulting from the rise in g will be split between r and Y, raising both to some extent. A weak Fisher effect will be observed – a positive correlation between g and r, but that’s all.
2. July 2009 at 05:13
so then we are saying exactly the same thing Kevin. Good that we agree.
The only difference is in our terminology with respect to the intermediate case where the central bank must increase the money supply in order to keep r from rising. You call that a weak Fisher effect, I call it a full Fisher effect combined with a monetary expansion. But we’re both actually saying exactly the same thing.
2. July 2009 at 05:17
Let me amend that slightly. If the central bank is so conservative that it refuses to allow an increase in Y despite the fact that there is unemployment, it must actually contract the money supply. That’s not quite the same thing as a vertical LM curve, but it produces the same result.
2. July 2009 at 05:20
Adam, sorry I was just posting my corrrection when you posted. So do we really agree or don’t we? I’ll think about it and revert later.
2. July 2009 at 05:57
Kevin, Adam,
IS-LM models I know about have simple expectations. People expect the trend from one Hicksian week to the next to be steady.
How does this dig Keynes out of any holes he dug for himself?
2. July 2009 at 06:08
Current,
Kevin’s argument has nothing to do with expectations formation. We are arguing the effect of a change in expecations, we do not need to explain why they changed.
Kevin,
the reason I think that in the intermediate case the bank needs to expand the money supply in order to prevent nominal rates from rising is becuase otherwise what kept the nominal rate above zero? If there are lenders who are willing to lend at lower than the market rate then the nominal rate should already have been bid down.
2. July 2009 at 06:10
Current, I actually agree with your view 100%. Sometimes I take a more pro-Keynes side and sometimes anti-Keynes, because I want people to see the other side. But you are exactly right, the GT is not a “general” theory.
Adam and Kevin, No comment on my hypothesis in the last comment?
Regarding your model, Is there any evidence Keynes thought we could be on the flat portion of the LM during a period where prices are expected to rise? I don’t recall any.
2. July 2009 at 06:27
Scott,
Here’s Keynes: “If the rate of interest were to rise pari passu with the marginal efficiency of capital, there would be no stimulating effect from the expectation of rising prices. For the stimulus to output depends on the marginal efficiency of a given stock of capital rising relatively to the rate of interest.”
Seems as though he does consider inflation and understand the implications of the Fisher effect but denies its empirical validity. So I don’t agree with your first statement.
On your second I basically agree, he doesn’t believe in the Fisher effect because with long term inflation expectations anchored by a gold standard he would not have clearly observed it for most of his lifetime.
On your third I can’t say too much because chapter 11 of GT is the only Keynes I’ve yet read so I can’t speak about what he said more generally.
2. July 2009 at 08:11
If I sound imprudent then I apologize, but arguing GT without mentioning chapter 17 seems beating around the bush to me.
Here is my interpretation of Keynes on the Fisher equation, using notation he defined in chapter 17.
interest on real goods : q1-c1
interest on money : q3+l3
expected inflation rate : a1
q1=MEC, c1=depreciation,
q3=interest rate, l3=liquidity premium
(It is assumed that goods has no liquidity premium, and money does not depreciate)
In equilibrium:
q1-c1+a1 = q3+l3
(attractions of goods and money become equal)
If we define interest on real goods as real rate, and interest on money as nominal rate, then this equation IS the Fisher equation.
However, what we can observe at the market is q3 only, i.e., l3 cannot be observed. Even the central bank lower q3 to zero, l3 remains positive, that is, money still attracts. And the problem is, l3 cannot be diminished no matter how much money is supplied. So if goods is not attractive enough, that is, if q1-c1 is less than l3, we need certain amount of a1, or inflation expectation, to fill the gap.
I think Keynes treated a1 as endogenous here. So I agree Scott in that perhaps he never thought about inflation targeting. But I don’t think he assumed a1 as some fixed number near zero.
2. July 2009 at 08:20
Adam P: “Kevin’s argument has nothing to do with expectations formation. We are arguing the effect of a change in expecations, we do not need to explain why they changed.”
To be honest, I’m not really what clear what Kevin’s argument is.
In an IS/LM model you can’t just say “and then there is a shift in expectations”. The model is built upon a specific concept of expectations. The LM curve is built on the liquidity preference theory of interest.
2. July 2009 at 08:40
Current,
I don’t want to put words in Kevin’s mouth so I’ll let him defend himself. What I had understood Kevin to mean is that inflation expectations were fixed exogenously in the initial drawing of the IS and LM curves and then they were exogenously shifted. After all, g in his model was expected inflation. He built the model for a given value of g and then said what would happen if g was raised.
