The header is a playful provocation: in truth I am probably closer to a value investor than any other type. But I place little credence on the historical superiority of value investing (which has faltered in recent years anyway), and I have little sense of affiliation or identity as a value investor. It’s not quite that I don’t want to be a member of any club which will have me; rather it’s that that I’m nervous about a club which is now so crowded and where all the members think alike.
The problem with value investing is that it is now a popular and widely followed investment philosophy. Analytical techniques which historically worked well when used by a small minority of investors will work less well as more investors use them. This is inherent to the nature of markets: a matter of arithmetic, not opinion.
The misleading header is useful if read as a prompt: are there any ways in which my thinking differs from value investing orthodoxy? Here are a few.
Scepticism about intrinsic value Many value investors place great stress on the concept of intrinsic value – the discounted future cashflows of a company, as distinct from its book value, liquidation value or market value.
As I explained in a previous post, I seldom write down an estimate of intrinsic value. I think mainly in terms of heuristics. Some heuristics are directly related to value, such as: P/E ratios, dividend yields, and price/sales ratios. Other heuristics have only indirect connections with value, such as: the presence of counterparties with non-investment motivations, director purchases, and patterns of affinity (ie previous associations of the management and advisors to a company).
Scepticism about deep research I generally do not want to read 20 years of past accounts, visit factories, or conduct interviews with customers, competitors and suppliers. I’m also lukewarm about meetings with management. I don’t deny that such deep research can further your understanding of a company, but I think it is subject to strongly diminishing returns. After a certain point, deeper research doesn’t make the risk sufficiently smaller to be worth the time. Rather than looking deeper and deeper into one investment (and often, convincing yourself to buy more and more of it), it is better to spend the time looking more broadly for new investments.
Small bets on large discrepancies, superficially understood When I find a price which looks very wrong for seemingly robust reasons, I sometimes prefer to make a small bet without fully researching the company. Although I might do deeper research later, the price often corrects before I do. This is fine – I just sell and go on to the next one. By keeping positions small I keep them easy to sell, that is I preserve options to change my mind as prices and my expectations change.
Summarising all three points so far: in a noisy information environment with more choices than I can process, I prefer to spend most time looking for new and obvious anomalies, rather than refining my assessment of anomalies I’ve already found. I aim to maximise the quality of my whole portfolio of insights, not my depth of insight on any particular companies.
Willingness to look foolish I actively look for investment ideas which are likely to seem silly, embarrassing or trivial to those who think of themselves as “serious” investors (eg value investors!). Apparent silliness doesn’t necessarily mean that the investment is a good one, but it does mean that fewer “serious” investors will be looking. One example of a silly investment is ASOS under 5p in 2003 – “a start-up website selling teenage fashion?!” (I actually bought ASOS in 2003, and I was acutely conscious of both its potential and its silliness at the time. I sold far too early. The current price is around £50.)
I also look for risks with the following characteristic: if the bad outcome materialises, it will seem obvious with hindsight, making those who took the risk look foolish. Again this doesn’t necessarily mean the risk is a good one, but it does mean less competition from professional investors with career concerns.
The Keynesian short-cut By “Keynesian short-cut” I mean the observation in Chapter 12 of Keynes’ General Theory that financial markets operate under a convention that “the existing state of affairs will continue indefinitely, except insofar as we have specific reasons to expect a change.”
In other words, we expect prices tomorrow and next month and next year to be unchanged (in real terms) from prices today, except where there is relevant new information. So if I expect some future change when other investors don’t, I can buy at current prices and wait for the change, without thinking much about intrinsic valuation in the current state.
If the current price is already far above intrinsic value – in other words the price is already a bubble – then this approach can lead to trouble. Implicitly, I do look out for and avoid situations where this might be the case. But my main focus is on looking for early indications of a knowable future change, rather than on intrinsic valuation in the current state.
This mode of thinking is particularly helpful for highly cyclical businesses such as shipping and house-building. High cyclicality makes discounted cashflow or accounting ratio valuations very difficult, so it may be better not to think about them very much. Better to look for early indications of change, such as freight rates or housing starts increasing.
Avoiding the three R’s Aswath Damodaran critiques value investors with three R’s: rigid, righteous, and ritualistic. Rigid, because they have firm views and rules about analytical techniques or preferred investment sectors (albeit the rules are different for different value investors). Righteous, because they believe theirs is the only true creed, and that success with other techniques is in some sense inferior or illegitimate. Ritualistic, because they see value investment education as a sequence of sacred rites: read The Intelligent Investor, read Security Analysis, read all of Buffett’s shareholder letters, attend a Berkshire Hathaway annual meeting, etc. I have followed many of these rites, and I don’t deny their usefulness; but I try not to be too reverent (a fourth “R”!) about them.
