Applied philosophy - Decomposing equity cycles

Are global equities in a recovery or still in the midst of a downturn? We analysed how valuations and earnings progressed through previous market cycles to get closer to an answer. While our findings provide some clues to the future path of equities this year, we think it hinges mostly on how inflation and economic growth develop. 

Is this the real thing or another false dawn? Can we start believing or should we continue to hunker down? We believe that these are the main questions on the minds of investors after a positive end to 2022 and a strong start to 2023. Timing investments is a notoriously difficult process, in our view especially for risk assets, such as equities, whose returns tend to be cyclical. Anything that can help determine where we are in the market cycle is a big aid to our asset allocation process. 

That process usually begins by working out where we are in the economic cycle, which we believe has a major influence on equity returns. However, we think equities tend to reflect the future state of the economy, hence we need to forecast 6-12 months ahead (otherwise we end up driving by looking in the rear-view mirror). Also, it seems to us that the direction of future economic growth is more important than its level. We think that this could give us a good indication of where earnings growth is heading, for example. 

The other major component of equity capital returns is the change in valuations, which can have the biggest impact in the short term, though diminishes as the investment horizon lengthens. There are many factors that can be reflected in valuations, perhaps one of the most important being the discount rate used to calculate the present value of future earnings (or cash flows). Nevertheless, the lower those valuations, the higher the returns we would expect from the asset class over the long term. 

How do earnings growth and valuations change throughout the market cycle? To answer that question, we use the five major market expansions and downturns (including the current one) identified for our analysis of the cyclicality of sector returns (see Figures 9 and 10 and here for more detail). We will use this framework to analyse how earnings growth and price/earnings (P/E) ratios changed throughout these cycles. This, we hope, will contribute to our understanding of where we are in the cycle and how it might develop.  

The main driving force in bear markets appears to be valuations. P/E ratios compressed in all five end-of-cycle bear markets that we analysed, while earnings kept growing in the first 6-12 months in three of them before declining (1981-1982, 2000-2002, 2007-2009), and they stagnated in one of them (1990). What we already know about the bear market that started in the autumn of 2021 is that earnings have yet to roll over, which can be explained by the resilience of the global economy in the face of inflation and a rapid rate hike cycle (Figure 1). Thus, it seems to us that bear markets tend to be valuation-driven events and the price index reaches its trough at or close to the bottom in P/E ratios. 

The dislocation between earnings and valuations continues into the early cycle phase, which starts with a burst of hope. Valuations expand after the equity market trough and investors look through deteriorating economic data (Figure 3). This triumph of hope over fundamentals has also earned it the unflattering name “dash for trash” (i.e. the rally of the most beaten-up stocks in the previous bear market). It can be accompanied by central bank rate cuts lowering the discount rate used to value future earnings. 

Meanwhile earnings may continue to decline in this period eventually reaching a trough during the early- or mid-cycle phase. We can observe this dynamic in all but one of the five market expansions: in the 2002-2007 bull market, earnings troughed during the preceding market downturn and started growing almost six months before the market trough (using the Datastream World Total Market price index in US dollar terms).  

In the mid-cycle phase valuations tended to stabilise or retreat slightly and earnings growth took over as the main driver of equity market returns. We think that this is the phase when fundamentals are the most closely aligned with the performance of equities and the growth in earnings reflects the strength of the economy (see Figure 11). The only cycle when valuations expanded during this phase was during the 1982-1990 market expansion, perhaps driven by the rapid decline in long duration sovereign yields (and equity the discount rates). 

Figure 2 highlights the strong correlation between economic growth and earnings growth. We used 12-month trailing real GDP growth in local currency in the nine markets that have been consistently among the largest by market capitalisation since 1973 in the Datastream World Total Market index (United States, Japan, United Kingdom, Eurozone, Canada, China including A+H-shares, Australia, Switzerland and South Korea). We then calculated a weighted average annual GDP growth measure using those market capitalisations in US dollars and compared it with global earnings growth derived from the price index and P/E ratio of the Datastream World Total Market index. Based on these series, the direction of GDP growth has lead earnings growth by about 6 months since 1975. 

