John Cochrane endorses futures targeting
Last year I had an email conversation with John Cochrane about futures targeting. He seemed intrigued, but I wasn’t sure if he was just being polite. Now he has endorsed the idea as a way of escaping liquidity traps:
But a commodity standard is impractical for a modern economy. If gold can double in value relative to other goods in the CPI, as it has done recently, then other goods can deflate to half their value if the government fixes the price of gold.
Instead, the Fed can target the thing it cares about – expected CPI inflation – rather than the price of gold. To do it, the Fed can target the spread between TIPS (Treasury Inflation Protected Securities) and regular Treasurys, or CPI futures prices. Here’s a simple example. Investors buy a CPI-linked security from the Fed for $10. If inflation comes out to the Fed’s target, they get their money back with interest, $10.10 at 1% interest. If inflation is 2 percent below target, the Fed pays $2 extra — $12.10. This pumps new money into the economy, with no offsetting decline in government debt, just like the helicopter drop. If inflation is 2 percent above target, investors only get back $8.10 – the Fed sucks $2 out of the economy at the end of the year. If investors think inflation will be below the Fed’s target, they buy a lot of these securities, and the Fed will print up a lot of money, and vice versa.
One might object that these markets are small and undeveloped. I answer that is exactly why the Fed needs to start doing it now, so the markets are large and developed when the Fed really needs them. And of course the details will be more complex than what I have outlined.
Of course this is an issue that I am especially interested in, having first presented the idea at the AEA in 1987, and I have since published numerous articles on the topic. And he’s right that the details are more complex that his short description suggests–it must be set up in a way where the Fed induces the market to forecast the optimal instrument setting, in order to avoid the circularity problem. Not all proposals did that. (I’d be curious as to what Garrison, White, Bernanke and Woodford think of Mankiw’s approach.)
A number of economists have discussed futures targeting, including Earl Thompson, David Glasner, Kevin Dowd, Bill Woolsey, and Aaron Jackson. (I apologize to those I forgot. Robert Hall developed an analogous idea that would affect money demand, not the money supply. But John Cochrane is the highest profile economist to endorse futures targeting. Great to have him on board. I don’t know if he recalled our email conversation, but even if not I might have planted the idea in his subconscious mind.
HT: Adam P
PS. Comment responses will continue to lag
Tags: Futures Targeting
8. October 2010 at 08:21
Would targeting the nominal level of a commodity basket be a second best policy here? I know that the relative price of commodities and finished good and services can drift, but how far apart can they really get in the long hall?
8. October 2010 at 09:02
I think he should have credited you – and I think it is really shameful of him that he didn’t.
But – as you say – this is a big time name – hooked into Fama and the big guns – so this must mean good things. He has thought a lot about markets and asset pricing – so his support is obviously very welcome.
8. October 2010 at 09:02
The funny thing is that no matter what monetary policy you chose, there is some targeting going on. Gold bugs believe that the gold standard would get the government out of monetary policy. But that is ridiculous. The government would just target the price of gold in conducting its monetary policy like it has done before.
So, the question is do we want monetary policy to revolve around how much of a shiny, yellow industrially useless metal is pulled out of South African mines each year? Or do we want monetary policy to revolve and target something more meaningful?
8. October 2010 at 09:44
And I thought John Cochrane was a dead lawyer.
8. October 2010 at 09:58
As long as it is a basket of commodities and not just CPI this seems like a great idea. Frankly it sounds like the King Dollar wet dream.
I hate that rents / housing would be counted in CPI… maybe we can get them to deflate before this market is big enough for the Fed to make the switch.
Another question: If it was a meaningful market, couldn’t Wal-Mart game it?
Even then, as long as the target is 2%. It sounds genius.
Even better the Fed ought to let investors buy direct without a brokerage in the middle.
8. October 2010 at 10:25
Concerning the gold standard this article really rocked my world: http://www.ces.org.za/docs/what%20is%20money.htm
8. October 2010 at 10:37
This seems like a great way to get the central planning out of monetary policy. It still seems like a second-best solution to free banking with commodity redemption, but it’s not a bad interim I guess.
8. October 2010 at 10:46
That is a fascinating historical piece, woupiestek.
8. October 2010 at 12:00
@scott_sumner: Great idea, this would do wonders for the economy. I know tons of businessmen who base their decisions on what they expect NGDP to be, and for them to expect to grow at 5%, guided by futures markets, would definitely help in them making better decisions. I know they talk about NGDP with each other all the time.
@woupiestek: Interesting article, I guess, but what about your world did it exactly rock?
8. October 2010 at 12:46
So, is Cochrane now a “left winger”?
8. October 2010 at 13:36
He is a commie Greg. A commie.
8. October 2010 at 13:37
@Silas,
Businessmen base their decision on how much demand they see. So, are you in support of fiscal stimulus then?
