Different asset markets are offering unusually disparate prospective levels of risk-reward, with bonds looking more attractive than stocks for the first time in 15 years. Our shorter-term market temperature tools are flashing bright green (buy) for developed market bonds, and dark red (sell) for developed market equities.
Two of the oldest adages of investing are that every risk has attached to it a price, and it is the risk case, not the base case, which tends to dominate forward-looking market returns. At the moment, different asset markets are offering unusually disparate prospective levels of risk-reward, with bonds looking more attractive than stocks.
The most striking disconnect is between developed market government bonds and equities, where bond premia are above developed market equity risk premia for the first time since the Global Financial Crisis. That is, for the first time in 15 years, investors look to be better compensated for owning bonds than equities (Exhibit 1).
Forward earnings yields are falling from the peak
Higher bond premia have been spurred by higher long-term rates — both real yields and inflation breakevens — as corporate spreads have broadly tightened. By contrast, forward earnings yields peaked last October and have been falling ever since, thus moving in the opposite direction to bonds and underscoring their relative appeal.
We see this from a 20 000-foot perspective in Exhibit 1, but also in more granular ways, for example, by valuing equities using a dividend discount model and comparing this to various measures of spread risk. They all paint a remarkably similar picture: bonds are cheap and equities are rich in comparison (and in some cases, in their own right, too).
An important explanation for lower equity premia (and higher earnings estimates) has been the expectation that companies will continue to pass on cost increases to consumers, with corporate margins commensurately resilient.
Although US margins have declined by 175bp since last year, they are expected to rebound smartly; for eurozone corporates by contrast, faced with a similar increase in input costs relative to sales prices, margins have been far more resistant, dropping by just 70 bp (see Exhibit 2).
Whither margins and valuations?
Recent work by our multi-asset team seeks to quantify where margins and valuations might fall to, based on historical relationships between input versus output prices captured in purchasing managers’ indices (PMIs) and estimates for corporate margins and valuations.
For example, in Europe the relationship between estimated EV/EBITDA1 and the prices gap in PMI manufacturing has a correlation of -0.6.
The evolution of 12-month forward EV/EBITDA estimates appears to be much loftier than might be expected, at 3 rather than 1 on the y axis in the chart below (see Exhibit 3). We anticipate a meaningful correction towards the line of best fit in the coming months, although we are mindful of the increase in recent dispersion (red dots) compared to the past (other dots).
Valuations look high given where PMIs suggest margins (and hence profits) are likely to go
12-month first differences for EV/EBITDA and output minus input prices (margins)
Notably, the medium-term signal provided by risk premia is also supported by our shorter-term market temperature work, which seeks to capture sentiment, volatility/skew and positioning in a systematic way across the main markets. Our tools are flashing bright green (buy) for developed market (DM) bonds, and dark red (sell) for DM equities.
Too much complacency
Our macroeconomic framework, meanwhile, continues to point to complacency given the consensus DM growth and corporate earnings ‘Goldilocks’ narrative — and much more so now than in the earlier months of 2023.
On the one hand, establishing a reasonable ‘base case’ has felt harder than normal this year, with exceptional volatility in analysts’ macroeconomic forecasts that have variously called for all four quadrants (expansion, slowdown, contraction, and recovery) of a typically multi-year business cycle to be delivered over a few short months.
But on the other hand, with 525bp of rate hikes still working their way through the pipes to slow US economic activity (see Exhibit 4), and now-depleted Covid-related savings, the risks to a Goldilocks-like macroeconomic ‘base case’ seem easily skewed towards weaker actual outcomes.
The latest US payrolls data, for example, revealed that non-seasonally-adjusted numbers were so soft that only a favourable swing in the seasonal factor prevented a much weaker reading.
As one of our favourite sell-side analysts reminded us, if the Bureau of Labor Statistics had used the same seasonal factor as in July 2022, the private payroll data for July this year would have been only 60 000, less than one-tenth of the 2021 peak of 650 000 and consistent with near-recessionary labour conditions.
The sharp slowdown in hours worked, household jobs and temporary employment in recent months are all rowing in a similar direction, and this is before student loan repayments restart in October, which is expected to shave 0.3%–0.5% off disposable incomes.
Pull all this together and the most lagging part of the real economy seems poised for a long-awaited slowdown, as leading lending and credit data has been pointing towards for some time.
Not a quiet summer
Anticipating this evolution, there has been no summer lull for us. We have taken advantage of sharp moves in bond markets to deepen long duration positions, seeking to capture the attractive and asymmetric risk premium on offer. At the same time, we are sticking to our caution towards equities.
In recent days, we have leaned further into long-dated US TIPS positions, as yields soared through 2% in volatile summer conditions; we have also fully unhedged our European Investment Grade Credit exposure, turning it into a high-quality total return trade from a pure carry trade previously.
We have also moved moderately into long US 10-year nominal bonds, where yields of over 4% are not to be scoffed at, particularly in the current growth and inflation setting. Long duration is our largest risk position across our actively managed multi asset portfolios today.
Slower growth (or even a recession) will likely drive earnings estimates lower, by 15%–30% depending on the market and period in question. Yet the analyst community collectively envisages double-digit compounded earnings growth in the US to be delivered over the next 24–36 months, and high single digits in Europe.
This growth comes after a period of exceptional profit gains already delivered in the aftermath of Covid. Put differently, the base from which earnings are expected to grow looks lofty, and is a particular worry in Europe, where equity index returns year-to-date have not been far behind those in the US even as growth is slowing far more quickly.
Valuations are an additional worry. The forward price-earnings (P/E) ratio for the NASDAQ this year has recouped half of the valuation decline from 2022. The increase in multiples in the first part of 2023 was at least partly justified as markets begin pricing in cuts in policy rates. But as the year has progressed, economic growth has beaten expectations and the Fed has maintained a hawkish stance.
The degree of the anticipated Fed pivot has consequently decreased. This should have lowered multiples, but enthusiasm around artificial intelligence (AI) has pushed tech P/Es higher. There has been some moderation in valuations recently, but a poor growth outlook or higher inflation could easily sap P/E ratios.
Overweight emerging Asia
Amid broader caution on equities, EM Asia is our chief overweight. The market appears to have a lot of bad news baked into both earnings and forward valuations.
The forward P/E ratio of the MSCI China index, for example, is as low as it has been relative to MSCI World since 2015 (see Exhibit 5), despite a much better earnings outlook than in developed markets. China equity index returns so far this year have disappointed, though the recent bounce following the latest stimulus measures announced by Beijing illustrate the scope for gains if there is more to come.
Thibault Bougerol contributed to the research for this note. The enterprise value to earnings before interest, taxes, depreciation, and amortization ratio (EV/EBITDA) compares the value of a company—debt included—to the company’s cash earnings less non-cash expenses.