- US multiples have fallen, but are still above their long-run averages; Europe is not obviously cheap
- Emerging markets and Japan are at multi-year low P/Es
- In the US, P/E-to-growth ratios look more attractive, but are based on overly optimistic earnings forecasts
- There is a potential opportunity in US small caps, where the premium to large caps is very low
Exhibit 1 below shows the z-scores for several different valuation metrics: P/E = price to forward earnings, P/B = price to book, P/S = price to sales, P/CE = price to cash earnings, PEG = P/E to growth, DY = dividend yield, ROIC = return on invested capital. Note that it is not appropriate to compare one metric against another as the time periods are different. For example, the P/E is calculated from 1987, whereas the P/B is from 1974.
You can, though, compare one country/region to another for a single metric. Hence it is fair to say that the US P/E is relatively higher than the P/E for Europe. You’ll note that I’ve split the US market into Tech+ (made up of the IT sector plus Internet Retail (essentially Amazon and Netflix)) and the rest. It is more appropriate to compare US ex-Tech+ to Europe as the European technology sector is small.
Exhibit 1: US Price/earnings to growth ratios attractive, but growth estimates very high
Source: IBES, MSCI, FactSet, BNP Paribas Asset Management, as at 26/10/2018
There are asterisks above the P/E and P/S metrics as they have had the highest predictive power for subsequent market returns (at least in the US). Unfortunately, they are giving different signals for US tech. On a P/E basis, the sector is averagely valued, whereas it still seems expensive on price to sales. The P/E multiple for the US relative to long-run earnings growth expectations (PEG) is quite low, however; two standard deviations below the long-run average.
The reason this metric is less compelling as a signal to buy is because earnings growth expectations have risen dramatically this year following the US tax cuts. The average since 1995 of the long-run EPS growth rate forecast is 12.3%, but the latest value is 17%. While in the short-term earnings growth will be higher because of the tax cuts, that won’t necessarily be the case over the longer term.
The only markets that appear absolutely cheap are Japan and emerging markets (EM). To put the discount into perspective, we can look at how things have changed since the start of the year. Exhibit 2 shows the forward P/E relative to the long run average, both today and as at January 2018. For example, the P/E for the S&P today is 15.5x, which is 6% above the long run average of 14.7x. All markets have obviously become cheaper since the highs in January, but the US is still at above-average multiples.
EM, by contrast, are well below and the last time multiples were this low was in August 2015. For Japan it was November 2012. EM returned 12% over the subsequent 12 months (in US dollar terms). Japan returned 33% (65% in yen terms).
Exhibit 2: Relative forward price-to-earnings ratios (latest P/E ratio relative to long-run average - today and as at January 2018)Note: Time period for US sector calculation is different than for broad index, so values will not average. Source: Bloomberg, BNP Paribas Asset Management, data as at 26/10/2018
Across US sectors, the only ones with somewhat below average P/E multiples are materials, media, pharmaceuticals, banks and semiconductors. Yield proxies like utilities and telecommunication services are rich. This suggests that ‘bargain hunters’ are not necessarily going to drive a rebound in the market. Either good earnings news or a reduction in trade tensions will likely be necessary.
Exhibit 3: Autumn colours
Source: S&P, IBES, BNP Paribas Asset Management, as at 26/10/2018
A better opportunity may be in the US small cap segment. The poor performance of US small caps relative to large caps has been one of the surprises of the second half of the year. Despite tax cuts, trade restrictions and a stronger dollar, the Russell 2000 fell by 2% between the middle of June and 3 October, while the S&P rose by 6%. Since the sell-off began, small caps have underperformed by a further 2.1%. These results have helped to shrink the valuation gap between the two markets.
The forward P/E for the Russell 2000 small cap index is normally about 40% higher than that for the S&P 500. Today it is only 27% higher, which historically puts the ratio at more than one standard deviation below the average premium. The last time relative multiples were at these levels was in January 2016. Over the following year, the Russell outperformed by 13 percentage points.
Exhibit 4: Relative forward price to earnings ratios (Forward PE relative to S&P 500 vs. long-run average; average=100)
Source: Bloomberg, BNP Paribas Asset Management, data as at 26/10/2018
Another view of the US market: overall market multiples are still above their long-run averages as I explained, but at the stock level, there are, as always, opportunities. The median forward P/E for the market has fallen to 15.8x, which is as low as it has been over the last five years and equal to the average median P/E over the last 20 years. So on a shorter-term horizon at least, stocks are relatively inexpensive. The spread between the multiple for the 25th and 75th percentiles, representing the dispersion of valuations, is also still reasonably wide (see exhibit 6), suggesting there are opportunities for shorts as well as longs.Exhibit 5: S&P 500 forward multiple percentiles Source: Bloomberg, BNP Paribas Asset Management, data as at 26/10/2018 Exhibit 6: spread between 25th and 75th percentile Source: Bloomberg, BNP Paribas Asset Management, data as at 26/10/2018
To read more on equity markets, click here. More articles by Daniel Morris.