However, it is not true that you need “simple” expectations for the IS-LM model. You can build that model on rational expectations, or adaptive if you prefer. Any specificaton of expectations will let you build the model.
2. July 2009 at 09:11
Adam: “the reason I think that in the intermediate case the bank needs to expand the money supply in order to prevent nominal rates from rising is becuase otherwise what kept the nominal rate above zero? If there are lenders who are willing to lend at lower than the market rate then the nominal rate should already have been bid down.”
Bear in mind that my IS-LM diagram has the nominal interest rate, r, on the vertical axis. Are you saying that the initial inflation shock must shift the LM curve? Obviously an actual increase in prices will shift it since it changes the real money supply. But does it shift in response to the expectation of higher future prices? I can see how you could argue for that and I’m not arguing against it, I just want to be clear as to what your question is.
Hick’s LM curve equation (in my notation not his) is simply: L(Y,r) = M. He doesn’t even bring the price level into it. Later writers made it L(Y,r) = M/P.
If Keynes had written his own specification for the LM side of the model, what would it have looked like? Judging by Chapter 15, it would have consisted of transactions demand for money h(Y), with dh/dY>0, and speculative demand q(r) with dq/dr<0. Both functions give the quantity of money demanded in wage-units. That’s because output prices are endogenous in Keynes’s model, so dividing by the wage-rate is the handiest way to work in real terms. For better or worse, that’s what Keynes usually does in the GT when he wants to work with real variables. So the LM equation is:
h(Y) + q(r) = M/W
Obviously the (r,Y) locus has a positive slope, though I’m not sure exactly what form the functions h(Y) and q(r) have to take if they are to give the sort of LM curve Hicks presented. But I guess they could be engineered to fit. My main concern is whether we have all the relevant variables included.
When I referred upthread to this discussion being a case where I think the IS-LM approach represents Keynes faithfully enough, I was mostly thinking about the IS curve. There’s a lot more room for argument about the LM curve. I hate to tell you this but Chapter 11 is one of the least controversial in the book!
Scott: “No comment on my hypothesis in the last comment?”
I think Adam has raised a sufficiently difficult question to keep me busy for a while.
Current: “IS-LM models I know about have simple expectations.”
You haven’t lived. John Geanakoplos has some papers on his web page that could make your head explode. The site seems to be down at the moment so no point linking, but check it out some time.
2. July 2009 at 12:34
I have just been reading up on the literature on Keynes and the Fisher Effect, and I think that the best piece out there (despite being 15 years old) is by Allin Cottrell in the Scottish Journal of Political Economy (1994) 41:(416-33). Cottrell is himself a post-keynesian but he provides a very careful review of the different arguments made on all sides. I take issue with some of what he says, but his discussion is thorough and insightful.
2. July 2009 at 21:31
Kevin,
yes, that’s basically what I’m saying. If you have nominal rates on the vertical axis then both curves shift when inflation expectations change.
My thinking is that, if the nominal rate is greater than zero then the money market has cleared so supply and demand are equated at the current nominal rate. Now along comes the higher g to increase investment demand , thus credit demand, at that nominal rate. So where does the extra supply come from at that (original) rate?
On reflection now I guess this doesn’t really contradict what you’re saying, you’re saying the nominal rate rises but not one-for-one. I guess that could well be true, it’s a question of elasticities or something.
2. July 2009 at 21:43
Kevin, just to explain myself a bit better.
I tend to think the LM curve should also be drawn with respect to the real rate. This view comes from the idea that people hold the minimum amount of money needed to fund expenditures and no more, thus higher Y needs higher r because people loan out less money/borrow more money because they need it to fund the higher expendiutre.
It also means that a higher nominal rate, with unchanged real rate, does NOT generate a movement along the LM curve. Why doesn’t a higher inflation rate make people hold less money? Because they can’t hold less without reducing consumption, they already hold the minimum real balance necessary to fund real expenditures.
3. July 2009 at 01:26
Gents, I still don’t understand what exactly you folks are arguing.
Earlier you were defending Keynes in Chap 11 part III of General Theory. I still don’t understand why or what all this stuff about IS/LM has got to do with it.
3. July 2009 at 02:35
Current, earlier we were only defending Keynes against this statement of Scott’s: “In Chapter 12 of the GT (p. 142) he completely fails to understand how the Fisher effect can raise nominal interest rates.”
