Follow by Email

Wednesday, 18 December 2013

Fed to market : If you think we screwed up last time, just watch this...

It's interesting to compare the situation now with 2003/2004. If the drop in unemployment persists as it has for the last 3 years, we'll reach 6.3%, the level of peak unemployment in the dot com recession, around the end of next year :

How should eurodollars be priced if the Fed subsequently reacts exactly as it did last time ? Well, they made their final cut to 1% in Jun03 as unemployment peaked, kept rates on hold for a year and then hiked in a straight line by 425bps over the following 2 years. Like this :

Matching the two dates of identical 6.3% unemployment suggests rates stay at 0.25% till Jan16, then rise to 4.5% by Jan18. Allowing 30bp for the funds/Libor spread, that gives 95.20 (4.8%) for Dec17 euros. EDZ7 currently trades at 96.80, ie 160bps too high (interestingly that's exactly where I'd expect the next selloff in EDZ7 to stop from a technical perspective : down 200bps from 98.10 to 96.15 over last summer, a 100bp bounce to 97.15 this autumn and then another 200bps getting you to 95.20-ish. Funny coincidence.).

Of course, there are some differences :
- Fed Funds are 75bp lower now than in 2003 : easier policy now
- the Fed is saying they'll keep rates low for longer than Jan 2016 : easier policy
- the Fed has ~$4trn of bonds on its balance sheet : easier policy
- the stock market is at an all time high instead of recovering from a crash : easier financial conditions
- GT10 is 3% instead of 5% : easier financial conditions

One thing that isn't very different is inflation. Here's core cpi :

Inflation is actually higher today than in 2003, so real rates are even easier than nominal rates. In fact current 10yr break-evens are almost identical now to their levels in 2003 as well.

Obviously the Fed made a colossal mistake in 2003-2007. They kept rates too low for too long and inflated a monumental bubble which nearly destroyed everyone when it burst. So how can the current forward guidance, committing them to even lower real rates for even longer, be credible ? It all comes down to inflation I think. If the core cpi numbers turn as they did in late 2003, then the Fed will start hiking rates soon. Within a few weeks everyone will be dusting off charts like these and drawing the obvious conclusion that the steady-as-she-goes sequence of rate hikes from 2004-2006 is a better-than-best case scenario, and that to prevent the risk of bubbles they really ought to tighten even faster. Which all points to EDZ7 and its friends being 200bps or more lower than they are now. 95.20 on EDZ7 strikes me as a best-case scenario.

Personally I think there's a significant chance of this realisation dawning on the market in Q1, as we post stellar growth numbers and everyone races to upgrade their forecasts against the backdrop of tapering. Low delta puts on the golds are the obvious risk/return trade. Ideally equities will fail again here at 1810 on spx, giving a risk-off trade with a great risk/return to set against the rates trade. The combination could be the trade of the year if it lines up.


  1. Good analysis. Already short gold and treasuries and like your idea of shorting 2017 eurodollars, but think q1 is too early to short us equities or go long USD. Yellen isn't going to start backtracking on fwd guidance until it's a blatant lie, so I think q2 or 3 is when the cookie will crumble.

    1. Thanks. I temporarily covered my gold short (I'd like a squeeze to re-enter) and got stopped on the ESH4 short too. Perhaps I'm too early on stocks, but they feel overbought, investors are v bullish and company announcements seem to be turning less positive. I'll try it with a tight stop a couple of times anyway. Selling the middle and back of the Tsy curve (and avoiding a head-on fight with Yellen) is the trade that's working at the moment obviously though !