Okay, that’s a big statement for a decision support platform, even one incorporating machine learning. So let’s begin with the observation that the best performing stocks from the top 100 UK names started this year with high valuation, high yield, low short interest, narrow CDS and low implied volatility. The worst performing stocks showed the opposite characteristics. Which stocks tick those boxes today?
Before we check that, a little theory. The latest GMO quarterly letter argues that QE and zero interest rates favour longer duration assets, such as equities, by pulling future returns into the here and now. On its own, this would suggest outperformance by higher value, lower payout stocks, because the cash flow from these names is more distant and yet is afforded a smaller discount as long term interest rates fall. While the UK has all the old QE money sloshing around, but no fresh printing, long rates have fallen throughout 2014. Yet what we observe is that while high valuation stocks outperformed, so did higher payout stocks, which have a higher yield that shortens the duration of the investment. So far, so mixed a message.
The other factors are easier to explain. We expect higher prices to coincide with lower short interest, CDS and implied volatility and have set OTAS up to flag as divergent the extreme examples of this not happening in each sector. Our expectation is based on extensive back testing, but has theoretical support too. As short interest, credit cost and volatility rise, they all increase the uncertainty about when you are going to get your cash back from an investment. We should expect this uncertainty to be reflected in a lower share price and, throughout 2014, this is precisely what we observe.
Jan 1, 2014
|
Winners
|
Losers
|
Top 100
|
---|---|---|---|
P/E x | 16.0 | 14.4 | 14.6 |
Div Yield | 3.1% | 2.8% | 3.0% |
SI % | 0.27 | 0.80 | 0.32 |
CDS bps | 67 | 73 | 73 |
Imp.Vol. | 15.1 | 21.9 | 20.6 |
The winners referred to in the table above are the top 20 best performing stocks this year, while the losers are the 20 worst performers. The average for the index is shown in the final column. The figures all refer to the average level of each factor on January 1, 2014, or what your OTAS Stamps would have shown you back at the beginning of the year. P/E and Yield are based on 12 months forward values, while short interest, CDS and implied volatility are right now.
Yield is much of a muchness, which may explain why the duration argument has not worked so well in 2014. However, on all the other measures, the best performing stocks showed clear blue water between themselves and both the losers and the average stock.
Here is the same analysis for these stocks today.
Nov 26, 2014
|
Winners
|
Losers
|
Top 100
|
---|---|---|---|
P/E x | 19.1 | 13.7 | 15.2 |
Div Yield | 3.2% | 3.6% | 3.4% |
SI % | 0.35 | 0.88 | 0.43 |
CDS bps | 60 | 76 | 62 |
Imp.Vol. | 22.2 | 25.9 | 23.1 |
The valuation gap is now more pronounced. Our winners have average EPS Momentum of +0.2%, while the losers average -4.2% and the index as a whole shows -0.4%. The re-rating of the top performing stocks is in recognition of superior growth prospects, which are being reflected now in analysts’ upgrades.
The difference in yield is only slightly more pronounced than it was, with payouts having risen. The premium yield on the losers owes much to the double digit fall in the price of the shares. The winners retain an advantage on the other factors, but the gap to the average share in particular has fallen. It may be time to freshen up the portfolio.
To rank the top 100 stocks today, we have sorted the names by each of the factors and accorded the highest score to the top ranked share in each category. For CDS and implied volatility, where the data set is limited, we have accorded the shares that have no value, the average of those that do have one. The impact of this is to change three stocks in the top 20 list, and to shuffle the order for the others a little, favouring larger stocks, frequently consumer staples plays, with low credit risk and volatility. However, as our starting point was an equally weighted portfolio of winners, constructed purely from absolute price rises YTD, the shuffling of ranking does not matter.
Top 20
|
Bottom 20
|
|
---|---|---|
P/E x | 18.6 | 13.7 |
Div Yield | 4.0% | 2.4% |
SI % | 0.16 | 2.37 |
CDS bps | 60 | 139 |
Imp.Vol. | 20.6 | 26.0 |
The top 20 stocks have a marked valuation and yield premium to the bottom 20, are barely shorted, have far lower credit risk and 20% lower three month uncertainty measured by implied volatility. This looks like a portfolio might be expected to, if it were constructed for the risk of interest rates rising. However, this has not entered into our thinking as this is a bottom up process and the QE comments were largely for fun.
Of course our methodology falls foul of exemplary statistical practice, not least by a type of survivor bias from not taking into account index rebalancing. However, it has the significant positive of being able to be performed by anyone with an OTAS and an hour to spare. The analysis may be done anytime a manager wants to refresh a portfolio and may be used to build a portfolio gradually, a few stocks at a time.
And now, drum roll please, the top 20 stocks of 2015 (maybe). These are in ranking order, Vodafone, GlaxoSmithKline, Astrazeneca, British American Tobacco, Compass Group, Hammerson, Friends Life, Land Securities, United Utilities, Diageo, Centrica, Reckitt Benckiser, Legal & General, Unilever, Imperial Tobacco, HSBC, British Land, Direct Line, Saint James’s Place and SAB Miller.
If you have suggestions about how we might refine our methodology, do please let me know. If you would like to know the names of the stocks to avoid, do likewise. One small disclosure, I own shares of RBS and am pretty happy with the performance of the stock against the other banks this year. However, it is near the bottom of my ranking, which is making me seriously consider taking my profits. And this is precisely what OTAS is designed to do; ask the questions that make you think. A decision support platform.