Those are some of the ways in which I believe my thinking differs from value investing orthodoxy. A difficulty with this exercise is that it’s hard to distinguish useful differences from capricious contrarianism or mere idleness. Some of the points above could be characterised as reflecting my lack of diligence. Although I worry about this, I don’t think diligence is an end in itself. My aim in investing is to make money, not to burnish a personal narrative or sense of identity as any particular 'type' of investor.
Perhaps a more accurate version of the header is to say that I don’t care whether or not I am a value investor.
Behavioural finance is a fashionable genre of academic research, and a productive strategy for writing academic papers. But it was not mentioned as a resource by any of the interviewees in Free Capital, and I have never found it much help in my own investing. There are several reasons for this.
Many explanations, few predictions The comprehensive menu of alleged behavioural biases offers an ex-post explanation for almost any decision which hindsight renders sub-optimal. Investors who borrowed money to invest in shares in September 1987 suffer from “over-confidence”; those who failed to do so in March 2009 suffer from “myopic loss aversion.” Investors who respond quickly to new information overweight “availability”; those who respond slowly are “anchored” by prior beliefs. And so on for every other ex-post mistake.
This descriptive charm and versatility is palatable both to academic story-tellers and to casual readers. But to be scientifically or instrumentally useful, a paradigm needs to make some specific predictions. Behavioural finance seems more like Freudian psychology: it can be contorted to explain anything ex-post, and therefore predicts nothing ex-ante.
Poorly defined allegations of “irrationality” A common trope in behavioural finance articles is to document some observed behaviour of investors and assign to this the pejorative label “irrational.” This is an over-used and often unwise epithet, because “irrationality” is usually defined against a narrow normative standard. On closer examination, all that can usually be said is that large groups of people who follow the observed action indiscriminately over long periods of time will lose money compared with large groups of people who don’t. It does not follow that all (or even most) instances of the action are individually irrational.
Data mining and publication bias Any behaviour which can be labelled “irrational” (often dubiously – see above) against some normative standard is newsworthy and publishable. Null results where people behave “rationally” are less exciting, and more likely to be filed in a drawer. The suspicion of data mining for manifestations of “irrational” behaviour is increased by the disparate and often contradictory nature of the claimed biases.
Politeness misconstrued as probabilistic judgement Many behavioural finance articles assert that investors make invalid probabilistic judgments. In the archetypal example, it is said to be a “conjunction fallacy” that Linda, a 31 year-old with biographical data suggesting liberal social views, is “more likely” to be (a) a bank teller and a feminist rather than (b) just a bank teller.
But the answer (a) which most people give is not a probabilistic judgement, it is a socially appropriate response to cues. This type of response is learnt both in educational settings (did you ever see an exam question where you were not expected to "use all the information provided"?), and also in everyday life (the social costs of ignoring cues are usually larger than the social costs of over-responding to them).
The so-called “conjunction fallacy” is generated only because the question is a sort of word-trick: a probability test masquerading as a cue-response test, or a signal detection test. If the normative standard is signal detection or polite response, the so-called 'wrong' answer is correct. And with a slight change in the question wording to focus attention on numerical frequencies rather than cue-responses, most people give the “correct” probabilistic answer (see Gigerenzer).
Detecting dishonesty is more useful than simulating truth Intuitions about truth and falsehood are often more usefully directed towards detecting liars, rather than simulating causal relations. For example, apropos the “Linda” example above: when considering reports, it may be a good heuristic to trust the insinuations of people who provide many details - people like Linda's acquaintance - because people who provide many details tend to be truthful witnesses. (This heuristic may invert when considering predictions, where people who provide many details tend to be charlatans.)
More generally, behavioural finance focuses mainly on failures of cognition and calculation, but largely neglects social phenomena such as trust and deception. A visiting Martian might expect the title "behavioural finance" to encompass the study of frauds, stock promotions, and pump & dump operations - all endemic behavioural phenomena in financial markets - but in fact the subject never goes there.
Few normative prescriptions Behavioural finance invites us to gawp at all the foolish mistakes other investors make. This financial freak show is superficially entertaining, but it doesn’t necessarily make us better investors: it doesn’t tell us what to do. To catalogue all your biases and then do nothing much about them is not humility, it is boasting of your modesty.
Some emotions are good emotions In the absence of explicit normative prescriptions, the implicit prescription of behavioural finance appears to be that investors should try to suppress all emotion. This is probably neurologically unfeasible, and in any case undesirable. A better aspiration is to aim to have the right emotions : to hope that your delusions are benign, and your compulsions have utility. A good recent book on this concept is The Emotionally Intelligent Investor.
Update (16 March 2014): How knowing about biases can hurt you.
(And now a short break from our normal programming....)
In a comment linked from the Wikileaks Twitter feed on the recent PRISM disclosures, Bernard Keane characterises the surveillance state with the term information asymmetry:
“Information asymmetry is how your government wants to know everything possible about you — where you are, whom you called, what you searched Google for, what was in that gmail you sent, etc etc — while trying to prevent you from knowing about its activities as much as possible by using national security (and other excuses, like “commercial in confidence”) to hide information.”
