Frontrunning The Fed: Weekly Update

Now that Morgan Stanley has officially thrown in the towel in fighting the Fed (and, paradoxically, deflation), and has abdicated the steepener bandwagon (which can only mean it is time to start selling bonds), the firm's rates analysts are once again focusing on what they do best: provide advice on how to front run the Fed. As we pointed out, Igor Cashyn was virtually spotless in his first two attempts to predict just which Treasuries the Fed was going to purchase. We present his latest cheat sheet for the upcoming POMO operations for Tuesday and Thursday of this week, and for the September 1st operation. In a nutshell for those who want to pick some virtually 99.99% risk-free money, here are the suggested trades: Buy 1.375% 02/15/2013 / Sell on-the-run 3y - this positions investors for the August 24 Fed operation; Buy 8.0% 11/15/2021 / Sell on-the-run 10y - this positions investors for the August 26 Fed operation; Buy 1.375% 01/15/2013 / Sell on-the-run 3y - this positions investors for the September 1 Fed operation.

But before we get into the details of these particular trades, we present Jim Caron's full mea culpa for the firm's persistently wrong stance on Treasurys, driven almost entirely by its unfailingly bullish economic outlook, and its view that stocks are massively underpriced. So much for these two myths.

We got our rates call wrong this year and missed a great opportunity to be long. The market is currently rife with relative value opportunities and that’s what we will focus on going forward. Previously, we did not give enough weight to a scenario of sub-2% growth in 2H10, a threshold, if reached, we  think could generate a deflation scare that might bring the UST 10y into the 2.00%-2.25% range. This was wrong and we are adjusting for that.  Should such a move occur, it might be led by a decline in inflation risk premiums. In addition, we can’t help seeing the Fed’s surprise decision to buy longer-term Treasuries as a risk for further actions to keep rates lower and avert deflation. We’re shifting gears and will become more tactical, playing for rate moves in either direction, rather than having our ideas hinge on longer-term macro themes. But first, let’s review what we got wrong:


What Went Wrong with Our Rates View

  • We thought the rise in rates would happen earlier in the year and trigger selling to hedge the substantial duration risk in credit bond portfolios.

We got this wrong. ICI data indicates that a record-breaking $186Bn has poured into bond funds this year despite the decline in yields. Lower real growth risks trumped economic recovery prospects.

  • We thought a recovery in equities would force people out of bonds to chase performance in equities. Instead people remained in bonds.

Equities maintained a steady pace of outflows. Investors cared less about cheap equity valuations and more about capital preservation in
fixed income products.

  • We were optimistic on the economic recovery; we thought the market would thus start to price an increase in private credit creation, which in turn would crowd out UST bonds to higher yields.

This never materialized. 2Q growth came out much weaker than we expected and set the growth trajectory lower as we head into 2H11. Private credit creation remains impaired and the risk of crowding out USTs is very low, given there was such low net issuance in other products.

  • We thought the above mentioned economic recovery would spur higher inflation expectations that would cause nominal yields to rise.

Rising inflation expectations fell by the wayside as the market became consumed with low-growth expectations.

  • We thought the Fed would be more likely to start hiking rates in 2H10.

The growth we were expecting did not materialize, and the market has pushed rate hikes into 2012.

  • We expected the unprecedented amount of UST supply to weigh on bonds.

This never materialized. In fact, bond funds bought more USTs as the year went on because they started from an underweight UST position.
Again, this goes along with the record inflow into bond funds.

Those who wish to read more, can do so at the link below. In the meantime, here are more details on the only sensible trade left - frontrunning the Fed.

What was noteworthy about the Fed’s operations this week was just how much of the purchases appeared to be driven by relative value on the curve. Exhibit 5 is a list we published in our last week’s publication of the top ten cheapest bonds the Fed is likely to target in its two operations this week – with the cheapest bonds appearing at the top, updated the day before each operation:

What’s noteworthy right away is that the top three cheapest bonds according to our Treasury spline accounted for 71% and 75% of the Aug 17 and Aug 19 buybacks, respectively. In both cases, the top two cheapest bonds on our list also witnessed the largest single allocations of any bonds for that respective buybacks. Therefore, it would appear to us – and this was also the case in 2009 – that the Fed is targeting bonds that appear cheap on the curve. As long as the SOMA allocation of a bond remains well south of the 35% SOMA limit, we would anticipate that those bonds which appear the cheapest to be targeted by the Fed in the future. Exhibit 6 provides an updated list of the cheapest bonds within the other sectors that the Fed will be targeting in the future.

One other thing that was also interesting is that in its 2016-2010 sector buyback on Thursday, the Fed did not buy any original-maturity 10s as it solely targeted the original maturity 7y sector. This is a bit different from 2009, when original-maturity 10s accounted for 35% of the Fed’s buybacks in the 7y sector (Exhibit 7):

The way we make sense of this is by comparing the OAS asset swap valuation of original-maturity 7s on the curve in 2009 versus today, where we find that originalmaturity 7s were not notably cheap on the curve in 2009 as they are today (Exhibit 8). Therefore, the Fed was not particularly inclined to target only the original-maturity 7y series, and consequently mixed in some original-maturity 10y series in 2009 as well. Today, however, the original-maturity 7y series stands out as outright cheap on the curve, and therefore it would make sense for the Fed to continue to target this series until it richens up on the curve versus the original-maturity 10y series. Applying this logic to the Aug 24 operation in the 2013-14 sector, we would therefore expect the Fed to focus exclusively on original-maturity 3y paper.

As a way to position for the next set of Treasury buybacks, we recommend investors use the relative ranking of which bonds we think are the cheapest per Exhibit 6 to:

  • Buy 1.375% 02/15/2013 / Sell on-the-run 3y – dv01-weighted (and TED hedged if investors are able, to reduce the curve exposure of this trade and take advantage of the recent cheapening of this point versus the 3y on asset swap); this positions investors for the August 24 Fed operation
  • Buy 8.0% 11/15/2021 / Sell on-the-run 10y – dv01-weighted (all on asset swap in investors are able, to reduce the curve exposure of this trade);  this positions investors for the August 26 Fed operation
  • Buy 1.375% 01/15/2013 / Sell on-the-run 3y – dv01-weighted (and TED hedged if investors are able, to reduce the curve exposure of this trade and take advantage of the recent cheapening of this point versus the 3y on asset swap); this positions investors for the September 1 Fed operation

In terms of the second trade above, Exhibit 9 – top – demonstrates where the cheapest Treasuries are for the 10-30 year sector buybacks (highlighted with red circles), with the Nov21s being the left-most red circle. We can see that in 2009, of the sectors the Fed targeted was the 2026-2028 sector, which appeared quite cheap on an OAS asset swap at the time, and is yet another demonstration that the Fed was sensitive to relative value in executing its buybacks (Exhibit 9 – middle / bottom).


As always, the risk to the above trades is that our rankings may deviate from those of the Fed in terms of which bonds the Fed ends up purchasing. This will keep the cheap bonds we target trading cheap on the spline. However, as Exhibit 5 demonstrates, the MS spline hit ratio for the top three cheapest bonds has done well so far, and we would consider these trades attractive regardless of the demand from the Fed.

So why waste time on stupid things like fundamental analysis and jumping on board hedge fund hotel trades (as we will show later there are 185 hedge funds invested in Apple): go for the sure thing, which is to do what the Fed does: rape and pillage future generations of Americans, just so Wall Street can make some more extra bonuses this year, before it all comes crashing down.

Full note here.