The UK newspaper the Telegraph had the financial community buzzing with the headline "US Money Supply Plunges at 1930s Pace as Obama Eyes Fresh Stimulus." According to the article—and the many subsequent reports that echoed this storyline—M3, a measure of money supply, fell from $14.2 trillion to $13.9 trillion in the three months ending in April, signaling an all but certain downturn in the US economy.
But what exactly is M3, and does it really signal another downturn?
First, note the US Federal Reserve stopped reporting M3 in March 2006 because "M3 does not appear to convey any additional information about economic activity that is not already embodied in M2, and has not played a role in the monetary policy process for many years." The M3 stats cited by the Telegraph and others are estimates from Dr. John Williams' website Shadow Government Statistics and are attempts to reconstruct the Fed's M3 metric-based on a combination of available data, financial models, and educated guesses. You can view a graph of Dr. Williams' estimate of M3 here. The accuracy of this data is questionable, and other third-party estimates vary significantly.
Still, for argument's sake, let's take the most recent M3 guesstimates at face value.
Federal Reserve's definition of M3 (as well as Dr. Williams' estimates) includes M2—mostly currency, checking accounts, savings accounts, CDs, and retail money market funds—and adds in institutional money market funds, repurchase agreements, large time deposits, and Eurodollars.
While these M3 instruments may once have been a valid measure of broad liquidity, due to financial innovation over the last decade, most large financial institutions probably now have many more options for cash-management and short-term investment than the now fairly arbitrary assortment of instruments included in M3, . For example, there are many swap-based instruments that may now provide cash-like exposure potentially with a higher rate of return, but they are simply not captured in M3.
It's also very likely that with interest rates so low and money markets and large time deposits yielding next to nothing, institutions may be in search of higher yields—such as asset-backed securities, bonds, or even equities (none of which are included in M3). There is some evidence to support the hypothesis: Over the last year, money market assets have fallen in relatively close sympathy with the rise in equity prices. (See Exhibit 1.) This could imply M3 is falling simply because institutions are committing more capital to riskier financial assets—hardly a negative for the economy or the stock market looking forward.
Exhibit 1: S&P 500 & Institutional Money Market Fund Assets
Source: Thomson Reuters, US Federal Reserve
Additionally—while it may be simply coincidence—since January, when the SEC approved rules forcing money market funds to invest in more liquid assets, institutional money market assets fell by nearly $300 billion (accounting for the majority of the decline in M3). However, commercial paper (CP) outstanding—which has traditionally been the staple of money markets—only fell by about $45 billion. Thus, it's plausible that instead of investing via money markets, institutions are now just buying CP directly. Although it wouldn't seem to make much fundamental difference whether institutions are investing directly in CP or through a money market, it could have a significant impact on estimates of M3 since money markets are included in the metric, while CP outside them is not.
It's also worth noting that since the Federal Reserve started paying interest on bank reserves in October 2008, the flow of money through the financial system has changed dramatically—more broadly calling into question the value of normal money supply measures. Traditionally, US banks have kept their reserves at the regulatory minimum—since excess reserves held at the Fed paid no interest. However, with the Fed now paying a 25 basis-point "reserve rate," excess reserves have increased dramatically. (See Exhibit 2.) Although these excess reserves are included in the monetary base (M0), they do not flow into M1, M2, or M3. Thus, much of the monetary stimulus over the last 18 months hasn't impacted traditional measure of money supply in the same way it usually would. Of course, if the Fed reduces the reserve rate, this could change rapidly—and all of the broader measures of money supply could expand significantly.
Exhibit 2: US Bank Reserves—Required & Excess
Source: US Federal Reserve
Ultimately, while money supply and the velocity of money should be, in theory, positively related to nominal economic growth, M3 no longer appears to be a particularly valuable monetary measuring stick. And even if it were, it historically hasn't necessarily told us much about the direction of the economy.
Even a simple visual analysis prior to 2006 shows a downturn in M3 has been (at best) a coincident indicator of economic growth, and not a reliable predictor. The last time annual growth in nominal M3 turned negative, it was the first quarter of 1993—in the subsequent few years, US economic growth accelerated and the stock market rallied. (See Exhibit 3.) Conversely, in 2001, M3 growth accelerated through the recession, before decelerating through the robust expansion during the middle of the last decade. This suggests a decline in M3 growth—or even negative M3 growth—may still be followed by strong economic growth.
Exhibit 3: Year-Over-Year Change in M3 and NBER Recessions
*Although the NBER hasn't officially declared an end to the current recession, we've assumed the second quarter of 2009 (the last quarter of negative GDP growth) marked the end.
** Fisher Investments estimates of M3 are based on the sum of M2, institutional money market funds, large time deposits, and repurchase agreements. Although our estimates vary slightly from those used by Shadow Government Statistics (due to differences in methodology), the general trend is the same.
Source: US Federal Reserve, National Bureau of Economic Research
The recent dust-up over M3 is more likely a powerful example of the wall of worry. When pessimism is this potent, we refer to it as the "Pessimism of Disbelief." This is when investors (frequently aided by the media) interpret most all economic data as negative, no matter what the trend. For example, when money supply is increasing, investors fear higher inflation. But when it's falling, it's interpreted as a sign of imminent economic slowdown.
While a rigorous search will always find some economic indicators showing signs of weakness (even in the best of times), currently, fundamentals overwhelmingly point to a robust economic recovery in the United States. GDP has shown its biggest three-quarter growth streak in six years, and most key indicators are showing very positive trends. US corporations posted the biggest year-over-year gain in profits since the 1980s, business spending is surging, personal income is on the rise, global trade is picking up, and banks are more willing to lend—just to name a few. And all the while, inflation remains benign, interest rates low, and investor sentiment overly dour. Together, this is the recipe for a continued bull market.
Sources: Fisher Investments Research, Federal Reserve, Shadow Government Statistics, Thomson Reuters, National Bureau of Economic Research