by Hilary Natoff, co-Portfolio Manager of the Fidelity Global Demographics Fund
November 2013
For most people, the challenge of investing at the outset is to accumulate wealth over a relatively long time. The job of the investment industry is to build strategies for these investors that generate attractive returns over the long term. While numerous studies show the value of long-term, buy-and-hold approaches to investing,1 these approaches clash with the short-term behaviour evident in today’s equity markets.
That greater focus of equity investors on the short term is shown by the decline in the average holding period for stocks on global exchanges. In the US, the average holding period of a stock on the New York Stock Exchange was around seven years in 1940.2 By the time of the tech bubble in 2000, it had fallen to about one year. Now the average holding period globally is less than three months.
A key issue that feeds this short-termism is the habit of most sell-side analysts to focus on short-term earnings projections. In fact, while the overall number of analysts covering large-cap global stocks is growing, the focus remains disproportionately on near-term earnings forecasts. Few analysts covering global stocks forecast greater than three years ahead.3
With the greater short-term focus of market participants and sell-side analysts, coupled with guidance from companies, the equity market has become relatively efficient at pricing near-term earnings expectations (as measured by a declining error rate in one-year IBES forecasts 2006-11).4 The corollary to this is that the market is less effective at evaluating longer-term earnings, thus neglecting the longer-term value of companies exposed to structural growth drivers. This represents a clear opportunity for investment strategies that can exploit this market failing.
Stock outperformance is driven by either superior earnings growth or a higher valuation multiple being applied to the earnings profile of a company. Over short holding periods, changes in valuation multiples are the key driver of returns since earnings expectations do not typically change by large amounts in the short term. However, in the long run, the opposite is true. Performance is increasingly explained by changes in earnings growth and the importance of the valuation multiples at which stocks are traded diminishes, as the chart of the 500 of the world’s largest listed companies below shows. With the knowledge that earnings are more important than multiples in the long run, how should we go about assessing the long-term value of companies?
Earnings multiples are commonly used to make snapshot comparisons between similar companies within industries or to measure value versus sector or market averages. For companies sensitive to the business cycle, these measures run the risk of overstating the value of the business based on peak earnings in periods of strong economic activity.
Discounted-cash-flow analysis is an alternative method of valuing a company where the value of an asset is defined as the present value of its estimated and discounted future cash flows. Instead of trying to project a company’s cash flows to infinity, however, a terminal value is applied to cash flows beyond the next few years (at which point forecasts drastically peter out in any case).
In these models, 60% to 75% of the value of a company is typically determined by this terminal value and great care must be attached to its calculation. For many high-quality businesses, the calculation of this rate can be sensibly informed by their exposure to structural growth themes, which, in turn, can justify higher growth rates than GDP.
Bucking reversion
Evidence is mounting that discounted-cash-flow models are useful for establishing a valuation that reflects long-term drivers for a certain type of company. Such companies are high-quality earnings growth compounders exposed to structural growth themes and whose earnings are less sensitive to the business cycle.
Exposure to structural growth (such as an industry-leading position in a strongly growing market) allows a company to generate a steady stream of cash that can be reinvested into a growing business. It is the ability of these companies to reinvest that cash into the strong structural growth opportunities in their markets – by increasing capital expenditure, which in turn enables stronger sales and profits growth – that provides the compounding engine for sustained growth in earnings.
Danish healthcare company Novo Nordisk is one of these companies. It is exposed to strong and sustainable growth in the market for diabetes via its insulin products. It is also a company whose value has been better judged by discounted-cash-flow analysis than earnings multiples. The stock price has risen from 102.25 kroner (A$19.75) at the end of 2002 to 945 kroner by mid-2013 – a compound annual growth rate of 24%.5 Diabetes affects more than 370 million people worldwide. By 2030, this is expected to rise to 552 million, a 49% increase.6 Novo Nordisk has a commanding 49% share of the global insulin market. Having built a reputation for innovative treatments, the company is spending around 15% of its sales on research and development of new products.7 Yet it’s clear that three-year, forward-earnings forecasts have consistently failed to keep up with actual earnings growth. Earnings growth estimates at three years and beyond tend to be subject to a degree of earnings fade or mean reversion, meaning the analyst community broadly fails to account for the sustainability of structural growth drivers.
The interesting thing about the market’s consistent under-estimation of Novo Nordisk’s earnings growth is that it was actually well known in the investment community throughout this period (2005-12) as an “earnings growth compounder”. In spite of this, the market still underestimated the ability of Novo Nordisk to sustain its earnings growth.
Why does this happen? Specifically, it is the failure of analysts to properly account for the reinvestment of cash into a compounding business franchise in three-year-plus models. And, the reason for this seems to lie in a flawed but widespread belief in ‘earnings fade’ and earnings mean reversion. While mean reversion is a valuable concept in the context of valuations, it is much less useful in the context of earnings growth. Consensus estimates regularly forecast a decline in returns of companies with high return on equity and an increase in returns for companies with low return on equity. They assume that competition will trim the excess returns of high achievers while low achievers will install remedies to fix their businesses. However, the expectation for mean reversion in earnings is in contrast to the sustained bifurcation evident in many industries, where the strong get stronger and the weak get weaker.8
The combination of the market’s short-termism and its expectation of earnings convergence creates opportunities for investors who can identify the companies that over the longer term can avoid mean reversion. It’s an investment formula for outdoing the market average that will help people achieve their long-term financial goals.
Important information
References to specific securities should not be taken as recommendations.
1 Dimson, Marsh and Staunton (2002) showed dividend paying stocks outperform. Fama and French (1992) showed stocks with high earnings/price ratios earn higher returns. 2 New York Stock Exchange. 3 This can be partly explained by availability bias (mentioned above) as there is a wealth of information available at any point to inform short-term forecasts. 4 IBES. 5 DataStream, as at 29.07.2013. 6 International Diabetes Federation, 2012 Update 7 Novo Nordisk Annual Report 2012 8 Research papers from Goldman Sachs, (The Die is Cast April 2011) and Credit Suisse (‘HOLT Wealth Creation Principles: Was Warren Buffet Right: Do Wonderful Companies Remain Wonderful?’, June 2013) support the point.