I think we’ve accomplished that, we are now finished defending Keynes and have moved on to trying to understand
1) what exactly Keynes was in fact saying;
2) whether or not Keynes was theoretically correct;
3) what actually is theoretically correct in general (regardless what Keynes was thinking).
3. July 2009 at 02:41
Adam P,
In what posts have you succeeded in defending Keynes?
3. July 2009 at 03:03
Current,
http://blogsandwikis.bentley.edu/themoneyillusion/?p=1652#comment-4315
http://blogsandwikis.bentley.edu/themoneyillusion/?p=1652#comment-4234
Keynes writes: “Although he does not call it the ‘marginal efficiency of capital’, Professor Irving Fisher has given in his Theory of Interest (1930) a definition of what he calls ‘the rate of return over cost’ which is identical with my definition.”
Keynes clearly means that marginal efficiency of capital = real return to investment = real rate prevailing in the market.
Keynes then writes “The mistake lies in supposing that it is the rate of interest on which prospective changes in the value of money will directly react, instead of the marginal efficiency of a given stock of capital. “
Keynes doesn’t think the nominal rate does all the adjusting to an increase in expected inflation. He thinks the nominal rate only partially adjusts and so the result is a lower real rate.
Here’s Keynes again: “If the rate of interest were to rise pari passu with the marginal efficiency of capital, there would be no stimulating effect from the expectation of rising prices. For the stimulus to output depends on the marginal efficiency of a given stock of capital rising relatively to the rate of interest.”
Clearly he does understand the implications of the Fisher effect but denies its empirical validity.
Thus, this statement of Scott’s is not accurate: “In Chapter 12 of the GT (p. 142) he completely fails to understand how the Fisher effect can raise nominal interest rates.” (Scott means chapter 11 of course).
3. July 2009 at 04:29
Adam,
Thanks. Obviously you treat the transactions motive for holding money as the important one. For Keynes the speculative motive is crucial and that depends on the nominal rate. Unfortunately it also depends on whatever variables the speculators happen to be influenced by at the time! He doesn’t provide the makings of a neat model – his animal spirits are too strong for that. But his reasoning does imply that the Fisher equation will hold in periods when income doesn’t vary much relative to potential. When the economy is shrinking or expanding the movement in the nominal interest rate will only compensate partially for the movement in expected inflation.
Ye olde IS-LM is at least specific, provided we take the trouble to specify it. AFAIK the most common textbook version has the real interest rate in the investment function and the nominal interest rate in the demand for money function. (I can’t say I surveyed many textbooks to reach this conclusion.) Accepting that for present purposes, the shock to expected inflation lifts the IS curve and leaves the LM curve where it is. Then as you say it all comes down to elasticities. In general 0<dr/dg<1. The more interest-elastic the investment function (flat IS) and the less interest-elastic the demand for money (steep LM) the stronger the Fisher effect. I think Keynes would have accepted that as a reasonable formalisation of his model, not that he encouraged anyone to formalise it.
Scott,
Unlike Adam, I don’t think we’ve settled anything because I don’t think you will accept that your charge is groundless. But there’s no more I can do. As to your three propositions:
1. False. Keynes wrote extensively about inflation and corresponded with Fisher. He can’t have been ignorant of such elementary points and you’ve no evidence that he was.
2. Keynes doesn’t need your pass, he observed the effects of inflation at first hand in WW1 and discussed them in How to pay for the war, which drew an admiring response from Hayek and many others. He also read quite a bit of history.
3. The GT does not assume fixed prices, that’s true, but it’s pointless to go looking for ingenious ways to defend Keynes since his view is reasonable and is widely accepted. Mundell and Tobin refined it slightly but they never denied that his analysis was basically sound.
himaginary,
I think Chapter 17 gives us some idea how Keynes would have approached the task of building a model with a wide choice of assets, but I don’t think it gives us any ammunition to use against Scott’s idea that Keynes didn’t understand the Fisher effect. But then I don’t think we need any. It’s up to Scott to make the case. His claims that Keynes had a limited imagination and couldn’t think outside the gold standard are just too easy to refute.
3. July 2009 at 05:57
Adam,
“Keynes clearly means that marginal efficiency of capital = real return to investment = real rate prevailing in the market.”
I agree on “Keynes clearly means that marginal efficiency of capital = real return to investment” part.
But as for “real return to investment = real rate prevailing in the market” part, I think Keynes regarded this equality as equilibrium condition, not identity.
