Tuesday, October 14, 2008

About a chart

Here is a recent chart-of-the-day I found quite interesting:



It's interesting in the sense that, in 1929, the stock exchange actually fell much more abruptly over a short period, but, as shown on this graph, on the longer timescale of a year, it actually fell less than in the current crisis. So it seems to point out the existence of various kind of volatility, one short term, and the other long-term.

One may propose a few remarks to explain this difference:
- There is much more volume today, implying some kind of inertia in the market
- The traders, despite all their shortcomings, seem more aware than their ancestors of economic forces and less liable to panic moves, but more liable to early reactions anticipating the worse.
- The way the authorities have managed this crisis is much better than what was done in 1929, and it seems to have at least spread the fall on a longer time-period, which, in itself, is a very positive achievement.

On the contrary, it may also indicate that, while the digestion of bad news is more progressive , this one still has to run its course fully. And that may be a teaching of the relative powerlessness of existing institutions which are basically only reactive in absence of proper regulations.

On a more technical aspect, all this indicates that the two volatilities are intimately connected and that basically, what is taken from one goes into the other, the final distribution being a result of current psychologies and existing institutions.
Despite the uncontroversial nature of that remark, it is a bold statement if one takes the time to extend it to all kinds of assets, and in particular to currency pairs. To perform this extension is however nowhere near obvious. We all heard of the wild variations of currency value in the 29 crisis, but the monetary system was then very much different from what it is today. In particular, there was a gold standard, and the arbitrage system we have today was not in place. In absence of a careful analysis, it would be specious to conclude anything detailed.
However, without that analysis, and given the current, more "efficient" system of currency pairs, it may already be possible to conclude that very high volatily in the currency market is to be expected in the coming months.

The weight of news

The following article from the Federal Reserve addresses the matter of the effects news announcements have on some assets price (taken in a general sense):
http://www.ny.frb.org/research/current_issues/ci14-6.html

It's a purely statistical approach and therefore lacks any model to really make sense of the data. In particular, the sample of data does not reflect the difference that may exist between a bull market reaction and a bear market reaction.
Some interesting comments however on the type of assets that are the more reactive, on the indices that elicit the most volatility, on the timing of the most significant reaction.