Curve jumps again as BOE fails to show
If people hadn’t got the message yesterday, then they have today. The BOE is not going to feel obliged to buy gilts just because GEMMs (and maybe their clients) offer them up for sale. As one seasoned gilt trader said this evening in London, “today nothing was bought, yesterday £20m or so was bought, last week the BOE was buying £1bn or so per day, so it’s nowhere close to the potential £5bn that it could be.”
“I thought it was interesting that the Chancellor approved £100bn for the BOE to buy but the Bank decided to cap it at £65bn. Now the BOE is not even doing that and it’s a tricky thing for GEMMs to manage because if you structure client trades on the premise that the BOE will be there to buy from you and they aren’t… well it’s quite tricky,” said the trader.
On the other hand, he said that by recent standards, the gilt future today traded in a fairly narrow range of about 98.4-100.9 ticks, and while today did see around £2bn of mostly 2044 gilt offers, he said “the subsequent sell-off was not very significant. 10y gilts sold off 10bps (after the results of the reverse auction but had already rallied 20bps.” The 2044, already suffering from a sense of double-rejection, did less well, selling off by around 12bps after the non-buyback to end 13bps higher on the day, at 4.18%.
In short, said the trader, “this is increasingly obviously not about QE but about promoting stability. If the market is both functioning and also rallying, as it more or less was today, then there is no need to buy. But if it sells off too much you know they’ll be there, so it does help to reduce volatility a bit while the market returns to something like functionality.”
Looking at the curve, the trader said that while 50y gilt (in the form of the 2071) price action has caught the eye since last week’s sell-off peaked with the BOE intervention (it has rallied about 100bps in yield terms and 35% in price since the BOE announcement but is up 50bps from yesterday's lows), the move, and 30s/50s flattening, has been in line with the broader market.
“If you look at the 30s/50s curve, it looks pretty well correlated with 10s/30s, flattening in line with the huge flattening in 10s30s. They are possibly both too flat but they are not out of step with each other,” he added.
Still, at the close today 30s/50s gilt only dis-inverted by about 6bps to -51bps after the non-buyback but 10s/30s jumped by 24bps, its largest daily move ever, according to Bloomberg, on the heels of a 20bps increase yesterday.
Huge ASW moves were dominant in the long end, driven by the late long gilt sell-off. The 30y ended the day -10bps at -35.5bps, while 10y was -1bps at 20.8bps.
Summing up the mood of the market today, the above trader said “this week is less mad, relatively calm… the market’s recalibrated to 20-30bps intraday moves in the gilts, and is hoping they narrow from here.”
The BOE even decided not to buy when the DMO was selling. Today’s quite lumpy £2.5bn tap of the 2061 gilt picked a fairly lively day to take place, but once again, the DMO managed to get supply out there in the most challenging of market circumstances. It attracted a respectable 1.97 times bid/cover and came at an average 3.371% yield, before BOE inactivity saw it sell off a further 23bps into the close to reach 3.60%.
RY steepens with longs ‘settling’ in positive territory
Offering a brief synopsis of the state of play in linkers today, one long-suffering RPI trader said that “today’s move was more of a real yield move but breakevens steepened led by RY steepening more than the nominal market.”
He said that “the inflation curve is steepening more than nominals because there’s been a lot of pricing-out of more aggressive hikes in policy because of the government shift on tax cuts. In the long end IL52 RYs are now at +35bps versus -10bps at their lows yesterday and the whole curve is shifting back up to positive from the 30-40y area.”
The busy inflation trader described the market as “thin and very choppy” again today, with little sign of a return to LDI buying in the long-end. Cash breakevens narrowed by around 10-20bps out to 20y.