The notion that interest rate on money is determined independently on its own, and does not necessarily correspond to interest rate determined by supply and demand of the economy, is the key idea of GT. Let me quote Krugman:
“One of the key insights in Keynes’s General Theory – actually, THE key insight – was that the loanable funds theory of the interest rate was incomplete. Loanable funds says that the interest rate is determined by the supply of and demand for saving; Keynes pointed out that the supply of saving is endogenous … To have a full (“general”) theory of interest, said Keynes, you need to add in liquidity preference – the demand for money …”
http://krugman.blogs.nytimes.com/2009/02/23/liquidity-preference-versus-loanable-funds-televised-wonkish-with-video/
Kevin,
I think that chapter 17 is the core part where he developed that key notion in the book, not sidenote about asset allocation. I strongly recommend everyone to read the whole chapter.
Yes, there is IS-LM model, and it treated liquidity preference. But it didn’t treat expected inflation explicitly, hence some contrivance is needed on user’s side. So why rely on this model to understand Keynes’ idea on the relationship between interest rate, expected inflation, and liquidity preference, whereas Keynes stated his idea in his own words much more simply?
3. July 2009 at 06:40
I’m going to go away and think about this. I’m not sure that any of us are really understanding each other.
Last week we had another interest rate related controversy. I’ve finally got around to posting a response to those who argued with me, if anyone is still interested:
http://blogsandwikis.bentley.edu/themoneyillusion/?p=1620#comment-4361
3. July 2009 at 07:09
himaginary,
I was using it as an equilibrium condition.
3. July 2009 at 10:13
I strongly recommend everyone to read the whole chapter [17 of the GT].
FWIW, I strongly recommend that everyone read the whole book and read chapters 1-16 before chapter 17. I agree that it’s a fascinating chapter. When I first read it I thought: this man didn’t just want to revolutionise macroeconomics, he wanted to revolutionise microeconomics as well. When you start looking at the implications of taking liquidity premia and storage costs into account, price theory gets a lot more complicated. But he was biting off more than he could chew and he knew it. In particular, he later regretted floating the idea that in certain circumstances land could be a relatively liquid asset. That idea isn’t ridiculous by any means. Possibly he was thinking of 19th century British aristocrats who preferred owning low-rent land to investing in illiquid industrial capital, simply because there would always be a ready market for land. But economists weren’t ready for the complications he was introducing and AFAICT they are not ready even now.
3. July 2009 at 12:09
Wow, I’m really behind. I’ll have to catch up here tomorrow.
4. July 2009 at 05:16
Kevin, it’s quite obvious that non-specific factors of production are more liquid assets than specific factors of production. Keynes did not originate that idea.
I’ll have a look at the other stuff folks are talking about.
4. July 2009 at 05:30
Adam, You said:
Seems as though he does consider inflation and understand the implications of the Fisher effect but denies its empirical validity. So I don’t agree with your first statement.
It seems to me that you are mixing up the Fisher equation and the Fisher effect. It seems to me that this merely shows that Keynes understood the Fisher equation. Again, The problem I have with Keynes is that he repeatedly denies that a higher expected rate of inflation would cause interest rates to rise in the first place. Thus he denies the Fisher effect exists. That is very clear in the quotations I provided. Even worse, he denies the existence of the Fisher effect for the wrong reason, he wrongly claims that if prices were expected to rise then asset prices would immediately rise to the future expected price level. But that is clearly wrong as higher nominal interest rates increase the carrying costs of real assets, and thus increase the expected rate of appreciation of real assets held for purposes of inflation hedging.
Himaginary, Excellent summary of Chapter 17, thank you. regarding your last sentence, I think we are both right. If you asked Keynes “Do you always assume zero inflation expectations?” he would have said no. But I still believe that the most useful way to think about the GT is to assume that Keynes implicitly assumed inflation expectations were always roughly zero. If you make that assumption then a whole lot of assertions in the GT that look crazy at first glance, suddenly make sense.
Another one I haven’t mentioned is that in the GT Keynes assumed that when the money supply increases, prices should generally increase by a smaller percentage. But that is only true for a one time increase in the money supply. If the growth rate of the money supply increases, inflation expectations rise, and prices rise faster than money. In the German hyperinflation prices rose 100X faster than money.
AdamP, Regarding your reply to current, Yes, you can build IS-LM with Ratex, but keep in mind that when you do an expansionary monetary policy will often raise both real and nominal interest rates, by shifting the IS curve, so the normal “Keynesian” assumptions about monetary policy no longer hold. And indeed we observe this empirically.