8. October 2010 at 14:10
Oneeyedman, What matters is the “short haul” and unfortunately they get very far apart in the short run.
JimP, I don’t have any problem as long as he isn’t claiming the policy as his own idea (which he didn’t.) Others have also published articles on the idea. He may well have forgotten where he got the idea, as people like him are bombarded with lots of ideas every day.
Liberal Roman, Exactly.
Morgan, It should be NGDP, obviously.
Benjamin. I didn’t know he was dead.
woupiestek,. Sorry, it’s waaaay to long for me. Are there any key paragraphs you can quote?
John Papola, I think it’s even better than a commodity exchange.
Silas, I agree. Business confidence in “the economy” (which basically means NGDP or AD) is a key factor in investment decisions. If car companies expect higher NGDP next year, they are much more likely to build new factories.
8. October 2010 at 14:11
Liberal Roman, it’s “communism with Chicago characteristics“!
8. October 2010 at 16:13
Liberal Roman,
Not to mention that gold itself has its origins as a fiduciary item. A gold coin is just a fiat currency as a substitute for barter.
There is a claim in certain quarters (I’m thinking of the Austrian School) that there is a constant marginal utility to gold. But any number of things have served as money for prolonged periods of time (from cattle to little statues) which clearly have no constant marginal utility; like gold, an item may have slowly diminishing marginal utility for one or other reason, but one should not confuse this with constant marginal utility.
Historically, the utility of gold has fluctuated erratically. Basing a currency on the value of gold, as you say, means that monetary fluctuations are decided by supply-side factors that have absolutely nothing necessarily to do with the real economy in the country that uses the currency. Just look, for instance, at what happened in Europe as a result of the discovery of gold in the Americas. There has been research that suggests that Tulip Mania was at least partly the result of monetary instability caused by an influx of gold and silver into the Dutch economy.
The value of gold is because it had value yesterday and the value it had yesterday is because it had value on the day before yesterday and so on, right back to gold’s original use as fiat money. If the gold maniacs, who are Austrians, actually appreciated the implications of their axiology (the Regression Theorem et al), they would realise that the value of gold is a product of past extrinsic instrumental value, just like paper currency.
But what about the Gold Standard as a system of monetary stability? One could achieve the exact same system, without distorting the market for gold, by using Random.org to get a random integer within a certain range and having that much inflation/deflation for a set time period. Then you’d get a slightly more stable version of the Gold Standard.
Futures targeting makes much more sense. Again, using Austrian terminology (simply because discussing commodity values usually involves dealing with people who have no more than a basic grounding in pop Austrian economics) the Extended Order of the market will produce a better response than either Federal Reserve designs or an inflexible commodities rule.
8. October 2010 at 17:13
Scott,
Why do you even care what John Cochranes’ opinion is. Wasn’t he the one (the fool) quoted as saying (about Keynesian economics):
“It’s not part of what anybody has taught graduate students since the 1960s. They [Keynesian ideas] are fairy tales that have been proved false. It is very comforting in times of stress to go back to the fairy tales we heard as children, but it doesn’t make them less false.”
I always start with any class about Macroeconomics with Keynes. Anyone who fails to do so is not providing proper historical perspective and is, IMO, simply a political demagogue.
8. October 2010 at 18:51
Scott,
You had a web debate with Cochrane in July 2009. Nineteen and a half minutes into the discussion you propose targeting ngdp futures contracts. It’s worth watching for sure.
http://www.finreg21.com/channel-21/interviews/monetary-policy-discussion
8. October 2010 at 19:11
tom s.,
I saw that debate. At the about the 19:30 minute point my eyes notice the following body language. I observe Scott blinking his eyes as though he is reflecting in a deep manner. I observe Cochrane shifting his body as though he is extremely uncomfortable at what Scott is saying. Take it as you will.
8. October 2010 at 23:04
The above “future targeting CPI linked security” idea strikes me as useless. First, only a small proportion of households directly hold government securities. To that extent you might as well boost an economy via a payroll tax reduction, but limit it to people whose names begin with the letters A-E.
Second, as regards the large chunk of government debt held by foreigners, I don’t see them (the Chinese government in particular) running out and buying U.S. produced goods and thus creating U.S. jobs.
Third, as regards the large chunk of government debt held by banks and insurance companies, I don’t see an immediate effect on demand coming from this source either. Witness the recent astronomic increase in bank reserves which has had no effect.
8. October 2010 at 23:22
Scott,
There’s a lot of talk about ‘all the liquidity’ out there pushing gold and other assets higher. Do you think that is wrong or could you have ‘high liquidity’ and ‘tight money’ at the same time?
