In part 1, I walked us through the basic macro model of aggregate supply and aggregate demand to show that either increased demand or decreased supply, or both, could lead to higher prices. In this post, I want integrate money into the picture, since I agree with Milton Friedman’s famous dictum, “inflation is always and everywhere a monetary phenomenon.”

Milton Friedman (1976 Nobel Laureate) and Anna Schwartz published their seminal work, A Monetary History of the United States, in 1963, and it was a complete tour de force. In this work, Friedman and Schwartz carefully documented the linkage between the growth of the money supply and nominal GDP for the prior 100 years, with an almost 1:1 linkage between money supply growth (above real economic growth) and inflation. Further, their work overturned the by-then conventional wisdom that monetary policy was impotent during the Great Depression (i.e., that monetary policy had done all it could, and that the forces of depression simply overwhelmed it). Instead, Friedman showed the Fed’s policies had allowed the money stock to drastically shrink during the early 1930s, leading to the deflationary result. Friedman and Schwartz’s insights, although not universally agreed to, changed the focus of macroeconomic policy from fiscal to monetary policy. As Federal Reserve Board governor (and later chairman) Ben Bernanke was to say at a dinner honoring Mr. Friedman, “I would like to say to Milton and Anna: Regarding the Great Depression. You’re right, we (the Fed) did it. We’re very sorry. But thanks to you, we won’t do it again.”

Friedman’s Monetary History was an outgrowth of a previous, highly regarded paper on the quantity theory of money, which shows the linkage between money and output.* As this is related in most textbooks:

MV = PY

In this identity, money (M) times the velocity (V) of money (how many times the unit of money is exchanged in a given period of time) is identically equal to the nominal value of output, which is equal to the price level (P) times real output (Y). In the simple exposition, velocity is relatively stable and Y is determined by real variables such as labor and capital, so that an increase in the money stock will lead proportionally to an increase in the price level. And it is this relationship that Friedman showed to be consistent for a hundred years in his monetary history. Friedman would go on (along with his many adherents, the monetarists) to demonstrate this in many other countries and historical epochs).

To illustrate the quantity theory, consider the citizens of Meekoville**, a society where 100 individuals consume one taco every day, and there are 100 taco dollars in circulation (which can only be spent on tacos, no other goods). This is what models are for–unrealistic but simple! In Meekoville, since we have 100 taco dollars, M=100, and we have one taco purchased per day, V=1, and the real output level (Y) is 100 tacos, the price of the tacos will be $1/ea. Now imagine that Scrooge McDuck decides that taco dollars stuffed into a pillowcase would make a nice pillow for him to put on top of his pile of gold to allow him to nap on, and he steals half the taco dollars one night while the village of Meekoville sleeps. When the people of Meekoville wake up the next morning, they will have only half the taco dollars with the same number of tacos. It will probably take some time (but less time than you’d think) to realize that each taco dollar is worth twice as much as before, and the taco dollar will be bid up to be two tacos instead of one. When M goes down by 1/2, the price level must go down by 1/2. Alternatively, we could imagine a different scenario, one where the taco union is tired of low wages and they agree to a work slowdown, with workers calling in sick, so that the output of tacos declines to 50. In this scenario, each taco dollar will only be able to by half what it used to, and it will take two taco dollars to buy a taco, so the price level will have doubled. Note how that works in the quantity equation above: MV is constant, so that a reduction in output (Y) by one-half must be matched by an increase of the price level by two. A third scenario to consider would be to imagine that there is a panic about the future (doesn’t matter the reason) so that people decide they probably shouldn’t eat a taco today, but save their taco dollar for tomorrow. Keeping the same total # of taco dollars at one hundred, with only half being exchanged for tacos, the price level will have to fall by 50% to keep PY equal with the new, lowered MV. Finally, imagine a rogue central bank counterfeiter, who prints 100 additional taco dollars in the middle of the night to support a new Meekoville stimulus program. As these additional taco dollars work their way through the system, we eventually will have twice the amount of taco dollars as tacos, and each dollar will decline in purchasing power by one-half. The winners in this case are the beneficiaries of the new taco dollar creation (who received ability to consume without having to produce), while the losers are the prior holders of taco dollars, who will have produced and then found their consumption cut in half.

Each one of these scenarios is simplistic, assuming only one thing is changing at a time (the economists’ assumption of ceteris paribus), and in the real world the complexity occurs because many of these are changing at one time. These examples were static, with “stock” variables (what you have is what you have!) whereas the real world is one of flows (where what you have grows as productivity changes and new money is created***). Further, you can loosely think of the quantity theory as an illustration of aggregate demand and aggregate supply, with AD being reflected in MV and AS with PY. The problems with using the quantity theory as a practical policy guide quickly became apparent in the late 70s, early 1980s, when the Fed (at least rhetorically) tried to follow the quantity theory to get inflation under control. First, in an era of monetary innovation in response to high inflation and interest rates, exactly what should be money was not quite so clear. Clearly little green pieces of paper with dead presidents on them, but what else? Friedman and the monetarists had two measures that they used, first a narrow form called M1 (e.g., checking accounts and currency as the main components) but later used a broader measure M2 (which included some small savings which could readily be converted to cash). They had to do this because Friedman’s assumption that velocity would be stable started to break down in an inflationary environment.**** Indeed, just as Friedman’s monetary rule started to break down in the inflationary environment as new money alternates emerged, so too did the need to have as much money in the most liquid form. With the advent of new innovations like the money market mutual fund, which featured check writing privileges for larger transactions, narrower forms of money were less necessary to facilitate consumption choices.

Unfortunately with the breakdown of the precise relationships that Friedman had shown, economists and policy makers in general threw the baby out with the bath water: if we can’t find a precise relationship between money and inflation, then keeping track of monetary aggregates and careful control of money didn’t seem to matter at all. This we do at our peril, because inflation is still always and everywhere a monetary phenomenon, even if we can’t precisely show the complex process that it operates in. I still consider the quantity theory as truly the beginning of all monetary wisdom, even if it is not the end.

In the next blogpost, I will explain a more Austrian view of inflation, and we’ll begin to see why monetary expansion does not necessarily lead to consumer price inflation, at least in the near term.

* The content of the quantity theory had been understood for centuries; David Hume had written about it, and more recent to Friedman was the outstanding treatment by Irving Fisher. Friedman’s contribution was to turn this theory into a theory of money demand, and in so doing, brought it back into the economic policy discussion as not something that is merely an accounting identity, but something that can be used for macro policy.

** Meeko was our dearly loved dog who passed away a few years ago. We did not give him tacos, but he would have loved them. So in honor of Meeko, he gets an imaginary town named after him! And I bet he steals a taco occasionally off a plate (like he did with chicken and velvet red birthday cakes!) in this town.

*** New money would be created even without a central bank (e.g., a gold standard) but could only be created with a real resource cost (i.e., the labor to get gold out of the ground).

**** As I wrote years ago, the so-called monetarist experiment was attempted at possibly the worst time ever because of the inflationary environment.