Whilst the term “information asymmetry” appears highly apt on dictionary definitions, the phenomenon being described is interestingly different from – and wider than – the typical meaning of the term “information asymmetry” in economics.
In economics “information asymmetry” typically signifies that a seller has better information than a buyer about the quality of the goods being exchanged.
In the canonical example, the seller of a used car has better information than a potential buyer about the quality of the car (ie whether it is a “lemon”). Similarly an employee – that is a seller of labour – has better information about their productivity than an employer.
In both these examples – and typically in economics – the information asymmetry is at one level and about one property only, the quality of the goods being exchanged. This property is exogenous, that is it is not changed by the interaction between the buyer and seller.
State v. subjects: two levels of information asymmetry
In the adversarial context of a surveillance state and a monitored population (call them “subjects”), there are at least two levels of information asymmetry.
(1) Knowledge about the state’s thoughts: The first level of asymmetry is knowledge about the state’s thoughts. This asymmetry is mutual: the state knows more about its own thoughts, and the subjects know more about their own thoughts.
(2) Knowledge about the state’s knowledge of your thoughts: The second level of asymmetry is knowledge about the state’s knowledge your own thoughts. In other words, knowledge of the state’s surveillance capabilities. Again the asymmetry is mutual: the state does not know how much the subjects know about its thoughts, and the subjects do not know how much the state knows about their thoughts.
We can conceive of asymmetry at higher levels: what the state knows you know the state knows about your thoughts, and so on. But at higher levels the significance of asymmetry probably decays (unless the lower levels are exactly symmetric). Two levels are enough to make the following observations.
At both level 1 and level 2 the uncertainty is mutual, that is there is uncertainty in both directions.
At both level 1 and level 2, the uncertainty is endogenous, that is the property to which uncertainty pertains may be changed by the interactions between the parties.
These features – mutuality and endogeneity – make the uncertainty here more subtle than the classical economics set-up, where “asymmetric information” pertains only to the exogenous quality of goods to be exchanged.
Freedom under asymmetric information
The freedom of a society is influenced by the asymmetry of knowledge. To a first approximation, asymmetry in the state’s favour tends towards totalitarianism, and asymmetry in the individual subject’s favour tends towards anarchy (but this is qualified below).
An increase in your level 1 knowledge – your knowledge of the state’s thoughts – always increases your own freedom. However the effect of an increase in your level 2 knowledge – your knowledge of the state’s surveillance capabilities – may be less clear, as I explain in a conjecture below.
I conjecture that freedom of subjects as a function of their knowledge of the state’s surveillance (that is level 2 knowledge) follows a U-shaped curve:
If the state’s knowledge of your own thoughts is completely secret, it will not inhibit your actions. Under blissful ignorance, freedom is high (the left hand side of the curve). However this is not a stable scenario: as soon as the state uses the knowledge against you, its actions betray the knowledge and so the power of surveillance is unlikely to remain completely secret.
On the other hand if you know everything about the limits of your state’s knowledge of your own thoughts – know all the details of its surveillance capabilities – then you adjust your plans and take steps to avoid the surveillance. This gives at least a modest amount of freedom, albeit perhaps at great cost in terms of surveillance-avoiding work-arounds (the right hand side of the curve).
It is in the middle of the curve – where you are acutely aware that the state has surveillance capabilities, but have incomplete knowledge of their workings or extent – that your freedom of thought and action will be most inhibited.
Prescription: create ambiguity about your powers of surveillance (or sousveillance)
Applying these thoughts to the surveillance state and its subjects yields the following prescription. To maximise its freedom of thought and action, each party will maximise uncertainty about its powers of surveillance (or for the subjects, sousveillance).
A surveillance state will not fully detail its surveillance capabilities, because to do so would delineate the subjects’ problem and facilitate their search for surveillance-avoiding tactics. A surveillance state will instead publicise only vague knowledge of surveillance capabilities coupled with a lack of details, thus maximising the subjects’ inhibition and self-censorship, and hence the state’s power.
A subjects’ leaking platform such as Wikileaks will not fully detail its capabilities in obtaining, anonymising and promulgating leaks, because to do so would delineate the surveillance state’s problem and facilitate the search for leak-plugging tactics. The leaking platform will instead publicise only vague knowledge of its capabilities coupled with a lack of details, thus maximising the surveillance state’s apprehension about what might be disclosed.
The value of a leaking platform is not to reveal all secrets of the surveillance state (an impossible and probably undesirable aim), but rather to increase uncertainty about which secrets may be revealed. Or in the vernacular: if you keep the bastards guessing, you can keep them honest.
ADVFN, Interactive Investor, Moneyam, Motley Fool, t1ps.com. Financial website entrepreneurs don’t seem to be very good at choosing names. This post does some of the thinking for them. It is not really an investment post, except insofar as a start-up with a good name may have a slightly better chance of success than one with a poor name.
