Unlock the Editor’s Digest for free
Roula Khalaf, Editor of the FT, selects her favourite stories in this weekly newsletter.
FT Alphaville provides news, analysis and occasional snark. Anyone looking to find how high Japanese government bond yields are would normally be well advised to check Bloomberg, LSEG, FRED, TradingView, MarketWatch, or even the FT’s own Markets Data page. But not here.
HOWEVER, in the course of market browsing, the Japanese curve struck us as so weirdly kinky that we went down a rabbit hole and learnt something that only Alphaville readers (those who don’t already know, at least) might appreciate: while Japanese government bond yields have soared, they haven’t soared quite as high as they look like they’ve soared. Or, at least, 15-year benchmark Japanese government bond yields haven’t.
Here’s the Japanese government bond yield curve — a snapshot of the yields you might lock in if you bought bonds of varying time to maturity:
Some content could not load. Check your internet connection or browser settings.
The 15-year JGB yield just makes our eyes hurt. Not because a yield approaching 4 1/4 per cent is pretty high, but because it’s higher than each of the 20-year, 30-year and 40-year JGB yields. In a steeply upward-sloping yield curve supposedly freaking out about inflation and risk premia more generally, that is deeply weird.
We’d expect the yield on the 15-year to sit happily around halfway between the 10-year and the 20-year. And grabbing the historical data, this looks like what it has done for the past pretty much ever. Until now:
Some content could not load. Check your internet connection or browser settings.
What’s going on? Sometimes the maturity of whatever bond happens to be selected to be as the “generic” government bond of the moment isn’t actually that close to the generic maturity. Could the 15-year JGB be an off-the-run 19-year JGB in drag? Not this time. The actual bond currently serving as 15-year is the JGB 0.5% Sep-2041, which matures pretty much as close to 15 years from now as we might reasonably hope.
Maybe it’s a natural habitat thing? Domestic banks tend to buy the shorter end while life insurers own the long end. Could this have left the 15-year as Billy-No-Mates? While this might be true, it’s surely not new.
Chris Scicluna, head of research at Daiwa Capital Markets Europe, told us that the MoF cut new issuance at the super-long end of the curve at the start of the fiscal year. This is new! And we can see how that might richen the 30-year versus the 15-year.
And Masayuki Nakajima, senior strategist at Mizuho, told us that life insurers have been switching out of low-coupon low-price 15-year bonds into newly issued super-longs as a way to book large unrealised losses that offset the accounting impact of humongous equity gains. This is also vaguely new(ish). And we can see why this too might sting the 15-30-year bond spread.
Which all suggests that the 15-year JGB yield can perhaps be understood as the actual clean and honest price of long-dated JGB risk, unflattered by forced buyers, tax-optimisers and clever-clever government debt managers. If so, you can see even more clearly why Scott Bessent might be worried about rising JGB yields.
But then we got in touch with Mike Riddell, a fund manager at Fidelity, for his take. “You have to be wary of generic JGB indices.” Yeah, yeah — maturity mix-ups, right?
No.
“For some bizarre legacy reason they are usually shown using ‘simple yield’, not compound yield like everyone else,” he told us.
😳😳😳
To be clear, this is stuff you learn in bond markets 101. But it is sufficiently angels-on-pinheads on a day-to-day basis that it’s easily forgotten while scanning the screens. Let’s un-memory-hole it.
A simple yield is just what it says on the tin: simple. To work it out you first sum the coupon rate and the pull-to-par (redemption value minus price, divided by number of years to run), and then divide the whole thing by the market price. This is far, far easier than working out the yield to maturity, a metric telling you the total return you can expect if you hold a bond, erm, to maturity, for which there is no closed-form formula.
We’re not entirely sure why the Japanese market converged on simple yields as standard. We can see that back in the days before Excel, when a basic bond calculator cost more than a top-of-the-range MacBook does today, any measure that could be calculated on the back-of-an-envelope and that usually provided a decent approximation would be pretty handy.
But when you do the work to match every bond that sat behind each Bloomberg generic JGB benchmark every month back to 2014, and then recalculate yields as proper grown-up yields to maturity in Excel, you can see quite how divergent the approximation has become from what the rest of bond-world calls reality:
Some content could not load. Check your internet connection or browser settings.
The same disconnect is not apparent at other points of the long end. This is because benchmark bonds for the 10yr and 30yr tenors are in a near-permanent state of being freshly minted, so the whole low-price/low-coupon pull-to-par approximation doesn’t interfere with the heuristics of simple yield calculations.
Taking all that into account, it looks like we’ll have to wait a bit longer for a benchmark JGB to trade consistently north of 4 per cent.
Further reading:
— But wait, do French OATs REALLY trade at a spread over Italian BTPS? (FTAV)
