Value: finding fault

There is a rule in seismology that expresses the relationship between the magnitude and intensity of earthquakes in a given area. Put simply: the longer the wait since the last movement along the fault line, the bigger the effect when it finally happens.

As people are naturally programmed to weigh recent experience over old, this can often lead to the most impactful outcomes being the most disregarded.

A lot of value investors think along these lines and can remain sanguine in the face of their style’s underperformance, confident that eventually and inevitably the market mood will change to one more in keeping with their fundamental principles.

Where are we now?

Value has underperformed the market and it has heavily underperformed growth, especially in the past six months.*

For the value-biased index, the year-on-year drop has been in line with the forward earnings expectations – i.e. the forward price-to-earnings ratio (P/E) has remained roughly the same.

More than this, though, the flat ratio suggests that value stocks’ earnings for ever after will be affected in the same way as next year’s. This seems unduly harsh, especially when you consider that among the largest constituents of the European value index are household-name healthcare, industrial, and financial companies.

I appreciate that there is an argument for the energy industry suffering from a permanent impairment, but it was only 7% of the index at the most recent quarter-end. To merit a 20% drop in valuation, we would have to lose all of the energy sector’s contribution and also lose a further 13% of earnings. To put that another way, we would have to allow for an undiscounted 1.4 years’ worth of earnings (as the index is at 11x P/E).

Given 2021 earnings are expected to be -20%, that means that the ‘scarring’ (or post-coronavirus depressed levels of earnings) would have to continue for at least another seven years.

Interest rates?

Low interest rates are often considered the drivers of growth’s outperformance over value, which makes sense due to the higher ‘duration’ of growth stocks. In essence, future corporate profits are being discounted less quickly because of these low rates, favouring growth companies with potentially better earnings prospects over value stocks with steadier present earnings but less exciting futures.

But all stocks should be buoyed by this effect, though, so where is the boost for the value index? It is at the same P/E level, despite a >4% drop in the discount rate that should in theory be adding more than 6x to a 10x stock.

The table below sets out what we would view as justified P/E ratios, using 1/(r-g) where ‘r’ is the equity discount rate and ‘g’ is the perpetual growth rate. The green swathe represents the current multiple for the value index, yellow the growth index in 2015, and red the current growth index.

So if we are to use the idea of low interest rates to explain the difference between growth and value’s trajectories, it at least partially clarifies the current growth multiples but does not help account for the current value rating.

If I was to stick rigidly with this model, keeping the P/E the same and lowering the ‘r’ by 4% would necessitate lowering the value index’s ‘g’ by 4%. Clearly this is a judgement call, but I cannot believe that the long-term growth prospects of the constituent companies have fallen by that much at the same time as the growth ones have stayed constant (or even increased).

To sum up, market movements which have differentiated the value and growth styles are not justified on an aggregated basis, in my view, unless it is believed that the fundamental properties (i.e. earnings growth rates) have changed considerably, permanently and divergently in the past five years, and especially in the past six months.

It is difficult to predict when the markets will turn, just as you cannot predict the timing of an earthquake. However, the underlying pressure is stronger now, which implies a more forceful event to come given the low rating of value stocks, even if value’s outperformance seems like a distant memory.

*For the purposes of this article I am using ‘value’ and ‘growth’ as described by the MSCI style-biased indices.

Appendix - performance of indices cited