A CASE STUDY ON THE FED’S PERMANENT OPEN MARKET OPERATIONS
Courtesy of The Pragmatic Capitalist
On Friday I posted a story highlighting the market’s outperformance when the Fed performs its Permanent Open Market Operations (POMO). POMO is nothing new for the Fed so a one month data set is really nothing more than datamining. If we look back at the data over the course of the last 5 years we obtain a much more realistic (and potentially disturbing) perspective of the market’s performance on days when the Fed performs its POMOs.
Since October 2005 there have been 205 operations. On the day the operation was performed the market finished negative 41% of the time, positive 53% of the time and finished flat 6% of the time. The total return on these days was +27.28%. This is equivalent to a +48.6% annualized gain. A look under the hood provides a more useful perspective on the data, however.
Of the days that were positive 63% of the total gains occurred on just 3 days in March 2009. If we remove these three days the total gains equal +9.98%. This is equivalent to a +18% annualized gain. If we remove the best AND worst three days from the set the total return surges to 18.1% or a 33.1% annualized gain.
Perhaps the most interesting perspective in all of this is just looking at the market’s long-term performance when the Fed is conducting these operations. As you can see below the market has performed dramatically different when the Fed is conducting POMO’s. The Fed ceased POMO’s in May of 2007 after a fairly steady schedule. The market declined almost 20% in the following year and a half. They did not initiate the program again until September of 2008 when the economy was melting down. Technically, the program began on September 19th 2008 just days before the Lehman crash. This skews the beginning point of the credit crisis set of operations enormously. If we take that exact starting point the market fell -2% between then and the last operation on March 24th 2010. Of course, one could easily argue that the September 2008 operations were largely useless as the market was already in meltdown mode.
Between March 24th and August 17th of 2010 when the program was halted the market declined -6.5%. Since restarting the program in August the market has risen 8.3%. The following figure provides a visual of the operations and their (potential) market impact:
For the period of POMO’s from 2005-2007 there were 50 operations and the market advanced a grand total of +0.35% on these days. Since the market collapse, however, there have been 155 operations and the market has advanced a total of +26.95% on these days. So, the outsized returns could merely be a function of coinciding with one of the greatest bull markets in history.
What’s so interesting about all of this is the real world impact, however. These operations don’t alter net private sector financial assets. Therefore, it’s really just asset shuffling. The Fed is not printing new money when it conducts these operations. They’re simply asset swaps. They don’t add to the private sector’s income, they don’t create jobs, they don’t make the economy better off (aside from a highly debatable and marginal interest rate effect). There is, however, an obvious argument that there is a high correlation between market response and POMO’s. So while there is no reason to believe that these operations actually make us all better off there is considerable evidence supporting the idea that these operations correlate with periods of assets being “higher than they otherwise would be” – in other words, assets tend to be disconnected from their fundamentals during these Fed operations.
I’ll be honest with the reader. When I ran this data I was really hoping that I would find evidence showing that the POMO’s have no impact on market direction. The conclusion is unsettling for obvious reasons. And while this might be nothing more than a case of datamining the evidence is convincing that the Federal Reserve is helping to boost equity
