Three Strawman Investing Strategies

Regular reader and commenter Strawman left an insightful and interesting comment on a recent post about cash in investing strategies.  It was such a good comment, and so central to the theme of this blog (the search for the ultimate investing strategy), that I thought that it deserved it's own post.

Strawman starts:

I also accept that the strategy of buying into quality companies when the price is good, but holding cash is only one way to make this happen.

All good so far. Strawman continues:

Basically your strategy seems to be:

STRATEGY1: 'Move my wealth from cash into X when the (real long term) P/E for X is really good'.

That's correct.  But let's just clarify a few things.  Firstly, the P/E in instantaneous, and changes over time.  Every time the market presents you with a price (P), you can calculate the corresponding P/E (based on either last years earnings (E), or sometimes, projected earnings).  Therefore, I don't really know what you mean by "real long term P/E".  Also, I think that you are trying to use "P/E" as a pure indicator of value, which is not quite right, as it depends on the growth prospects of the company.  However, let's assume that you can generate a "value indicator", and, to keep the language consistent within this post, let's call that value indicator "P/E".

Immediately you have a problem - ie how to define 'really good'.

Yes, although I'd phrase it as the problem of finding the optimal point of cheapness.  Too low, and nothing will ever be that cheap.  Too high, and you'll run out of cash too quickly and miss out on the real bargains.

Okay, let's consider another strategy:

STRATEGY2: 'Move my wealth from where it currently is, to X when the P/E for X is much better than the P/E for what I am currently holding'.

This has the same problem as STATEGY1, but why is it an inherently worse strategy?

Good question.  Let me explain.  Firstly, let's start with some realistic nominal P/Es.  In the last decade or so, equity P/Es have averaged around 16 (which equates to a return of 6%).  Longer term, however, equity P/Es seem to average around 10 (and hence a return of 10%), and there have certainly been plenty of times over the last 50 years when equity P/Es have averaged around 6 (return of 16%).  Let's also say that, today, "Cash" has a return of around 6%, and therefore a P/E of around 16.

It is not clear from your above definition whether X could be "Cash", or not.  If X can be cash, then the two strategies could be considered equivalent if Cash was one of the X's (in which case, why not just use Strategy 1), or sub-optimal if Cash wasn't one of the X's.  Likewise, if X cannot be Cash, then I would consider Strategy 2 to be worse than strategy one.

The reason that using Cash is a better strategy than a pure equity strategy is that it the correlation between Cash and the equities is lower than the correlation between the various equities.  That is, during crash events, nearly all equities go down, but cash doesn't (during the crash).  This low correlation creates more P/E or value arbitrage opportunities.  This necessarily makes the investing strategy that includes Cash the better strategy.

Now consider a strategy that avoids the 'how good is good?' problem:

STRATEGY3: 'always move my wealth into the form with the highest P/E' (minus the transaction costs, so you don't squander your time and wealth on E*TRADE brokerage fees).

Will this perform worse than STRATEGY2? You will probably say 'Yes, because you never pick up a bargain - you will never effectively double your P/E ratio, because you already own all the stock in X you can afford.'

Firstly, I'll assume that you meant "lowest P/E" rather than highest, as the lowest P/E represents the best bargain.  Secondly, it's not about moving your wealth into best (lowest) P/E, it's about maximizing the transitions between high P/Es (overpriced assets) and the low P/Es (cheapest assets).  That it, maximizing the number of "Sell High, Buy Low" transactions that you execute.  Technically, it's actually about maximizing the cumulative gains across these transactions, which is essentially maximizing the product of the number of transactions and the size of the transactions, given fixed thresholds.

In other words (as I don't think I've explained it that clearly), using Cash, there will be both a) more gaps between "high" and "low" P/Es present, and b) wider gaps, presenting both a) more opportunities and b) bigger opportunities to ratchet up ownership, and hence wealth.

But be wary - as soon as you make that claim, you are admitting that you are gambling on short-term price fluctuations - ie playing the day-trading game, which this strategy is supposed to avoid.

You can only make money by waiting for a hot-stock to go into gloom, and then buying it. Just because you don't have a 'sell' strategy doesn't mean you aren't playing the zero-sum-game of the day trader.

I don't see how arbitraging the value "gaps" if and when they occur in statistically optimal way is gambling on short term price fluctuations.

And even if you are correct, and this one dimension of the strategy is a zero sum game, that just means that is has a 1x multiplier against the other dimension.  I, however, believe that both dimensions are positive sum, and therefore have a multiplier effect of the overall performance of the strategy.

Let me explain.

I could just as easily buy these quality companies "now" (rather than waiting for "cheapness"), and the expectation value would be positive.  What it wouldn't be was optimal, and neither would be as high as I'd like.  As I've argued before, investing in quality companies is a positive sum game because money flows from the companies' customers as revenue, and, after subtracting costs, the remaining profit flows through to the shareholders as either shareholder equity or dividends.  So, over time, there is more money for the shareholders that there was initially, hence it is a positive sum game.

So, if your still believe that I'm "gambling on short-term price fluctuations", and "trading a zero sum", then you are saying that this dimension is only providing a 1x multiplier.  I, however, believe that both dimensions significantly improve the strategies performance, and together, they create a great - dare I say - near optimal - investing strategy.

The reason that I don't have "sell" in my strategy is simply because the "what to buy" question, and the "when to buy" question are higher priority, and not yet complete.

.. and now I really should do some work, and make some money instead of just dreaming what I will do with my fortune once I have made it .. 

I see this problem as fortune creation, rather than fortune maintenance.  It's about building the snowball slowly and steadily, as fast as possible, if that makes sense...