I'm on my way to Manchester to participate in a panel discussion at a conference on portfolio construction, organised by Dealer's Group. This is the second event of the week, following a similar session on Monday in London.
I think it's fair to say that the market conditions of the last eighteen months have thrown the financial advisory industry into a well-deserved bout of navel-gazing; both the topics on the conference programme, and the debate I witnessed at the Monday event, reflect this. Rare are the advisors who took protective action for their clients ahead of the slump in the markets, and many are those who are now trying to prevent the same clients from dumping their equities after a 60% fall in the averages.
It's unsurprising that the "buy and hold equities for the long term" argument is now under serious attack, and justifiably so. The past ten years have been one of the worst decades on record for owning shares - you'd have done far better in bonds, and gold has outperformed the Dow by a factor of 7 since 2000. Unfortunately, justifications for owning equities like the following one are still prevalent (this came from a firm recently voted "IFA of the year", by the way) - "with the average fall in a bear market 29%, and the average gain in a bull market 92%, the long-term investor is usually rewarded". There's a problem with that calculation, somewhere, and not in the scale of the recent bear market fall...
Yet I know from personal experience the risk of becoming progressively more bearish as market valuations move lower, and missing out on great opportunities when they present themselves. Is there no yardstick we can use to give us guidance on when to buy and hold shares, and when not to?
The author of the recently-released Barclays Equity-Gilt Study 2009, Tim Bond, makes the observation that the investment industry pays enormous attention to forecasting corporate profits and the economic cycle, but very little to aggregate valuations. The study shows, conclusively, that where you start from in valuation terms is by far the most important factor influencing subsequent equity returns. In other words, there's a great deal of wisdom in the apocryphal Irishman's response, "if I wanted to get there, I wouldn't start from here."
In his study Bond charts two valuation measures, Shiller's trailing real price/earnings ratio, using a ten year moving average of earnings, and Tobin's Q, the ratio of corporate equity market value to corporate net worth at replacement cost, over the period from 1900. The results are striking, and both measures reveal an equity market that moves over very long-term cycles between periods of excess valuation and periods of cheapness.
The real P/E ratio chart shows that equities were cheap (one standard deviation or more below the long-term average) in 1920, 1932, 1942, and between 1974 and 1982. They were expensive (one standard deviation or more above the long-term average) in 1928/29, 1965/66, and the whole period from 1995-2008.
The Tobin's Q chart shows equities as cheap in 1915-22, 1932, 1942, 1945-52, and 1973-83, and as expensive in 1928/29, 1966/67, and 1995-2008.
The good news for equity investors is that, as at end-2008, the real P/E ratio had fallen back to the long-term average, and the Tobin's Q measure to slightly below. The falls in equity market values in 2009 so far will have improved these valuation measures still further.
The bad news is that the 1995-2008 overvaluation of equities stands out on both charts as the most extreme in stock market history, and that, in the past, once equity valuation measures retreated from a period of overvaluation, they always moved to a period of substantial undervaluation, sometimes lasting several years.
In other words, although equities have become substantially cheaper in the last eighteen months, there's a good chance, if history is any guide, that they are going to become cheaper still, and stay cheap for quite a while. Bond goes on in his study to model P/E ratios from demographic trends, and implies that the price/earnings metric may fall further, hitting single figures early next decade.
If you have a long-term time horizon - say 20-30 years or more, for example in a pension plan - then a policy of regular investment in equities over the next decade looks sensible. If you have nearer-term liabilities to meet, then the decision is a much trickier one. In fact this seems to be what is preoccupying many advisors at the moment, and understandably so. Cash produces almost no income, Barclays' study shows bond valuations at or near a peak, and threatening capital losses from here, and the equity averages may provide little in terms of capital return for a few years. Many equities and corporate bonds offer apparently highly attractive income yields, but with the risk of dividend cuts and defaults.
So, looking longer ahead gives more reassurance than the near-term outlook. We'll return to the theme of income generation in forthcoming features and blogs...