A brief update on M2

This post is an update of a thesis I have been working on for many years. It's based on a fundamental tenet of Milton Friedman, who taught us that inflation happens when there is an oversupply of money relative to the demand for it. Newer readers might want to read my description of how I learned to understand this while living in Argentina, and how this helped me predict the big jump in US inflation in the 2021-22 period.

For the past two years money demand has stabilized as M2 growth has picked up moderately. The "bulge" in M2 money supply has been almost completely absorbed by economic growth, with the result that there is no longer a monetary source of inflation that the Fed needs to stop—higher interest rates are not necessary. M2 money supply is once again growing at a 5-6% rate, which is similar to what we saw in the period from 1995 through Q1/20—a period when CPI inflation averaged about 2%. Given 

Chart #1

Chart #1 shows the level of the M2 money supply, which grew at a fairly steady rate of about 6% per year from 1995 to just prior to the Covid panic. M2 subsequently exploded as the government "printed" some $6 trillion to pay for an avalanche of Covid-related checks to the population. The Fed began to reverse this process in mid-2022, by raising interest rates and slowing the growth of M2, which has grown by only $500 billion since its 2022 peak. M2 today is only about $1.4 trillion above where it would have been if the 1995-2020 trend growth had continued.

Chart #2

Chart #2 is my way of measuring money demand: M2 divided by nominal GDP. The ratio is a proxy for the amount of readily spendable cash that people want to hold as a percentage of their annual income. For many years (50s, 60s, 70s, and 80s) money demand was fairly stable. That is what led Milton Friedman to assume that the velocity of money in his famous equation (M*V=P*Y) was relatively constant. But as we have learned in recent decades, that is a highly questionable assumption. In my way of looking at things, the demand for money should be the focus (money demand is the inverse of money velocity), since it ties in with the theory that the price level changes when the demand for money exceeds the supply of it. 

Sharply rising money demand in the wake of Covid effectively offset the sharp increase in money supply, as evidenced by the fact that inflation didn't pick up until 2021, when money demand started to plunge but money supply remained relatively constant. 

Today, Chart #2 tells us that money demand has ceased falling and is picking up just a bit. Combined with the fact that money supply is growing at a "normal" rate, we should expect to see inflation decline modestly, which in fact it has if we exclude shelter costs.  

Chart #3

Chart #3 makes a very important point: the growth of private sector jobs has decelerated significantly in recent years. However, the growth rates shown are most likely overstating the true growth rate, given the relatively large downward adjustments the BLS has made after the fact, and especially considering the magnitude of deportations in recent months. True jobs growth might well be only slightly more than half the rates shown in the chart. That amounts to a significant headwind to economic growth, and it has also contributed to depress confidence (most of measures of which are at relatively low levels). Any loss of confidence is likely to increase the public's demand for safety—and money hoarding is a natural response to that. Weak economic growth is likely fueling at least a modest increase in the demand for money. The Fed should respond to this by relaxing policy. And they have, reducing the federal funds rate by a quarter point recently. More cuts are likely in order and I hope they follow soon. 

Inflation is not a virus, and it's not going up

Inflation is not like a virus that spreads through a population. A rising price in one part of the economy cannot "infect" a price in another part of the economy. A currency doesn't just "catch" inflation because oil prices go up or a drought causes wheat prices to rise. Inflation is the result of an imbalance between the supply of money and the demand to hold it. It's a monetary phenomenon, as Milton Friedman taught us. And it generally results in a rising price for most goods and services.

The reason inflation appeared to rise last month, as the August CPI report implied, is not because of a monetary imbalance. As the charts below will show, the bulk of the apparent rise in inflation can be traced back to the way the BLS calculates inflation in the housing sector. There's another factor at work as well: the BLS can and does make mistakes, and numbers are typically volatile from month to month. Seasonal adjustment factors can be off and in need of revision. Consumer preferences for many goods change as prices change, but the BLS is slow to pick that up. The change in one month's number does not reliably mark a change in trend.