In the late cycle phase, expanding valuations tended to boost returns exceeding the contribution from earnings growth in three out of the five market expansions in our sample: 1974-1981, 1990-2000 and 2009-2021 (see Figure 11). We suspect that in this period investors chase the winners of the cycle at all costs and there seems to be a collective hope that cyclicality has been extinguished. Earnings growth may still be strong in this period, underlining the view that increasing valuations are correctly foreseeing a more profitable future. There was only one period within our sample – the final stage of the 1974-1981 market expansion – where earnings growth rolled over before the price index peaked. The 1990-2000 and the 2009-2021 expansions were two examples that included intra-expansion bear markets, which preceded earnings declines, but where growth reaccelerated in both cases before the market expansion ended. These market events were also triggered by shocks that may have been difficult to foresee: the default of Russia contributing to the collapse of Long-Term Capital Management and the lockdowns during the COVID-19 pandemic. 

One of the most interesting insights we gained from this analysis is that equity markets are a fascinating reflection of collective human behaviour. We are correct in focusing on economic growth to explain how company earnings develop much of the time. However, out of the 13079 trading days from 1st January 1973 to 16th February 2023, only 6776 days (about 52%) fell into what we would consider mid-cycle periods, when the relationship between earnings growth and market returns is the strongest. The rest of the time, equities are pulled higher or lower by the changes in how much investors are willing to pay for those earnings. This may partly explain why there is so little correlation between economic growth and equity returns. It also highlights to us why turning points in markets are hard to pinpoint in real time. 

Nevertheless, Figure 3 illustrates what tended to happen around the past four global equity market troughs (using the Datastream World Total Market price index in US dollar terms). Valuations fell during bear markets, and then quickly expanded after the market bottomed and remained the main driver of returns during the early-cycle period. At the same time, earnings may have fallen somewhat, but remained quite stable throughout on average. So far, our current experience is similar to what happened in the past, even if we assume that the market downturn has not ended. The big difference is how Treasury yields have behaved this time. In the past, they tended to fall before the trough in the bear market, mainly driven by investor expectations of an impending recession, in our view. Thus, we think that de-rating tended to be caused by a prospective deterioration of earnings. However, in 2022, bond yields rose rapidly mostly driven by sharp tightening by the US Federal Reserve. Therefore, unless the cyclical trough is already behind us (as of mid-October 2022), falling Treasury yields would not be unusual at this point in the cycle. 

What does this mean for where equities are headed in the next 12 months? One of the possibilities is that valuations are yet to reach their lowest level for this cycle and may turn lower yet again. This would not be unusual based on previous experience, especially if we are in a multi-year bear market, similar to those in 1981-1982, 2000-2002 or 2007-2009. We envision this would happen in the “persistent inflation” scenario outlined in our 2023 Outlook, if inflation proved to be stickier than we currently expect and perhaps even reaccelerated prompting central banks to raise rates higher than reflected in rate futures at the moment, while keeping them elevated for longer. 

Another possibility is that the early cycle phase started when P/E ratios troughed in mid-October. This would be closest to the experience in the 1990s when valuations recovered well before earnings. We think this is only possible if inflation falls rapidly as we currently expect, and the global economy avoids a deep recession allowing valuations to rise. Recent economic data releases have increased the probability of this scenario, in our view. 

We think the recent consolidation in equities reflects concerns that markets may have got too optimistic too soon about inflation and growth. Further de-rating is a possibility, though earnings could remain resilient in the first half of 2023. In our view, the interest rate cycle must turn before we can be fully confident that the valuation adjustment is behind us. When looking ahead to 2023 (in November) we forecast positive equity returns but felt the risk-reward comparison favoured credit and gold (Figure 7). We also maintained a balance between early-cyclicals and defensive sectors in our model sector allocation (Figure 8). However, based on our analysis, we think that global equities are in the early phase of an expansion, which implies strong returns for the rest of the year.