Frédérick Dodard looks at European stocks through the 1970s, 80s, 90s and 2000s and finds that the market will often pay much less for companies - or much more - than the fair price suggested by their earnings

Within the equity world, two key items that can change dramatically at inflection points are earnings and valuation - that is, what the market will pay for those earnings. We have observed powerful evidence that both risk aversion and risk seeking can drive stock returns, not only in the short term but also in the medium and long term. There is an old mantra in the financial world that ‘stock prices follow earnings', and while this may well be true over very long cycles it clearly is not always the case: the market price of these earnings also matters.

This potential divergence of earnings and the mispricing of stocks by the market are of significant interest for investors. The following is a summary of a study of the impact of the price/earnings (P/E) multiples on European equity returns over the past four decades, conducted using data provided by Ned Davis Research, Inc. on the European equity markets, as proxied by the MSCI Europe index. It decomposes the MSCI Europe return between earnings growth and the P/E multiple impact, using cumulative data from January 1970 to December 2011. We also found it informative to look at decade-specific data from January 1970 to December 2011, as this highlights the different fortunes of European equity investments over these time intervals.

The information in figure 1 should remind every investor of the importance of dividends - they are a significant component of total return. In fact, since 1970 the price return of MSCI Europe was 5.7%, even though the total return was 9.6%, meaning that 3.9% of the total return came from dividends. Interestingly, the recent past has been so turbulent, following the 2007-09 financial crisis and the euro sovereign debt crisis (ongoing from 2010, at the time of writing), that the valuation of the MSCI Europe ended 2011 well below its 1970 level. All in all, the de-rating observed on European equities contributed to a negative P/E effect over time of -0.6 per year.

This study will focus on the price return component of the MSCI Europe return and will detail the contribution from earnings and the contribution from P/E expansion or contraction. For example, since 1970, 6.4% was derived from earnings growth, versus 5.7% in price return, with -0.6% coming from multiple contraction. Said differently, 112% of the price return was driven by earnings growth. However, other regimes indicated a significant impact from P/E multiple compression or expansion.

We also analysed specific decades since 1970 to measure whether the contribution between earnings growth and P/E multiple was stable or varied over time. Each of these periods was unique. The 1970s period (figure 3) was marked by a difficult equity market, low economic growth and high inflation (or stagflation). The inflation effect is typically assumed to be a negative for P/E multiples, and in this period that was indeed the case. While earnings grew by around 7.4% annually, the price appreciation of the MSCI Europe grew just 0.3% a year because P/E multiples contracted significantly. At the end of the decade, the P/E ended just above seven - not a very rich multiple.

The 1980s (figure 4) saw a reversal of the 1970s. Not only did earnings grow around 9.5% annually, which was very strong but, in addition, P/E multiples expanded significantly, leading to a very strong average annual return of close to 21%. This multiple expansion can be explained by a number of factors: first of all, the 1980s saw a structural decline in inflation and a decrease in long-term interest rates. Secondly, after the severe recession that struck the developed world in the early 80s, European economies recovered and economic growth accelerated. This led to a more favourable environment for companies' profitability. This was the best decade for European equity returns and it is striking to observe how much P/E expanded.

The 1990s (figure 5) continued on this very positive note. It was a very strong decade in terms of annual MSCI Europe returns. While this decade had low average annual earnings growth at nearly 4%, it was marked by huge P/E expansion. In this decade, the P/E expansion accounted for 70% of the price return (ie, 9.1/12.9). MSCI Europe ended the decade with a P/E close to 30, the highest in our study. This was the end of the Golden Age of equities but, sadly enough, it took several years for investors to discover that the price of earnings they were paying was too high.

The 2000s (figure 6) represented a volatile period and was disappointing for equity investments. Markets entered the new millennium so expensively priced that they deteriorated significantly during the TMT crash, coming back strongly from 2003 to 2007, only to crash once again. During this time, earnings growth was actually positive, although well below long-term averages. Yet, price return was very negative, indicative of a severely contracting P/E multiple.

So where does this leave us today? It is clear that we have lived through a contracting P/E cycle for at least a decade, as any number of poor-performing equity benchmarks will show. This followed a 20-year long decade of price appreciation and multiple expansion that could not continue forever. We have seen that stock price multiples eventually mean-revert. Today, the market consensus sets a P/E close to 12.75 for 2012 earnings, which is neither too demanding, nor too cheap. But after all, this was the level where the market was paying earnings for almost half of the time.

The good news, though, is that in these regimes, the annualised rate of return for European equities in the past was close to 11%. We do not forecast anything close to that for the next 10 years, since we do not believe that there will be any significant multiple expansion, given the strong headwind from the euro-zone crisis. A weakening economy, along with rampant pessimism, could continue to drive P/E multiples down further. While it is always difficult to predict which direction the short term will take, as P/E shrinks, equities become better and better value, and their future returns are not only driven by improving earnings but also by expanding multiples.

The intent of the ‘stock prices follow earnings' statement is often at the micro level - that is, if you find individual companies that can grow their earnings faster than the market, then their stock will reward you. But earnings are only half the equation. When the current price of a stock already reflects that earnings optimism, future stock prices may not follow earnings, as the 1970s and 2000s' downside showed.

On the positive side, the 1990s showed that stock prices can benefit greatly from expanding P/E multiples. In periods like these, it might be more accurate to say that ‘stock prices follow P/E multiples.' Investors should consider not only the future earnings expectations but also the price that the market is willing to pay for those earnings.Attempting to assess both of these factors - a difficult task to say the least - is a more rational approach than focusing solely on earnings growth and helps investors navigate through the perils of excess pessimism or unsustainable euphoria.

Frédéric Dodard is a managing director and head of multi-asset class solutions at State Street Global Advisors