By Mike Riddell
The moves in gilts in August were truly exceptional. Volatility in the gilt market (based off 10-year gilt futures) has soared to close to the highest levels seen this millennium, on a par with the eurozone debt crisis of 2011/12 and behind only the global financial crisis of 2008/09.
The first distortion to note is that gilt yields have been driven sharply lower relative to other government bond markets. The chart below illustrates the changes in core government bond yields from 3 August, immediately before the QE announcement, to the end of August. The outperformance of gilts at a time when UK data has, if anything, slightly exceeded expectations suggests that QE has driven UK yields down by as much as 40 basis points.
As can be seen from the charts above and below, QE has also caused a significant flattening of the gilt curve relative to other markets, with the 25-30-year part of the curve experiencing the biggest fall in yields. Ultra-long dated gilts have also performed extraordinarily well and, to put the scale of the move in context, the 36 basis plunge in 50-year gilt yields since QE was announced is the equivalent of a price return of almost 12 per cent.
The BoE’s recent continuation of its QE programme has also caused a huge move in relative value between individual gilts on the curve, which is due to the BoE’s rule that it will not exceed a maximum of 70 per cent ownership of the free float of individual gilts.
Pre-QE, gilts with a small free float (that is, the older, higher coupon gilts) tended to trade at a large premium due to their relative scarcity, which made their valuations unattractive. However, post-QE there has been an enormous reversal in relative valuations, where the lower-coupon, more recently issued gilts have substantially outperformed on the basis that the BoE has a greater ability to buy these as part of its QE programme.
Two examples of this are highlighted below. The first two charts (Figures 3 & 4) show the yield difference between UKT 2.75% 2024 and 5% 2025. Historically, the 5% 2025s had yielded slightly less than the 2.75% 2024s, owing to the 5% 2025s having a slightly shorter duration (despite having a slightly longer maturity) and being a scarce, old, high-coupon gilt. In July, the yield on the UKT 5% 2025s crept above the UKT 2.75% 2024s for the first time, perhaps because investors started to price in some possibility of QE, but since QE started the UKT 5% 2025s have hugely underperformed due to the 5% 2025s no longer being eligible for BoE purchase. The difference has fallen slightly, but you can still get over 6 basis points of extra yield for a gilt that is of a similar maturity and shorter duration. This 8 basis point swing in the yield difference between these two gilts in the past two months equates to about a 0.7 per cent underperformance in terms of the price of the UKT 5% 2025 relative to the UKT 2.75% 2024, which is a big difference for what is a fairly short-dated gilt.
Another gilt that has dramatically cheapened is the UKT 4.25% 2027, which is also no longer eligible for the BoE’s buybacks since the BoE owns 69.7 per cent of free float. Prior to the announcement of QE, the duration-neutral switch of buying 2027s versus selling 2024s and 2030s had historically been fairly stable at about a 0-5 basis points yield pickup. This changed post-QE announcement as 2024s and 2030s hugely outperformed, while 2027s massively underperformed, which meant that suddenly it was possible to gain 20 basis points of yield through a duration-neutral switch.
These kinds of distortions represent great opportunities for active fund managers who can exploit them. QE has caused, and will continue to cause, various distortions between gilts on the curve, and across different parts of the curve. This generates a lot of possibilities to add alpha versus the benchmark.