As we head for the fateful FOMC announcement on September 18, US data have continued to moderate. Accordingly, the consensus seems to be converging on a $10-15 billion initial reduction in monthly purchases (mostly focused on the Treasury side and less so on MBS) with any 'tightening' talk tempered by exaggerated forward-guidance discussions and the potential to drop thresholds to appear more easy for longer, since as CS notes, assuming Fed policymakers have learned anything in the last four months, they must know that the markets view “tapering” as “tightening,” even though they themselves for the most part do not. Thus, they are going to need to sugar-coat the message of tapering somehow. But as UBS notes, political risks have grown and there is little clarity on the Fed's thinking about the housing market. This leaves 3 crucial surprise scenarios for the FOMC "Taper" outcome.
On Size & Composition (via Credit Suisse),
On the subject of size, the trajectory of mortgage and Treasury issuance and outstandings is in our view the best and least-tortured reason for tapering. Mortgages outstanding are shrinking as household deleveraging continues to outpace new housing activity, and gross MBS production is now slowing as the refinancing pulse burns out. Meanwhile, the budget deficit is expected to be sharply-reduced for the next few years from the trillion-plus annual deficits of the last few years.
From the perspective of optics, at least, and possibly market functioning as well, $85bn per month of securities disappearing into the Fed’s vaults is simply too large. Indeed, in the case of MBS purchases, the Fed risks buying so much of the supply of particular issues that large market disruptions may re-emerge, as we saw near the end of the QE1 episode and for several quarters afterwards.
Taper Surprise Scenarios (via UBS),
Political risks have grown
Key Fed players have indicated time and again that a reduction in monetary accommodation is not a done deal. In addition to economic data, the Fed may consider political risks. We have discussed these risks at length in our recent publications, so we just highlight a few key items here. The debate over the US debt ceiling is approaching, the US government needs stopgap funding, the German election could roil markets, and the situation in the Middle East is dicey at best. Each of these issues could alter the Fed’s calculus on whether/when/how much to taper.
Too little clarity on the Fed’s thinking about the housing market
It is no secret that the Fed views the housing market recovery as a key to sustaining economic growth. In that context, the 100bp jump in mortgage rates must be worrisome to policy-makers, but we do not know how worrisome. Moreover, we have noted that the traditional buyer base for MBS seems to have lost its appetite. Consequently, the MBS market could respond erratically to a taper surprise.
Surprise I: delay
We think this scenario is the most likely surprise. Bear in mind that it is still a “surprise”, so therefore unlikely in our view. The Fed decides to postpone reductions until a vague date. It reiterates the policy that tapering is not a done deal.
What may cause it: A US political impasse is the probable catalyst. We recognize that a stand-off in Washington could well lead to a Treasury rally. The Fed might, in turn, see the rally as a chance to reduce its Treasury purchases. However, we suspect that the Fed would hold off tapering if Congress and the Administration are hamstrung. The Fed would want to appear apolitical, and initiating tapering could invite a lot of criticism from Capitol Hill.
Likely market reaction: US bull flattener. Five- and ten-year Treasuries and Agencies outperform, MBS tighten, US swap spreads widen initially, USD vol sells off sharply. Treasuries outperform Bunds and Gilts.
Surprise II: aggressive tapering plan
This scenario could take two forms. The obvious one is simply to reduce purchases by more than $15 billion. Second, the Fed could use aggressive forward-looking language to describe its future tapering plans.
What may cause it: Chairman Bernanke and others may push to accelerate tapering, in advance of considerable turnover slated for the Board early next year. The FOMC may ramp up tapering to put pressure on Congress. Mr. Bernanke and other officials have argued for months that monetary policy has its limits, and fiscal policy has to help bolster the economy.
Likely market reaction: US rates jump. The 7-10y part of the curve leads the way, but 20-30y catches on shortly thereafter. Off the run issues underperform OTRs. Gamma catches a strong bid. Treasuries lag Bunds and other high quality sovereigns by a wide margin.
Surprise III: mortgages bear the brunt of the taper
The broad consensus is that Fed will “go easy” on mortgages to avoid rocking the housing market recovery. Reducing MBS purchases as much as Treasury buys would be a major surprise to most investors (note above on size and composition)
What may cause it: the Fed appears to be much less comfortable owning MBS than Treasuries. Furthermore, numerous Republicans are apoplectic at the thought of the Fed maintaining a central role in the US housing market.
Likely market reaction: sharp underperformance of MBS, wider swap spreads in 5-10y area, initial 5y/10y steepener followed by 7y/20y steepener, as mortgages extend and convexity flows materialize. US gamma catches a moderate bid. Treasuries outperform Bunds, Gilts, etc.
On Pace and Sugar-Coating (via Credit Suisse),
A spoonful of sugar may ease the commencement of QE3 tapering, especially given the generally disappointing August jobs report. Assuming Fed policymakers have learned anything in the last four months, they must know that the markets view “tapering” as “tightening,” even though they themselves for the most part do not. Thus, they are going to need to sugar-coat the message of tapering somehow. We expect FOMC participants will insist that short-term interest rates remain anchored near zero for a long time, even after tapering is completed and net asset purchases have reached zero, and that reinvestments will continue for a long time to cover maturities and amortization.
Moreover, even after the short-term interest rate eventually in the future starts to go up, the Fed will not sell assets but rather will begin to allow assets to run off in the normal course of maturities and amortizations. We think the FOMC may also suggest that the new level of asset purchases announced at the upcoming September meeting might well be adequate for a time, signaling that there is no built-in expectation for a further step in October and perhaps not even in December. This could help the market absorb the first tapering shock and the uncertainties surrounding the identity of the next chairperson.
As Exhibit 4 indicates, we expect a $20bn taper in September, evenly split between MBS and Treasury purchases. Our baseline scenario includes another couple of $20bn taper moves in early 2014, at the January and April FOMC meetings, with a final $25bn cut at the June meeting to end the program around mid-year.
If our scenario holds, QE3 will have totaled just over $1.3 trillion, more than double QE2’s $600bn and a few hundred billion short of QE1, which expanded the Fed’s balance sheet by $1.725tn. Under these assumptions, the balance sheet will be nearly 50% larger by mid-2014 than it was when QE3 commenced in September 2012.
We have drafted an imaginary press release that embodies many of the ideas discussed above. Also, in the interest of shortening and simplifying the FOMC statement, we changed some of the arguably superfluous language that has been employed by the Committee over the past few meetings. As a consequence, Exhibit 6 below is one part forecast, and one part recommendation: