David Glasner has an excellent new post on Keynes’s views on the Fisher effect. I’m going to come at this from a slightly different angle, which I hope will help illuminate Keynes’s thinking.  But I’ll end up in a similar place to David.

Here’s Keynes in the General Theory:

The expectation of a fall in the value of money stimulates investment, and hence employment generally, because it raises the schedule of the marginal efficiency of capital, i.e., the investment demand-schedule; and the expectation of a rise in the value of money is depressing, because it lowers the schedule of the marginal efficiency of capital. This is the truth which lies behind Professor Irving Fisher’s theory of what he originally called “Appreciation and Interest” – the distinction between the money rate of interest and the real rate of interest where the latter is equal to the former after correction for changes in the value of money. It is difficult to make sense of this theory as stated, because it is not clear whether the change in the value of money is or is not assumed to be foreseen. There is no escape from the dilemma that, if it is not foreseen, there will be no effect on current affairs; whilst, if it is foreseen, the prices of exiting goods will be forthwith so adjusted that the advantages of holding money and of holding goods are again equalized, and it will be too late for holders of money to gain or to suffer a change in the rate of interest which will offset the prospective change during the period of the loan in the value of the money lent. For the dilemma is not successfully escaped by Professor Pigou’s expedient of supposing that the prospective change in the value of money is foreseen by one set of people but not foreseen by another. (p. 142)

David talks about how this conflicts with Keynes’s more sophisticated treatment of the Fisher effect in the Tract on Monetary Reform.  In my view that’s because Keynes was (perhaps subconsciously) addressing different monetary regimes in the two books.  The Tract really should have been called “A Tract on Fiat Money Regimes,” as it was written in the wake of the big European hyperinflations and has a real monetarist slant.  The GT should be called “The General Theory of Business Cycles under a Gold Standard and/or Bretton Woods-Type Regime.”  I once published an entire paper arguing that the GT was essentially a gold standard book, as the expected rate of inflation was implicitly assumed to be near zero (which was true of the actual gold standard.)

Here’s the difference.   Under a fiat money regime you can have changes in the trend rate of inflation that are in some sense “equilibrium” cases.  Thus if trend inflation rises from 0% to 10%, then all nominal prices wages and assets prices eventually adjust to that new reality, and nominal interest rates also rise by roughly 10%.  Nothing real changes in the long run, except real cash balances.  In contrast, the trend rate of inflation never really changes (much) under a gold standard, where the price level follows a random walk with near-zero trend.  Thus actual price movements are mostly unexpected, and there’s not much of a Fisher effect.  But that doesn’t fully explain Keynes’s comments; we need to add one more factor.

Under the gold standard there were two very important groups of prices; flexible prices and sticky prices.  (Today sticky prices are more dominant.)  The flexible prices included the prices of commodities (then a far larger share of GDP), real estate, and stocks—which are claims to the ownership of real corporate assets.  A positive monetary shock would immediately raise the prices of the flexible price assets, and that is clearly what Keynes is thinking about in the quote above.  At the same time the CPI inflation will play out gradually over time, and thus there will be some expected inflation at the CPI level.  But in those days most economists focused on the WPI, which was dominated by flexible prices.  Furthermore, and this is a point modern economists sometimes overlook, the Fisher effect applies more to expected flexible price inflation, rather than CPI inflation.  That’s because commodities, real estate, and stocks are closer substitutes to holding money that are toasters and haircuts.

This interpretation makes Keynes’ statements somewhat more understandable, but still not correct.  First Keynes:

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. The prices of existing assets will always adjust themselves to changes in expectation concerning the prospective value of money. The significance of such changes in expectation lies in their effect on the readiness to produce new assets through their reaction on the marginal efficiency of capital. The stimulating effect of the expectation of higher prices is due, not to its raising the rate of interest (that would be a paradoxical way of stimulating output – insofar as the rate of interest rises, the stimulating effect is to that extent offset) but to its raising the marginal efficiency of a given stock of capital. 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 stimulating effect depends on the marginal efficiency of capital rising relatively to the rate of interest. Indeed Professor Fisher’s theory could best be rewritten in terms of a “real rate of interest” defined as being the rate of interest which would have to rule, consequently on 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. (pp. 142-43)

And then Glasner:

Keynes’s mistake lies in supposing that an increase in inflation expectations could not have a stimulating effect except as it raises the marginal efficiency of capital relative to the rate of interest. However, the increase in the value of real assets relative to money will increase the incentive to produce new assets. It is the rise in the value of existing assets relative to money that raises the marginal efficiency of those assets, creating an incentive to produce new assets even if the nominal interest rate were to rise by as much as the rise in expected inflation.

