UK Value Investor

This is just a minor update to my previous post about allocating assets to stocks depending on the current value of CAPE compared to its long term average.  In the graph using Shiller data I plotted a straight line at 16.35, the current long term average of CAPE.  However, it would perhaps have been better to show the CAPE average as it would have looked at the time, therefore removing the benefit of hindsight and showing what information investors would have had at the time.  And here it is, with the long term CAPE average shown in black:

I've mentioned my tactical asset allocation efforts in a couple of previous posts.  Both of those have been somewhat vague about how I actually decide on the stock/bond split, although not deliberately so.  If Ben Graham can shout out the Net Net method to the world over 50 years ago and not have its effectiveness affected, then surely my tiny wispers on the web can do my approach no harm.  In fact it may to some miniscule degree make the markets more efficient.  Like the proverbial fly stopping an oncoming supertanker.  Perhaps I may even win a Nobel Prize, but I doubt it.

Both Shiller and Smithers and others have shown that it is possible to value markets and that market valuations are bound by an invisible elastic thread to both earnings and assets.  More importantly, these valuations allow you to say something about the expected future returns.  Higher valuations mean lower expected returns and lower valuations mean higher expected returns.  Also, average valuations mean average expected returns.
Shiller's CAPE (Cycically Adjusted Price Earnings) is my starting point when looking at earnings related valuations.  This is the ratio of current market price to the market's average real earnings over the past decade.  For the S&P500 the long term average CAPE is currently 16.35 and is shown below as the horizontal line.  The data for this can be found here.

After reading a post on The Div-Net, I started to think about how I differ from dividend investors and why.  The first point to make is that I'm not a dividend investor, in fact I consider myself a trader rather than an investor.  An investor to me is someone who buys something with no explicit intention to sell it.  This typically means they are either buying it for the income (dividend investors, landlords, Warren Buffett etc) or perhaps they are buying it to let the capital appreciate for decades or to let the kids inherit. 

So why would I trade rather than invest?  Well, there is some evidence that the returns are better if it's done properly (which I'll try to cover at some point), and more importantly it's a better fit with my personality.

Since I'm buying with the intention to sell, how long do I expect to hold my stock? 

I recently read "Painting by Numbers - An Ode to Quan" by James Montier and Dresner Klienwort via a link from Richard Beddard to Greenbackd.  This paper, and the papers it refers to, have helped strengthen some of my existing opinions about stock picking and investing in general. 

My opinions are also those of Ben Graham towards the end of his life, that "I am no longer an advocate of elaborate techniques of security analysis in order to find superior value opportunities. This was a rewarding activity, say, 40 years ago, when our textbook "Graham and Dodd" was first published; but the situation has changed a great deal since then. In the old days any well-trained security analyst could do a good professional job of selecting undervalued issues through detailed studies; but in the light of the enormous amount of research now being carried on, I doubt whether in most cases such extensive efforts will generate sufficiently superior selections to justify their cost" (www.bylo.org/bgraham76.html).

I've long been fiddling around with various mechanical methods of adjusting an almost passive index investing stragety to improve the risk/reward ratio.  This is sometimes known as Tactical Asset Allocation (TAA).  I thought I'd put up some charts of my efforts.

The lines in the charts are for four portfolios:  Cash, with the returns calculated using the average instant access interest rates borrowed from the rather excellent Swanlopark;  The FTSE 100 with dividends reinvested;  A 60/40 FTSE 100/cash split rebalanced each year; Another FTSE 100/cash split which is rebalanced annually using my asset allocation function which is fed with the FTSE 100 real earnings over the period in question.

I'm going to record an analysis of each of the trades that I make so that I can learn from each trade.  I'm sure that sometimes there may be nothing to learn, but that's not always going to be the case and it certainly wasn't with my first value stock back in 2008.

I had come to value investing from a more typical mindset where I was trying to predict the future in order to see where it was going to be most profitable to invest.  I had been heavily invested in energy stocks through unit trusts back in 2005-2008 and they'd done incredibly well, almost doubling my money.  But I had no idea how to value these unit trusts nor the stocks within them.  When oil went to $146 I thought I was pretty smart.  But we all know what happened next.  I lost about 50% and that's a really big drawdown, one that made me almost physicall sick.

I'm a big fan of CAPE (cyclically adjusted price earnings) and Tobin's Q as tools for understanding expected future risk and returns from a stock market.  After reading Wall Street Revalued: Imperfect Markets and Inept Central Bankers, I'm an even bigger fan. 

The logic is simple.  Market valuations must be tied in some way to earnings (the discounted cash flow that I hear so much about from earnings based investors).  Earnings for an entire market, over the long term, are somewhat predictable using past earnings.  These earnings are generated by assets and so market values are tied in some way to assets.  CAPE seeks to value markets using earnings and Tobin's Q does it with asssets (or net assets to be more precise).

I think it can be difficult to learn anything without actually living it. So handily Ennstone, my first purchase of a value stock using not much more than price to book, has fallen over into the abyss. This is good for a number of reasons, although of course not so good for the staff.

The collapse of Ennstone has helped me re-think my approach to value investing and most importantly helped me to clarify to myself what it means to me to invest at all.

I want to run some back testing on a couple of value portfolios. They're both based on picking relatively small cap stocks with a low PB, preferably below 0.5.

The first portfolio holds the stock for a year and if the criteria are no longer met (cap too large or PB too high) I sell, otherwise I hold for another year.

The second method buys the same stocks but holds them until the PB reaches 1, regardless of cap. Will this outperform the first method or will I just end up holding a bunch of stocks that are going nowhere, i.e. never reach par? Only back testing and time will tell.

