EFTA Document EFTA01362019

4 September 2015 US Fixed Income Weekly Excess liquidity indicator vs. output: OECD !Excess 30% 40% -Global excess iipuldity yin +1 m lead OECD 20% —OECD output momentum / 30% 10% 0% -10% -20% -30% -1 -20% 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 soya arocentregrepanne elagt liquidity indicator vs. output: China —China excess liquidity yoy +12m lead —China output momentum China 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 Soiree tibesibeig end Dant** Pint The more sinister undercurrent is that as the relationship between negative rates has tightened with weaker liquidity since the crisis, there is a sense that policy is being priced to "fail" rather than succeed. Real rates fall when central banks back away from stimulus presumably because they "think" they have done enough and the (global) economy is on a healing trajectory. This could be viewed as a damning indictment of policy and is not unrelated to other structural factors that make policy less effective than it would be otherwise -- including the self evident break in bank multipliers due to new regulations and capital requirements. Of course our definition of "failure" may also be a little zealous. After all why should equities always rise in value? Why should debt holders be expected to afford their debt burden? There are plenty of alternative viable equilibria with SPX half its value, longevity liabilities in default and debt deflation in abundance. In those equilibria traditional QE ceases to work and the only road back to what we think is the current desired equilibrium is via true helicopter money via fiscal stimulus where there are no independent central banks. One step at a time... Gm I 2 y2 v as a carry tiflic'ent flattener We recommend a 6mly-2y2y flattener as an optimal carry proxy for USD 2s5s. The 6mly-2y2y can be thought of a leveraged version of the 2s5s spot: it has a 98% correlation and a beta of 1.74 with the latter over the last 12 months. Because of the 1.74x leverage, the beta-adjusted 3m carry is -2.0bp instead of -2.9bp for 2s5s, a 31% improvement. This flattener takes advantage of a recent 2.3 standard deviation decline in the (negative) roll for the 6mly paying leg, which compares to a 1.7 standard deviation decline in the 2y spot. The positive roll for receiving the 2y2y is also more attractive; it had just a 0.5 standard deviation reduction compared to a 0.7 standard deviation reduction in the 5y spot. Historically, 2s5s flatteners have performed well going into a tightening cycle, with nearly 70 percent of trades put on within three months of the liftoff beating their ex-ante forwards and thus being profitable. The market clearly thinks Friday's mixed jobs report was not enough to take a September liftoff completely off the table. DEC15 Fed funds future sold off 1.5bp after payrolls, and the implied probability of hiking in September rose slightly from Thursday to 34% at the time of writing. The 2s5s slope also flattened 2.5bp to 71.5bp, but still remains 7bp+ above its 2015 lows. A policy error by the Fed (i.e. hiking more than once this year in spite of declining global liquidity and falling inflation) can easily flatten 2s5s to 50bp or below. The risk to this trade is if the Fed relents in September but we think they will more likely than not do a Deutsche Bank Securities Inc. Page 15 CONFIDENTIAL - PURSUANT TO FED. R. CRIM. P. 6(e) DB-SDNY-0051316 CONFIDENTIAL SDNY_GM_00197500 EFTA01362019