So regular is this occurrence that it's tempting to suggest the axiom that, if Antofagasta's shares are a main mover on day one, then on day two they will be a main mover in the opposite direction. Okay, I know that won't actually work out, if only because today's rise (Tuesday's) ends a run when the shares dropped heavily four days out of five. Nevertheless, it does seem that the shares have a peculiarly bouncy tendency - up one day, down the next.
Maybe this is simply because Antofagasta is a mining share and we all know that such stocks are supposed to be a geared play on the price of the commodity that affects them most. This has come to seem normal - that commodity prices will be volatile and that shares in mining companies will be even more volatile, even though only limited logic says it must be so.
Yet as I contemplate today's 2.2 per cent rise following yesterday's 2.3 per cent, I wonder: are Antofagasta's shares as volatile as casual observation suggests? Is there a discernible pattern to their price movements? Can money be made out of them?
Certainly, if there is lots of volatility, money could be made from traded options in the shares. The obvious tactic would be to use a 'straddle' - simultaneously buying contracts to buy and sell the underlying shares at the same price. Then, if the share price is volatile enough, a punter can make profits both from the calls and the puts. The problem is that market spreads ensure this is easier said than done. Besides, there is this troubling notion that changes in Antofagasta's share price have little to do with the copper price.
What influences Antofagasta's shares
Average daily price changes (%)* | Versus All-Share | Versus Copper† | |||||
|---|---|---|---|---|---|---|---|
| Antofagasta | All-Share | Copper | Beta | R2 | Beta | R2 | |
| 2014-15 | 1.32 | 0.55 | 0.82 | 1.12 | 23 | 0.05 | nil |
| 2015-16 | 2.29 | 0.85 | 1.05 | 1.62 | 33 | 0.44 | 4 |
| 2016-17 | 2.00 | 0.55 | 0.97 | 1.19 | 13 | 0.39 | 4 |
*See text †S&P GSCI Copper index
Take the data in the table. The three columns on the left show average daily price movements for each of the past three years, ending early this month. It's important to explain, however, what is meant here by 'average'. To emphasise which price is moving most between Antofagasta, the All-Share index and copper all negative daily moves have been switched to positive (a 2.1 per cent drop has been flipped to a 2.1 per cent gain, for example). So the figures tell us the average daily change regardless of its direction. To compare volatility using a conventionally-calculated average risks implying stability where it does not exist. For example, in 2016-17, the actual average daily price change for both the All-Share and copper was 0.07 per cent. Yet copper was clearly the more volatile, as shown by the adjusted data in the table - 0.97 per cent versus 0.55 per cent for the All-Share.
Clearly, though, in each of the three years Antofagasta's shares were the most volatile of the three, with average daily changes always more than twice the other two, with the exception of copper in 2014-15.
So why, one wonders, don't Antofagasta's share price movements produce a higher beta coefficient, where beta - in a conventional least-squares regression analysis - shows the degree to which the shares move for a given change in an underlying factor (in this case, either the All-Share or the copper price)?
Sure, against the All-Share Antofagasta's beta is quite lively - especially in 2015-16 - but not prodigiously so. Meanwhile - and mysteriously - copper barely seems to influence its daily price movements. That point is illustrated by the data in the table for 'R-squared', the statistic that measures how well a slope-intercept equation joins up the scattered data points between two variables (in this case, the price of copper and Antofagasta's shares); on a scale of zero to 100, the higher the figure, the closer the fit. Thus, R-squared figures of nil, four and four (see table) indicate no correlation between copper and Antofagasta worth the name.
Granted, the daily price movements may be filled with too much noise. Take the monthly price changes between copper and Antofagasta over the past five years and the relationship becomes a bit more stable. Antofagasta's beta to copper is almost 1.3, indicating that the shares are a geared play on the copper price. The R-squared is 44, implying that copper is the biggest single influence on its price, although maybe not that dominant. And not for the first time I play around with beta and wonder if this ubiquitous tool for gauging share price returns is really much use at all.
