Defining market ups and downs

Dr Ron Bewley explains what volatility means, and whether reports of ‘volatile markets’ in the past couple of years have been accurate.

It seems fairly clear to me that the lay person and finance analysts use the term volatility to characterise different concepts.

From my reading, the popular definition of volatility is to do with bad things happening and the market not going up.

Analysts, on the other hand, typically use a statistic called a ‘standard deviation’ that weights good and bad share price movements to the same extent.

This measure allows analysts to put some notional range on what might happen to a stock price over a year due to the random nature of news affecting the stock.

For this reason, we usually use the expression annualised volatility to demonstrate that a one-year period is being analysed.

For example, a volatility of 10 per cent in a flat market is taken to mean there is a one in three chance of the price in a year’s time being more than 10 per cent higher or less than 10 per cent lower.

Two ‘standard deviations’, or 20 per cent in this example, implies that there is a one in twenty chance of the price at the end of the year being outside ± 20 per cent.

Understanding volatility and what drives it is important in both constructing and monitoring the performance of an equity portfolio.

The volatility of the ASX 200 index was 13 per cent in 2010-11. That is normal by historical standards. The average for 1985 – 2011 is about 12.5 per cent.

Interestingly, this is the average volatility for the S&P 500 over the last 50 years.

I suspect non-specialists find it hard to believe that 2010-11 was normal because the market didn’t seem to go anywhere and there were a number of modest corrections.

That happens to be life.

Indeed, the total return (including dividends) on the ASX 200 was 11.7 per cent – again about the average for the last quarter of a century.

Furthermore, the return in 2009-10 was 13.1 per cent – a little higher than average – and volatility too was just above average. Two normal years on the run – but few are happy or confident.

To put volatility into context I calculated the return for 2010-11 but starting at any day two weeks before or after 30 June 2010 and finishing any day in the two weeks leading up to 30 June 2011.

The best (approximate one year) return from this set was 13.8 per cent and the worst was 0.7 per cent – a huge range in returns by adding or subtracting a few days.

That is why I think it is important to measure things properly because our memories can easily distort our impressions of what has been going on.

Of course the ASX 200 is made up of 200 stocks – from 11 major sectors – and these stocks do not move in unison.

However, there is some degree of market volatility shared across the 200 stocks.

In chart 1, I show in blue, the market volatility on the right and the volatility of the 11 component sectors to the left of it.

The IT sector had the highest volatility at 21 per cent but even Materials – including BHP, RIO and the little miners – had a volatility of only 18 per cent.

That is, because BHP and RIO have such a large weight in the Materials index that they swamp the little miners. BHP and RIO happened not to be that volatile last financial year.

Investors possibly think more in terms of the volatility of the individual stocks they hold, rather than of some sector index. I have added the red bars in chart 1 to show the median (middle) value of the component stock volatilities from each sector and the broader index.

The diversification within each index and weighting towards the larger cap stocks tends to produce a lower volatility for the index than the average component.

For the ASX 200, half of stocks had volatility above 26 per cent – double that of the index – and half below 26 per cent.

There is not much difference between the volatilities of the median stock and the index for IT and Telco because there are only a few stocks in each of those two indexes.

This difference between the volatility of individual stocks and indexes highlights the importance of not only holding an adequate number of stocks overall – but a sufficient number of stocks in each sector.

In chart 2 I show the volatility of each stock in the ASX 200 plotted against the market capitalisation of the company - with colour and shape coding to identify to which sector the company is classified.

The vertical dotted line separates the stocks in the top 100 (to the right) from the rest.

There is much more variation in the volatilities of smaller cap stocks than larger.

The red diamond at the bottom right represents BHP and the four green triangles next to BHP represent the four major banks.

Although the chart is too busy to track each individual stock, I hope that chart 2 clearly shows that lots of the high volatility stocks are typically red and blue diamonds (materials and energy).

It just so happens that BHP is a relatively low volatility stock. 

The implication of chart 2 is that investors should be prepared to hold more small cap stocks for diversification purposes than if they were to confine themselves to larger cap stocks.

But not all volatility is bad. The high red diamond just to the right of the dotted line represents Lynas – one of last year’s ‘market darlings’.

Bathurst Resources – with the highest return for the year – is another high red diamond but to the left of the dotted line.

To separate the good from the bad volatility, I have plotted stock volatility against the returns over 2010-11 in chart 3.

The vast majority of high volatility stocks also had high returns.

This trade-off is the standard risk-return trade-off.

Of course, not all high volatility stocks had high returns, but all of the high returns stocks had high volatility.

Managing a ‘high octane’ portfolio of high return stocks is difficult. Stock prices do not go up in a straight line so how does a manager differentiate between a temporary dip in price from a serious downturn? With difficulty is the only honest answer.

But is volatility predictable?

The Australian Securities and Investments Commission implores us all to say past performance is not a reliable indicator of future performance, but Rob Engle got the 2003 Nobel Prize for his work on volatility.

His ‘elevator speech’ version of his lifetime contribution was along the lines of “I found some averages of past volatility to be useful predictors of future volatility”.

I show a highly simplified presentation of this inertia in volatility in chart 4.

I calculated the volatility for each stock in each of two years: 2009-10 and 2010-11. Since the composition of the index changes over time, and some stocks are offered or delist in a given period, I have only included the stocks that were in the ASX 200 on 30 June 2011.

I excluded all stocks that did not have prices for the two-year period. The red line in the scatter diagram is a line of best fit.

Since 2010-11 was slightly less volatile than the previous year, the red line is a little flatter than a ‘450 line’ that would be the case if volatility was the same in both years.

From my perspective, the blue dots are reasonably clustered around the red line for low-volatility stocks but less so for higher volatility stocks.

That is, other than for major market movements as we saw in the GFC, low volatility stocks are reasonably likely to stay that way.

Importantly, no high volatility stock suddenly became low volatility – that would require blue dots in the bottom right hand corner. Some moderately volatile stocks got a whole lot more volatile in 2010-11.

The take-away from this analysis for investors constructing and monitoring their own portfolios is:

1. Stocks are a lot more volatile than indexes and so stock selection (and the number of them) within a sector is extremely important.
2. Small resource stocks tend to be a lot more volatile than other stocks. A portfolio with just a couple of small or mid-cap miners representing the materials sector could produce a wild ride. Including BHP and/or RIO or a larger number of other juniors might be needed to slow down volatility in that sector of an investor’s portfolio.
3. Large cap stocks are typically a lot less volatile than small caps.
4. Stocks with a volatile past are not particularly likely to become low volatility stocks.

Technical note

When analysts calculate volatility or returns they usually use what is known as the ‘log (1+r) return’. Certain problems could arise without this transformation. A simple return, r, cannot be less than 100 per cent but it can be 1,000 per cent or more.

Also, if a stock price goes down by 50 per cent then up by 50 per cent it doesn’t finish up from where it started. The log return approach is necessary for proper calculations of annualised volatilities, but for small variations there is little difference between the two measures.

Dr Ron Bewley is the investment consultant at Infocus Money Management.