Peak Rates – finger on the trigger
The current business and interest-rate cycles have been idiosyncratic for a number of reasons, with COVID-19-related goods and labour shortages coinciding with unprecedented monetary and fiscal stimulus, driving the highest global inflation readings since the 1970s. In the same way that inflation exploded to the upside, certain readings are showing an equally rapid descent while other measures, most notably core inflation, are proving more sticky than desired.
Either way, global investors are currently sitting with one finger on the trigger, ready to get into fixed-rate bonds once again after a rather rough period. Has anybody been listening to the likes of Australian Super, UniSuper, or even Warren Buffett recently? Why is this the case?
Mostly it is related to the historical precedent that bonds tend to rally aggressively once a central bank concludes its rate hiking cycle. In more recent hiking endeavours, central banks have form in hiking too aggressively and then having to cut just as aggressively as the delayed impact of prior rate hikes rattles the system (usually in the form of USD shortages). This is despite the narrative currently building in markets - the Fed will be able to stop hiking rates and then hold them at an elevated level for an extended period of time.
Let’s have a look at the Federal Funds rate compared with US bond yields since 1985:
Figure 1: Fed Funds Rate vs. 2yr and 10yr UST's in % terms
Clearly, what the Fed does has a very important impact on bond yields. The shorter 2-year bond is impacted more directly (given its competition with repo and bank deposit rates), but the impact is also clearly felt on the 10-year bond.
It then follows that if the Fed was to start cutting rates, we should expect yields across the curve to fall and deliver positive capital returns to investors. One distinction worth noting is the differing impact of a given fall in yield, depending on the term of the bond. Longer-dated bonds are more sensitive to interest rate changes (to the upside and the downside) compared with shorter-dated bonds. If yields fell by 1 percentage point across the curve, 10-year bonds would deliver a capital gain of around 8%. Combined with the coupon of around 4%, that is a total return of 12%.
We are unlikely to see interest rate cuts in the near future. Market pricing (as of 28 August 2023) expects the Fed to start cutting rates by May 2024, with two standard-sized cuts (50bps) priced in by September 2024.
Even though headline inflation has fallen rather quickly, the employment market in the US remains very healthy with a low unemployment rate and still fairly elevated roles per applicant. However, the chart below indicates that the US unemployment rate has and can jump by up to 2% over a year or two (or quicker in the event of a crisis) and this is historically correlated with Fed Funds cuts of up to 5%:
Figure 2: US Unemployment Rate vs. CPI and Fed Funds Rate
Perhaps the late 1990s / early 2000s is the best historical analogue for the current environment. The unemployment rate had declined in a straight line to a trough of around 4% (currently 3.5% but some COVID-19-related worker shortages are persisting) and stocks were highly valued on account of internet stock mania (AI, anyone?). US consumer price inflation also peaked around 3.7% year-on-year, auspiciously similar to the last print having declined nearly as fast as it went up. As the cycle turned, the Fed Funds rate peaked at 6.5% before being cut to 1% over the ensuing few years.
The US 10-year Treasury yield fell from the mid 6% level to the mid 3% level. Using rough numbers and the duration discussed above, capital returns of 20% were in the offing (with coupon returns coming in on top of that). These are the sorts of returns that real money investors are preparing for in their long positions in duration. Anecdotal survey evidence suggests that real money investors (i.e. fund managers) are long bonds (anticipating the outcome detailed above) while hedge funds and other so-called ‘fast money’ are short bonds in aggregate. As their name suggests, ‘fast money’ tend to unwind losing positions in a hurry given the use of leverage in their strategies. They are also more likely to use trend-following strategies, and the recent trend has been bond price weakness.
The bonus for local investors is that here in Australia the yield curve is still shaped normally (i.e. upward sloping) so you are still getting paid to hold duration compared with many global developed markets where yield curves are inverted (long-end yields lower than short-end yields):
Figure 3: The spread between developed market 2-year bond yields and 10-year bond yields
Source: Bloomberg, pricing as of 28 August 2023 (we can debate the inclusion of Italy in a list of ‘developed markets’)
While the US economy keeps humming along, there will be heightened market commentary regarding de-globalisation (higher costs), greenification (higher investment), and higher levels of inflation volatility demanding a higher term premium on bonds. These arguments have, not coincidentally, become louder after a 4% surge in interest rates.
While these arguments certainly hold some weight, the jury is still out on whether they can negate structural issues such as demographics (ageing populations), increased debt loads (essentially bringing forward future growth at the expense of the medium to long-term), and the newfound funded status of global pension funds (on account of higher discount rates lowering the future value of liabilities – which ultimately incentivises duration immunisation via buying long bonds).
In AUD bonds, given the upward-sloping shape of the yield curve, investors are still being paid for lending for longer.
Therefore the risk of taking longer duration bonds vs. the US for example is somewhat mitigated by the higher coupons on offer for a particular tenor.
Gradually increasing portfolio allocations to duration on value reduces the timing risk associated with the next turn down in rates, and in AUD at least, still earns higher income than shorter-dated bonds.