Our 2008 call to invest... written by Luke Howard Taylor and edited by Chris Broadhurst:

The current mood of Mr. Market

At times of extreme turbulence it is often best to look back at history and for wisdom in the writing of those who have prevailed over long periods in the past. On the 16th October Warren Buffett wrote a column in the New York Times ‘Buy American, I Am.’ in which he stated that he was now buying equities on his personal account and that if prices remained attractive his non-Berkshire Hathaway net worth would soon be 100 percent equities.

The piece is available on the New York Times internet site for those who haven’t read it:

http://www.nytimes.com/2008/10/17/opinion/17buffett.html

Why is Buffett so confident about investing in equities now? Apart from the adages about ‘Be fearful when others are greedy, and be greedy when others are fearful.’ there is quantitative data indicating that valuations are extremely attractive. The volatility of the Dow Jones Average in the last two months is approaching the levels last seen in the Great Depression and smaller capitalised stocks are in some cases changing hands at levels we haven’t seen for a several decades. At times such as this returning to the work of Benjamin Graham, a mentor to Buffett and a guide to some of the most successful investors in history can’t be a bad thing.

Volatility as a proxy for fear and uncertainty

Throughout this piece we will generally be referring to the Dow-Jones as a proxy for the equity market, other developed country indices are experiencing similar volatility & recent declines but have not been around as long – nor have the level of consistency as this index of 30 leading US companies. As we have discussed, smaller companies here in the UK have fared far worst than the Dow-Jones but there is a good level of justification for using the DJIA as a proxy for equities as an investment class (at least in the developed world).

Put simply the Dow hasn’t experienced this level of volatility on a sustained basis since the Great Depression. The 1987 crash remains the largest extreme event but is tempered by the fact that the Dow had appreciated 50% in the 12 months up to the crash and ended the year slightly up. Putting aside this event (but not ignoring it as so many Efficient Market professors would...) Figure 1 shows that the level of recent daily changes in the Dow have been of magnitudes not seen since the 1930s. A better representation of the same data is shown in Figure 2, the 90-day volatility on the Dow over the last 108 years, which incorporates the movements of the index over the preceding 3 months. This figure clearly shows the currently level of volatility (just below 50%), which can be attributed to both uncertainty and fear in the market, has not been these levels since the 1930s. However, it does not yet exceed it, despite certain commentators claiming the contrary – the 90-day volatility exceeded 40% for 2 full years from 1931-33, whilst we have only thus far seen two weeks in this uncomfortable territory.

Daily changes in the level of the index are an indication of volatility, but can be distorted by the appearance of significant one-day events – such as the 1987 crash. Additionally, for long-term investors a 10% drop followed by an 11% gain the following day, whilst perhaps interesting, is of little consequence – unless he should panic and sell before the gain! Sustained volatility at higher rolling level is of more practical importance such as the 3-month volatility shown below:

The central level of the Dow

The principal part of Graham’s work focussed on the appropriate ways in which analysis could be conducted to indicate if individual securities represented sound investments with appropriate margins of safety or were speculative in nature. He also made reference to the ways in which investors could attempt to establish if equities as a class were either high or low. In the 1951 edition of Security Analysis Graham & Dodd have this to say:

‘We have made some hindsight calculations of results from the use of a “central value method” of purchasing and selling, as a group, the stocks in the Dow-Jones Industrial Average. This method involves an actual appraisal of the Dow-Jones Unit, together with the decision to buy at a fixed discount below and to sell at a fixed premium above such value. In effect however, it is not far different from a simple effort to wait for historically indicated low levels to buy and high levels to sell. On paper the results are rather attractive.’

The ‘mechanical method’ referenced is contained in the appendix of the work and although they caveat that ‘It should not be assumed that the method is either theoretically sound or practically dependable under all circumstances’ we consider that it remains one of best methods available for gauging the level of the market since it is based on earnings over time and compared to the available return on AAA bonds.

