Double or quits: valuing Wall Street

Receive free Advice & Comment updates

We’ll send you a myFT Daily Digest email rounding up the latest Advice & Comment news every morning.

Time to get to work — as my generation used to say but no longer can for fear of being accused of bullying. So don’t worry. I will be sweating the hard stuff on this week’s topic, which is whether to raise my portfolio’s US equity exposure.

I wrote previously that I want more risk. This will require buying some weirder assets. A friend, for example, is keen on Venezuelan debt right now. Lots of readers think me an idiot for having no crypto (they’re half right).

It also means fewer funds. Which is why — as flagged in May and reiterated a fortnight ago — I have taken profits on my Amundi European banks ETF. A handy 15 per cent return in three months. Annualised that’s 75 per cent. See what nonsense extrapolating returns is?

Hence the £58,000 burning hole in my board shorts. My impulse is to purchase more US shares. It seems plain wrong to have the same allocation to Japan (where money has historically gone to die) as the US (which spawns trillion-dollar companies before breakfast).

Who wouldn’t prefer to buy more cheaply tomorrow if they could? It comes down to market timing

And I have twice as much again in UK stocks! Sure, my liabilities are in sterling and the FTSE 100 is cheap versus history and other equity markets. But come on. The MSCI All Country World index has the US at 15 times the weighting of the UK.

No, my proportions are all over the place — as my children constantly remind me, pushing their fingers into my midriff. And an underweighting to US equities seems a priority to address. Is now a good time to buy, however?

The question is topical. Hedge funds have cut their exposure to US shares to the lowest level in a decade, it was reported this week. Of course, the headline could also have read: “Bullish investors pile into US stocks as hedge funds take profits.” There’s a buyer for every seller.

There are always conflicting views — and where they meet is the price. Fine. But who wouldn’t prefer to buy more cheaply tomorrow if they could? It comes down to market timing, therefore.

Can it be done? Should it even be attempted? Let’s start with the latter, because if you’re happy with a negative answer, we can simply load up on US stocks now and be done with it.

The likes of Warren Buffett would tell us just to look at a very long run chart of the S&P 500. It goes from the bottom left to top right, on average by 6.5 per cent annually in real terms. Don’t think about it too hard. Just buy.

Statistics back up the folksy wisdom, with the volatility of US equity returns declining more over time than can be accounted for by simply a longer investment period. The risk of losing money falls by holding on.

This phenomenon relates to our first question of whether it is possible to time the market. If returns were random around an upward drift, we wouldn’t observe this decline in volatility. That it happens is because returns fluctuate around a long-run mean.

In other words, stocks tend to do better after periods of below average returns, and worse when returns have been strong. Sounds obvious. But until recently economics and finance was dominated by so-called random-walk theory.

For readers interested in this, Andrew Smithers goes into the numbers in detail in his excellent book The Economics of the Stock Market. But the implication of returns which revert to the mean is that I should be able to time my purchase of US stocks.

How? By trying to calculate when future returns are most likely to be above or below average. To do that we need to value the equity market. There are lots of methodologies out there. Luckily it is simple — if laborious — to back-test them to see if they have predictive power or not.

The least worst either compare prices today with some measure of normalised earnings or the replacement value of aggregate company assets. The famous Shiller price-to-earnings ratio is an example of the former. Tobin’s Q the latter.

Using data going back a century or more, both have been remarkably similar when flashing red if the S&P 500 has deviated too far in either direction from where mean-reverting returns would suggest is fair value. Subsequent returns have justified buying or selling, on average.

And both have roughly twice the success rate of a standard price/earnings ratio, which remains the most popular valuation tool out there for some reason. Whichever you choose, timing works best when valuations are at extremes — which by definition isn’t often — as Howard Marks eloquently explained in this newspaper on Monday.

Where is the S&P 500, then? Perhaps you don’t want to know. The current Shiller PE ratio (which averages the past decade of earnings in its denominator) is above 30 times, compared with a mean of 17 since 1870. That suggests the index has roughly to halve to return to normal.

Likewise, US companies are trading in aggregate at a level equivalent to almost double the replacement value of their net assets — based on Federal Reserve data. The last time the Q ratio was fair value was during the financial crisis.

Therein lies a problem. Shiller and Q have indicated “sell” for the past 15 years as US stocks have been shooting the lights out. Therefore I need to decide whether they are sound but early or somehow broken and to be ignored.

Many believe the latter. And who cares anyway, if the S&P 500 always goes up? Then again, if I can buy it for half price, I’ll wait. A huge decision — one requiring a deeper dive into valuations. More to come.

The author is a former portfolio manager. Email: [email protected]; Twitter: @stuartkirk__