TEUN DRAAISMA WAS THE HEAD OF EUROPEAN EQUITY STRATEGY AT MORGAN STANLEY. VERY WELL KNOWN AND RESPECTED AS A STRATEGIST HE WAS  ALWAYS KNOWN TO BE A GREAT STRATEGIST. HE LEFT RECENTLY TO JOIN TT INTERNATIONAL, AN OLD AND QUITE FAMOUS EUROPEAN HEDGE FUND. HE RUNS AN EQUITY FUND UNDER A MACRO STRATEGY.

Below is his commentary for February 2012, he notes that European stocks are extremely cheap and despite decades of underperformance there is a way to find value and make money in Japanese equities. He believes QEIII could be the catalyst which drives Japanese stocks and that stable growth companies should be the best performers.

Three multi-decade extremes

Executive summary

  • We agree with consensus view that developed market economies face a continuation of low growth, deleveraging and shorter cycles;
  • Despite this sub-trend outlook, developed market equities will exhibit rallies while policymakers insert liquidity, for instance through quantitative easing programmes;
  •  The correlation between stocks reached an 85-year high in 2011, which tends to be followed by strong markets unless there is policy failure;
  • High dividend yield stocks have reached their most expensive level of the last 60 years (except for a few months in 2008), suggesting that stock pickers would do best looking elsewhere;
  •  European equities’ valuations have reached record cheapness versus US equities, based on more than 30 years of data, and the risk-reward for their out performance is excellent.

The backdrop

It has now become consensus that the medium-term outlook for economic growth in developed markets will continue to be dominated by deleveraging, low growth and shorter cycles. I always try to challenge consensus views, but in this case, I tend to agree. Of course, Japanese equity markets have been in this type of environment for over two decades now, so one can learn some lessons from their equity market behavior (see chart 1). I found that there are three such lessons:

  1.  There are occasional large rallies lasting several quarters, and the trigger for these rallies is invariably policy action. Recently, QE1 and QE2 have provided such rallies, and we are currently witnessing the LTRO rally, while later this year a QE3 rally may well unfold.
  2. The best stock picking factor in Japan has been 1 month return reversal – this is a quantitative finding rather than a tradable strategy of course, given the huge turnover involved if applied in practice.

3. Despite the second lesson just mentioned, stocks with the right characteristics can be consistent outperformers and become very expensive, as was the case in Japan for exporters with strong brand names. Of course, Japan’s deleveraging period occurred during a time of prolonged expansion in the rest of the world, which is quite a different context from developed markets’ current predicament. Stable growth names should be the structural outperformers.

Signals from the dashboard

My dashboard is populated with a wide variety of indicators. Frequently these indicators are not at extremes and give no strong message. I get excited, however, when indicators do reach multi-decade extremes, as these are often the basis of good investment ideas, as and when reversion-to-the-mean takes hold.
Reversion to the mean is the most powerful force in finance. The precise timing of the reversion is a tricky and crucial little detail, hence Keynes’ famous quote about markets remaining irrational for longer than you can remain solvent, or equally one of his other quotes that in the long run we are all dead. But still, economic, technical and sentiment factors eventually revert to the mean, and determining where we are with respect to that mean is a crucial starting point in investing.
My approach to investing is to take discretionary decisions based on a disciplined quantitative framework. To make well informed investment decisions, I find it valuable to discuss ideas with colleagues and peers, in order to test hypotheses, learn new insight and gauge where the consensus lies.
At a recent roundtable discussion, organised by a broker and attended by a select group of distinguished investors, one of the distinguished investors felt compelled to state, with some conviction, that reversion to the mean is dead! It made my contrarian alarm bells ring.
So, here are the three extremes that I am excited about.

If one takes all the 15 readings that were above 50, one also finds that markets have always risen in the next 6-12 months, except for in the 1930s. Therefore, history suggests that record high correlation is a bullish sign for equities, unless there is policy failure. But in the absence of policy failure, this evidence suggests that the outlook for equities in the 6-12 months after record correlation is good, for instance if policymakers embark on all kinds of ways to insert more liquidity in the system (however bad the long-term consequences of these actions may be).

In addition, the high correlation suggests that stock pickers should spend time trying to identify the stocks that have become too cheap as a result of the indiscriminate selling across the board. The second and third multi-decade extremes mentioned below make strong suggestions in that respect.

Extreme 1: correlation between stocks at an 85-years high in 2011

In 2011, the correlation between stocks reached its highest ever level, based on calculations by Empirical Research, using US data (see chart 2), based on data since 1926. This record high correlation is a result of all stocks having a strong tendency to move up and/or down together, which typically happens when volatility rises. Another reason for higher correlations is that ETF trading and indexing have become more popular in recent times, but in my judgment that is only a very minor element to the high correlation story in 2011. Interestingly we found that high correlations happens in the most extreme way when macro events dominate (recession, depression or other), when stock picking takes a back seat, stocks are being sold or bought together, and the proverbial baby is being thrown out with the bathwater. In fact, extreme levels of correlation are reached towards the end of a bear market, most of all.

Also, note that this phenomenon does not occur at the end of bull markets. Things are quite different at that point, as bull markets have a strong tendency to become more and more narrow, thus resulting in low – and not high! – correlation between stocks. Remember for instance the fact that TMT stocks peaked in 2000, while the rest of the market had gone nowhere since 1998.
What are the investment implications of this first multi-decade extreme observation? I have sliced and diced the data in a few ways. If one identifies peaks with perfect hindsight, markets have always risen in the next 6-12 months, with one exception, which was the period following the peak in Dec-1929.

offer good risk-reward from an absolute return perspective. It is even possible that the valuation of some of these high dividend-yielding stocks mean-reverts through the valuation multiples, not through stock price performance. And there

1, 2  - View Full Page