Five principles from the ‘lost decade’ of value investing
The decade to 30 June 2020 may be remembered as the ‘lost decade’ for investors who favour a value style of Australian equity investing. Value investing targets companies priced below what their net assets (book values) or other financial fundamentals suggest the companies are worth.
The investment thesis is that markets are efficient and will eventually push the stock price up to fair value (‘mean reversion’) prompting investors to buy the stock while cheap and sell once fair value is restored. Benjamin Graham established the first principles of value investing in 1928 and – of particular interest to large, fee-constrained institutional investors – academics Fama and French famously identified in 1992 that the value (‘high minus low’) style factor could be captured in an equity portfolio in a low-cost, rules-based way using quantitative techniques.
Value investing has become a painful experience
Targeting risk premia (outperformance) from style factor risks in this way has come to be known as ‘factor investing’. Since the release of the Fama-French factor investing framework, the three stand-out factor performers in the Australian equity market have been momentum, low volatility (low beta or ‘safety’) and value.
Leading into the last decade, value, with its intuitive thesis and empirical support, seemed to follow the script in Australia. Before fees and taxes, a hypothetical value portfolio of Australian equities (nominal returns) would have grown by 31.9% over the decade to 30 June 2010, easily outstripping a momentum portfolio (17.9% compound growth) or a diversified portfolio spread equally between these two style factors (25.2% compound growth).
But the past decade to 30 June 2020 has proven extremely painful for value investors, and not just in Australia. The performance rankings of these three strategies, in fact, reversed: it has been momentum’s decade (72.1% compound growth), with the diversified style portfolio growing 46.3% and value bringing up the (distant) rear at 23.9% compound growth. This ranking has not changed through the coronavirus crisis period (measured from March-June 2020).
Value is known to be a long-cycle play, but a decade (and counting) of waiting for value’s expected mean reversion to eventuate has tested the faith of even long-horizon investors. There is even speculation that Warren Buffett, an exponent of value investing, has abandoned the approach.
What should we make of all this, especially during the financial market stress brought on by the coronavirus? Historically, value stocks have not benefited from ‘flight to safety’ market environments, but they tend to rebound strongly from recessions. So, value’s long-awaited recovery can’t be ruled out and plenty of institutional money is flowing into value again … but plenty, of course, is not.
A reminder of five basic principles
There are, at least, some principles we can remind ourselves of as we reflect on the experience of value investing over the past two decades.
First, value investing can require a long time horizon for its thesis to play out. For large institutional super funds, many stakeholders (trustees, investment committees, product teams) need to commit to this journey and not panic or abandon value’s investment thesis during protracted periods of under-performance. A fund can pursue other well-supported equity factor risk premia with shorter payoff cycles (like momentum) if it doesn’t have the fortitude to persevere with value investing.
Second, a good factor investing strategy should not rely on getting the timing right. Like timing markets in general, factor timing is tricky at best. Over the two decades of our analysis, a value strategy that made a contrarian switch into momentum just before the ‘lost decade’ of value commenced would have experienced compound growth over the two-decade period of 127%. Sounds good, but a momentum strategy that switched to value at that time (lured by value’s stellar record during the 2000s) would have forgone 2/3 of this portfolio value at period end (46% compound growth), making factor timing a high-stakes game.
Third, as our hypothetical 50-50 portfolio’s performance shows, tried-and-true principles of diversification provide a safe course between these portfolio outcome extremes. If the investor’s conviction is in the idea of factor investing in general, then a well-constructed multi-factor ‘core’ portfolio which does not rely on timing may be the portfolio-of-least-regret.
Fourth, whether we have really just experienced the ‘lost decade’ of value depends on the investor’s objectives and the role the value portfolio is designed to play. For example, comparing our hypothetical portfolios over the decade, momentum delivered higher volatility; and, intuitively, value as a style is thought to lower drawdown risk (loss of capital) because stocks are already cheap to begin with. Those who use risk-adjusted return metrics, or custom risk definitions, may have a different view of value entirely.
Fifth, and finally, we counsel investors to evaluate their factor investing options with an after-tax investing lens. In 2017 research for funds, we estimated the long-term annual capital gains tax drag on a value factor portfolio to be 84 basis points (0.84%), versus 69 basis points (0.69%) for a momentum factor portfolio. The value of franking credits widens this gap.
The longer it takes for an investment thesis like value investing to play out, the more it makes sense for an investor to ‘control what can be controlled’ (implementation frictions like taxes and transaction costs) in the meantime. This allows the investor to take the small, steady wins until the big investment bets pay off.