Like Krugman and many other new Keynesians, Keynes insists on thinking of stimulative effects via the transmission mechanism of interest rates.  But as David points out, the rise in existing asset prices is expansionary, in and of itself.  My only suggestion would be to talk about a rise in the value of existing assets relative to nominal wages, not money.  But since nominal wages are sticky in money terms, it amounts to essentially the same thing.  March to July 1933 is a great example.  Asset prices soared, nominal wages were steady, and firms responded by producing lots more commodities and investment goods.

PS.  This analysis may also relate to the following observation by Glasner:

I think that one source of Keynes’s confusion in attacking the Fisher equation was his attempt to force the analysis of a change in inflation expectations, clearly a disequilibrium, into an equilibrium framework. In other words, Keynes is trying to analyze what happens when there has been a change in inflation expectations as if the change had been foreseen.

In my view Keynes saw unexpected shocks leading to unexpected changes in flexible prices, and then subsequently to expected changes in the broader price level.

PPS.  Paul Krugman recently wrote the following:

Just stabilize the money supply, declared Milton Friedman, and we don’t need any of this Keynesian stuff (even though Friedman, when pressured into providing an underlying framework, basically acknowledged that he believed in IS-LM).

Actually Friedman hated IS-LM.  I don’t doubt that one could write down a set of equilibria in the money market and goods market, as a function of interest rates and real output, for almost any model.  But does this sound like a guy who “believed in” the IS-LM model as a useful way of thinking about macro policy?

Low interest rates are generally a sign that money has been tight, as in Japan; high interest rates, that money has been easy.

I turns out that IS-LM curves will look very different if one moves away from the interest rate transmission mechanism of the Keynesians.  Again, here’s David:

Before closing, I will just make two side comments. First, my interpretation of Keynes’s take on the Fisher equation is similar to that of Allin Cottrell in his 1994 paper “Keynes and the Keynesians on the Fisher Effect.” Second, I would point out that the Keynesian analysis violates the standard neoclassical assumption that, in a two-factor production function, the factors are complementary, which implies that an increase in employment raises the MEC schedule. The IS curve is not downward-sloping, but upward sloping. This is point, as I have explained previously (here and here), was made a long time ago by Earl Thompson, and it has been made recently by Nick Rowe and Miles Kimball.

I hope in a future post to work out in more detail the relationship between the Keynesian and the Fisherian analyses of real and nominal interest rates.

Please do.  Krugman reads Glasner’s blog, and if David keeps posting on this stuff then Krugman will eventually realize that hearing a few wisecracks from older Keynesians about various non-Keynesian traditions doesn’t make one an expert on the history of monetary thought.

The tone of western reporting about Chinese high speed rail has been fairly negative.  Stories of safety concerns, lack of ridership, excessively high ticket prices that the Chinese cannot afford, inability to compete with the discount airlines, “white elephants,” etc.

The safety concerns always mention the Wenzhou accident, which means the evidence has a sample size of precisely one.  And as we just saw in Spain, accidents can happen anywhere.  Maybe China’s high speed rail is dangerous, but I don’t see the evidence.

When I took a high speed rail trip from Beijing to Shanghai last year, the system seemed far more crowded than during my previous HSR trip in 2009.  And this Credit Suisse article suggests that some of the ridership fears may be overdone:

China is already home to the world’s longest high-speed rail line and some of its fastest bullet trains, and the government is pouring billions of yuan into building new capacity. In fact, China plans to add more than 3,100 miles of high-speed rail track by 2015, bringing the total length of the system to about 9,000 miles. Meanwhile, China’s airlines are also growing rapidly, with nearly 100 new airports planned by 2020.  .  .  .

The major concern of the market is that the airlines will cut airfare to protect their market shares, and that load factors (a measure of how full flights are) could be substantially lower than the levels before the launch of high-speed rail,” they wrote. However, the analysts go on to say that such fears are probably unfounded. China’s fast-growing middle class is more mobile than ever, and should do enough traveling in the future to keep both sectors healthy. In addition, the competition will be most heated on short routes, a realm in which airlines have already shown a willingness to cut their losses early and avoid costly price wars.  .  .  .

