Today I've taken some guff from folks that are a bit upset about my skepticism about the Fed's estimates of the money supply. Its not that big a deal some said... banks know what they have on deposit.. say others.
oh... oh well that's comforting. Well then I shouldn't be bothered by the fact... that the FED steals ideas on estimation... from freaking biologists.
What follows... comes directly from a fed white paper on currency over-seas. The first crazy thing we learn when reading this paper... is that the Fed has no damned clue how much US currency is actually out there... or where it is. Are you getting this? They don't know where the damned money is... or how much is gone... and apparently... they think this is one good way to try to take a guess at it. I give you... The Fish Method:
"The biometric method, also known as the “fish” method, applies a method developed by
Petersen (1893) to estimate fish populations to cash processing data to obtain estimates of the
“populations” of notes in the United States and the rest of the world. In the biological
application, populations are estimated by capturing some animals, tagging them, releasing them,
and then recapturing another sample of animals later. Assuming that both samples are
representative, the share of tagged animals in the general population should be the same as the
share of tagged animals in the second sample, and the population can thus be estimated. More
formally, suppose M animals out of N total are captured and tagged. Next, suppose that in the
second sample, m tagged animals are found out of n captured. Assuming that both samples were
representative, the share of tagged animals in the second sample, m/n, should be equal to share of
all tagged animals, M, in the general population, N, or m/n = M/N. Since M, m, and n are known,
N can be estimated as N=(n/m)*M"
Seriously... these people are supposed to be traders. They are supposed to have degrees in economics... which... oh look... is a science. That means... they should've gone to school... with other science majors... where they should've learned... that biologist suck at math. People! You're better off getting help with your differential equations assignment from hungover elementary ed majors.
But hey.. that's ok... they have another method... which they call the seasonal method... which apparently works great... as long as you assume there isn't much US currency in Canada. No... no I am not making this up... and I did not rip it out of Python skit... this is actually in the white paper... which from what I can tell is not actually a parody.
Modern economists are so bad at modeling... they are actually stealing ideas from biologists.
That should be the punchline of a nerd joke. Not reality.
I want to close by quoting the paper directly again... from the conclusions...
"In sum, much as in earlier work, the currently available data do not allow for precise
estimates of foreign holdings of U.S. currency, and the available estimates are somewhat