Join our community of smart investors
OPINION

Understanding investment in 50 objects: a medal for a model

Understanding investment in 50 objects: a medal for a model
August 11, 2016
Understanding investment in 50 objects: a medal for a model

It's surprising we haven't used this object before. It's the medal presented to the winner of a Nobel Prize in economics; or, specifically, the Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel. Surprising because, among the people who have featured so far in our tour of investment, we have stumbled across three Nobel Prize-winning economists - Daniel Kahneman (the Cornell University coffee mug, 10 June 2016), Harry Markowitz (the portrait of Porter Boy Resting, 17 June) and Franco Modigliani (a packet of M&M's, 24 June).

So maybe this week's choice of object is chiefly an excuse to ramble through some tasty little factoids about the Nobel Prize. For instance, the medal - in 18-carrot gold, gilded with 24-carrot plate - that's awarded for economics is different from the one presented for the other five disciplines. That's partly because the economics prize is the nouveau arriviste. Whereas the first Nobel Prize was awarded in 1901 - in physics to Wilhelm Röntgen for his work on X-rays - the economics prize only dates from 1969. And the appearance of the medal for economics is distinct. Like the Nobel Prize medal, it is dominated by a bust of Alfred Nobel, but the pose is different. In addition, Sweden's central bank, Sveriges Riksbank, gets a mention - rightly, since it sponsors the prize.

However, the prize money is the same as for the other disciplines - 8m Swedish krone, a bit more than £700,000. That prompts the speculation whether the prize is more valuable than the medal and there has been a second-hand market in medals to grease the debate. Value will vary according to who is selling. In 2015, when Leon Lederman, a physicist, auctioned his medal to meet medical bills, he raised about £510,000 at prevailing exchange rates. But a year earlier, James Watson - he of 'Watson and Crick' fame - got £2.6m when he auctioned his medal.

Most likely, economists don't have the public profile to get near that figure. But that hypothesis has never been tested - none of the 76 economics laureates recognised since 1969 has sold his or (in the case of Elinor Ostrom, the sole female on the list) her medal.

Maybe that's surprising since economists like to formulate hypotheses then test them. In a sense, that's all they do. William Sharpe - the subject, so to speak, of this week's object - tested the hypothesis that, in liquid financial markets, prices were determined by many factors but, in practical terms, all these could be distilled down to one - the relationship between price movements in the whole market and price movements in the asset under review. In so doing, Sharpe formulated the capital-asset pricing model, or CAPM (pronounced 'cap-em') and for his efforts won a joint share of the 1990 Nobel Prize.

The prize was deserved since CAPM is a cornerstone of modern financial theory. It picks up on the mean-variance analysis of portfolios by Harry Markowitz then simplifies it brutally but effectively. CAPM says that investors are rewarded for two things. First, for choosing to save rather than spend, but that's a low-risk activity for which their reward is a risk-free return or, to use some jargon, 'the price of time'. Second, they are rewarded for taking on risk. Prospectively, the returns from taking on risk must be greater than the returns from risk-free saving, otherwise no one would buy risky investments. Sometimes it transpires that risk is priced wrongly and investors suffer. That has to happen otherwise there would be no risk. But CAPM's hypothesis is that investors in risky assets get a risk-free return plus a premium for taking on risk.

From that starting point, the challenge was to identify and quantify risk. Markowitz had done that in his mean-variance analysis, but in a hugely complex way. Sharpe's insight was to see that the job could be done just as well by contrasting price movements in the investment under review with those of its "most important single influence". And that could be achieved by way of a standard regression model, using lots of data points on a graph with Cartesian co-ordinates.

So the most important single influence would be the independent - or predictor - variable and the investment under review would be the dependent variable. Risk, therefore, became, on average, how much the price of the dependent variable changed for a given change in the price of the predictor. A lot of change meant lots of risk; little change meant little risk. As in a standard regression, the slope of the line of best fit - how much the line rises or falls for a given horizontal movement - quantifies risk. For example, two upwards units for one unit across would mean a risk factor - or beta coefficient - of 2.0.

Put all this together and CAPM can produce a likely return for a security or a portfolio and from that its value can be derived. The detailed workings for shares in blue-chip beverages group Diageo (DGE) are shown in the box below. True, in the 55 years since William Sharpe formalised the basic CAPM it has been tweaked, refined and improved. It has also been criticised for being simplistic and for getting the wrong values. Maybe. But its merit lies in pointing to the truth that primarily investors are rewarded - and occasionally punished - for taking on the risks associated with the markets in which they invest. Like it or not, investors can rarely escape that influence. Providing a simple model that points up that truth was probably worth a Nobel Prize medal.