Airlines Adapting Rapidly

Chinese airlines have already showed a canny knack for maintaining profit margins by quickly responding to increased competition. Last December, when China’s newest high-speed rail line opened – a 572-mile [900 kilometer] high-speed line connecting the northeastern cities of Harbin and Dalian – most airlines simply left the market, cutting routes that overlapped with the rail corridor and shifting planes to other, less competitive routes. “We see this swift capacity adjustment, instead of cutting their airfares and profit, as a good sign that airlines can be nimble in competing [with bullet trains],” Credit Suisse analysts wrote. “While rail will remain a major competitor in the domestic market, its impact has largely been reflected in airline fleet planning strategies.” .  .  .

Taiwan Setting a Precedent

The caveat: Chinese consumers have shown themselves to be very price-sensitive, and a high-speed rail fare cut could pose a serious risk to airlines, Credit Suisse warned. Taiwan offers an interesting precedent. In 2007, the year high-speed rail lines opened for business in Taiwan, air carriers pared back capacity on the critical route from Taiwan city to Kaohsiung, the territory’s second-largest city, by nearly 44 percent. In an effort to undercut train prices that were 15 to 32 percent lower than airplane tickets, airlines cut fares by an average of 36 percent. Not to be outdone, high-speed train operators responded by slashing their prices an additional 20 percent. At that point, the airlines cried mercy: By the end of 2008, every airline with the exception of Mandarin Airlines had stopped flying the route. And Mandarin eventually conceded defeat in 2012.

Lowest Rates in the World

For now, a major move to cut rail ticket prices doesn’t seem likely, the analysts said. For one thing, though bullet trains are significantly more expensive than China’s highly subsidized “slow trains,” the rates are still relatively affordable – an average worker’s monthly income buys 24 tickets on a fast train in China, higher than the 23-ticket global average. In fact, Chinese high-speed rail fares are the lowest in the world, with an average price that is 36 percent lower than the cost of an airline ticket.

I’ve argued that the Chinese HSR network was built a bit too soon, especially in some of the less developed areas.  Perhaps they should have focused on establishing rights of way.  But I see no reason to believe that the massive infrastructure investment in Chinese HSR won’t be a major success, especially compared to Europe, which is less densely populated than eastern China.  Chinese incomes are rising rapidly, they’ll grow into the system.

We used to read lots of stories about how the Chinese economic boom was only in the coastal areas, and that the interior has been left behind.  That wasn’t even true in the old days, although the interior was and is still poorer.  When I arrived by train at the central Chinese city of Wuhan in 1994 I felt like I was in Calcutta, swept along by a mob of people in a crowded, dirty and confusing train station.  Here’s Wuhan’s HSR station today, in the “backward” part of China:

PS.  Earlier I did a story that discussed Chinese subway development, an even bigger success story.  One report I read suggested the crowding problem is due to a lack of cars, not a lack of line capacity.  And that can be traced to protectionism. China can’t build subway cars fast enough, and domestic content rules prevent imports.  It was in a comment section, so I don’t know if it’s true.

The title doesn’t sound very promising.  “Ad hoc” is supposed to be bad and interest rates are not sticky.

But ad hoc doesn’t actually mean bad, at least not all the time,  And the fed funds target might well be sticky, even if market rates are flexible.

In my “magnetic hill” post I suggested an ad hoc theory of the last three busienss cycles.  Long term real rates have trended steadily downward since 1983, and the Fed did not realize this was occurring in real time.  Hence they cut rates too slowly in recessions, and the recoveries tended to be slower than normal.  On the other hand the Taylor Principle did do a good job of stablizing inflation and NGDP growth.  So the recessions were mild—recall that unemployment peaked at only 6.3% in the 2001 recession (actually in the recovery.)

Evan Soltas has a very thoughtful new post that comes at this issue from a different perspective, but dovetails nicely with my theory.  You should read the whole thing, but I’ll provide an extensive quotation on interest rates, investment and the business cycle.  I won’t comment, I’d be interested in what others think.  Do these two theories fit together?

It’s possible that an investment has a positive net present value at a certain interest rate (a low one), but has a negative net present value at another, higher rate. So if all else is equal, Summers is right: If the central bank holds interest rates artificially low, it can in fact induce malinvestment — the central bank is creating an incentive to invest in projects which have negative net present values at normal interest rates. If the cost of capital was to fall below the internal rate of return, then that may generate a misallocation of resources.

That’s a bad thing, and it’s not “Austrian” to fear that. Nor is it the irrational response of a private sector to malinvest under those conditions.  (I’ll concede that the Austrian version, in which the malinvestment theory leads to eventually higher inflation, higher interest rates and a vicious bust, relies on a wildly implausible assumption of basically zero rational expectations — a collective delusion. But this isn’t my concern, though Summers seems to be saying as much.)

