Monday, September 12, 2011

Miscellaneous afterthoughts


A big day for the blog, yesterday.

I have long thought that Milton Friedman's argument -- "inflation is always and everywhere a monetary phenomenon" -- was a good argument, but that his evidence was shoddy. So it is a big deal for me that in comments here yesterday, three of us expressed similar doubts about Friedman's evidence, and expressed them strongly.

The best response I had before, on this topic, was from the Secret Economist. But nothing ever came of it, that time.

In my noontime post yesterday I wrote:

You cannot convert actual-price output to "real" output without bringing the price-trend into the numbers. After you do that, anything you do with those numbers will show the price trend.

Less than an hour later, Clonal was reporting "a correlation coefficient of close to zero" for the "correlation between the annual percent change in the CPI, and the annual percent change in M1".

And then, "very little correlation between GDP or Real GDP and M1".

And then, "M2 shows a high negative correlation with CPI".

And then, "MZM is even more negatively correlated".

Wow.


Clonal also describes the meaning of the negative correlation: "increases in the money supply appear to moderate CPI increases".

Knowing what it means is one thing. Assimilating the information is another. In the noontime post I wrote:

The excess of money-growth over and above real-output-growth gets burned off as inflation. Makes perfect sense to me. But when you try to show it using numbers and maybe a graph, there is a problem.

"Makes perfect sense". That's where I'm coming from. I reject Friedman's evidence. But I accept his argument. This has been my position for twenty years or more.

For me, more than twenty years. For Clonal, less than an hour. I'm impressed.

Clonal is saying that, far from causing inflation, increases in the quantity of money may actually subdue inflation.
Hard to accept, isn't it?
I don't accept new ideas quickly. But I am always open to considering them. Or so I like to think... But look at my response to Clonal in yesterday's exchanges.

1.
Clonal said

To me it appears that the solution to the oil crisis would have been to have engaged in deficit spending and increased the money supply.

I said: "Dunno..."

2.
And again, later, Clonal attempted to explain the effects of these negative correlations he was finding:

The strong negative correlations between changes in MZM and M2 and changes in CPI, particularly before 1984 lead me to believe that higher inflation leads to lower savings, and vice versa. After 1984, that relationship would have weakened because of increasing household debt...

He even concluded that thought by referring back to my favorite theme:

(too much debt!)

Did I open up to his analysis? Nah. But at least I didn't try to bludgeon him with my obstinate doubt this time.

The best I could eke out was my earlier

Of course we HAVE been talking today about how inflation does not correlate to the quantity of money...

But we hadn't been talking about that at all. We had been talking about a negative correlation. Not *no* relation, but a relation that is *opposite* to what we think.

Wow. Sorry, Clonal. I didn't realize my mind was so set in its views.


Clonal took up the idea quickly and saw where it leads. He drew graph after graph, complete with correlation coefficients, showing that money must not be the cause of inflation.

I still have trouble with that thought, of course. But I'm trying to at least consider it, as opposed to simply reacting by reflex from my own personal set of stock answers.

I think that if you are going to reject the view that money growth causes inflation, then you must find replacements for widely accepted reflex reactions like these:

Printing money causes inflation.

Too much money chasing too few goods.

Inflation is always and everywhere a monetary phenomenon.

I think that if we are going to reject those views, then we must find popularizable replacements to use in their stead. And besides that, we need good, strong, simple arguments that show that the reflex responses are wrong.

That's a daunting task.

Some people are willing to crumple up the whole existing story-line and toss it into the wastebasket. Robert Lucas did that, for example, with his call for microeconomic foundations for macroeconomics. And it seems to me that MMT does it as well, with its reversals of everything that mainstream economics teaches.

Obviously, some people accept that approach. I have difficulty with it. For better or worse, my mind is surprisingly set in its views. I was happy for a long time, rejecting Friedman's evidence but accepting his argument.

But anyway, if one is going to reject a popular concept like "printing money causes inflation," then one must offer both an equally catchy form of the alternative concept, and an equally simplified explanation showing why the old concept is wrong and the new concept is right.

Or else one must find a version of money for which the old saw holds good. Like maybe the money is all the credit currently in use, which of course is the same as "debt".

And then, the new saw would be: Lending money causes inflation.

8 comments:

Jazzbumpa said...

And then, the new saw would be: Lending money causes inflation.

I, too, am somewhat resistant to new ideas. But you have a coherent narrative that is getting harder to resist.

