This post is a sequel to one I wrote a few weeks ago about the ‘external perspective’. As I said then, I think it’s important to challenge yourself by looking at investing from different perspectives. I’ve found that this can be a useful way of gleaning insight. I want to follow up my previous post by looking at a more fundamental question for which there are also multiple perspectives: ‘how should you think about your investing strategy?’

A sensible active investing strategy needs some basis to support why it should beat the market. It needs to start with the idea that markets are not fully efficient and that share prices often do not accurately reflect the true value of businesses. The aim of the strategy should be to take advantage of these inefficiencies to deliver outperformance while minimising risk.

Investing strategies start to diverge in how they go about identifying and exploiting these inefficiencies. Broadly speaking, you can put them into two camps:

• A valuation perspective: identify mispriced shares directly by assessing the intrinsic value of the business against the price the market is offering. Make decisions based on conviction in these assessments.
• A behavioural perspective: identify mispriced shares by assessing how and when the market tends to misprice. Make decisions using rules designed to exploit these tendencies while controlling your own behaviour.

The valuation perspective

I see the valuation perspective as the most common, or default, perspective that most investors use for thinking about their investments. Certainly all my favourite quality investors, from Buffett to Terry Smith, think in this way. The idea is simple. The most direct way to find businesses that the market is undervaluing is to assess the valuation yourself. You don’t need to understand why something is undervalued to take advantage. You should treat the erratic short term behaviour of the stock market as essentially random and instead rely on the idea that share prices should gradually converge towards their ‘true’ values over the long term. If you are confident in your assessment that the business is fundamentally undervalued, as long as nothing unexpected happens you should eventually be proved right.

This way of thinking is captured well by Benjamin Graham’s analogy of Mr. Market, in which the stock market is personified as a neurotic business partner. Mr. Market is prone to bouts of extreme pessimism, when he offers to sell you his share of the business for a pittance, and extreme optimism, where he’s prepared to pay almost anything for your share. Mr. Market’s moods change randomly and you should react appropriately to them when they happen rather than attempt to predict them. Part of the idea behind this is that a true ‘investor’ should think of themselves as the owner of a business, and not feel compelled to deal with Mr. Market unless he offers what is objectively a good deal. You shouldn’t think of yourself as a speculator or player of the stock market game.

The directness and simplicity of the valuation perspective has warranted appeal. By asking the most direct version of what is fundamentally the only question that matters, how can you go wrong?

Very easily as it turns out. While the question is simple the answer is not. The value of a business depends heavily on its long term prospects, which in most cases are highly uncertain. I don’t think it is a good idea to account for valuation and growth prospects separately, as seems to be standard practice. The relationship between future growth prospects and valuation is direct – the true value of a business is directly determined by its future cash flows. It is also complex, as future growth both compounds and is uncertain. This means simple mental accounting can go awry. If your investment decisions are based primarily on your assessment of whether a business is undervalued you need to tread very carefully in balancing everything out correctly. If a business looks cheap but has very uncertain prospects then it’s generally actually expensive.

Because of this complexity, directly identifying whether businesses are undervalued tends to work best at the extremes. At one extreme, it’s sometimes possible to identify clear value in ‘deep value’ situations, where the valuation is primarily dependent on a binary outcome, such as whether or not the business can avoid imminent bankruptcy, rather than on more nuanced speculation about its long term future. At the other extreme, high quality businesses with more certain future growth can appear obviously undervalued just from taking a sufficiently long term perspective. In between these extremes, the appropriate valuations of ‘average’ businesses are a lot harder to judge because of the uncertainty inherent in their long term prospects. Because of this many investors who properly apply a valuation perspective would tend to head to these extremes, either focusing on ‘cigar-butt’ deep value or certainty in long term quality.

Another drawback of only looking at things through a valuation lens is that mispricings in the stock market can take a long time to resolve themselves. If they were likely to resolve themselves quickly then how likely would they be to exist in the first place? As John Maynard Keynes famously said: ‘the market can remain irrational a lot longer than you can remain solvent.’ Because of this a valuation-based approach to investing goes hand in hand with a long time horizon and a buy and hold strategy. A long time horizon is inherently weaker than a short one that can be applied again and again, unless it yields sufficiently more certain results. The need to take this long time horizon is another reason why certainty in long term prospects (i.e. quality) is so important when your approach is based purely around valuation.

Most importantly, you need to rely on the conviction that you are right in your assessment of valuation while the market is wrong. Success relies on you having superior information or analysis to others. Because of this, a valuation-based approach is often coupled with the idea that you should do very detailed research to understand your investments better than other investors. It can therefore make sense to concentrate your portfolio on a small number of investments that you know well. However, relying heavily on detailed research is difficult and can be a dangerous approach – I’ve written previously about how excessive research can expose investors to mistakes.

More generally, thinking solely in terms of valuation doesn’t provide a direct way of tackling the behavioural biases that we know most investors suffer from. Without this the danger is that you will suffer from them yourself. Committing to a buy and hold strategy can go a long way to protecting yourself but I wouldn’t say it was the best way.

