Wednesday, May 27, 2015

Asset Price Inflation

Larry White mentioned that Alchain and Klein showed long ago that the measure of the purchasing power of money should include asset prices.

I don't agree.

I strongly agree that it is a mistake to measure the purchasing power of money solely by the prices of consumer goods and services--the CPI or CEP.

But I don't think that the prices of financial assets should be included in a measure of inflation.

As for real assets, I think that only the newly-produced ones should count.

In other words, I think that the GDP deflator, or something like it, is the least bad approach to measuring the purchasing power of money.

First, consider equities.   If we assume that a share of stock is a claim to a fixed quantity of goods, then any increase in the price of a share would imply a lower purchasing power of money.   However, suppose that a company is expected to become twice as profitable.   The price of  a share would rise, but it would not be a higher price for the same quantity of future goods.   It would be a higher price for a higher quantity of future goods.

Now, suppose the market interest rate should fall, and the lower discount of future returns results in higher prices of equities and existing long term bonds.   Superficially, the purchasing power of money is less.   It is necessary to pay more for the same quantity of future goods.

However, suppose we lived in a world where the only saving instrument was a saving account.   There is no possibility of capital gain or loss on financial assets.   If the interest rate should fall, then it ia true that money put into a saving account will provide less future consumption.   But is that a lower purchasing power of money?   Saving or accumulated wealth now generates a lower nominal and real income.    Is that a higher price level?    I don't think so.

I guess my thinking on the subject is very much influenced by some sort of presumption that  inflation is bad and the monetary regime should be stabilizing the purchasing power of money.   And so, I would ask if a monetary contraction would desirable to force down the prices of consumer goods and services enough to offset the increase in long term bond prices.   I don't think so.

The coordinating role of the lower interest rate is to raise the demand for both  consumer goods and services and capital goods.

For example, suppose there was an increase in saving supply.   The direct and immediate effect is a lower demand for consumer goods and services.   The coordinating role of the lower interest rate is to increase the quantity of consumer goods and services demanded as well as the quantity of capital goods demanded.    The result is what returns saving and investment back into balance.

I certainly grant that it would be inappropriate for this reallocation of resources to occur with stable prices for consumer goods and services.   Lower prices of consumer goods and services along with higher prices of capital goods should be expected.  

Now, the lower interest rates should result in higher prices of all the existing capital goods--not just the newly produced ones.   This doesn't require anyone to spend any money on all of these capital goods.   It is just that entrepreneurs will be willing to offer more money for them and those using them now will insist on a higher payment to part with them.

Should the weighting of capital goods in the price level depend on all of the capital goods or just the newly produced ones?

Suppose consumption is $8000 billion and investment $2000 billion.   The capital stock, however, is $20,000 billion.    There is an increase in saving and a $200 billion shift in demand.    Suppose that this 2.5% decrease in demand for consumer goods results in 1% lower prices.   However, the 10% increase in the demand for capital goods requires 4% higher prices.    If we look at only currently produced goods, the price level remains the same.    But if all of the existing capital goods count, then there has been a very substantial inflation--about 2.5%.

Would it be better for the prices of  consumer goods to fall more and the increase in the prices of capital goods to rise less?    What would this be signalling?   That the production of consumer goods and services should be reduced by more?   That the expansion in the production of new capital goods should be less?

Presumably this would require that wages be reduced to return to equilibrium.   Why?   Is this signalling that people should work less?

In my view counting stock prices is more or less the same thing as counting the prices of existing capital goods--in theory.

In reality, is it really the best that if stock prices go up,  there is a deflation of consumer prices along with money wage cuts?   What is the point?   Do we need to get people to work less?   Or is it just to free up resources from the production of consumer goods and services to print up more shares of stock?

As for bonds--as old bonds mature and new ones are issued at par with lower coupon rates, does that count as deflation?    The "prices" of bonds are falling.   Should consumer prices and money wages rise to offset that deflation?   Clearly they should not.

Now, perhaps Alchain, Klein, and White have no presumption that the purchasing power of money should be stabilized.   If more saving supply or reduced investment demand results in a  lower natural interest rate, then that just lowers the purchasing power of money--and there is nothing bad about it.   Maybe.

And I must admit, I don't favor a stable price level of any sort, but rather a stable growth path for spending on currently produced output.   However, I do favor a growth rate of spending that is consistent with the expected trend growth rate of potential output.   And so, this would tend to result in a stable price level for currently produced output.  (And I know White does not favor stabilizing the price level, especially when there are changes in productivity.)

But it wouldn't tend to stabilize the prices of current output and existing capital goods and financial assets all together.

And it shouldn't.   Or at least, I don't think so.