Try timing the market but accept your limitations

There are limits on an ability to judge future probabilities and never pay too much.

  • By John Authers,
  • Financial Times
  • Markets
  • Markets
  • Markets
  • Markets
  • Facebook.
  • Twitter.
  • LinkedIn.
  • Google Plus
  • Print

Timing the market is a mug's game. Trying to spot when a market has hit bottom or top and leap to another is prohibitively difficult, and it costs money. In the long run, all that happens to market-timers — unless they are very, very lucky — is that they spend much more in trading costs and end up worse off than if they had stayed put all along.

That is where I stand. And that is an issue because, as some alert readers have pointed out, much of this column implicitly aims at showing readers how to time the market. Is this month's correction the beginning of a bear market? That is a question of market timing and I have written a lot about it.

Strategies for volatile markets

With the right preparation, you may be able to handle a market drop.

Not only that but the investment industry is now geared up to help you time the market. Exchange traded funds tracking indices and allowing you to trade in and out of the market each minute make no sense unless they are vehicles for market-timing. And they are the hottest investment product.

And the industry of "advice" has now refocused on helping clients do market-timing. Rather than commission-charging brokers picking stocks for you, now fee-charging professionals move you from ETF to ETF.

If market-timing is such an obviously bad idea, then, why are so many people trying to do it? One answer lies in temptation: get it right, and you get rich.

I tried a simple model based on switching between two ETFs: long SPY (the S&P 500 (.SPX)) and short TLT (long bonds), and vice versa, since the Iraq invasion in 2003. From then to now, an investor needed only to switch five times to turn $100 into $5,888. Buying and holding the S&P would have netted $440.

This would have been madly risky. Nobody could time it so well without hindsight. But hindsight is so tempting.

A better case for market timing comes from Victor Haghani of Elm Partners in London. "Market-timing" is a pejorative term, implying all or nothing switches from 100 per cent in one market to 100 per cent in another. Instead, he suggests "expectation-driven sizing" can underpin "active indexing".

He points out that we can alter exposures to different asset classes over time according to reasonable expectations for the future. Winning is not guaranteed; we are talking of probabilities, not certainties. But mathematical literature on gambling problems gives precise guidelines on how much we should stake given different probabilities.

He says: "The proposition that investors should not change their exposure to the stock market requires investors to assume the expected excess return is constant over time. But is this a reasonable assumption? Most researchers and practitioners would tell us no, it isn't."

He is right about this. Models based on the idea that markets follow trends that revert to a mean produce clear guidelines.

The cyclically adjusted price-earnings multiple (CAPE) advanced by Robert Shiller of Yale University is scarily accurate over 10 years. If stocks are in their cheapest decile as measured by CAPE, their performance 10 years hence will be best; as stocks grow more expensive, so their expected return over the next decade proportionately falls.

Surviving a stock market bubble

When a bubble pops, you should be prepared. These options can help.

The odds shift as the CAPE changes, so why not adjust your holdings to reflect those odds? In finance, unlike many other sources of stress in modern life, you can hedge your bets.

So, Mr Haghani argues, use active indexing. Keep a portfolio of cheap ETFs in different asset classes and fiddle with their weights every so often as the odds on future performance shift.

Two serious arguments against this remain. First, the probabilities may shift in your favour but reversion to the mean is hard to implement in real time.

The academics Elroy Dimson, Paul Marsh and Mike Staunton show that relying on mean-reversion "in sample" (without hindsight) is perilous.

When stocks deviate widely from the trend, we are inclined instead to re-draw the trend. So, when they tested a model that left stocks whenever the forecast real return was negative, based on their valuations compared to the long-run trend at the time, they found that, in all the 15 countries they attempted it, buying and holding" equities would have worked better. That leaves us having to ask the question: reversion to which mean?

The other problem is fees. Mebane Faber of Cambria Investments tested a range of asset allocations that all relied on rebalancing — they reverted to a fixed percentage of each asset class each year, so that they took profits in those doing well and bought more of those that had fallen, a type of market timing.

All had differing allocations to gold, bonds, cash, real estate and so on. But over 40 years, the gap between the best and the worst was only from 4.12 per cent per year in real terms to 5.96 per cent. A 2 per cent fee per year would have transformed the best strategy into the worst.

Timing the market is not a dumb thing to do. It only makes sense to adjust to changing probabilities. But we should accept the limits on our ability to judge future probabilities and we should never pay too much for the privilege.

  • Facebook.
  • Twitter.
  • LinkedIn.
  • Google Plus
  • Print

For more news you can use to help guide your financial life, visit our Insights page.


© The Financial Times Limited 2018. All Rights Reserved.
Votes are submitted voluntarily by individuals and reflect their own opinion of the article's helpfulness. A percentage value for helpfulness will display once a sufficient number of votes have been submitted.
close
Please enter a valid e-mail address
Please enter a valid e-mail address
Important legal information about the e-mail you will be sending. By using this service, you agree to input your real e-mail address and only send it to people you know. It is a violation of law in some jurisdictions to falsely identify yourself in an e-mail. All information you provide will be used by Fidelity solely for the purpose of sending the e-mail on your behalf.The subject line of the e-mail you send will be "Fidelity.com: "

Your e-mail has been sent.
close

Your e-mail has been sent.