I think I first heard about a macro-inspired, short JGB trade in 1996 and it has been one of the great widowmakers of global markets ever since. In the context of the global secular trend to lower yields pre-pandemic, JGB yields were given added downward impetus by the BoJ’s tight monetary policies. Notice I said tight and not easy, because JGBs have been Exhibit A in Milton Friedman’s case for tight monetary policy leading to lower, not higher interest rates. Misperceptions of the stance of Japanese monetary policy have been a key driver of the widowmaker trade.
The other driver of the short JGB trade was a view that Japan’s high gross government debt to GDP ratio would eventually force some kind of reckoning, a reckoning that never came. Japan is a standing reproach to the public debt overhang threshold literature. More recently, the rest of the world has made Japan’s government debt seem less much salient on a relative basis than it was in the 1990s and 2000s, but the short trade is back on in the context of rising global yields.
Some market participants are betting that the BoJ will be unable to maintain its yield target in the face of rising yields globally, which has seen dollar-yen to near quarter century highs and the US dollar to record highs. The latest iteration of the widowmaker trade takes some of its inspiration from the collapse of the RBA’s 3-year bond yield target last year (see this post for details).
But there are some important differences between the RBA and BoJ approaches to yield curve control. The RBA’s 3-yr bond yield target was designed to reinforce its forward guidance on the cash rate. It was also intended as a mechanism by which the RBA sought to avoid expanding its balance sheet, at least prior to November 2020. As soon as the RBA was forced to start buying significant quantities of the targeted bond lines, it let the peg go. At that point, the RBA was already rhetorically qualifying its forward guidance, which only served to reinforce market perceptions the guidance wasn’t credible. Once the RBA started long duration QE from November 2020, the 3-yr bond peg no longer served its original purpose of avoiding balance sheet expansion. Technically, the RBA could have maintained the peg for longer, but only through acquiring most, if not all, of the targeted stock.
Unlike the RBA, the BoJ is not averse to expanding its balance sheet or owning a significant share of the stock of JGBs. With the yield target under attack, mainly though JGB futures, the BoJ has already doubled-down, increasing the quantity of bonds bought, but also expanding its target for fixed-rate operations to JGBs with a duration as short as seven years, where yields have risen above the target.
Japan’s inflation rate is at seven-year highs, just above its target of 2%, but Japan, like much of Asia, is upstream of many of the supply problems plaguing the global inflation outlook. With the yen at multi-decade lows, much of this inflation is imported.
Whether the BoJ sees this as a feature or a bug is what the latest iteration of the widowmaker trade is betting on. The case for it being a feature is that it approximates Lars Svensson’s ‘foolproof way’ of escaping a liquidity trap by pegging the exchange rate below its equilibrium level until a price level target is met. In this case, the intervention is via the bond market rather than the fx market, but in an environment of rising global yields, this has much the same effect on the yen, inflation and the price level.
The ‘foolproof’ part of the foolproof way is that there is no in-principle limit on the ability of the BoJ to weaken its currency, both domestically and internationally. Of course, there is a point at which this could become more inflationary than the BoJ would like. Japanese policymakers also have a sorry history of snatching defeat from the jaws of victory when it comes to monetary policy and inflation. That record might be worth betting on. But the point is that this is a policy choice, it is not something they will be forced to do. There is probably no better time for the BoJ to achieve and maintain its inflation target. If not now, then when?
Fed media signalling
The Fed raised rates by 75 basis points this week, the largest nominal rate hike since 1994. The move was well flagged by the WSJ’s Nick Timiraos, not least because of the last-minute change in story from 50 bp to 75 bp, which clearly came from the Fed:
It is remarkable how the pre-FOMC announcement media drop has become institutionalised in the US. We had something similar in Australia until about a decade ago when a few people, myself included, made an issue of it. I did a Lateline Business interview on the issue which aired only after the handful of other economists they had lined-up for the story pulled-out on instructions from their employers. The producer, Richard Lindell, deserved considerable credit for pushing ahead with the story through some high-level opposition.
In my view, the media drop raised issues about procedural fairness, but also potentially compromised the independence of those who were on the drip. The accuracy of media RBA rate calls has taken a noticeable hit since then, but the point was to incentivise the RBA to do a better job of communicating directly with markets rather than via selected journalists.
RBA media signalling and QT
Fast forward to this week, and RBA Governor Lowe made an unprecedented TV appearance on the 7:30 Report (at least if we discount Governor Stevens’ appearance on Kochie, which doesn’t really count):
Right out of the gate, Lowe said Australians had to prepare for higher interest rates, so his appearance was clearly intended to soften us up for further tightening. He also let drop a revised inflation forecast of 7% by year-end. While dropping forecast revisions into a TV interview is not ideal from a procedural standpoint, the level of transparency and direct communication is at least a big improvement on where it was in days of yore (you can all thank me later):
We previously noted Governor Lowe’s claim that QT was already in the price of bonds, in contrast to market perceptions of the Fed’s QT program, for which there is much more uncertainty, not least on the part of the Fed itself. Here is how it was reported in the FT this week:
The official and market consensus is that the Fed’s QT is good for at least one or two 25 basis point rate hikes, which takes some of the pressure off the Fed’s policy rate.
This suggests an interesting policy option that the RBA could pursue in lieu of at least some additional prospective tightening in the cash rate, which is to move from passive to active QT by selling-off some of its longer duration bond holdings. This would serve to tighten monetary conditions via the long-end of the curve and the exchange rate, alleviating at least some of the pressure on the short-end of the curve, a more aggressive unwinding of its post-November 2020 QE program. This is analogous to the Fed’s Operation Twist, but with a view to bear-steepening the yield curve.
I doubt the RBA would ever go down this route. It remains overly-wedded to a conception of monetary policy transmission via the short-end of curve and the mortgage lending channel, which is one of the reasons it was so reluctant to embrace longer-end QE in the first place, until it was forced to act because its balance sheet expansion was lagging its G10 peers on a relative basis, putting upward pressure on the exchange rate. There are also obvious concerns about bond market liquidity and volatility which they would not want to amplify by aggressively selling into a bear market. But if those concerns are valid, then you can’t also maintain that QE/QT has no effect on yields.
And no one would envy Phil having to explain QT on The 730 Report.
ICYMI
Remember when people argued RBA rate cuts hurt consumer and business confidence? I’ve got the receipts!
Your tweets: