The market fever index

patrickcallioni | January 7, 2010

In my forthcoming book, Waves of Change: Managing Global Trends in the Financial Services Industry, due out in May 2010, I suggest that it would be useful for investors, especially unsophisticated investors, to have access to a market fever index.

 

My proposal is based on the same logic that underpins bushfire warnings: to let people know that while they are free to invest as they please, they ought to be aware of the risk that is inherent in the environment. It is almost impossible for retail investors to have any such awareness as things stand. I am hypothesising that a useful indicator of risk would be the number of items that appear on the World Wide Web that refer to activity in the global financial markets.

The underlying assumptions are that the number of items would be a good proxy for the level of activity and that investment bubbles are accompanied by a significant rise in activity.

 

To test that hypothesis, on the first Friday of each month I am going to publish my market fever index. The index is constructed by reference to a search using Google, as the dominant search engine, for news items on the WWW that mention one of these words: shares, stocks and bonds. The search will cover a period of seven days, concluding on the day preceding the day when the index is published. The use of the terms shares and stocks is to accommodate diverse varieties of English.

 

I shall make today (8 January 2010) the baseline, noting that at the moment we are still in the relatively quiet waters that follow a flood due to a bubble burst – in this case, the global financial crisis of 2008-09. The baseline will be expressed as 100 points. Shifts in numbers of items will be tracked and the index will be adjusted up or down accordingly.

 

I know that there is an underlying trend that will inherently push up the number of items reported – this is the reality of the internet world. However, for this purpose I will assume that the underlying trend is smooth and that the rising tide of information will lift all boats equally. Relying on this assumption, the trend patterns should still be reliable as an indicator of change, although the underpinning numbers might not be.

I am also aware that there are confounding factors likely to be associated with the choice of a particular day. I am assuming that since mine is a global search, local or regional factors will, over time, have relatively little impact – the bumps will smooth themselves out over time.

 

The first search – which covered the period from 30 December 2009 to 8 January 2010) produced 1,939 items, which now becomes the marker for our baseline of 100 points.

 

Let us see what happens on the first Friday of February.