A good company or product name should be SUMPIER:
and (most distinctively)
The first four criteria SUMP make the name easy.
Short A short name facilitates word-of-mouth marketing, forestalls unwanted abbreviations (see next point) and generally increases salience. Three or four syllables are good; two are probably better.
Unique A name should to be unique in two senses: clearly different from all relevant rivals, and not susceptible to unwanted variants, abbreviations or acronyms. Checks on the first should include a trademark search. Checks on the second are less easily systematised. One can quickly rule out rule out Celtic Region Aluminium Products or New United Trading Services, but some variants are not so easily anticipated (see the anecdote about Exxon below).
Memorable This helps salience and word-of-mouth marketing. Memorability is a know-it-when-I-see-it property, but one useful device is alliteration. Toys for Tots, not Toys for Young Children; Leaping Lizards, not Jumping Reptiles.
Phonetic A name is phonetic if spelling and pronunciation are mutually implicative. A person who first sees the name in print should be immediately able to say it out loud; a person who first hears the name in speech should be immediately able to write it down. Motley Fool is phonetic; ADVFN and t1ps.com are not. Odd spellings, capitalisation and even punctuation are quite common nowadays (del.icio.us, flickr, tumblr…), but a phonetic name is always more effective.
The next two criteria IE make the name pleasant.
Inoffensive A name should have no unhelpful associations for product or the target customers, either in English or any other relevant language.
Note that this “technical” definition of “inoffensive” does not necessarily exclude risqué names – it depends on the target customers. There are many risqué names, but also many indications of difficulties with them; I suspect that for most products this game of “niche appeal through offence” isn’t worth the candle.
Under this “technical” definition, most of the financial website names are OK, but Motley Fool is weirdly offensive to its own users. Yes, I know the etymology: the founders were English majors with a penchant for ironic Shakespearian allusions. But Americans don't do irony. I think the name is weak.
Inoffensiveness can be hard to verify. When Esso and related companies became Exxon in 1973, the new name had been chosen from computer-generated sequences of letters and exhaustively verified as inoffensive in dozens of languages. Within days Exxon had acquired an alternative moniker throughout the oil industry: it became the double-cross company.
Euphonious A name should sound pleasant – or at least, not unpleasant. It should be easy to answer the telephone. Motley Fool is euphonious; ADVFN is not.
The final criterion of resonance makes the name effective. This criterion is more subtle than all those above, and distinguishes great names from merely good ones.
Resonant A name should resonate with – that is describe, evoke or emphasise – either or both of:
(a) the benefits of the generic category (product resonance)
(b) the comparative advantages of your specific offering (brand resonance).
Resonance can come either from the name’s literal meaning (denotation) or associations (connotation).
Examples of names with product resonance: eBay, Match, Rightmove, Twitter, uSwitch, Youtube.
Eamples of names with brand resonance: Betfair, Costco, Dulux, Easyjet, Topshop, Valu-mart.
As these examples show, product resonance tends to be more important when a company is creating an entirely new category, and brand resonance when a company is entering an existing category. None of the financial website names has either type of resonance.
The ideal is to have both product and brand resonance in one short name, but this is almost unachievable (I find it hard to think of examples, although perhaps Betfair is close). More achievably, a name with product resonance can be supported by a strapline or slogan with brand resonance, and vice versa.
EVALUATING FINANCIAL WEBSITE NAMES
For financial websites, the following table evaluates some extant names against these criteria, and also some hypothetical names (tick = positive, X = negative, ~ = neutral or hard to say).
Morphological versatility For some products or (especially) services, it helps for the name to satisfy most of the earlier criteria (unambiguous, phonetic, euphonious etc) when used as a verb. The dominant search engine got this right (“I Googled him”); the ones which fell by the wayside didn’t (I AltaVista’d him”? “I AskJeeves’d him?”). Other morphological variations can sometimes be anticipated eg to describe a user of the product or service (“eBayer” for users of eBay works; “tumblrer” for users of tumblr doesn’t.)
Category name versus brand name The category name is the generic description; the brand name (protected by trade-marking) is your specific product. Generally you want these to be clearly different; and so if your product is entirely novel, you may need to invent a distinct category name.
Vacuum cleaner – Hoover
Personal organiser – Filofax
Betting exchange – Betfair
If you don’t create a distinct category name, customers may create one you don’t like; and your own brand name is at risk of being corrupted by customers to encompass competitor products. (Obvious exception: if you are a latecomer copycat, you may prefer the dominant brand name to be corrupted this way.)
Narrow versus broad names A narrow name alluding to the company’s initial product or geographical location may give good initial resonance, but also act as an obstacle to later expansion. Think carefully before being too specific.
Domain names The discussion above assumed a unique and hence new name, for which a corresponding domain is likely to be available; but obviously this needs to be checked. Any similar domains (for example slight misspellings of the name) also need to be checked (and acquired, before somebody else does).
