We have extensively discussed the size (here - must read!) and growth (here) of the Fed's largesse in soaking up massive amounts of the primary and second Treasury (and now MBS) markets with the ongoing theme of 'what about the exit strategy?' among other things. The onset of QEternity likely means the Fed's balance sheet will grow to over $4 trillion within the next year and, as UBS notes, although the Fed has suggested that it will not begin an exit strategy until 2015, the magnitude of the excess balance sheet argues for considering whether the Fed has the ability to unwind their balance sheet. We, like UBS, believe that the Fed will find it far more difficult to exit than they have found it to enter given the limitations of the exit tools frequently cited. There are three main tools for reducing the Fed’s balance sheet: asset sales/maturation (bad signaling), reverse repurchase agreements (size constraints), and interest on reserves (inflationary).
- Asset sales/maturation. The portfolio shift to a longer average maturity means that the Fed is unable to reduce its balance sheet only by letting securities mature. There would be no material reduction until at least 2016 and even then the reduction would likely be under $250bn. Outright sales face a different problem – expectations. Unlike purchases where announcing a certain amount of purchases reinforces the Fed’s goal of lowering rates via expectations, any sales would likely result in the market pricing in a fully normalized balance sheet. As such, an initial sale program of just $200bn would not be credible as the expectation would be that the sale announcement signals a desire to return to normality requiring an addition $2.5tn in sales at some point.
- Reverse repurchase agreements. This tool is swamped by the magnitude of the drain required. At present money fund assets are roughly $2.5 trillion, $200 billion less than the excess balance sheet we anticipate by the end of 2013. However, this statistic does not tell the full story as reverse repurchase agreements only make up just over 20% of money fund assets, or just $500 billion. The other primary counterparty the Fed would rely on, the Primary Dealer community, are unlikely to be able to participate in anything close to that size. While these figures do not prevent the Fed from using this tool, they do suggest it can only be a part of the overall solution.
- Interest on reserves. Although interest on reserves theoretically creates a floor on rates in the interbank market, it does not prevent banks from using funds to make loans. Loans eventually end up as deposits somewhere and, as such, the overall level of deposits at the Fed provide little guidance as to whether the funds are circulating in the economy. The only way to prevent lending would be for the Fed to raise this rate sufficiently to make banks prefer depositing the money at the Fed to lending to their client base.
QE3 and QE4, if enacted and continued until the end of 2013, will leave the Fed with excess balance sheet of roughly $2.7 trillion.
Fed Treasury holdings by coupon maturity date (2013-2042) expected at the end of Operation Twist at the end of 2012.
Money fund assets have been falling and the total amount of repos done by money funds does not appear sufficient to allow the Fed to rely solely on reverse repos from this financial segment to drain liquidity.