Unlock the Editor’s Digest for free
Roula Khalaf, Editor of the FT, selects her favourite stories in this weekly newsletter.
As a clever-clogs analyst a long time ago, I remember ripping a friend’s new business venture apart. Why would you touch such a horrible industry? The profit margins are barely in to double figures! But when revenues passed eight digits, his earnings were a million pounds per year. He realised before I did that money is absolute, whereas my financial models were full of percentages and ratios.
Relativism is not only a cultural problem. It is also rife in finance — as my friend demonstrated — as well as economics and politics. We are constantly bombarded with comparisons which make no sense at all.
Take the news that the UK plans to increase its defence budget to 2.5 per cent of gross domestic product by the end of the decade. What on earth does military spending have to do with the final value of goods and services produced in a country in one year? Even Nato believes its 2 per cent of GDP guideline “serves as an indicator of a country’s political will to contribute”.
No it doesn’t — the denominator has nothing to do with politics. If anything should be calculated bottom-up on an absolute basis it’s national defence, assessing risk versus capability. In any rational world, how many 155mm artillery shells are needed should be independent of consumer confidence (the biggest driver of GDP). Armies should not starve if artificial intelligence fails to boost low productivity.
Carbon intensity metrics used to judge corporate greenness are another example of relative hogwash. Our atmosphere needs lower emissions, period — not having them increase less than revenues. Plus you can end up with a ridiculous situation in which companies are praised when sales outpace emissions solely because of inflation.
The hot topic of unemployment also suffers from wayward denominators. America has its jobless claims data, but the rest of the world rarely mentions the actual number of real human beings out of work. Mostly we talk about the unemployment rate, expressed as a percentage.
But these are a fiction — usually dividing an arbitrary definition of unemployment by a worse estimate of the number of people who are economically active. If someone feels worthless the day a survey hits their door mat, they may tick the “currently not actively seeking work” box. Hey presto, they aren’t included in the unemployment rate.
And therein lies the appeal of ratios and percentages, I suppose. They are one step removed from having to face unpalatable truths in either the numerator (how many tanks do we actually have compared with Russia) or the denominator (better optics on defence spending are merely due to a recession).
The inequality debate is another case in point. Those on the left prefer to throw around comparisons between the rich and poor. Meanwhile the right tends to focus on the numerator only, showing how much wealthier the poor are in absolute terms.
You may think markets are above spurious relativities. And to their credit (and, often, my shame) they frequently are. For example, no one cared that Amazon or Tesla had negative margins for years. Investors knew the revenue line was the only thing to watch.
Likewise, for decades markets have not been spooked much by soaring public debt ratios. Rich country public sector liabilities are now 100 per cent of GDP on average. This may yet prove disastrous. However, so far government bond owners have been right not to panic.
That said, investors still do things every bit as silly as comparing defence budgets to national incomes. We rank companies by their price-to-earnings ratios, say, despite the fact every company and sector measures their net profits differently.
Strategist reports are also devoured. These mostly consist of charts of variables which seem to move together (such as solar flare activity and year-on-year Nasdaq returns). Causation is claimed when the two numbers are merely correlated — if even that.
Perhaps our most common investment mistake is comparing today with the past. To be sure there are ratios that do revert to a mean — returns on equity for example. But extrapolating historical relationships is often ruinous, as many a failed hedge fund — Long-Term Capital Management springs to mind — will tell you.
Absolute numbers can mislead as well, of course. Sure, Alphabet’s market cap rose by a headline grabbing $250bn on Friday — but it’s a big company. Similarly, it was wrong during Covid to quote cases and deaths by country with no reference to population size.
We must all be more careful not to take relative or absolute facts as given. But the former are especially dangerous; inaccuracies can lurk in two numbers. And very often these shouldn’t even be compared at all.
stuart.kirk@ft.com