The Number Doesn't Matter!

A lot of the investment industry is based on trying to look smart, rather than trying to be smart – whether it’s to your existing investors you want to keep, or new investors you want to attract. You might even spend time trying to convince your colleagues you’re smart, in order to get your ideas into portfolios.

Here are a few of the more common tricks I see used:

• Fact bombardment. Overwhelm your audience. Recall statistics or facts verbatim (or pretend to do so) at such length and with such vigour that those on the other side of the table are ground into submission; “if they know this much information, they must know what they’re talking about”.
• Jargonise. Use meaningless but fancy sounding words as substitutes for things real humans say (a favourite of the more quantitatively-inclined investor/ bankers). Instead of saying “change”, talk about “delta”. Instead of saying “if markets go down this thing will probably lose a sh*t load of money”, say “negative convexity”. Whoever you’re speaking to will either think you’re fantastically clever (see above), or be so tired of translating that they throw in the towel and give you whatever you want. Then you get out of their office, quickly.
• Be accurate. Predict the future with outrageous accuracy. Assess the risk of your positions to five decimal places. Have five different screens with fifty different models running at any time. Know the answer!

The first two can perhaps be explained, or at least, forgiven slightly. After all, people working in finance spend a huge amount of time and money preparing for industry qualifications with fancy acronyms. That makes analysts a) good at recalling useless facts, b) teaches them lots of meaningless but fancy sounding words and c) makes them want to USE their knowledge.

However, when it comes to being accurate, that does feel like it’s somewhat the job of those working within investment finance. Surely that precise number is important??

No.

There’s too much uncertainty in the investment environment to rely on one number. Ever. Trust me! However, the opposite, which tends to be a commonly held view, isn’t true either. I’ll often hear that in relation to investments “setting expected returns is a waste of time”.

In the hands of the retail or untrained investor, point estimates can be dangerous. We often go to great lengths to emphasise that, however important, return estimates for financial planning purposes are guideline returns, as much as expected returns. Maybe that’s a subtle distinction. However, it does serve to emphasise the uncertainty. For professional investors, if we know those caveats, what’s the issue? Why do swathes of professional investors stray away from putting a number to how much money something could make? Just because finding the exact number is likely to be impossible, it doesn’t mean that trying to find a number isn’t important.

As I first realised when working on an investment risk desk, very (very) frequently the actual number you calculate for risk doesn’t matter that much. That's partly because there are normally several numbers that could all be right at the same time. What actually matters, much, much more than a number, is understanding the model you used, and how it might fail or be flawed or less accurate. After all, what use is a risk number of you don’t know what drives it, or how to change it?

When you become a decision-maker or “risk taker” within an investment organisation, you have even more things to estimate. At that point you also have to formulate a view on the likely return from an investment, as well as the risk characteristics. Again, in this instance the number you estimate doesn’t matter so much as the process or model. That will force you to define assumptions, methodology and subsequently gives a framework for discovering more about your investment idea. Much more than is possible in the simple “good vs. bad” dimension.?

Sebastian Page’s book, ‘Beyond Diversification’ (which I would highly recommend), opens with an anecdote from a conference. Someone is speaking on stage about financial modelling.

Some smart arse (my words) in the audience puts his hand up and asks something like “surely if the expected returns as an input are wrong, all optimisations are useless - garbage in garbage out”. The response was “if you don’t think you can estimate expected returns, you shouldn’t be in the investment business”. Crushing. A tour-de-force rebuttal.

I agree with the presenter, but with the important nuance that the number doesn’t matter. The process to come up with the estimate is likely very worthwhile though. Asking yourself what the expected return of something is and trying to answer it, whether or not you think that’s possible, is nearly always going to make your investment case more robust. This whole mindset applies to business in the broad sense, as much as simply investment finance. Now, done properly, that should make you look smart.

Matthew Yeates, April 2022

The views in this piece are absolutely my own and absolutely nothing to do with my employer (although occasionally I bore some of them with these opinions).

If you have opinions/ thoughts/ questions on this article, please comment/ share or ask questions!