The quantity of money was a big issue in the background of my childhood. There were no money machines or credit cards (at least in our house). So every Monday, my father would write a check for cash, usually for less than 100 pounds, and hand it to my mother for housekeeping for the week. She would then drive to the bank, cash it, put the notes in her purse and slowly dole them out as the days went by and she tried to keep a family of seven fed. Throughout the week, as she shopped and worried, she would know how much, to the penny, was left. The quantity of money was a binding constraint in her life.
We live in a different world today. There is, of course, a huge gap between rich and poor—wider today than 50 years ago. But for the vast majority of Americans, how you hold your money has ceased to be important. For small transactions, cash is hardly used any more, with people waving cards and phones at check-out counters. For bigger transactions, you could sell some mutual funds, if you have them, or take out a loan. Either way you can get hold of your money almost instantly. Lifestyles are still dictated by wealth and income. But the means by which you hold or access that wealth or income has ceased to be relevant.
The week ahead should be dominated by fresh assessments of the U.S. economy. For most economists, the key data will be contained in the first quarter GDP release, due out on Thursday.
However, some observers will pay equal attention to the monthly report on the money supply, due out at 1 p.m. tomorrow. For many decades, it has been economic dogma that “inflation is always and everywhere a monetary phenomenon” and the monetarist school of economic thought rose to prominence in the 1970s in part because of a very strong relationship between the growth in the money stock and the subsequent growth in nominal GDP.
For those who do worry about the money supply, recent data have been very concerning. Through February, M2 has fallen for seven consecutive months and is down 2.4% from a year ago. The March data, due out tomorrow, could show an intensifying contraction.
However, the truth is that the tight positive relationship between money supply and nominal economic growth has evaporated in recent decades. This collapse in correlation partly reflects a proliferation in alternative, easily accessible financial stores of value and means of payment. It also could reflect a collapse in the negative correlation between policy interest rates and nominal economic activity, which is the rationale for any activist monetary policy.
Whatever the cause, the fact that money no longer sends clear messages about the pace of economic growth provides little reason for complacency. Other indicators are pointing towards a worrying slowdown in economic activity, albeit with a steady decline in inflation. Moreover, the diminished predictive ability of money in the economy adds to the suspicion that the Federal Reserve is entirely wrong in claiming that it has the tools to restart the economy, should the economy indeed fall into recession.
Why M2 Is Falling
Data on the money supply, or described more accurately, the money stock, are contained in the Fed’s H.6 report, published at 1 p.m. on the fourth Tuesday of every month. The report provides monthly average data on the major monetary aggregates, M1 and M2, and their components. In rough terms, M1 consists of currency in circulation and various kinds of liquid checking accounts at banks and credit unions. M2 equals M1 plus money held in CDs and other time deposits below $100,000 and retail money market funds.
Both monetary aggregates have fallen in recent months with M2 peaking at $21.7 trillion in July 2022 and declining to $21.1 trillion by February of this year. We expect this Tuesday’s data will show a further sharp drop to roughly $20.8 trillion.
There are at least three big reasons behind this recent decline. The first is simply quantitative tightening.
In its most recent quantitative easing campaign, initiated to reduce the economic and financial strain of the pandemic recession, the Fed boosted its total assets from $4.1 trillion in February 2020 to almost $9 trillion two years later. This was achieved by buying huge quantities of Treasuries and mortgage-backed securities in the open market. Much of the cash received by the sellers of these securities ultimately found its way into deposit accounts and money market funds, causing M2 to vault higher from $15.7 trillion in February of 2020 to $21.7 trillion last July. This process is now firmly in reverse with the Fed ramping up quantitative tightening last summer and currently reducing its balance sheet at a pace of up to $95 billion per month.
However, beyond this obvious force in reducing the money stock, two other factors may be in play.
First, recent changes in the macro environment may have reduced the desirability of holding cash or low interest checking accounts. First, physical cash obviously pays zero interest which isn’t a big deal when inflation is 2%. However, with CPI inflation hitting a peak of 8.9% year-over-year last June and still running at 5.0% in March, there is now an incentive to conserve on cash holdings. Second, even with the Fed raising short-term interest rates sharply, average bank deposit interest rates remain well below 1%, increasing the incentive to move deposits out of the banking system. That being said, money market mutual funds, which are part of M2 but not M1, pay much more competitive interest rates and this is very likely contributing to the sharper decline in M1, which has fallen by 5.8% over the past year compared to a 2.4% decline in M2. Moreover, in addition to higher U.S. inflation and interest rates, the exchange rate of the U.S. dollar has been falling recently, with the dollar index now down 11% since last September. Fears of a further dollar decline give global investors a reason to transfer liquid assets out of physical U.S. dollars and dollar-denominated accounts.
Second, a decline in M2 holdings could reflect economic deceleration. If, due to worsening economic prospects, businesses are more reluctant to borrow and banks are less willing to lend, commercial and industrial loans should fall or grow more slowly, and this very much appears to be the case in recent months. This, in turn, leads to a diminished flow of money into corporate checking accounts. This reluctance to lend or borrow has worsened in recent weeks in the wake of the closure of some smaller regional banks and could contribute to an expected M2 decline in March.
