Investment performance comes primarily from having a good and stable process within which you can execute your strategy and perform your analysis. If your process flip-flops or you make exceptions 'just this once' you are drifting - and your results will eventually show.
So what do we mean - exactly - by this bold headline statement? Let's take an example not from investing but the very close cousin of capital allocation in companies:
In a landmark study of 1048 large, global companies two Harvard professors, along with McKinsey, concluded in 2009 that only about 8% of financial performance of the companies over a five year period was attributable to the volume of data and the sophistication of subsequent analysis concerning capital allocation.
A further 39% was due to various relatively non-controllable factors such as the predictability of the given industry, availability of capital and how many options for capital allocation they had.
The remaining 53% was attributable to what the study called 'Processual factors' that covered 9 specific variables on how the companies worked with the available analysis and the given non-controllable factors. This could be the inclusion of opposing views in management discussion, team diversity, specific handling of large unknown factors and so on.
To us the results are in a sense both striking and so predictable as to be almost boring. Because when you think about it it makes perfect sense that if you are faced with the choice of having either a good and rigorous process for thinking through something, despite only having basic data to run on, you're far more likely to avoid catastrophes and make effective decisions than the opposite, which means large volumes of elaborate data and modelling, which is then processed in an off-handed, biased and overconfident manner.
The first is like having your 35-year old religious aunt drive you to church in a family sedan on a lazy Sunday while the latter is like having your drunk nephew give you a joyride in his new porsche while you're worrying that he seems to have forgotten his glasses. Sure the Porsche is a far better car than the sedan - it's just really not what matters.
The same common sense insight applies to investing: you can fail - spectacularly - even with a very good strategy and very good data if your process is biased or otherwise flawed.
That doesn't mean that strategy and data can be dispensed with - far from it. Both play crucial roles in high level investing done right. But they are, and remain, secondary to good process.
So what is a good investing process?
This is the core theme of this site, and you should head over to the articles section where you can find our writing on that and subscribe to the email service (if you want to of course).
In the mean time the abbreviated version is that a good investing process is the one that deals the best with risk. Some would argue that what we should be dealing with is return, but for reasons that we'll not cover here we'd advice you to keep your eyes fixed on risk - if you do the returns will come.
The best concept of risk we've been able to come up with is the notion that risk is the chance that something bad might happen. There are serious flaws in all of the more technical definitions which you'll see us go on about elsewhere.
For investing purposes we can unpack the concept a little bit more, however, and say that risk is a function of certainty, character, time and price.
In other words risk is a factor of:
- How well you know what you are dealing with.
- How disciplined you are in order to follow up on that knowledge: which covers both your personal strength of character and preparation but also, if you invest on behalf of others, what I call Institutional Character.
- Random chance events - both those predictable and unpredictable, good and bad - as time goes by.
- What you pay for a stake in all of the above.
We simply define an investment as the purchase of a security where you've deliberately and adequately dealt with all four aspects of risk above.
The process we use
The process we use follow these 4 steps in order:
- Identify: this is the process where you think through the business and try and understand it. How it makes money. How it functions in the marketplace. Why customers buy and so on.
- Quantify: we try to quantify as much as possible, including future forecasts and estimates. We use classic financial analysis as well as probability functions and
- Falsify: we try and tear our case apart. Look for evidence that contradicts our hypothesis. Ask other people what they think could go wrong and why. Predict what will go wrong if this fails.
- Testify: the last step involves logging our investment thinking so that we may later revisit it and learn from it. Memory is a fickle thing and we can not count on it to give us an honest assessment of what we thought and how we really viewed a situation when we now look back. So we write down our theses in a formal structure that both forces us to touch all bases of our thinking and allows us to compare the thinking from different investments.
The investment decision is eventually made by asking the following question: "Is this investment better than my other available alternatives - including staying where I am and waiting for better times"