HONTI #1: How Not To Invest

Rule #1: Bias the odds in your favor: start by avoiding the avoidable mistakes.

Introduction

There aren’t many good ways to invest, but there are countless bad ones. Yet the knack to successful investment, in the end, is formidably easy: learn how not to invest – and then do the opposite. Perhaps the biggest problem for investors is that the very nature of investment tends to cause automatic behaviours that damage our returns. It’s not hard to find the underlying causes – we evolved to cope with a very different environment, where decisions were a matter of life and death. When these decision making mechanisms are translated into the modern world they cause us to behave in ways that economists regard as irrational, and which are generally damaging to our personal wealth.¹

In truth our apparently peculiar reactions to all sorts of situations aren’t irrational at all, they’re just inappropriate. Unfortunately, as these behaviours are very deeply buried in our brains, and largely unconscious, they’re very difficult to eradicate when something triggers them – and almost everything about investment tends to fire off these ancient risk management mechanisms.² A rich irony of this evolutionary legacy is that more knowledge is not always an advantage for investors. In fact more information can actually lead to worse decisions, because in an environment as uncertain as investment you can usually find some data that supports any decision that you might prefer to make – and generally we bring our preconceived preferences to every decision that we make.³

The classic answer to these problems is to look at the past in order to predict the future. As Winston Churchill once said: “Those that fail to learn from history are doomed to repeat it”, and it’s not surprising that we find a lot of investment analysis is dedicated to picking over past data in order to predict the future. While this may seem like a rational way of overcoming our behavioural biases in fact it exposes us to a whole new set of issues about how we select the data we analyse and whether the past is truly representative of the future.⁴  Indeed it’s possible to claim that the purpose of the investment industry is to analyse past data to present the most favourable vision of the future. Most small investors fail to recognise that the industry is a machine dedicated to transferring money from their pockets to its. This failure too is a problem caused by deeply buried neural mechanisms.⁵

At the heart of these analysis problems is the issue about how we make decisions. Most economics analysis operates in terms of statistics, but most human analysis doesn’t. There’s a good and simple reason for this – historically people had to make decisions based on the information they observed around them, not on data gathered by researchers from unseen events. So our decision making processes are heavily biased towards our own experiences.⁶  More than this, our data gathering tends not to be methodical – we will almost always prefer a good story to a set of raw data. This is how newspapers and the media work – a dry recitation of numbers doesn’t sell, but a human interest story, no matter how unrepresentative, is good for business.⁷ When we come to make investing decisions these aren’t just biased in favour of our own experiences, no matter how unrepresentative they are, but are weighted towards a good narrative – the inexorable rise of China and India, the infinite potential of dotcom stocks, the unstoppable momentum of Japan, the eternal dominance of the Nifty Fifty … all the way back through history to the South Sea Bubble, Tulipmania and beyond.⁸

Here, though, lies a clue to a possible approach to sensible investing because this is history in its rawest and most accessible form. Bubbles form and burst and do so with remarkable regularity, yet investors manage to ignore this seemingly inevitable cycle. In this they can be forgiven because nearly everyone else does as well: economists, regulators, investment analysts, politicians and expert commentators to name but a few.⁹  This almost wilful ability to turn a blind eye to the ever-present possibility of collapse is also a problem seated deeply in our brains. Psychologists call it disaster myopia, short-sightedness when it comes to the probability of a disaster. This, like most of these so-called behavioural biases, isn’t just a problem for investors, it affects us in all sorts of ways: building nuclear power stations on a flood plain in an earthquake zone, or failing to put tsunami warning systems in place are simply recent examples of a long-standing problem.¹⁰

Disaster myopia extends beyond individual investors to infect the whole investment industry. In 2008 when the sub-prime crisis struck almost the whole of the financial industry from regulators to the credit rating agencies to the investment banks was using a statistical model of risk that showed that the chances of a market crash were negligible, which proves only that statistical models are merely as good as the data you give them. Of course, the data is provided by humans, who sift and select, and are affected by their own preconceptions and behavioural biases.¹¹

In this particular case we saw a common investment problem rear its head: people change their behaviour dependent on situations, and are strongly affected by incentives. Many subprime mortgages were given to people who couldn’t afford a plastic sheet, let alone a five bed condo, because the sales agents were incentivised to do so by commission. The people running the models believed that their models were protecting them from problems but failed to realise that the models only covered risks that had been common in the past.¹²  The net result of this was that the behaviour of the sales agents changed to garner more commission, leading them to introduce new risks that the models didn’t capture. In effect the network of sales agents probed the risk models by attempting to make sales until they found holes in the protection. Meanwhile the heavily bonused risk modellers and their bosses were happily unconcerned until the roof fell in.¹³

This idea that people change their behaviour when their situation changes is known as reflexivity and because stock markets are really only the sum of people’s behaviour if that behaviour changes so does the markets. This leads to feedback, because as markets adapt to new behaviour so they present participants with a new situation, and reflexivity dictates that people change their behaviour once more. We can summarise this: people are reflexive, markets are adaptive and they go around in a dance which never ends. There is no perfect way of predicting what will happen in stock markets, there is no specific, definable outcome. In the jargon when we invest we do so in conditions of uncertainty.¹⁴

