Fed Paper on Asset Purchases

The Federal Reserve recently published a noteworthy paper on the Large Scale Asset Purchase (LSAP) programs.  A little background: when the Federal Reserve reach the effective lower bound of traditional monetary policy in December 2008 (Fed Funds rate between 0 and 25 basis points), they had a strong conviction that the economy was in need of more stimulus than the ultra-low Fed Funds rate was providing.  Specifically, Fed staffers were looking to lower long term borrowing rates, a goal which cannot be achieved by manipulating the short end of the yield curve.  This forced the Fed to resort to unconventional monetary policy.  The desired effects were ultimately obtained through 3 LSAPs which, combined, totaled over $1.7 trillion.  The purchases were divided as follows: (1) $300 billion worth of Treasuries concentrated in the 2-10 year term (2) $175 billion in agency debt and (3) $1.25 trillion in agency mortgage backed securities.  As the latter two of these facilities are scheduled to wind down at the end of the month (the Treasury purchase program ended in October) the Fed published a timely paper discussing the execution and cumulative impact of the LSAPs on long term interest rates. 

The Fed figures that the impact of their purchases were twofold.  Initially, most of the impact can be attributed to increased liquidity in the targeted markets, thereby reducing the enormous liquidity premiums present in these markets in early 2009.  The second impact was what they termed the "portfolio effect", the mechanics of which are essentially as follows: by purchasing such an enormous volume of securities currently held on private balance sheets, the supply of said securities is meaningfully reduced, increasing their price and reducing their yields.  The purchases also create more reserves in the system, and force investors to turn to other markets in the search for yield, thereby bringing down interests rates (and borrowing costs) in a wide cross-section of markets.  The sheer size of these purchases is put in perspective as

"22 percent of the $7.7 trillion stock of longer-term agency debt, fixed-rate agency MBS and Treasury securities outstanding at the beginning of the LSAPs...We believe that no investor - public or private - has ever accumulated such a large amount of securities in such a short period of time"

I had not given it much consideration, but I was quite surprised by how large this proportion was.  Armed with this knowledge, I am (even) more willing to accept the conclusions of the study.  The qualitative discussion is brought to a close with the statement that, as a result of the portfolio effect, "the winding down of LSAPs need not cause a meaningful rise in market interest rates". In other words, the effects of these programs are seen as largely permanent (a point of much contention in the markets) by the Fed.

Also included are number of statistical analyses (primarily event studies) designed to determine the cumulative impact of the operations on long term yields.  A number of classes of securities are examined.  The results are as follows (for the baseline 8-event set): 2y Treasury, -34 basis points (bps); 10y Treasury, -91 bps; 10y agency debt, -156 bps; Agency MBS, -113 bps; 10y term premium, -71 bps; 10y swap, -101 bps; Baa index, -67 bps.  I was not surprised by the conclusions of their study - in fact I fully agree - but proving it statistically is no enviable task.  There is just too much noise.  All told, the discussion preceding the statistical analysis, as well as the charts (two of which are below) are definitely worth a read.  The statistical analysis, on the other hand, was a bit of a slog.