The advantages of such a method are numerous – we have already mentioned the consistency of composition of the Dow-Jones, there is also the fact that there are only 300 data points for the calculation of the earnings figure (far more relevant before the advent of computers of course). The data for earnings on the Dow-unit are published by ValueLine from 1920 to the present and it is therefore possible to update Graham’s table and examine how the Dow has performed relative to this central value in the last 57 years since he drew it up.

The immediate observation to be made from the graph is that based on this original method in recent decades the DJIA has been significantly over-priced, indeed the peak of the bull market in 2000 was of far greater magnitude than the heights of 1929 market before the Great Depression. However, this fails to capture an important change since Graham constructed this original model. The yield on AAA bonds has reached heights which were unimaginable in the 1950s – this is shown in the figure below:

The original formula used by Graham divided the 10-year average earnings by twice the yield on AAA-bonds – at the peak in 1981 at 14.2% this implied a 10-yr P/E of only 3.5x as the central value for the Dow-Jones. Clearly the inflationary policies of monetary authorities have necessitated a change in the formula to adjust for the structurally higher inflation in the current system of fiat currency.

Adjusting the formula for inflation

During periods of high inflation, or perhaps more accurately stated as ‘periods of monetary devaluation’ the bond yields of even the most secure corporations represented by AAA bonds must rise to compensate investors for the erosion of their purchasing power. It seems logical therefore to eliminate the inflationary element of the higher yield and estimate the real yield on AAA bonds.

Having adjusted for changes in purchasing power of the dollar the real yield available on AAA corporate bonds is far more volatile. In particular during WW2 bondholders saw meaningful erosion of their purchasing power. It is obvious that we cannot simply exchange the real yield for the nominal yield in Graham’s formula – zero yields would result in infinite central values for the Dow-Jones – which is clearly not a satisfactory result.

Similar to the process that Graham has used in averaging the earnings over 10 years we can adjust the AAA-bond yield to take account of the prevailing levels of inflation. We contend that this allows more meaningful comparison to Graham’s original work. If we take the period over the original analysis 1924-51 then we see that US CPI averaged 1.8% whilst the average AAA-bond yield was 3.6%, so that the real yield on AAA bonds over the 27 years was 1.8% per annum.

The original model in Figure 4 begins to meaningfully diverge from the central value around 1964 by which time the average of the 27 previous years of CPI had increased to 3.0% per annum. The peak value for the 27 year average is in 1993 at 5.6% encompassing the period from 1966-93. If we use this rolling 27-year inflation figure we can adjust the original model for periods of higher inflation as experienced since the 1950s. We can now see how the adjusted central value provides a more realistic trailing 10-year P/E in recent decades and avoids the clearly ridiculous low P/E’s that the original model implied in the 1970s and 80s.

Is the Dow-Jones cheap or expensive?

We originally stated that there is quantitative data that stocks are very low, which serves to back up the comments of Warren Buffett that this is an excellent time to be invested in equities. Indeed we reiterate to our own investors the following lines:

‘Today people who hold cash equivalents feel comfortable. They shouldn’t. They have opted for a terrible long-term asset, one that pays virtually nothing and is certain to depreciate in value... Equities will almost certainly outperform cash over the next decade, probably by a substantial degree.’ 

We can’t distil the thousands of hours research into individual investments we’ve done over the years that brings us to the conclusion that the general price level of equities is very low. We can provide specific examples of companies trading at irrational levels than no private businessman would ever consider selling his firm for – and yet – some people are selling (forced or otherwise) and we’re happy to be buying from them and we think that you should be too.

What else we can do is show you Figure 4 from earlier, this time adjusted for the prevalence of inflation on which we could wax lyrical for many pages but will resist the temptation. The recent low reached by the Dow on the 27th October at 8,176 was 80% of its adjusted central value, below the level offered in 2003 at 82% and the best value since 1974 or 1978.

Could equities go lower? Certainly. In the Great Depression the Dow reached 40% of its central value in 1932, which would be a further 50% decline from this level – but from its lows it then appreciated 266% in a year. By doing just the opposite of the investor who takes out his money after a 10% decline and misses an 11% gain – if your investment horizon is measured in years and not days or months then this is an excellent time to be out of cash and invested in the stock market.