Book names Similar criteria can be applied to choosing book titles. I think the title Free Capital satisfies the criteria above, except for one drawback: it isn’t resonant – or even comprehensible – until after you’ve read the book. Hence the sub-title How 12 private investors made millions in the stock market, which I hope gives some prospective resonance.
Finally, I acknowledge that some companies do succeed despite poor names against the above critieria. Yahoo!, GoDaddy, Digg…these all seem to me to have no resonance, and foolishly indulgent spelling, capitalisation or punctuation. But a poor name makes life unnecessarily difficult. Why not try to get it right?
“The scarcest resource for successful investors is not money but attention: how to manage the trade-off between time and rationality to best effect. There is not time in life to find out everything about every potential investment. Investment skill consists not in knowing everything, but in judicious neglect: making wise choices about what to overlook.”
...that's the first paragraph of the first profile in Free Capital. There is nothing casual about this placement: I think allocation of attention is the most fundamental choice an investor has to make. Some relevant dimensions of attention as follows.
SOME DIMENSIONS OF ATTENTION
Acute observation versus judicious neglect In speaking of judicious neglect, the above extract really gives only half of the recipe. You also need acute observation. The critical inputs to an investment decision are often relatively obscure to an unskilled observer: something in the notes to the accounts, the background of the CEO, recent stakebuilding by an activist, etc. Investment skill consists of being able to spot these critical points, and neglect much else: a mix of acute observation and judicious neglect. Pay sufficient attention (due diligence). But not too much, because that would waste attention which could better used elsewhere (undue diligence).
Depth versus breadth Is it better to look very closely at just a few possible investments, or more briefly at a larger number? My preference is generally the latter. This doesn’t mean researching all possible London-listed investments: at any time only a minority with some combination of good financial metrics or other idiosyncratic attractions are worth investigating at all. When you do focus in on a particular company, ir's usually more useful to look for more sources of information (another form of breadth), rather than doing more detailed calculations. If you need a calculator, it’s too close.
Local versus global The phrase “London-listed investments” above highlights another dimension for attention. I generally stick to UK-listed companies, which gives the huge advantage of familiarity: I know the parameters of accounting and corporate governance and market dealing. But I often wonder if I would do better by scouring the world continuously for the cheapest markets – Bolivia one year, Botswana the next, Bangladesh the next, etc. (All these countries appear to have stock exchanges. I know nothing about any of them.)
Defence versus offence This is similar to the depth versus breadth trade-off, but worth highlighting separately. Defensive attention means keeping up with news about what you already own. Offensive attention means searching for new ideas. As the number of investments you hold grows, defensive attention cand easily swell to occupy most of your time. It takes conscious effort to allocate sufficient time to looking for new ideas.
Focus versus serendipity Focus is a self-directed and structured agenda: monitoring news on what you already own, daily checking of new market lows, quantitative screens for new ideas, daily or weekly reading of particular sources, etc. Serendipity is pursuing an idea suggested by a friend, or an article on another investor’s holdings, or an RNS headline which catches my eye. I think focus produces better choices most of the time. But the very best investments are necessarily found through serendipity, because the idiosyncratic features which made them great investments don’t shown up on any quantitative screens.
First order versus second order thinking Spend a small but non-negligible fraction of your time thinking about how to think: how to allocate attention (eg writing this blog post!), how to make decisions, how to embrace truths which I dislike, and so on. Almost every chapter in Free Capital says something about the investor’s record-keeping and filing methods – not because they spontaneously talked about this, but because it was of keen interest to me. (My real interest was how they think, but direct enquiry along those lines invited abstract or flippant responses; asking about filing systems kept the discussion concrete and serious.)
BETTER ALLOCATION OF ATTENTION
How can you improve your allocation of attention? The first step is simply to be consciously aware that attention is your scarcest resource, and that it’s worth thinking about trade-offs such as those above. Some further suggestions are as follows.
Develop domain expertise How do you know which elements of a scenario require acute observation, and which can be judiciously neglected? I think this ability – “good judgment” – flows mainly from domain expertise.
For example, I said above that notes to the accounts are often important. This doesn’t mean plod through (or even skim) all the notes. I skim only the following points (parenthesis gives where they’re usually found):
- the totals in the remuneration report (usually separate from the notes, in a dedicated section before the income statement and balance sheet);
- then turn to related parties (towards the end);
- then pensions, if there’s a defined benefit scheme (3/4 way through);
- then contingent liabilities (towards the end)
- and then anything particularly suggested by the company’s line of business or its financial situation. For example if the company has significant debt I would be very interested in the debt notes – anything on the interest margin, maturity and covenants.
Different investors operate in different observational domains. Some (like me) pay attention to the notes to the accounts; others pay attention to what management says in person; others pay attention to price charts. The important thing is to notice something which is relevant, and preferably not noticed by most investors.