The Collapse Of The Relationship Between Money And GDP
However, while it is possible to rationalize the recent decline in the money supply, the ability of the money supply to drive or predict economic activity has collapsed in recent decades.
In the early decades after World War II, the growth in the money supply was an excellent leading indicator of economic activity. Indeed, between 1953 and 1983, 42% of the change in year-over-year nominal GDP growth could be explained by changes in the year-over-year growth in M2 two quarters earlier. A surge in the money supply led a surge in economic activity in a very predictable way.
Milton Friedman described the process very eloquently. A surge in the money supply meant that people had more money than they wanted in their pockets or their checking accounts. They would try to reduce those balances by buying goods and services, thereby boosting economic activity. However, the recipients of those dollars would also then find themselves carrying too much money and would consequently spend them down, further boosting activity. Eventually, there would be an increase in real output or prices or both to a level that was sufficient to make money holdings appropriate again. Moreover, if a central bank increased the money supply when real output couldn’t accelerate any further, inflation was inevitable. Hence, inflation was a monetary phenomenon.
Strangely, however, as soon as the monetarist doctrine had become the dominant line of economic thought in the early 1980s, the relationship collapsed. From 1984 to 2019, year-over-year M2 growth explained less than 1% of the year-over-year change in nominal GDP two quarters later. So why did money stop talking?
The main reason appears to be that a lack of liquidity ceased to be a binding constraint on the operations of the economy. For centuries, a lack of reliable money did act as a binding constraint. In the 19th century, a lack of gold in a fast-expanding U.S. economy led to recurring bouts of deflation. In deflation, money becomes more valuable just sitting under the mattress, so deflation actually feeds on itself. Even in the 20th century, a lack of currency or money in checking accounts could constrain the economy.
However, today, the proliferation of alternative payment methods and competitive interest paid on money market mutual funds and ETFs has made monetary aggregates, such as M1 and M2, much less relevant to the economy. If, for some reason, the use of cash and checking accounts was somehow constrained, the economy would seamlessly move to using other means to complete transactions.
This is likely the main reason for the decline in the relationship between the money supply and nominal economic growth. However, there is another trend that is worth considering. A time-series of rolling correlations between M2 and nominal GDP doesn’t just show a decline from positive to zero over the decades—it shows a decline from strongly positive to mildly negative, with the zero correlation since the mid-1980s being an average of small positive numbers at the start of period offsetting small negatives in this century.
But how could an increase in the money supply actually predict a decline in economic activity? This may have something to do with the Fed’s use of quantitative easing as it cuts interest rates. Prior to the Great Financial Crisis, the whole idea of quantitative easing hadn’t been tried since the 1950s. However, in recent years, it has become part of the Fed’s standard toolkit. When the economy gets into trouble, the Fed both cuts short-term interest rates and engages in quantitative easing, which boosts the money supply.
However, starting from already low interest rates, it may well be that further cuts in interest rates actually cause the economy to slow, by reducing the income of savers, increasing public worries about recession and encouraging people to wait for even lower rates before completing transactions. Meanwhile, the interest sensitive parts of the economy, including home-building and capital spending have diminished in importance and are also highly sensitive to economic expectations. If, over the decades, the negative-lagged relationship between policy interest rates and nominal GDP growth has actually turned positive, and the Fed boosts the money supply via quantitative easing at the same time as it cut interest rates, a surge in M2 may reflect a supposedly stimulative monetary policy that is actually slowing the economy down.
On The Edge Of A Swamp Without A Monetary Raft
This last issue may become important in the year ahead. While the money supply data does not provide a reliable guidepost for the direction of the U.S. economy, there are plenty of other signs of trouble ahead.
In particular, while first-quarter GDP, due out on Thursday, is likely to show positive growth, unemployment claims have been rising in recent weeks. In addition, survey data show a continued tightening of lending standards both as perceived by senior loan officers and small businesses. Delinquencies on consumer loans are rising and an expected restart of student loan repayments in the next few months could force consumers to retrench elsewhere. In short, the economy remains on the edge of a swamp—not in recession yet but close to one.
In recent press conferences, Jay Powell, has stressed that the Fed has the tools to restart the economy should it fall into recession. However, it is not at all clear that this is the case. Cutting short-term interest rates proved very ineffective at boosting economic growth in the last long expansion. And if the Fed decides to resort to quantitative easing to boost the money supply, the history of recent decades suggests that this won’t work either. For this reason, the Fed would be wise to stop raising rates now to avoid adding further stress to the banking system, which could, indirectly, topple the economy into recession.
However, if they instead boost the federal funds rate to above 5% next week and hold it at that level throughout the rest of 2023, a return to monetary easing in 2024 would do little to boost the economy in 2024. This would be a painful outcome for many American households. However, by ushering in a new era of low inflation and low interest rates, a succession of monetary mistakes could help boost the value of both stocks and bonds.
David Kelly is chief global strategist at JPMorgan Asset Management.