But all else isn’t equal, and I’ve been meaning to point out this error for a very, very long time to others. So here’s the thing about recessions: They reduce businesses revenue. They do so for the short run, and perhaps the medium run. That, by definition, reduces internal rates of return across prospective investments. If expected future revenues fall and the central bank doesn’t adjust the interest rate, in other words, it’s misallocating resources, but just in a different way.

This isn’t an idle fear. I’ve sung a lonely chorus for a year or so now arguing that economic commentators have underappreciated the role of expectations. (See here and here and here.) As I document in the last link, investment volatility now explains 77 percent of all volatility in GDP over the last decade, as compared to 47 percent from 1950 to 1960. Our recessions are increasingly about shocks to expectations of future revenues — they’re not about downward shocks to consumption. So many people just don’t get this.

This evidence suggests Summers has this backwards. Recessions appear to be about internal rates of return dropping suddenly below the cost of capital, and everyone trying to exit their investment and deleverage at the same time. I also present expectations data in those links. If you think about a roster of investments ranked in order of their internal rates of return, recessions seem to create a sudden jump in how many of those investments would be “underwater.”

Here’s a simple example. Suppose that you’re starting a business. It requires a $10,000 upfront investment in the first year. You’ll make no revenue then, either. Then you expect it to bring in $1,000 in positive cash flow for the next 29 years. That’s an 8.4 percent internal rate of return.

But, woah, here comes the recession! Now you’re expecting the first three years that had positive cash flow to be zero instead. Now the net present value, at the current rate, is -$2,475. Yikes! You’ll liquidate the investment rather than take a loss like that. Trouble is, if everyone does this, it’s a problem. So the central bank should cut the cost of capital from 8.4 percent to 6.4 percent. (And had it made clear it would have done this in the beginning, none of this would have happened in the first place.)

Further evidence that this is what happens in recessions, and that an NPV-related malinvestment is nothing to worry about, comes from the fact that businesses founded during recessions are more likely to last and be successful. It implies a past discrimination towards higher-NPV ones during recessions, not negative ones. (I don’t have the data in front of me on this; I remember seeing it in a Kauffman Foundation report.)

The central bank’s solution, of course, is to bring down the cost of capital when net present value falls due to fears of weak future demand. Stabilizing that, so that entrepreneurs can know the sign of the net present value of a prospective business regardless of the point in the business cycle, is really the way you need to think about monetary policy. It’s how you kill the business cycle — by making investment more insensitive, and less of John Hicks’ “flighty bird.”

I don’t have anything profound to say about the Detroit bankruptcy, but I’ll throw out one possible factor.  A few years ago I did a post discussing the striking population loss of large American cities after 1950.  The top 4 continued to do pretty well, but almost all of the rest of the top twenty dropped completely off the list.  Detroit barely hung on, dropping from #5 to #18, and Houston and San Francisco also stayed on the list.

Detroit isn’t unique, just bigger.  These cities also lost well over half their population since 1950:

St. Louis:  857,000 to 319,000,  Cleveland:  915,000 to 397,000, Buffalo: 580,000 to 261,000

All three now have fewer people than Mesa.  So why did Detroit go bankrupt?  I’m sure there are lots of reasons, but one might be that Detroit is in Michigan, and Michigan used to be very rich.  When I was young I saw a list showing that Michigan had the highest income in the nation.  I can’t find that list, but this link shows that in 1969 only Alaska, Hawaii, Maryland, Connecticut and New Jersey had higher incomes.  And all five are sort of special cases.  Alaska and Hawaii have a very high cost of living, and the other three picked up wealthy suburban spillover from NYC and DC.  (This was before the gentrification movement.)  So Michigan was the richest “normal state.”  It was richer than New York or California or Illinois or Massachusetts.  (OK you east coast intellectuals, I grew up in Wisconsin.  So I can define normal any way I wish.  A normal state looks like a mitten with a thumb on the right.)

Back in 1969 Michigan was governed by an ultra-liberal.  Someone named Romney.  Matt Yglesias once voted for his son, but the dad was far to the left of Yglesias.  Rich liberal states build up a costly overhang of government spending, taxes, high wages, high costs of unemployment insurance, union featherbedding rules, expensive pensions, etc.  That’s fine if the productive sectors keep churning out the wealth (think Wall Street, Silicon Valley, Harvard/MIT, etc) but if your economy tanks you’d actually have been better off with a more modest government, say Texas, or even an Indiana.  

