In search of the sector effect
Prior to the launch of the euro in January, investment banks began reorganising their analysts into sector as opposed to regional groups. The belief was that the new currency block would mean that stocks would begin to set behavioural patterns within individual sectors.
Six months on, are there any signs that the theory has been borne out?
The banking sector seems a good place to begin an examination of trends. Easily identifiable, solid, and presumably following set patterns in response to changes in the economic climate.
Most analysts are certainly treating banks as a sector, but find it difficult to come up with any concrete proof that they are behaving in the manner that allows them to be viewed as such.
Tim Harris, equities strategist at JP Morgan in London, says the evidence is there. “We definitely feel that it is moving towards a time when we will be able to view all sectors individually, not just banking,” he says. “We are monitoring their behaviour from a correlation point of view.”
Over a 10-year period JP Morgan strategists monitor a sector such as banking, creating correlation pairs from a group of 10 companies. Such a group would provide 45 correlation pairs. They then examine the results to see how closely the pairs behave, and calculate a degree of correlation.
“Over the past three years, the banking sector has shown modest correlation, but this has increased from less than 10% in 1996 to 45% in 1998. This latter figure is really quite high, and with the euro imminent I think everyone felt it could only increase,” says Harris.
As well as banking, there has been strong correlation in telecommunications, despite some controversial deals in southern and eastern Europe, as well as construction and the retail sector.
Then, however, something strange happened. Last autumn the banks began to drift their own way. Other sectors too showed falling correlation in the wake of the Asian crisis, as stock exchanges reacted violently to past news and predictions. Inevitably there was a recovery but again it surprised strategists.
“By last September banking was back to 24% correlation, but since the introduction of the euro we have seen it climb to an all-time high of almost 50% All other sectors have behaved similarly since September, and it now means that we can look at sectors in just the same way that we would look at country groupings such as the DAX 30,” says Harris.
It looks then, as if we can expect to see correlation inside Euroland sectors to increase steadily over the second half of the year, and prove to be a reliable indicator. Clearly, however, shifts in economic policy around the globe, not just Europe, is likely to have an effect, but not more so than on regional groupings.
There is a danger, however, that simply following sector analysis will have a dangerous downside on pricing. It could mean that there may be buying of the market leader as proxy for the sector. If the sector is moving as one, it is possible that individual pricing may become slipshod and careless in the view of a number of analysts.
But to a certain extent, this is no bigger danger than regional groupings posed in the past. Managers will simply have to reset their sights for the tell-tale signals.
All indications are then, that sector analysis and movement has rightly been chosen as the way to predict equity movement in Euroland, and the predictions are that we shall see all stocks beginning to behave as sectors by the year-end.