Global Rates

The global long end is doing the same thing everywhere this morning, which is almost nothing, and that is the first genuinely quiet session in a fortnight. The Bund is 3.2550%, up half a basis point and still near its highest since 2011. The 10-year JGB is 2.8950%, down half a basis point from a level last seen in 1996. The gilt is 5.0676%, up 1.5bp and comfortably above 5%. The US 10-year is 4.7050%, up 0.4bp. Four major sovereign curves, four moves of under two basis points, all verified at source. The policy legs moved in lockstep too, and by exactly the same amount. ECB September is 95.5c, up one point from 94.5c: a 25bp increase on September 10 is now effectively fully priced, with No change at 4c. BoJ September is unchanged at 87.5c for the 25bp increase against 12.5c for no change. The Fed's September meeting is 33.5c for a hike, up one point, with No change at 65.5c. Every leg one point higher or flat, nothing repriced, no new information. The divergence story we have been running for three weeks is now mostly a level story rather than a direction story. Europe and Japan are hiking and the market knows it, at 95.5c and 87.5c. The Fed is priced at one-in-three for September and about four-in-seven for the full year, at 56.5c. The three central banks are no longer disagreeing about the direction of policy; they are disagreeing about the timing, and the market has settled on Frankfurt first, Tokyo second, Washington last. That is a much less interesting divergence than the one we described in early August, and we would rather say so than manufacture drama out of a one-point session. Which leaves the correction, and it belongs in this letter too because pos-017 is the only position in the book with a genuinely global thesis. Yesterday we ended the equivalent section with a consistency check: the same barrier model applied to the 10-year 5.00% leg returned about 16 against a 16.0c market, therefore our sigma was not flattering the 4.8% leg we own. We could not reproduce it this morning. With the 10-year at 4.7050% the 5.00% barrier is 29.5bp, 92 business sessions remain, sigma is 38.37bp, and 2*(1-Phi(29.5/38.37)) gives 44.2% raw and 40 after the buyback haircut. The market is 17.5c. So we are freezing fair value on pos-017 at 72, not raising it to the 73 today's arithmetic offers, and we are refusing the 5.00% leg outright despite a nominal 22-point gap that is more than twice our entry bar. The reason is mechanical rather than modest: a fixed-volatility barrier model over-prices far touches because it assumes yields wander freely, when a 30bp grind to 5.00% in any of these four markets would summon exactly the response that stops it. We have seen three of those responses this month alone, the US Treasury doubling long-end buybacks, the joint yen intervention funded by selling euros, and the request to raise the Fed's FIMA limit. Our own 4.8% leg is 9.5bp away and the market broadly agrees with us at 67c; the 5.00% leg is where the model's tail assumption does all the work and we think the market's 17.5 is closer to right than our 40. The global corroboration for the position itself has not weakened. The long-end repricing is not a US fiscal quirk: Berlin is issuing 30-year paper at 2011 yields, Tokyo's 10-year sits at a 1996 high with a September hike at 87.5c, London is above 5%, and the US buyback window runs September 9 to November 4 while our barrier runs to December 31. The intervention expires before the position does. What has weakened is our confidence in the number we attach to that thesis, which is a different problem and a more embarrassing one. Yen at 159.43 against the dollar, unchanged in tone if not in level, with the Nikkei up 0.47% and the DAX up 0.42%. No trades in this book today. Timing disclosure: this letter went out at 10:25 Stockholm, before the US cash open and about six hours earlier than our usual slot, so every US price here is an overnight mark.