I’m making a modest claim here.  The expensive overhang in Michigan helps explain why the Detroit area struggles even compared to other rustbelt cities.  I’m sure it’s not the only factor.  Alex Tabarrok has a related post.

It also might explain why upstate New York has done so poorly.  NYC is productive and appealing enough to survive NY state’s high taxes.  Upstate isn’t.  Places like Rochester used to be major centers of high tech—I kid you not.  They have no chance against Austin or Raleigh, or even against low tax cold/rainy states like New Hampshire and Washington.

PS.  Perhaps individual incomes in Michigan were higher in 1969–the link I found had household incomes.

PPS.  I have fond childhood memories of visiting rich Michigan to see my grandparents in the 1960s.

PPPS.  I’m a bit puzzled by the graph on unward mobility that Matt Yglesias and others are discussing.  I don’t have time to study the underlying research, so perhaps someone can help me.  I notice that the strongest upward mobility by far is the Great Plains and Utah areas.  And to a lesser extent the Kentucky/W. Virginia border area.  So that’s what?  Boom and bust farm area with low rainfall, fracking area, and coal country.  Is this really a story about economic opportunity, or is it about a casino-like economy where you can easily go from poor to rich and vice versa?  Or both (my initial view).  Are the studies based on annual income (volatile in commodity areas) or twenty year averages of income (less volatile in commodity regions.)

PPPPS.  As you know I’m a big fan of Yglesias, but unless this was a joke that went over my head, it seems out of line:

Detroit is everything that conservatives hate””labor unions, black people . . .

I hope Yglesias does not adopt the Noah Smith approach to conservatives.

Otherwise I agree with the post.

Mark Sadowski left the following comment:

I found something which may be of interest.

Larry Summers and John Taylor debated the implications of federal economic policy on April 4, 2012 as part of an event hosted by the Stanford Institute for Economic Policy Research (SIEPR). This was the second of two debates, after a February meeting by the pair at Harvard. They addressed the topic “Are Government Interventions an Important Cause of Our Recent Economic Problems?”

The video of the debate can be found here:

http://www.youtube.com/watch?v=fb8dO_D2b60

At the 16:30 minutes in you will find Summers saying the following during his introductory remarks:

“I will leave the question of monetary policy to John [Taylor] where he is an expert. I don’t think the question of whether the Fed was wise or was unwise during the noughts bears on the question of whether government caused our problems. The Fed’s job is to set monetary policy and it may or may not have done the right job. But I will speak to the question of fiscal policy in the wake of the crisis.”

Now ask yourself the following question; is someone who doesn’t consider himself knowledgeable enough to debate monetary policy qualified to be placed in charge of the monetary policy of the most important currency area on earth?

Let’s start with the assumption that Summers might have been just being polite, or modest.  (No sarcasm please.)  So I won’t take any cheap shots.  But there is a more serious question; how much expertise does the Fed chair need?  Here’s my answer:

1.  When not at the zero bound literally anyone could be Fed chair, even a janitor.  The staff will tell you when to raise and lower interest rates, and the Taylor Rule will keep things pretty stable.

2.  When at the zero bound we have one of two choices:

a.  NGDP futures targeting.

b.  Spend whatever it takes to hire the very best monetary economists.  Not just for the chair, but all 12 members of the FOMC.  Nobody should have any expertise in regulation, as that means they haven’t devoted their entire life to monetary economics.  Put regulators in other branches of the Fed–the bank regulation department.  Maybe Summers should head that group.

Paul Krugman once said something to the effect that “sometimes I feel like I’m the only one who understands liquidity traps”  (not exact words)  That sent a chill down my spine, as I suddenly realized that I’m just as arrogant as Krugman.  So maybe there are only two people who understand liquidity traps, not enough to staff the FOMC.

Of course I’m kidding, but I seriously believe there are a relatively small number of qualified people (and I’m not one of them, for unrelated reasons.)  I see Summers as a “tweener,” wildly overqualified for being Fed chairman during normal times, and not really qualified to handle the zero bound.  We don’t want a Fed Chairman who says; “not much the Fed can do at the zero bound, so let’s rely on fiscal policy.” Congress isn’t likely to do anywhere near enough stimulus to make a difference.  And recall the recent Romer and Romer paper that shows that all three major Fed failures were due to a belief that the Fed was powerless to stop the depression/inflation.

I think Summers would have done less QE than Bernanke, and I think he would have been overly anxious to “normalize” interest rates, although perhaps not as prematurely as Trichet.