The behavioural perspective

The behavioural perspective essentially takes the opposite approach. It acknowledges the difficulty inherent in valuing individual businesses and so treats it as a black box. Instead, it focuses on understanding the behaviour of the market. It looks for systematic rather than specific sources of mispricing. Instead of thinking of yourself as the owner of a business you think of yourself as a player of the stock market game. You ask: ‘how and why does Mr. Market get it wrong and how can I take advantage?’

This takes into account the fact that investing is strategic: the outcomes of your investments are not objectively determined but rather depend on the behaviour of other investors. Beating the market is not necessarily about being objectively ‘correct’. It’s more directly about predicting what other investors are going to do or exploiting the ways in which they get things wrong.

One implementation of a behavioural perspective is through factor investing. Factors are measurable attributes of stocks that have been shown empirically to persistently generate excess returns on average, implying they tend to be undervalued by other investors. Factors have their root in numerous empirical academic studies disproving or refining the efficient market hypothesis (EMH) – the idea that with a sufficient number of intelligent well-informed investors stock prices should reflect all information available. Factors arise because the assumptions underlying EMH do not closely reflect reality. The principal reason is the assumption that a sufficient number of investors are ‘intelligent’ (or ‘rational’), i.e. able to make the complex decisions that would arbitrage away excess returns, is ill-founded. In reality investors are biased to systematically undervalue shares with certain characteristics, such as those that have already recently risen a lot in price (momentum) or those with low valuation multiples (value).

While the conventional way to exploit factors is through complex quantitative models directly based on the empirical evidence, this isn’t really necessary. I would say that any strategy that uses rules to take advantage of systemic stock market biases is behavioural. For example, technical analysis aims to identify situations where there is a high expected payoff based on recent patterns in market behaviour (share prices and volumes). While a lot of technical analysis can seem whimsical or arbitrary when applied in practice, fundamentally the idea is similar to factor investing i.e. identify market inefficiencies from looking empirically at how investors have tended to behave in the past. A lot of technical analysis is basically more sophisticated interpretation and implementation of the momentum factor.

Approaches based on a behavioural perspective can address many of the disadvantages inherent in valuation-based approaches. They are simpler to implement, don’t require detailed research, can use a shorter time horizon and don’t require the same conviction in your ability to predict a business’s future prospects. They can also be applied systematically through rules that not only limit behavioural mistakes but specifically target those of other investors.

For all these advantages, following a behavioural approach certainly isn’t all sunshine and roses. The flipside of requiring less conviction in predicting the future of individual shares is that you need more conviction in the overall strategy. You need confidence that investors will continue to behave in the ways you are expecting – get this wrong and your whole strategy may fail.

The extensive statistical evidence supporting the efficacy of various factors can justify some of this conviction but it’s not clear-cut. This type of statistical evidence carries huge risk of data-mining – identifying patterns that are spurious artefacts of the specific data being interrogated rather than evidence of actual relationships that hold more widely. More fundamentally, the way in which investors behave can change over time. Indeed you would expect factors to stop working over time as they become more widely known and investors try to take advantage of them. In many ways it’s surprising that factors like momentum and value have worked as consistently as they have. While momentum seems to continue to work well, the value factor hasn’t worked for more than a decade now – it’s not clear whether this is temporary or not. Continuing to try to take advantage of it does require a leap of faith.

The limitations of empirical evidence mean it’s also important to have confidence in the underlying reasons why different factors should work. Factors should arise from systematic mistakes so there should be a credible intuition about why the mistakes implied are being made. The momentum factor seems to have a lot of strong reasons underlying it, as I’ve discussed before. The value factor I’m less sure about.

One of the drawbacks of factor investing as commonly applied is that little weight is placed on important qualitative information compared to information that can be measured. Qualitative information can be key in assessing some of the characteristics that are most predictive of a business’s success in the long term, such as whether it has a competitive advantage. This is quite a blind spot, as in my view probably the most significant source of inefficiency in the stock market arises because investors tend to be short-sighted, paying relatively little attention to certain characteristics that predict long term success.

Combining the two

Neither perspective is perfect. Both face major challenges. They ask complementary rather than mutually exclusive questions: ‘is the business trading below its intrinsic value based on my own assessment?’ and ‘does it have the characteristics of a share likely to be undervalued?’ It’s hard to see why you shouldn’t try to answer both. Combining the best of both perspectives is the aim of my strategy.

In fitting them together, I think the behavioural perspective is best suited for providing the rules and discipline you need to define a consistent strategy. It is an opportunity to put the odds in your favour before you get into the challenge of assessing individual shares in detail. There is still plenty of room for more detailed assessment of quality and intrinsic value within a behavioural framework. There is nothing to stop you using qualitative information in a systematic way or betting harder where you have more conviction that something is seriously undervalued. You’ll be glad to have rules you believe in when times are tough…