Kevin, I don’t want to get bogged down in IS-LM, but see my previous reply to Adam.
David, Thanks.
Adam P. You can’t be serious? I hope you are not arguing that inflation expectation don’t affect money demand. That is massively refuted by the data. What exactly are you arguing here? LM is a function of nominal, not real rates.
Adam P, You have in no way defended Keynes on the Fisher effect. I provided several other quotes that showed beyond the shadow of a doubt that Keynes denied the existence of a Fisher effect. All you have done is to show that Fisher understood the Fisher equation. Big deal, a 4th grader could understand the Fisher equation.
Again, He denies its empirical validity for the wrong reason, he denies that there ever could be an expected increase in asset prices because he says asset prices would immediately adjust to the new expected future price level. And that is false.
Kevin, Adam’s comment is wrong even if there is only a transactions demand for money, as Baumol showed decades ago.
You said:
“But his reasoning does imply that the Fisher equation will hold in periods when income doesn’t vary much relative to potential.”
I hope this is not true, as the Fisher equation is a tautology, despite what you and Adam keep saying. I have a higher opinion of Keynes than you and Adam do, as I believe he did assume the Fisher equation always held true.
You said:
“As to your three propositions:
1. False. Keynes wrote extensively about inflation and corresponded with Fisher. He can’t have been ignorant of such elementary points and you’ve no evidence that he was.”
You obviously didn’t read my comment, because I would never say something so foolish. Obviously he observed inflation, and made many comments about its causes. But none of these comments apply to secular inflation. For instance, he said inflation only occurred at full employment, when bottlenecks develop in the economy. That’s pretty clearly cyclical inflation under a gold standard, for which you won’t expect much of a Fisher effect.
BTW, A lot has happened since “Mundell and Tobin” and I can tell you that almost all the economics taught at elite universities today (new Keynesian included) assumes that Keynes was not “basically sound.” He completely ignored the effect of inflation expectations. Otherwise he would not have argued that expansionary monetary policy lowers interest rates, something that was overwhelming rejected by the data in the 1970s. There is absolutely nothing in the GT that could explain the 1970s.
You say I have never provided any evidence that Keynes didn’t understand the Fisher effect. My evidence is two-fold.
1. I have irrefutable evidence that Keynes denied the existence of the Fisher effect. Here it is plain as day:
“my contention is that the expectation of a change in price has no effect on the rate of interest, since it leaves unchanged the relative attractions of cash and loans.”
You and Adam have two choices. One is to claim there is no such thing as a Fisher effect. I don’t think you want to make that argument. Your second choice is to say that Keynes understood it, but simply made an empirical judgment that it was unlikely to occur in reality, because every time the expected inflation rate rose, some other factor would magically reduce the real interest rate by exactly the same amount at exactly the same point in time.
2. But even that won’t work, because when he tries to explain why there is no Fisher effect in the GT, he makes the bogus argument that there is no Fisher effect because if asset prices were expected to rise in the future, they would have already risen!
So my evidence is that Keynes said there is no Fisher effect, and he denied it’s existence for a bogus reason. Hence he didn’t understand it. Again, if he had understood it wouldn’t have worried about liquidity traps, rather he would have favored inflation targeting. Instead he favored a stable price level
4. July 2009 at 07:16
If I may muddy the waters of this already muddy thread some more….
I think part of the problem here is the difference between short-term rates and medium-term ones.
The Fed control the short-term rates in the majority of circumstances. However their power diminishes as successively longer term rates are considered. This is where the Fisher effect must come into play, which is what Scott is talking about.
If the interest rate for a year term is high then what the Fed can do with shorter term rates can’t change much for many practical debt arrangements. This has been demonstrated several times.
5. July 2009 at 06:58
Current, Good point, but unfortunately Keynes is rather vague on the question of long vs short term in these quotations. So I am not sure whether that explains things.
6. July 2009 at 01:03
Here is my point about the medium vs short run distinction.
What Kevin and Adam seem to be saying is that Keynes believed his liquidity preference theory of the interest rate. So he did not believe the Fisher effect to be empirically valid.
What Scott is pointing to is the sheer inevitability of the Fisher effect. In anything beyond the very short run it must play a role.
So, I think the mistake that Kevin and Adam are making is not making a distinction. Holding the liquidity preference theory of interest for the very short run in one thing. I wouldn’t agree with that, but it’s not silly. Failing to understand that the longer runs are different though is an error.