Chart #1

The Producer Price Index for Finished Goods (Chart #1) is a great illustration of the rather explosive inflation episode which the economy experienced from early 2021 to mid-2022. Prices were rising slowly from 2015 through 2020, then suddenly rose by more about 27% from early 2021 to mid-2022, only to once again resume a slow rise from mid-2022 through today. The fuel for that price explosion was a $6 trillion increase in the M2 money supply, a subject I've covered extensively in prior posts. From mid-2022 to August 2025, prices have increased at an annualized rate of only 1.4% with little or no sign of any meaningful acceleration. The return to low and relatively stable inflation was preceded by a dramatic contraction in M2.

Chart #2

The Final Demand version of the PPI (Chart #2) shows the same pattern. Although it rose at an annualized rate of 2.2% over the past 3 years, it is up at only a 1% annualized rate in the past six months.

Chart #3

Oil prices (Chart #3) show a rather sharp increase from 2021 through mid-2022, but they haven't increased at all since pre-Covid times. Since mid-2022, oil prices have actually fallen by 40%! Non-energy commodity prices have fallen by almost 10% since March '22. If inflation is on the rise, someone forgot to tell the commodity markets.

Chart #4

Chart #4 shows that the BLS's calculation for housing inflation today (which they call Owner's Equivalent Rent, or OER) is closely related to the year over year change in national home prices from 18 months prior. Today, home prices nationwide are barely increasing and are very likely to decline. Yet the BLS calculates that housing-related costs rose 4% in the past year. 

Chart #5

Chart #5 looks more closely at the behavior of housing inflation according to the BLS. In the past month, BLS calculates that OER rose at a 4.7% annualized rate, and is up at a 3.9% annualized rate over the previous 3 months. At a time when national home prices are on the verge of falling (and rents are flat to down), the BLS figures that housing inflation is accelerating! Trump is right to want a change in BLS leadership.

Chart #6

Chart #6 shows the 6-mo. annualized change in the CPI versus the ex-shelter version of the CPI. Without OER inflation, the CPI has been increasing at about a 2% annual rate for the past 3 years, and the rate of change has actually fallen to a mere 1.7% over the past six months. This means that the uptick in CPI inflation in August was driven by a big increase in the OER component of the CPI, as shown in Chart #5. 

Chart #7

Chart #7 shows the ratio of M2 to nominal GDP. I think that's a good proxy for money demand. Thanks to a big slowdown in M2 growth over the past three years, the amount of readily spendable money in the economy (M2) is almost back to pre-Covid levels. Excess money has been absorbed, and money demand is stabilizing. Thus, there is little or no reason to worry that inflation is coming back to haunt us.

California's Bullet Train Boondoggle

California voters approved the now-infamous high-speed-rail project in 2008; it was projected to cost some $33 billion and was to encompass some 800 miles, connecting Los Angeles to San Francisco, and the Central Valley to coastal cities, with an expected completion date of 2020. As I recall, the feasibility study for this project assumed that revenues from the LA-SF segment alone would exceed revenues from all other Amtrak lines in the country. After 17 years and over $15 billion in spending, not a single mile of track has been laid. Projected costs have tripled, and the scope of what is left of the project will deliver a 119-mile segment in the Central Valley, with no connection of SF or So. California. 

It is a manifest failure, and a monument to the fantasy that modern government can successfully and efficiently undertake complex projects. In just two decades, and for a tiny fraction of the cost of California's bullet-train boondoggle, Elon Musk's SpaceX has proved that point, going from zero to wildly successful and profitable, leapfrogging all other nation-sponsored space undertakings in the process. 

My good friend, talented artist, and fellow free-market advocate Nuni Cademartori has penned the cartoon which follows. May this boondoggle never be repeated, and may it swiftly be put out of its misery!

For more of Nuni's editorial artwork, check out my posts using this link. 

P.S. And may Gavin Newsom NEVER become president of the U.S.!