Why the notion “intrinsic value” isn’t very useful

This is one of a series of columns that were produced for Moneyweb Investor in which Stuart Theobald explores the intersection of philosophy of science and finance. This followed an earlier series for Business Day Investors Monthly on the same theme. This column was first published in April 2016. 

Financial analysts often talk of “intrinsic value” when it comes to analysing shares and other financial assets. But just what is that? I find it a rather perplexing notion. I know what the price of an asset is – an amount that someone will pay me for it or sell it to me for. I more or less also know how much I value an asset – so the subjective value that I place in it. But when we talk of “intrinsic value” we seem to mean something different to both these things, something that is a “real” feature of the asset, distinct from both my own subjective view of it and the market price.

The story goes that prices will converge to intrinsic value, because, we like to think, prices are rational and reflect all available information, at least in the long run. Prices can diverge for short periods, creating opportunities to buy or sell. Intrinsic value is often talked about by analysts who focus on the “fundamentals” and like to call themselves “value” investors. They try to ignore the price, which they consider to be the source of much distraction and panic, and focus on the fundamental operations of the company and come to a decision on what it’s worth.

I find this odd, because in one sense it is unscientific. Science is largely about empiricism (not entirely, but let’s not complicate things). Scientists disdain talk of invisible things that can’t be detected. Scientific claims should be about facts in the world that we can, at least in theory, go out and try to find evidence for. In the realm of human behaviour, economists have spent a century trying to shape their science in that same image. The famous American economist Paul Samuelson reviled speculation about what people were really thinking, arguing that the only thing that mattered was what they revealed through their behaviour. This could be measured; what’s in people’s heads can’t be. And when it comes to the value of assets, the only thing that reveals what people are thinking is the price.

Some analysts might respond that they are working off the basis of real facts, and then deduce the intrinsic value. This is true for certain kinds of assets. Some derivatives, like futures, are a mechanical function of the price of the underlying asset, the time to expiry, and the time value of money. While we often use the term “intrinsic value” for such calculations, it is misleading because such processes are just a mechanical extrapolation from the spot price of the underlying asset. This is based on a “no arbitrage” strategy, which works because we know that the price of the future must be such that it is impossible to earn risk-free profits. Such intrinsic values are, a philosopher might say, a priori, analytic facts. The prices are necessarily true, given the price of the reference asset.

Analysts may argue that this is basically what they are doing whenever they calculate intrinsic value. Instead of price, time to expiry and time value of money, they are determining cash flows and discounting these by a discount factor, or calculating a sum of the parts by looking at what prices are visible. But this is only a chimera of objectivity. Future cash flows are uncertain. The discount factor is meant to be a way of dealing with this – basically, that we should be compensated if the cash flows are uncertain. But this depends on a particular notion of uncertainty – that it is itself some measurable thing that can be determined. Well it isn’t. The best we can say of probabilities in this sense is subjective – the credence, or degree of belief, we have in the likelihood of a predicted outcome becoming true. This is an entirely different notion of risk than we typically have in finance – the volatility of prices, a measurable property of historic prices. When we are determining the probability of our estimated cash flows coming true, we are back at that subjective notion of value. It is that invisible thing inside our heads.

Some analysts try to avoid this repugnant conclusion by coming up with hard-wired models that mechanically take inputs and produce an output at the other end. There are many underlying strategies that can be employed – ones that rely on statistical regularities, others that rely on models of causes and mechanisms. However, again in such cases, the can is being kicked up the subjective road. The work is now being done at the level of calibrating the model and choosing the inputs.

What’s the point of all this navel gazing? Fundamentally (that word again) intrinsic value is not a very useful concept. When I see it being used in marketing literature for financial products, I have a flush of revulsion. It is a weasel term, used to obscure the fact that what is happening is nothing more than subjective judgement. It pretends to be a scientific notion but it cannot ever be that. It is ultimately based on the invisible beliefs of an analyst, as unscientific as one can get.

I’m all for wise judgement. But using the term “intrinsic” just obscures what’s really going on, serving as a cover for what in fact may be very unwise judgement indeed.