This could be mixed in with the sliding cash holding method of the last post, or I could just be strict with the criteria and if the market is overpriced then there won't be many if any stocks with a PB of 0.5... there certainly weren't many between 1999 and late 2001.

I had an idea today that I might do some back testing with. It's kind of a slant on MPT with a dynamic cash allocation. What if instead of having a fixed cash allocation of say 30% which you re-balance to each year, you instead use the p/e ratio of whatever indices you're invested in (S&P, FTSE100 etc) to determine that allocation?

Assuming a p/e of 14x is fair value, 7x is very cheap, 21x is very expensive and 28x is an insane bubble, what if you used double that number as the cash %? So each year, if say the FTSE100 had a p/e of 14x you'd set your cash allocation at 28%. If the p/e was 7x you'd have the cash allocation at 14% and if the p/e was 28x you'd have the cash allocation at 56%.

This seems reasonable at first glance since at 14x, 28% cash is a reasonable safe amount. At 7x the stock market is historically cheap (with a better than average probability of going up) and yields are high so you load up on it with only 14% cash. At 28x the stock market is historically very expensive (with a better than average probability of going down) with low yields so you don't want much exposure, i.e. 56% cash.

I think I'll do some back testing if possible and see what history says.

I saw a nice chart that illustrated the point of following the long term trends. This ties in with Long Valuation Waves where the p/e of a mature market tends to ride up and down over a time scale of one or two decades. Last time I wrote about this I mentioned the roughly inverse correlation between energy consuming and producing companies. The 1965-1983 period was good to energy producers and inflation bets (gold), the 1983-2000 period was good for energy consumers since energy was cheap.

Another way to view these long cycles is that there is always a decade long bull in something. If you invested $35 in gold in 1970 it was worth $627 in 1980. Put that into the Nikkei until 1990 and it was worth $3,548. Put that into the nasdaq until 2000 and it would be worth $35,105. Finally stick that in oil and it's now worth $159,591.

Those weren't insanely hard trends to spot. Of course, you'd probably be hedged to some extent on a couple of major plays, but in general I still love this strategy. Currently I think the trend for the 2000-2010 period is oil/gold, although emerging markets have been a pretty handy place to be. I think the trend for the 2010-2020 period is likely to be renewable energy and energy efficiency, but again China could be useful. For 2020-2030 the trend may be back to consumption once a new global energy infrastructure is in place.

I made up a little graph the other day to look at the house price/earnings ratio over the last 25 years or so. I was actually interested in creating some kind of affordability index based on prices, earnings and interest rates, but that didn't produce anything with clear trends. However, simply looking at average UK house prices against average earnings gave surprisingly smooth trend lines.

The p/e ratio was 3.05 in 1982 and 1983, a low. Then, EVERY SINGLE YEAR, it increased to a peak of 4.32 in 1989. Then it decreased EVER SINGLE YEAR, finally hitting 2.64 in 1996. Then, once again it increased EVERY SINGLE YEAR up to 5.69 in 2007. The question now is if it drops in 2008, as seems likely, then does that mean a drop EVERY SINGLE YEAR until we hit a low, somewhere around a p/e ratio of 3? If so then that means a real drop of about 47% which is something in the region of how much the International Monetary Fund said UK houses were overpriced by. Also, given that we've been so far over the long term trend, it doesn't seem beyond the bounds of reason that we may drop below a p/e of 3 for some time as part of mean reversion, which could easily result in a drop of more than 50%.

Unless there is a economic crisis or extreme interest rate rise I don't see how this is going to happen in just a few years (not even the 7 years of the last downturn). It seems more likely to me that we'll have a property downturn for a decade or more finally resulting in fair value, before we start the march up again.

Long valuation waves are the secular bulls and bears of the stock and commodity markets. The theory is that due to fundamental (for the last century at least) reasons, stocks and commodities become approximately inversely more or less expensive over long 30-40 year waves. General valuation measures such as p/e help indicate which stage of a long valuation wave the market is in and also what sort of returns you may expect from stocks or commodities.

This is of interest to me because my investment strategy is value based and long term, i.e. I don't like to buy or sell more than once a year and I expect to be investing until I drop dead.

Currently my knowledge of these waves only scratches the surface but here's the gist:

1. Let's say that currently we have the capacity to provide all the commodities that the world needs. Commodity prices are low because of that and so investing in new capacity is very limited as it isn't profitable and we don't need it.

2. Population grows, standards of living rise, partly because of low commodity prices. This increases demand.

3. Gradually this increase of demand, over a period of years, begins to strain the current capacity to supply the commodities. Miners, farmers and others start to make more money and begin thinking about buying more land to plant crops or dig more mines. However, these things require big injections of capital so the producers don't try to increase capacity straight away, they wait until they are sure that it's worth it, i.e. until commodity prices are high.

4. Producers start to build new mines, farm more fields and pump more oil. However, it can take years to get new mines up and running, and the same goes for oil and food production (although less so for food). During this phase commodity prices are very high and this hurts other sectors of the economy as consumers have less spare money to spend on things other than food and fuel, and the other things also cost more because of raw material costs. The rest of the economy starts to make their use of commodities more efficient or just cut back on their use due to the high costs.

5. Eventually the producers do increase supply. At the same time the consumers are consuming less (or not increasing demand so fast) because of high costs. The producers can now supply enough commodities to consumers and prices drop. Prices keep dropping as people don't want to invest in decreasing assets and also as it becomes obvious that supply exceeds demand.

6. We are now back at stage 1. All of this took 10-20 years to go from stage 1 to stage 5, then there may be another 10-20 years at stage 5 as money flows into other areas of the economy (technology or housing for instance) which very gradually increase demand for commodities again.