Seek what is not offered This heuristic is independent of domain knowledge. It helps to have a predilection to notice what is not said (the dog which does not bark), to seek opinions which are not publicised; and to seek disconfirming evidence. Things which nobody is discussing often have great value in investment. Companies with no analyst coverage, or little coverage relative to their size, are often good investments.
Control your own time Investment is an unusually open-ended activity. There is almost nothing you have to do, and no limits on what you could do; this open-endedness is what makes allocation of attention so important. It helps to keep control of your own time, and to be self-conscious about how you allocate it.
I prefer not to have a schedule of meetings, or even blocks of time allocated to particular tasks. I just work on whatever seems highest priority at every moment, balancing the trade-off between urgency and importance many times every day. By not allocating time in advance, I’m relatively free to switch attention, say to a share I was recently buying where the price falls (so I might want to buy more)
Focus on what is knowable Before pursuing a particular line of enquiry, ask yourself whether it is likely to lead to reliable and actionable conclusions. Generally I find that it is not a good use of scarce attention to think about macroeconomics, because such thinking seldom leads to conclusions which are reliable and actionable. Analysis of big banks, insurers and other financially complex businesses falls into the same “unknowable” category, so I generally ignore them.
My last post drew a distinction between two classes of shares: potential “moonshots” (those with fat-tailed and/or right-skew returns), and “mundanes” (those with symmetrical returns). I argued that when selecting potential moonshots for a portfolio, one should be more tolerant of false positives than when selecting mundanes.
However, I can also see a good case that it may not be worth spending time looking actively for moonshots. This is for the following reasons.
The incidence (frequency) of true moonshots in the entire investment universe may be too low. It may be high enough in an exceptional period like 1998-2000 (my formative investment experience), when there were many technology moonshots. But in more normal times, the incidence may be too low.
The losses on false positives are often high. Potential moonshots which fail often fail disastrously, with loss of most or even all of your investment. A smaller investor might be able to use a stop-loss, but this is not practical for a larger liquidity-constrained investor (and for a fundamental investor, I’m doubtful it is ever a good idea anyway – see p60 of Free Capital).
The large gains on the few true positives are hard to realise. It is hard to hang on to quoted company moonshots. Even when the moonshot takes off, you may not recognise the scale of the company’s potential; and the daily temptation to prudently realise part of your gains is hard to resist.
Incidence and classification accuracy are unknown parameters. You have no way of knowing the true parameters for the population incidence of moonshots, or your accuracy in classifying them correctly. Therefore when you have a long stream of costly failures, you cannot tell whether this is because of (a) bad luck (in which case you can reasonably expect a success soon), or (b) the true parameters are too low (in which case you may go bust before you have a success).
Moonshots give big deviations from index returns. If the deviations were modestly positive in most periods, this would not be a problem. But the more likely scenario is that a moonshots portfolio will produce mediocre returns in most periods, and (we hope) a few big hits to compensate in other periods. This irregular pattern of returns gives little reassurance of your ability, and so is psychologically difficult for most investors.
Moonshots are hard to recognise ex-ante. I held potential moonshots QXL and ASOS in reasonable size in 2004, but sold them far too early. Even with hindsight, I don’t see that either of these had features which made them easy to recognise as moonshots in 2004. For ASOS, one comparable was the privately-owned Figleaves, which was founded as long ago as 1998, and sold for slightly over £10m to N Brown in 2010. Considering the information which was available in 2004 with hindsight, I still see no way of telling in 2004 that ASOS was going to grow into a £1.5bn business, and Figleaves only into a £10m business.
For all these reasons, I don’t screen actively for potential moonshots. I just remember that anything can happen, including good things. Or in the words of poet Alice Walker: "Expect nothing. Live frugally on surprise.”
The main point of this note is to suggest that an investor should have a higher tolerance for false positive classifications when selecting shares with right-skew and/or fat-tailed returns (potential “moonshots”).
Statistical discussions of hypothesis testing commonly refer to "false positive" (Type I) and "false negative" (Type II) errors. The term “error” is value-neutral in statistical testing, but may be unhelpful in a portfolio selection context, because it might be misread as insinuating some analytical mistake on the part of the investor. To avoid this blameworthy connotation, I will instead use the (slightly) more neutral term “disappointment.”
In portfolio selection I can then seek to avoid ex-ante errors of analysis, whilst recognising that there is an optimal rate of ex-post disappointments. (I can't think of a crisp and widely recognised word pair for the distinction I am stressing here: on the one hand blameworthy ex-ante errors, and on the other hand blameless ex-post disappointments. This seems an interesting linguistic lacuna.)
Portfolio selection as a classification problem
Portfolio selection can be viewed as a classification problem, with “positives” being shares which are added to (or retained in) your portolio, and “negatives” being shares which are rejected (or sold). Normally in classification problems you seek to minimise the error rate, defined as a weighted sum of false positives and false negatives. The two weights – one for false positives, and one for false negatives – are set according to the cost (the payoff) of each type of disappointment.