In the PDF it was stated that single year data didn't correlate money supply and inflation, yet Clonal's graphs indicate actual negative correlations.

I now believe that PDF to be quite a dishonest piece of work.

Who ya' gonna' trust?
JzB

Clonal said...

Here is another graph for you

Annual % Change in CPI, M1, M2 and MZM

the graph has monthly observations starting at Jan 1 1960 and ending in June 30 2011.

The spikes in inflation are always associated with real events like the OPEC Oil Crisis and the Iranian Crisis. High inflation leads to lowered savings, and that depresses M2 and MZM. So my observation is that real shortages cause inflationary episodes which determines the excess income available for saving. The differences in MZM and M2 appear to be interest dependent. If interest increases .above a certain threshold, people will put more money in time deposits - willing to take the risk of interest penalties if withdrawn early.

However the entire 50 year history makes it clear (at least to me) that inflation is always driven by supply constraints, and not by the money in the system. Once an individual, household or business's consumption needs are met, any excess income gets saved. The saving decision is independent of the interest rates. Once the decision to save has been made, the interest rate and the risk profile determines the financial instrument in which to put the money.

From other data, it is apparent, that for the bottom 80%, improved living standards have come not from increasing real wages, but rather from increasing household "for pay" working hours, and trading future income for current spending (borrowing) The post 1984 low inflation period had nothing to do with "monetary policy" but rather probably was an artifact of improved technology and cheaper oil that made possible very low cost manufactured items. One of the costs of this was "outsourcing" that has led to stagnant real wages for the bottom 80% in the US. The net result has been increasing wealth and income disparities inside the US.

So Art, my take on inflation is somewhat different from yours. The 1980's stagflation issue was a different one. My proposed solution would have prevented the stagnation, but probably not the higher prices in the short run. In fact Reagan did just that by his large deficit spending. The large deficits were entirely Reagan's personal ideas - as stated by Pierre Rinfret - But they worked. I do not totally agree with Reagan's priorities, but the large deficits worked.


Clonal said...

Typo - I meant to say 1970's stagflation and not 1980's

Clonal said...

The Secret Economist would be Robert F. Martin

Clonal said...

Also from Steve Keen today via Lord Keynes

Quote:
Some interesting facts:

(1) M1 and M0 (the monetary base) do not lead the business cycle: they both lag it.

(2) M2 – M1 represents credit money. This leads the cycle. That is, changes in credit money precede the cycle.

(3) The velocity of money is quite volatile. It is not essentially stable, as assumed and required by the quantity theory of money.

These data do not support the orthodox theory of exogenous money or even the orthodox explanation of inflation via the quantity theory of money. Instead, they confirm the Post Keynesian endogenous money theory (Palley 2002; Moore 1988), and the view that credit dynamics are a major cause of business cycles, as argued in Hyman Minsky’s financial instability hypothesis.

The direction of causation to explain inflation as postulated in the quantity theory of money is in reality reversed: it is not growth in the money supply causing inflation and wage rises, but wage rises and rises in factor input costs financed by business credit/debt from banks that cause money supply growth.


But I do not think that Steve appreciates the negative correlation beween M2/MZM and CPI

The Arthurian said...

Clonal,
"The direction of causation to explain inflation as postulated in the quantity theory of money is in reality reversed: it is not growth in the money supply causing inflation and wage rises, but wage rises and rises in factor input costs financed by business credit/debt from banks that cause money supply growth."

I like Keen a lot.

But if the quantity of money is not permitted to rise, then "wage rises and rises in factor input costs" will hold each other down.

Costs ARE rising (and have long been rising). The Federal Reserve *accommodates* rising costs by allowing inflationary increases in the quantity of money. Not too much, but usually enough to keep the economy growing.

Our views differ, as you said.

Jazzbumpa said...

Clonal -

Lots of meat in your 12:37 post. I will take some time and thought to digest.

From other data, it is apparent, that for the bottom 80%, improved living standards have come not from increasing real wages, but rather from increasing household "for pay" working hours, and trading future income for current spending (borrowing)

This is true post 1980. I think not for 1950 - 197x.

In the 70's, women reentered the work force. Now, it takes my kids two incomes to achieve he standard of living that came from my income from 1968, through my long career.

What is the other data you mention?

Cheers!
JzB

Clonal said...

Jazz,

Yes I did mean the post Volcker era after 1978, and not before that.

for references to the other data, see the article by Trader's Crucible - Monetary policy and Human misery