6. July 2009 at 02:54
Scott: …the Fisher equation is a tautology, despite what you and Adam keep saying.
If the Fisher equation were really a tautology it would be logically impossible for me to enter into a contract whereby I supply a basket of goods to you now, in return for your promise to supply me with an identical basket of goods one year hence, unless the money interest rate for one year happened to be the same as the rate of inflation which I anticipate for the items in the basket. It is clear that such a contract is not a logical impossibility. It’s not even impossible that a vast number of such deals should be booked (though of course it’s not likely). Hence the Fisher equation is not a tautology. It is derived from assumptions about rationality.
Current: What Kevin and Adam seem to be saying is that Keynes believed his liquidity preference theory of the interest rate. So he did not believe the Fisher effect to be empirically valid.
Keynesian theory is compatible with a strong Fisher effect in some circumstances but not all. That is to say, if there is a one-point rise in expected inflation there will usually be an increase of less than one point in the nominal interest rate. A strong (one for one) Fisher effect is more likely if the level of output is reasonably stable relative to its potential.
6. July 2009 at 03:30
All the Fisher equation does is relate the nominal interest rate, the real interest rate and the price inflation rate.
nominal interest rate ~= real interest rate + price inflation rate.
It is an accounting tautology. Kevin, all that it shows in the example you give is the real interest rate which is being used for the transaction.
6. July 2009 at 08:18
It’s the ex-ante real interest rate we’re talking about – or at least I am.
6. July 2009 at 08:40
That’s fair enough, but isn’t the Fisher equation is about the ex-post interest rate.
6. July 2009 at 10:18
No, not where economic theory is concerned. It’s nearly always expected inflation which appears in the equations of an economic model.
6. July 2009 at 11:41
On reflection, I retreat somewhat from the line I took in my last few comments. I think of the real interest rate as the reward, in future goods, which an agent can get for giving up current goods. Hence I think of Fisher’s equation as a result derived from assumptions about rationality etc. But having looked at some of Fisher’s writings that’s not how he defines it. So the equation, in his work at least, is a tautology. Likewise in most other writings I expect.
6. July 2009 at 15:24
Kevin and Current, And just to be clear, both the ex ante and ex post real rates are defined tautologically. One uses ex ante inflation expectations, and the other uses ex post inflation.
14. July 2009 at 13:27
Don’t trust Cherry Pickers.
Harding was either dead last or damned close to it in every included poll since 1948. Oh, and he rose all the way to 37th – of 39 – in the WSJ poll of 2000!
So, yeah. Where does one even start?
14. July 2009 at 14:49
[…] Kevin Feasel @ 7:47 pm – Why is Woodrow Wilson considered better than Warren G. Harding? Scott Sumner theorizes that historians put a lot of prestige in “power” and Wilson was the premier “power” President, to America’s great […]
30. March 2010 at 15:18
Godaddy.com coupon codes news. Many of the current Godaddy.com coupons are not valid. Here are official promo codes that are valid. These coupons will give you a discount in 2010. .COM Domain Names or Manual Renewals for just $7.49 – Use Godaddy promo codes GOO3, OK9, or ZINE10. 25% discount on orders of $100+ – Use Go Daddy coupon code OK25. 30% Discount when you buy any com domain – Use Godaddy coupon OK30. $12.99 SSL Certificates – Use Godaddy coupons GOOSSL, OKSSL, or ZINESSL. Host Plan Promo Code – 20% Hosting Discount – Use Godaddy coupon codes OK20H, ZINE20H1 or GOO20H. 10% off any size order – Use Go Daddy coupons OK7, GOO1 or ZINE8. $5 Off $30 or More – Use Go Daddy promo codes GOO2 or ZINE9. 20% Off Any order of $50 or more – Use Godaddy promo code OK8.
17. August 2010 at 15:29
[…] figures behind slavery and its continuation (it’s debatable who is truly focal here), Woodrow Wilson, the Rosenbergs, and any number of assassins, domestic terrorists, and serial […]
17. October 2011 at 18:17
[…] Cottrell wrote a very interesting paper on the subject many years ago. A couple of years ago this subject came up on Scott Sumner’s blog, and Kevin Donoghue was helpful to me in understanding what Keynes was […]
30. March 2020 at 21:13
[…] industrial production,” writes economist Scott Sumner, “had fallen by 32.5% by March 1921,” when conservative Republican Warren G. Harding became […]