I'd like to add the money supply to the list.

The money supply numbers that are released each week by the Federal Reserve are crude estimates. They rely on accounting definitions that may be incorrectly applied by various banks. And the procedure the Fed uses to seasonally adjust the data is imperfect at best.

In other words, we should think twice -- or even thrice -- before reading too much into that data.

No, this column is not a repeat of the one I wrote a couple of months ago, in which I pointed out that the money supply is affected by many factors outside the Fed's control. (Read archived column)

That point remains valid, and we should keep it in mind as we consider the arguments many advisers make about how the money supply reveals the Fed's true intentions.

Instead, my column today raises questions about the data themselves.

One of those questions deals with how the Federal Reserve seasonally adjusts the data. This is crucial, because the seasonally adjusted data often paint a different picture than that painted by the unadjusted data. Yet I found almost none of the advisers who regularly pontificate about the money supply's meaning had even a clue about how the Fed goes about seasonally adjusting the money supply data.

Based on the seasonally adjusted data, for example, there was an explosion earlier this year in M3, which is the broadest definition of the money supply. That growth rate received lots of attention, and was taken as proof positive that the Fed was being particularly accommodative.

On an unadjusted basis, however, M3's growth appears to be have peaked in March, as illustrated in the accompanying chart.

There is another problem as well with the seasonally adjusted data, according to Madeline Schnapp, a senior research analyst at TrimTabs Investment Research, and editor of the TrimTabs Personal Income newsletter: The Fed is prepared to revise the seasonally adjusted data for years after the fact.

Earlier this month, for example, according to Schnapp, the Fed revised its seasonally adjusted money supply data all the way back to 1998, more than six years ago. So how can advisers be particularly confident in drawing conclusions based on the latest weekly release?

Schnapp has another reason to question the widely held belief that the money supply is growing at an abnormally high rate: She suspects that certain relatively risky items on banks' balance sheets - known as Variable Interest Entities, or VIEs - are erroneously making their way into the definition of M3. If those items were removed from M3, then it would be growing at what Schnapp calls a more "normal rate."

Schnapp hastens to add that she doesn't know for sure that VIEs are being counted. But she is "highly suspicious." Indeed, she writes that she suspects "the Fed is aware of the distortions that these entities cause and is attempting to devise some statistical mechanism to adjust them out of the data."

This picture that emerges doesn't inspire a lot of confidence in the data, does it?

Sausage anyone?