Explaining the Drop in the Weekly Money Stock Measures

Thursday afternoon the Wall Street Journal came out with a startling headline: “US M1 Fell $16.2 B In June 29 Week; M2 Fell $36.2 B.” (See http://online.wsj.com/article/BT-CO-20090709-714875.html#mod=rss_Bonds.)

Looking at the H.6 release that comes out at 4:30 PM on Thursday afternoon the Journal reported correctly. The H.6 release is titled Money Stock Measures. The seasonally adjusted M1 money stock measure averaged $1,669.1 billion in the week ending June 22, 2009 and averaged $1,652.9 billion in the week ending June 29, 2009, a drop of $16.2 billion. Please note that in the two weeks previous to June 22, the M1 Money Stock measured $1,630.9 billion and $1,656.5 billion, respectively.

Thus, M1 rose by $7.0 billion in the week ending June 15 and by $26.5 billion in the week ending June 22.

In terms of the seasonally adjusted M2 series, the weekly average for the week ending June 22 was $8,385.4 billion and for the week ending June 29 the M2 Money Stock averaged $8,349.2 billion, indicating a $36.2 billion drop. We can note that for the two previous weeks M2 averaged $8370.0 billion and $8,385.2 billion, respectively.

M2 rose by $15.2 billion in the week ending June 15 and by $0.2 billion in the week ending June 22.

Now let’s see what happened to the non-seasonally adjusted data. The M1 money stock rose $29.5 billion in the week ending June 15, by $52.0 billion in the week ending June 22 and rose another $47.9 billion in the week ending June 29. These figures are significantly different than the seasonally adjusted series.

In terms of M2, this series rose by $16.4 billion in the week ending June 15 but it dropped by $74.4 billion in the week ended June 22 and dropped again by $49.0 billion in the week ending June 29. Again there are serious differences.

There are two points to make here. Formerly the Fed did not put out weekly data on the Money Stock Measures because they jumped around so much. Such volatility can be unnerving to people watching the money stock. Obviously, people at the Wall Street Journal reacted very strongly to the weekly release.

Second, trying to seasonally adjust weekly data is only for the foolhardy or for the very brave. The only conclusion I can draw from the behavior of both the seasonally adjusted series and the non-seasonally adjusted series is that a lot of “stuff” is going on and the seasonal adjustment process is doing very little to capture what is going on. That is, what we are seeing here is white noise!

Is there any clue to what might be happening to banking accounts?

The answer to this is yes, there have been things happening that might help to account for some of the swings and because of the uncertainty of exactly when these things happen from year-to-year their movements can “screw up” the seasonal adjustment of the raw series.

To see what might be happening in the banking system, I go to the Federal Reserve release H.4.1, “Factors Affecting Reserve Balances.” This release also comes out at 4:40 PM on Thursdays. The account I am particularly interested in on the Fed statement is the line item called U. S. Treasury General Account. This is the account that the Treasury Department pulls in tax money from the private sector and then pays it out to the private sector. It is the “transaction” account of the Treasury Department, the one in which the Treasury deposits tax money and the one which the Treasury Department writes checks against.

This account is a “Factor that is absorbing reserves.” That is, when the Treasury draws funds in from the private sector it removes reserves from the banking system. When the Treasury writes checks to the private sector and these balances at the Fed decline, reserves are put back into the banking system.

The Fed and the Treasury work hard to coordinate their actions because they don’t want to incur large swings in the bank reserves. So, what happens is that tax payments and such are kept in the banking system until the Treasury is about to write some checks. When it draws funds from the banks, private deposits go down and the Treasury balances at the Fed go up. When the Treasury turns around and sends out checks to the private sector, bank deposits go up and the Treasury balances at the Fed go down. The Fed then manages bank reserves so that there are few if any dislocations caused in the banking system due to these transaction.

What we have here in June is a buildup in balances at the U. S. Treasury General Account and then a draw down as the Treasury writes out checks. What the Wall Street Journal caught was the building up deposits in the Treasury account. This comes out in the releases up to June 29.

However, we have not yet see the affect of the Treasury checks going out because we don’t have more current data on the Money Stock measures. We do have data for the Treasury’s General Account for the banking weeks ending July 1 and July 8.

And what do we see?

In the banking week ending June 10, the Treasury account averaged $31.4 billion. The next week the account averaged on $42.3 billion, but the account AT THE CLOSE OF BUSINESS on June 17 was a whopping $132.8 billion. Most of the money was drawn from the banking system at the end of the banking week so that the average did not move much.

But, for the banking week ending June 24, the Treasury General Account balance averaged $118.7 billion reflecting the growth in deposits, but the account AT THE CLOSE OF BUSINESS on June 24 stood at $78.8 billion. A lot of money passed though this account in a very few days.

The U. S. Treasury General Account then continued its decline. The average balance for the banking week ending July 1 was $72.0 billion and the average balance for the banking week ended July 8 was $34.2. This latter figure was right at the level of the average balances in the account in the first two weeks of June.

So, as the Wall Street Journal reported, we saw a massive decline in both measures of the Money Stock in the week ending June 29. However, the swings were caused by operational transactions within the government and should be reversed out in the data that are released for the weeks ending July 6 and July 13. But we won’t see those data for another two weeks.

Bottom line: the money stock is not collapsing! Whew!