To give an example from medicine: if a disease is potentially fatal but has a reliable and safe treatment, then false negatives are costly, and false positives are benign. Hence we should tolerate a higher rate of false positives (eg pap smear testing for cervical cancer in young women). On the other hand, if a condition is benign and treatment tends to be worse than the disease, converse payoffs would apply and we should tolerate a higher rate of false negatives (eg PSA antigen testing for prostate cancer in elderly men).
In portfolio selection, false positives and false negatives cannot sensibly be identified with just two payoffs. Instead each type of disappointment generates a return distribution, reflecting the underlying share returns.
The underlying shares can be thought of as two classes, with different return distributions – potential moonshots, and mundanes. The graphs below show the return distribution for an individual stock drawn at random from each class.
Potential moonshots are shares with the potential for very high growth: shares which might “go to the moon”. Say Apple in 1997, or ASOS in 2003, or QXL in 2005 (of course examples are easy to identify ex-post!). Potential moonshots are few in number. Most potential moonshots never take off (call these “duds”). Actual moonshots are very rare (call these “hits”).
Mundanes are (unleveraged versions of) housebuilders, manufacturers, engineers – reliable businesses, but not plausible moonshots. Mundanes are plentiful, and easy to recognise. The nature of their operations makes long-term high growth unlikely. Compared to potential moonshots, mundanes produce outcomes which are less extreme, and more evenly distributed between “hits” and “duds”.
Generally, any adjustment to our tolerance of disappointments involves a trade-off between false positives and false negatives: it is not possible to minimise simultaneously both type of disappointment. The optimal trade-off depends on the (dis-)utility of each type of disappointment.
Is portfolio selection more like pap smear testing (false negatives are costly), or PSA antigen testing (false positives are costly)? The answer depends on the type of share: false negatives have much higher (opportunity) costs for potential moonshots than mundanes. For potential moonshots, a false negative means we miss one of the very few Apple-like shares which could make a big difference to our portfolio return. For mundanes, false negatives are not much of a problem, because a single missed mundane won’t make much difference to our portfolio return, and there are always plenty more largely interchangeable mundanes we could include.
This implies we should apply a heavier penalty to false negatives (and therefore necessarily also accept more false positives) when classifying potential moonshots than when classifying mundanes.
The differing optimal strategies for portfolio selection from potential moonshots and mundanes can be be illustrated with toy examples, as follows.
Toy example: portfolio selection from potential moonshots
Suppose that the entire class of potential moonshots comprises 100 shares, with these payoffs: 90 out of 100 go bust, and the other ten can be sold for 3x their cost.
Assume we select 10 shares for our portfolio, and make equal investments in each of them (these assumptions are simplifying, but not necessary).
If we select at random, the expected portfolio return is a loss of 70% (0.1 x3 x1 + 0.1 x0 x9).
But if we can manage to select four true moonshots, the expected portfolio return becomes a positive 20% (0.1 x 3 x 4 + 0.1 x 0 x 6). With five true moonshots, it’s 50%. With 6,7,8,9, 10, it’s then 80%, 110%, 140%, 170%, 200%.
So in this (extreme) example, we have can have a false positive rate as high as 60% when selecting mooonshots, and yet still generate a positive portfolio return.
Isn’t it better to tighten our criteria, and so reduce the false positive rate below 60%? For example, if our criteria for potential moonshots include forecast sales growth of 30%pa, we could tighten this to 40%pa. Yes, that would probably reduce false positives – but it would also probably increase false negatives – that is, we exclude more true moonshots. Because moonshots are so rare, the combination of reducing false positives and increasing false negatives may produce a portfolio with a lower fraction of moonshots, and hence a lower expected return.
(Technical note The argument as stated here is implicitly in cross-sectional form, adding contemporaneous raw returns in one period to get portfolio return for that period. But it also applies in longitudinal form: for compound returns, you just add the log returns over time.)
Toy example: portfolio selection from mundanes
Suppose the entire class of mundanes comprises 500 shares, with these payoffs: 250 give a 20% loss, and 250 give a 30% gain. (Note that realistically, there are 5 times as many mundanes as potential moonshots.)
As before, we select 10 shares for our portfolio, and make equal investments in each of them (these assumptions are simplifying, but not necessary).
If we select at random, the expected portfolio return is 5%.
If we select mundanes with a 60% false positive rate – the same disappointment-tolerant strategy which produced a positive 20% return from the potential moonshots class above – then our expected return is nil (0.1 x 0.8 x 6 + 0.1 x 1.3 x 4). In this case, a 60% false positive rate produces a lower result than chance; the disappointment-tolerant selection strategy which worked for moonshots doesn’t work for mundanes.
