Sunday, June 24, 2012

Immorality of Inflation Targeting

Steve Waldman has an excellent post about the immorality of stabilizing the price level or inflation rate.    However, it is important to emphasize that his argument only applies to using shifts in aggregate demand to offset shifts in aggregate supply.   It isn't saying that it is immoral to have a stable trend price level or inflation rate.  And, as he emphasizes, there is nothing wrong with avoiding or reversing shifts in the price level or inflation rate due to changes in aggregate demand.

The argument about the undesirability of offsetting aggregate supply shocks was made (with less moral force, I think) by George Selgin in Less Than Zero.   While I don't find George's argument for a mild deflationary trend convincing, he did convince me long ago to give up on price level stabilization because of the undesirability of using monetary disequilibrium to reverse the impact of aggregate supply shocks on the price level.

David Eagle has also provided an analysis of the risk sharing implications of price level stabilization.  


This is one of the reasons shy Market Monetarists favor nominal GDP targeting rather than price level or inflation targeting.  

Wednesday, June 20, 2012

The Seventies

Scott Sumner  points out that since real GDP grew approximately 3% over the decade of the seventies, the reason for the high and rising inflation during that decade was the high and rising growth rate of nominal GDP.   In my view, Sumner's argument made perfect sense as a response to criticisms of Market Monetarism.   Market Monetarists argue that nominal GDP should be targeted and inflation ignored.   Some have responded by expressing concern that this would allow inflation to get out of control-- like in the seventies.     But the problem in the seventies was rapid growth in nominal GDP.   Like usual, Marcus Nunes provides helpful diagrams.  


Karl Smith argues that real GDP isn't important, it is the unemployment rate, which was quite high during all of the decade and reached 9% in the 73-74 recession.   Yichuan Wang argues that the accelerating growth of nominal GDP illustrated by Nunes should have resulted in more rapid growth in real GDP.    If real GDP only grew approximately 3%, there must have been some offsetting adverse supply shocks.   David Glasner provides an historical account of the decade, including accelerating inflation expectations and two increases in oil prices as significant adverse supply shocks.  

I decided to take a look at real GDP relative to the CBO estimate of potential GDP.



What is striking is that potential GDP looks very steady.  Other than that, there is a boom, bust, and slow recovery.    Not much room for supply-side decreases in productive capacity.    The growth rate diagram is below:



Like real GDP, as noted by Sumner, the CBO estimate of potential real GDP averaged close to 3 percent growth and really had very little fluctuation.

Here is the CBO estimate of potential output along with a linear trend from 1948 to 2012.


This shows that potential output was above its trend during the seventies.   Of course, it is obvious from the diagram that the growth rate of potential output is decreasing.    Still, if that trend is superimposed on real GDP and potential, real GDP remains above the trend of potential during the entire period.



If we had nominal GDP targeting during the period, at 5%, or better yet, at 3%, then the increases in the price of imported oil would have raised the price level and inflation rate.   The prices of imports don't directly impact nominal GDP.   Given the growth path of potential output, the prices of domestically produced goods would have grown slowly, 2% with the 5% nominal GDP growth path, or not at all, with the 3 % growth path.

Of course, maybe the CBO estimates of potential output are wrong.







Friday, June 15, 2012

Frankel on NGDP Targetting

Jeff Frankel has written some posts suggesting that inflation targeting has failed and that nominal GDP targeting is an attractive substitute.   See here and here. His citation of the economists who wrote about nominal GDP targeting is very helpful.   The only think missing is the importance of "level" targeting.  

Monday, June 11, 2012

Now That's a Model!

I was at the Canadian Economics Association meetings in Calgary this weekend and I went to hear David Laidler speak on Irving Fisher. (Laidler's paper is here.)   The other speaker at the session, Robert Dimand, mentioned that Fisher's doctoral dissertation was an independent discovery of general equilibrium.   While he did a good job,  the effort wasn't necessary since Walras and Edgeworth had already figured it out.

Still, Fisher didn't just do the math, he also designed and built a "machine" to calculate equilibrium prices.

My first thought was--how can they make fun of hydraulic Keynesianism?     What about hydraulic microeconomics!

A photo  


A diagram.



Another diagram.


I must confess that some of my micro intuitions have a hydraulic feel--the composition of demand changes like a fluid being sucked out of one sector of the economy and flooding into another--according to the preferences of households and firms.   But I must admit in my vision of the market system, there is a lot of continuous sloshing about rather some kind of exact balancing of the water levels.

I think it is common knowledge that Fisher's reputation was destroyed by his prediction that stock prices would rise in 1929.   The American economy was heading upward and ever onward!

I didn't realize that he put prohibition on his list of reasons for the fundamental strength of the American economy.   

