Beware of buy and hold

The last few decades of the 20th century represented a golden era for equity investment with an average compound annual return, including dividends, of 14% pa in the period 1973-2000 for the US, UK and Europe. In this century to date, the annualised rate of return has fallen to 5%.

Exhibit 1: Total returns for major markets – buy and hold worked better up to year 2000

To a hesitant investor in the latter part of the 20th century, when dividend yields were 3-5% and corporate revenues were growing at 8-9% pa, the idea of “time in the market not timing the market” was backed by the statistics, Exhibit 1. A 14% expected return when combined with market volatility of 17% pa meant that the likelihood of losing money over any 10-year period became vanishingly small in theory. We show in Exhibit 2 below that under certain assumptions this would have been an event more than 2 standard deviations away from the mean expected return of 270% over a 10-year holding period. In practice, for the UK, there was only one period between 1965 and 2000 when this happened, which was during the bear market of the mid-1970s. Buy and hold, looking through the month-to-month variations in share prices, was otherwise sound advice for a generation of investors.

Exhibit 2: “Time in the market” sound advice for the 20th century – but maybe not the 21st

Unfortunately, investing today is not nearly as simple. While investors may recoil from the sub-2% returns available on long-term government bonds and on the rebound may instinctively look to increase equity allocations, it is not, in our view, necessarily the right thing to do at this point in time. We continue to observe very high price/sales valuations in each of the US, UK and Europe and the evidence is building that we have entered a structurally lower growth environment for corporate revenues and profits, Exhibit 3.

Exhibit 3: Slowdown in consensus forecast US sales growth in this cycle, compared to 2006/7. Chart shows median sales forecast n-months prior to reporting date.

The combination of high valuations and slow growth in corporate profits is pointing towards a much lower rate of return for equities over the medium term. In the past, the yield available on an equity portfolio of 3.5%, but more importantly the average dividend growth rate of 8%, mechanically led to the high expected and realised returns of the latter part of the 20th century. For comparison, if corporate revenues continue to grow at only 2-3% and are combined with dividend yields of 3%, expected returns could now be as low as 5-6% pa. In terms of yields, we would also highlight that in the UK at least, the market average yield is skewed to the large cap sectors which have recently been cutting dividends.

A return several times that of low-yielding bonds may still look superficially attractive, but only before proper account is taken of equity volatility. Equity markets over the long run have delivered annualised volatility in the region of 15% pa. When returns were in the double digits, this was of relatively little consequence to the long-term investor who was almost guaranteed a significantly positive return in almost all scenarios due to the effect of compounding.

However, in today’s market with our expected return on equities close to 5-6% pa, the same maths shows that even over holding periods of 10 years there is a material probability of an absolute loss, (Exhibit 2) and a significant probability of underperforming the returns on risk-free government bonds. This is because this rate of return struggles to keep up with the increase in expected volatility over the holding period, even as volatility only increases with the square root of time.

There is unfortunately no easy answer to the pension trustee’s or institutional manager’s problem of delivering returns which satisfy benchmarks set in prior periods. The rate of return on risk-free bonds is very low and the risk/reward on equities appears unattractive. We would instead highlight niches of the market, which perhaps by definition are less liquid but offer returns only slightly lower than equities and at a significantly lower level of risk. At present for example, a combination of property investments, high yield bonds, senior loan and infrastructure debt will currently deliver a return similar to that on equities (ie 5-6%) with significantly lower volatility.

As a strategy, any rotation away from equities and into credit instruments would be a relative rather than absolute call; if there is a major recession ahead then both asset classes would obviously be expected to decline in value. However, credit would be expected to meaningfully outperform equity in that scenario. We believe the most likely scenario of sub-par but tolerable growth would be consistent with credit returns significantly above cash, while equities could easily underperform. It is only in the scenario where there is an unanticipated surge in corporate profits that equities are likely to outperform credit by a wide margin, which in our view remains a low-probability outcome given that non-financial profit margins are still well above average.

What is remarkable is that despite the slowdown in revenue growth and with valuations so extended, global markets seem completely focused on the US Federal Reserve’s policy actions. We can see numerous potential risks to the market and a dollar rally inspired by a newly hawkish Fed is just one.

For example, we see medium-term uncertainty in China’s growth trajectory, of which the recent rapid credit expansion is a symptom. The risk of Brexit may be diminishing, but the democratic and legitimate rise of populist political movements is a natural response to the EU’s failure to generate adequate GDP growth and may yet cause significant volatility. Recent survey and durable goods data indicate that a slowdown may be underway in the US and we continue to watch bank surveys which indicate that US credit conditions continue to tighten, Exhibit 4. Perhaps in these circumstances it is not a surprise that M&A activity has slowed down significantly during 2016 – and the M&A statistics would have been much worse had Chinese corporates not been so active.

Exhibit 4: US credit tightening – US banks’ senior loan officer survey (corporate lending)

In short, we see many risks and only a few positive triggers, by far the most important of which is that in any slowdown scenario there is a much higher probability compared to history that central banks will step into the equity market directly. We believe this continued perception of a central bank ‘put’ is responsible for equity markets holding up at levels which offer such low returns and to a degree explains the sensitivity of markets to the direction of monetary policy.

The expectation that investment strategy is about making all the right calls all of the time is at odds with the reality of the marketplace. There are times, such as now, when it is hard to discern which way the dice will roll. We cannot be sure for example that the next faltering of economic activity will lead to a further intervention by central banks into (equity) markets, or whether the Fed is about to finally call time on a strategy of ever easier policy. In our view, this is the time to follow a strategy of owning diversified and robust, rather than optimised, portfolios which can achieve an acceptable return in many scenarios and sufficient preservation of capital in all.

Quick conclusions:

1. The combination of slow growth and high valuations points to a period of low returns for US, UK and European equities over the medium term. Such a low expected return on equities offers no guarantee of outperformance over government bonds, even over 10-year investment periods.
2. In our view, this is the time to follow a strategy of owning diversified and robust, rather than optimised, portfolios which can achieve an acceptable return in many scenarios and sufficient preservation of capital in all.
3. The latter years of the 20th century appear to have been an exceptional period for equities where buy and hold or “time in the market” strategies fitted the then prevailing investment parameters. In the 21st century, the numbers are different.