Unlock the Editor’s Digest for free
Roula Khalaf, Editor of the FT, selects her favourite stories in this weekly newsletter.
When the Bank surveyed market participants back in September 2022 as to the likely past and present impact on 10yr gilt yields of selling down its stock of bonds — so-called quantitative tightening — the 40 respondents collectively reckoned maybe 14–35 bps.
We had to wait eight months before Dave Ramsden (then deputy governor) communicated the Bank’s own estimates: a more modest zero to 10 basis points.
Since then, a host of academics and market-types have come up with their own guesses — some backed by hundred-page peer-reviewed pdfs, some not. Also since then, Alphaville has published a succession of posts gently ribbing the Bank of England for undershooting pretty much everyone else’s estimates as to the impact of QT on market yields, despite periodic upward revisions.
But no more. Yesterday, as reported by MainFT, the Bank published a new estimate as to QT’s impact on gilt yields:
Some content could not load. Check your internet connection or browser settings.
No longer an outlier, the BoE reckons that the total increase in 10-year gilt yields from cumulative QT comes to 🥁 🥁 🥁 20–30 bps.
Sure, the latest red dots still look lower than a lot of the other dots on the chart. And, sure, this is just the latest in a series of estimate revisions over recent years. Moreover, almost all the blue dots on the chart show ranges from academic studies that the Bank cites as being “broadly consistent with Bank staff analysis” look a lot higher.
But, in fairness, the ranges we show for the academic studies have a bunch of caveats that don’t fit neatly on the graph. Joyce & Lengyel 2024, for example, reckon the impact on 10yr gilts would be around 20 bps in an unstressed environment, and 100 bps only in a stressed one. And Du et al 2024 warn that their estimates for the cumulative impact on government bond yields ranging from 44–70 bps across horizons of one year and longer “should be interpreted with caution as they are sums of insignificant coefficients with large error bands”.
Importantly for us, the midpoint of the new BoE estimates, 25 bps, is pretty much bang in line with the midpoint of the interquartile range from that survey of market types back in September 2022 (24.5bps). Victory goes to the wisdom of crowds! 🏆 (For some reason, the BoE never repeated the question via MaPS after September 2022.)
So how much more has QT cost HM Treasury in the form of higher yields than the BoE’s initial estimates? We’re not sure that this is a question to which we can do much justice on a Friday afternoon. For one thing, it might be totally moot.
If, for example, higher bond yields have informed the Monetary Policy Committee’s decisions to keep Bank Rate lower than it would otherwise be, higher gilt costs could have been more than offset by lower costs payable on the asset purchase facility’s short-rate obligations. And who really knows how the UK monetary policy transmission mechanism works in sufficient detail to make a call as to whether counterfactually lower short rates might have boosted the UK economy, and with it tax receipts — the proceeds of which may swamp any additional gilt interest costs?
For another thing, any impact from QT on term premia may fade over time. The BoE thinks impacts from QE start to fade after one to four years. If QT works a bit like QE, but in reverse, the inflation of gilt term premia may soon be yesterday’s news.
But pretending that we live solely in a world of cash flow costs, of constants, and where everything else is always equal, we can try throwing the difference between the current BoE midpoint estimate of costs to HMT, and the initial BoE midpoint estimate through the Office for Budget Responsibility’s 2023 ready reckoner. The result, for the fiscal period 2023-24 through 2026-27, comes to £3.4bn.
Would foreknowledge of this higher fiscal cost have changed anything? Honestly, we have no idea.
Further reading:
— How the Iran crisis undermines the Bank’s QT programme (FTAV)
— How much is QT driving up UK borrowing costs? (FTAV)
— When active QT makes sense (FTAV)
— When a ‘modest increase’ costs £16bn (FTAV)
— Britain’s quantitative tightening will hurt us for a long, long time (FTAV)