9. October 2010 at 00:14
[…] this link: TheMoneyIllusion » John Cochrane endorses futures targeting Related Posts:Book futures | john hawks weblog Book futures . An article about the future bookless […]
9. October 2010 at 04:31
If you googled “money” back in the autumn of 2008 you quickly found the same popular history of money all over the internet: barter – gold standard money – banks and fiat money – financial crisis. I imagined this was more or less correct, until I found this article. What shocked me is that this answer to all the non sense on the web was written and published almost a century ago.
Mitchel Innes starts by pointing out the difference in quality of Greek and Roman coins with the same nominal value, arguing that producing large number of coins with equal metallic content was not possible let alone necessary in those days.
Then he tells us that banks already existed in Babylonia centuries before the invention of money. Debts can be measured in arbitrary commodities and people invented borrowing long before inventing money.
He describes tally sticks which were a medieval kind of credit cards.
Then he exposes the gold standard as a primitive way of controlling inflation, points out what a waste a gold reserve actually is.
So like this blog, Mitchel Innes’ article has been a real eye opener for me.
9. October 2010 at 05:00
It is supply and demand, Mattias.
Read it this way… “The amount of liquid assets that people are holding at this time is large by historical standards.”
Monetary disequilibrium theorists are saying, the reason money expenditures are way below the trend growth path of the great moderation is that the quantity of money is less than what the demand to hold it would be if money expenditures were on target.
From our perspective, the demand to hold money appears to be very high by historical standards and while the quantity of money may be high, the demand for money is higher still.
You can say the same thing with interest rates. Interest rates on short and safe assets are very low by historical standards. I would say, that they are still higher than the natural rate on for those particular assets. The natural rate is at historically low levels, and while the current market levels are low, they are not as low as the natural interest rate.
Unfortunately, one point that Scott has emphasized, and while obvious enough in retrospect, wasn’t something I emphasized in the past, if there were a commitment to get the quantity of money high enough and market rates low enough to get money expenditures back to the growth path of the great moderation, then the demand to hold money would fall and the natural interest rate would rise. It is entirely possible, if not likely, that the actual quantity of money would need to fall and market interest rates would need to rise.
9. October 2010 at 05:02
P.S. I don’t like fancy security schemes, and I think they are wrong headed if the way the quantity of money changes is through payoffs on the securities. A basic and common error for those thinking about index futures targeting. I will grant that you could construct some security that makes it work, but I have my doubts.
9. October 2010 at 05:49
Mark, Well I suppose he could argue that he was talking about grad classes, by which time they should know the Keynesian cross and paradox of thrift. But I agree that more monetary history should be taught–even in grad classes.
Tom, I have trouble watching myself on camera–it makes me cringe. But thanks for pointing that out.
Ralph, He didn’t explain the idea very well in that quotation.
Mattias, Sure they could both occur at once, unless you define liquidity as monetary policy. ‘Liquidity’ is a very vague term.
woupiestek, Thanks for that info. The history of money is endlessly interesting.
Bill, All good points. The example Cochrane provides is not the best way to do it. The money supply change is delayed for a year.
9. October 2010 at 15:01
Scott,
You wrote:
“Mark, Well I suppose he could argue that he was talking about grad classes, by which time they should know the Keynesian cross and paradox of thrift. But I agree that more monetary history should be taught-even in grad classes.”
Well, I suppose it’s a fact I’ve never taught a graduate class in macroeconomics. But my dissertation adviser, who is in fact a Republican (unlike myself), always starts with Keynes, even when teaching a graduate class. Maybe he is a simpleton. But I hapen to think he’s one of the finest macroeconomists alive (James Butkiewicz).
10. October 2010 at 04:55
Mark, Good point. There’s is also the issue of which Keynes Cochrane was talking about. It’s true if he was talking about models like the Keynesian cross, which are being dropped from textbooks (for good reason.) But there is much more to Keynes than the Keynesian cross. I happen to agree with Hicks and Friedman that the GT is basically a liquidity trap model.
10. October 2010 at 16:44
[…] Perhaps me and the other futures targeting proponents were just ahead of our time. One famous macroeconomist endorsed futures targeting just recently. A sign of things to […]
11. October 2010 at 07:57
Obviously it should be NGDP?
Why? He seems to say that what the Fed cares about is “expected CPI inflation.”
I understand your NGDP plan, but what is the monthly metric that allows us to have feedback for expectations?
And man, WHY DOESN’T everybody focus on the important part: That very soon, we’re running above 2%, unemployment is WAY ABOVE 8%, and there is NO MORE EASING coming…. and Fiscal has zero, so the point of this really is to lock the Fed into a expectations path, that will not be deviated for political gain.
12. October 2010 at 05:06
Morgan, You need NGDP precisely to deal with a case where inflation and unemployment give mixed signals. They should estimate NGDP monthly, like in Canada.