I had always thought that Fisher gave the Federal Reserve too much credit, expecting them to listen to the world's greatest monetary economist (him,) and prevent the collapse in the quantity of money, spending on output, and the price level.   Fisher was shocked and appalled that they continued to apply their "real bills" nostrums, worrying about speculation and "inflation" as they put the economy through a deflationary wringer.

Come to think of it, Fisher's faith that the public officials at the Fed would do the right thing is similar to his notion that a legal ban on alcohol would create a nation of productive teetotalers rather than the reality of a nightmare of organized crime. 

Tuesday, June 5, 2012

Use the Force, Ben

Matthew O'Brien hits this one out of the park!

There is No Try,
            Ease or Ease Not.



(But target the growth path of nominal GDP!)

Andolfatto on NGDPLT: Beside the Point?


It is great when a smart monetary theorist like David Andolfatto takes a hard analytical look at nominal GDP targeting.    His recent post develops a overlapping generations model, considers expected productivity shocks to capital, and suggest that nominal GDP targeting would provide no benefit--I guess.

However, when output is given and prices are perfectly flexible, nominal GDP targeting will likely have few benefits.  That nominal GDP targeting provides few benefits in this model and also would provide benefits in the real world where output can change and prices (including wages) are sticky, should also be no surprise.

Anyway, supposedly young people expect their capital goods to be less productive when they are old, so they prefer to produce fewer of them and instead produce consumer goods to sell to the old people and accumulate money/bonds.  (That there is no difference between money and government bonds is a pretty serious problem for any monetary model, but I realize that doesn't phase "top tier" monetary theorists.)

The quantity of money is given, apparently, so this creates deflation. The old people get extra consumer goods, and the young people get less money per consumer good, with the prices they receive being lower.

Then next period, they will be able to spend the money/bonds on consumer goods produced by the next generation. It seems to me that this money for which they gave up more consumer goods will buy them less consumer goods once the next generation's willingness to supply consumer goods returns to normal.

But then, I guess the real return on capital is supposed to be negative too.  Andolfatto seems to identify the natural interest rate as the interest rate without the realization of the productivity shock.   I guess it would be the expected natural interest rate.   Because there is no growth, it is supposed to be zero.   And so,  if the shock is zero, then I guess a negative shock makes it negative absolutely. And so we are equalizing that negative return on capital with a negative across generation transfer of consumer goods using money.   Well, negative for the current young generation.  The windfall for the old folks suggest they get a positive return on their money.

From a market monetarist perspective, the quantity of money should rise. The young people want to hold more money, and so more should be issued, presumably in exchange for consumer goods. Obviously, the old people have to get the consumer goods since no young person wants them. But the government is creating all the money.  

So, how does the government get the new money to old people?

Suppose the government pays extra "interest" to the old people today, expanding the quantity of money. And gets it back by paying less interest to the current young people when they are old.

The old people buy more consumer goods, the young people get more money, but earn lower interest on the money/bonds.    Given the assumption of zero interest in "normal" times, this would be positive interest for the current old folks and negative interest paid to current young people.

I guess I just don't understand. Because it seems to me that when the current young generation expects a poor return from capital goods and they switch to selling more consumer goods so they get a lower return from money too, driving down money prices, until the expected inflation rate makes the real interest rate equal to the lower expected return on capital goods, it is just creating a windfall for the current old folks.

It seems to me that to provide more to the current group of young people, taxes must be increased on the next group of young people, making up for the unusually low productivity of capital this period.

And I don't see what this could possibly have to do with nominal GDP targeting.

Also, in this scenario, output is constant, and so constant nominal GDP is the same a constant price level. Isn't the live debate about which of those is better?

Fixed quantity of money, higher money demand, and deflation somehow being better because it prevents too much investment is beside the point.  (Doesn't this sound like the Austrian business cycle theory?)

So, how is it that a fixed quantity of money, inflation targeting, or price level targeting allows the next group of young people to be taxed more?

Or is it simply that nominal GDP targeting just doesn't help in this scenario?   Well, the argument in a perfectly flexible price scenario is that when potential output falls unexpectedly, a price level target shields creditors and puts all of the burden on debtors.   Nominal GDP targeting shares the loss by a higher price level for creditors.

Since in this model output is fixed, there is nothing to share.   And, the government is the only debtor, so any benefits would go the the government/taxpayer.    

It is a bit odd that output stays constant but the productivity of capital falls, but that is because the "productivity of capital" is the ability of capital goods to produce consumer goods for the old folks that own them.   It would be like the cars they purchased when young break down more than expected when they are old.   Or they buy fruit trees when young, and when they are old, low yields mean they don't eat much fruit when old.    Young people, hearing news that this will happen to them, leads them to buy fewer cars and instead accumulate money/bonds.   They get them by selling massages, or something, to the old folks today.   

Well, that is my take so far.   And I think that there is no substitute for adding a bit more realism to the analysis.