Uncertainty is the key term: when we read, anywhere, of someone confidently predicting the future we should immediately fire up our bullshit deflection system. One of two things is almost certainly happening: either the prediction is so general it has virtually no chance of being wrong – and is therefore worthless – or its only chance of being correct is serendipity. In a world where everyone is making predictions someone will predict everything, but there’s no need to reward them for picking a winning lottery ticket.¹⁵

In truth, success in investing is about hard work and biasing the odds in your favour. You don’t need to be a genius, just someone willing to understand the basics of investing and possessed of sufficient self-knowledge to recognise that there are times and places where sensible people should simply go sit on the fence and watch the circus unfold. We don’t have to juggle with flaming chainsaws just because everyone else says we should.

So this is HONTI: How Not To Invest. Over the coming months we’ll pick each of these destructive behaviours apart and flag up some ways of dealing with them.

>> HONTI Home Page

---

Notes to the article:

1. The difference between what economists and the rest of the world regard as "rational" is discussed in Unpredictably Rational. But essentially, economic irrationality means people not doing what economists' models say they should: so who's wrong, us or the economists?  Look also at O Investor, Why Art Thou Rational?
2. There's lots of research on the nature of  automated reactions to investment situations.  Take a look at Stocks Aren't Snakes, for instance, and also the research on how emotions are both necessary for our everyday health yet potentially damaging to our investment returns, such as that done by Antonio Damasio in Get An Emotional Margin of Safety.
3. The idea of information overload is now widespread, and the psychological term, known as psychophysical numbing, is explained in Risk, Stone Age Economics and the Affect Heuristic. Note again that emotion - supposedly the thing that rational investors should avoid - is absolutely key to survival, and is unavoidable for investors.
4. The dangers of people selecting their data to prove preconceived points is outlined in Sharpshooting the Investment Gurus: the Texas Sharpshooter Effect is everywhere.  Meanwhile the curse of data mining is explained in Twits, Butter and the Super Bowl Effect - do make sure you read the brilliant paper by David Leinweber.
5. The unsurprising tendancy of financial institutions to manipulate data in their favour is widespread.  Look at Survivorship Bias in Magical Mutual Funds or Finance: Where the Law of One Price Doesn't Apply or Freedom of Financial Choice is a Myth for examples.
6. The "problem" that people have evolved to deal with observed frequencies rather than doing complex statistical analysis of data they can't possibly know about is described in Investor Decisions - Experience is Not Enough. Related to this is the representativeness heuristic which is covered in Behavioural Finance's Smoking Gun, along with Gerd Gigerenzer's ideas about how we actually reason.
7. The idea that we're storytelling apes rather than statistic churning nerds was outlined in Fairy Tales for Investors. There's a related discussion in Memes, Money, Madness about how ideas form lives of their own and propagate.
8. The relationship between observed experience, a preference for narrative over data and the not infrequent failures of markets was described in In Markets Bad Stuff Happens - Frequently while the psychology behind market crashes was outlined in Why Markets Crash. How the media interacts with investors, and pretty much everyone else, to generate fear and panic is addressed in The Media, Fear and Stockmarket Manias, linking back to the issues about how we process information
9. The misjudgements of "experts" is a perennially amusing area to investigate - see You Can't Trust The Experts With Your Investments, for instance. Philip Tetlock's research is essential reading for the intelligent investor.
10. There's plenty of evidence around about disaster myopia: see Black Swans, Tsunamis and Cardiac Arrests.  Both Nicholas Taleb and Richard Posner have written extensively about this and it seems to be intricately bound up with the problem of Hindsight Bias.  You can follow through on the links in that last article for more details about a range of specific behaviourial biases.
11. The failures of risk management models by the financial industry are a favourite topic, although it's a bit like shooting fish in a barrel.  Have a look at It's Not Different This Time and Risk, Reality and Richard Feynman: especially read the paper from Kevin Dowd and colleagues on "How Unlucky is 25 Sigma?" for amusement and education.  Also see Credit Rating Agencies: A Market Failure? for an example of just how haphazard is the world of professional risk management.
12. The power of incentives to make people do the unexpected seems almost infinite.  Look at Perverse Incentives Are Daylight Robbery, Gaming the System and On Incentives, Agency and Aqueducts. 
13. The nature of incentives to cause misbehaviour across a whole range of organisations across the subprime crisis is covered in When Muddled Modellers Model Muddles.
14. The idea of economic reflexivity championed by George Soros, that people change their behaviour in response to changed conditions, is described in Soros' Economic Reflexivity while the concept of adaptive markets was covered in Capitalism Evolving: Be a Cockroach, Not a Dinosaur.
15. Uncertainty is a key concept for investors, there's an introduction at Ambiguity Aversion: Investing Under Conditions of Uncertainty while there's more detail at Physics Risk Isn't Market Uncertainty - which also starts to introduce the historical background to some of the fundamental problems that modern economics suffers from.
16. An introduction to how a lack of self-knowledge leads us into basic investment errors is given in Profit From Self-Knowledge.