By Jonathan Weil
In a feat that would seem to defy the odds, Goldman Sachs,
JPMorgan Chase and Bank of America this week each said its
trading desk made money every day of the first quarter. Goldman
said its daily net trading revenue topped $100 million 35 times
last quarter out of 63 trading days. JPMorgan and Bank of
America disclosed similar eye-popping stats. Citigroup, too,
recorded a profit on each trading day, Bloomberg News reported,
citing unnamed people who knew the results.
The intrigue is high. If a too-big-to-fail bank’s traders
were able to make money every day of a quarter, were they really
trading in any normal sense of the word? Or would vacuuming be a
more accurate term? What kinds of risks do such incredible
profits entail, for the banks and the rest of us taxpayers? And
are results such as these too good to be true?
There seems to be no satisfying way to answer those
questions, or even the more basic inquiry: How exactly do these
banks’ trading divisions make money? Reading the companies’
impenetrable financial reports is of little help. However they
did it, the data suggest it was as easy last quarter as hitting
the side of a barn with a baseball from three feet away.
This isn’t the way “trading” works in the real world. A
simple exercise in measuring probabilities is instructive here.