To achieve a positive return from mundanes, we need to penalise false positives more heavily. Say we tighten our selection criteria to reduce the false positive rate to 40%. Then our expected return is 0.1 x 0.8 x 4 + 0.1 x 1.3 x 6 = 10%. With the false positive rates of 30%, 20%, 10%, and 0%, the expected returns are 15%, 20%, 25%, and 30%.
By tightening our selection criteria, we also probably increase the false negative rate that is we reject some good mundanes. But we don’t care much, because no mundane makes a big difference to the portfolio, and there are hundreds more good mundanes to look at.
In advocating raised tolerance for false positives when selecting potential moonshots, I am not saying that we should set out to make careless judgments. We should strive to avoid ex-ante errors of analysis; but we also need to accept that even diligent judgement may lead to a high rate of ex-post disappointments, and we need to be comfortable with this pattern of outcomes.
A problem with advocating higher tolerance for false positives for selecting potential moonshots and for selecting mundanes is that shares are not labelled as belonging to one or other of these categories. The categorisation is itself a matter of judgment. I have no solution to this.
How do we increase the false positive rate for potential moonshots, and reduce it for mundanes? The most obvious way is just to be (a little) more credulous when assessing potential moonshots, and conversely for mundanes. To formalise this, one can use looser requirements for current financial metrics when assessing moonshots.
Another way might be to apply an inclusive checklist for potential moonshots, and a disqualifying checklist for mundanes.
By inclusive checklist I mean that the presence of certain positive features (say a management team with exceptional previous start-up success) guarantees inclusion in the portfolio largely irrespective of other any concerns. By disqualifying checklist I mean that the presence of certain negative features (say Debt > 3 x EBITDA, or large share sales by insiders) guarantees exclusion from the portfolio, irrespective of any other merits of the company.
One extant manifestation of my suggested strategy “fatter-tailed and/or right-skew returns => be more tolerant of false positives” is the tech start-up sector. For angel investors in tech start-ups, most investments are duds, but they hope to more than make up for this with a few runaway hits. Peter Thiel (of Paypal / Facebook fame) suggests that to a first approximation, an angel investor will achieve a positive return only if his single best investment ends up being worth more than all the others combined.
Update (21 October 2012): Paul Graham makes much the same point: angel investors in tech start-ups are Black Swan farming.
Why I don't use DCF models
One criticism of the investors in Free Capital which I have heard from more than one expert reader goes something like this: “The interviewees say they are making investment decisions, but none ever actually works out what a company is worth.” These readers then elaborate on the concept of intrinsic value – the discounted future cashflows (DCF) of the company, as distinct from its book value, liquidation value or market value. They suggest that “real investors” focus on this concept of intrinsic value.
I seldom write down an estimate of intrinsic value, and I’m not sure I’ve ever attempted a DCF valuation of a company. I think mainly in heuristic short-cuts: quick and dirty metrics like P/E ratio, dividend yield, price/sales, price/net current assets, price/net tangible assets, and so on. Of course, P/E ratios imply rates of capitalisation: if I think a P/E of 12 is ‘fair’, I’m saying intrinsic value is the company’s current earnings capitalised in perpetuity at 8½% pa. But in general, I don’t find it helpful to make this transformation.
There are several reasons why I find simple heuristics more useful than more rigorous analytics like DCF valuation.
Time is precious There are more than 2,000 shares quoted on the London Stock Exchange and AIM. Given the scope of the search space and the pace of change, DCF models simply take too long.
If you need a calculator, it’s too close A good buying opportunity shouts at you from the market. The cheapness should be striking enough that you can see it without detailed calculations. If you need a calculator – let alone a spreadsheet – you should pass, because it’s probably too close.
Robustness matters more than refinement Investment is about finding valid discrepancies in a noisy-information environment. Finding discrepancies is easy: there are always plenty of companies which appear to have extreme valuations. But most of these discrepancies are not valid: the company deserves its extreme valuation. When you think you've found something, searching for further independent insights which confirm or disconfirm the discrepancy is more useful than refining your estimate of its size.
In other words: when information quality is good, focus on quantifying and ranking your different options; when information quality is poor (as it usually is in investment), focus on raising information quality. (In a different but analogous context, Givewell give an explicit Bayesian justification for this.)
Non-financial heuristics are quicker Sometimes heuristics such as affinity – the class of people associated with a company – can be a quick and sufficiently accurate route to correct decisions. For example, John Hempton suggests finding stocks to short based on a company’s association with dodgy people, not dodgy fundamentals. He will short a stock (in very small quantity) based on association with one suspect promoter and one suspect lawyer, without any investigation of the fundamentals. If the stock rises (ie moves against him), he investigates the fundamentals; if it goes down, he just takes the profits and moves on to the next one.
The heuristic investor may make some mistakes the rigorous analyst does not make. But the heuristic investor works much faster, and is able to evaluate many more opportunities. This is usually a good trade-off.
Displaying entries 1-8 of 20 |
Next page »