Research published in the July 2005 Economic Journal proposes a new stock market
trading rule that uses analysis of speculative bubbles to generate higher returns with lower
risk. The study by Professor Chris Brooks and Dr Apostolos Katsaris uses a very long
monthly data series on the S&P 500 to indicate when to move from equities into bonds.
According to their analysis, this is a considerably less risky strategy than ‘buy-and-hold’.
What is a speculative bubble?
A speculative bubble occurs when the price of a financial asset deviates from its
‘fundamental value’. Fundamental values are defined as the amount that a security should
be worth given the future cash flow stream that is expected to accrue to its holder.
Fundamentals are usually measured using dividends or earnings.
During the 1920s, the 1980s and the late 1990s, stock prices rose considerably above the
values that could reasonably be justified by dividends or earnings, and these price
increases were all followed by spectacular falls. Market participants have agreed that a
bubble was present during those episodes.
Early theories of speculative bubbles were behavioural – that is, bubbles arose, grew and
collapsed as a result of irrational investor sentiment and herding behaviour. But many
academics and practitioners are uncomfortable with theories based on irrationality, since
they would expect such investors to be weeded out of the market through time as they lose
more and more money to rational participants.
Recent theories, including the one used in this study, propose ‘rational bubble’ models.
These suggest that investors could be entirely rational in continuing to pay ever-higher
prices for an over-valued asset because they expect to be able to sell it at an even higher
price to a ‘greater fool’ at a later date.
Thus, as a bubble grows, investors are being compensated for the ever-higher risks that it
will collapse by continually increasing returns. Eventually, the bubble will become
unsustainably large as the number of greater fools dries up, and then the bubble will burst,
taking asset prices back towards fundamentals.
The quantitative approach
Brooks and Katsaris develop a regime-switching model that has separate equations for
when the bubble is growing, when it is collapsing and when it is dormant. After this model is
estimated (using a monthly data series on the S&P 500 back to 1888), it can be used to
calculate the probability of the bubble collapsing at each point in time, and this probability
can then be used to inform a market-timing rule.
Within their framework, a bubble can grow and then lie dormant at its current size for some
time, before either growing again or collapsing. The model is better able to capture the
stylised pattern of stock prices over the past century than competing models.
The rewards
Using this approach in real time (in other words, using only information that would have
been available to investors at that time) from 1946 onwards, the study finds that returns can
be increased from 7.2% per annum for buying and holding stocks to 7.9%.
More importantly, the risk involved in the strategy, measured by the standard deviation of
returns over time, is only two thirds that of buy-and-hold. This arises because the model
has some ability to predict highly volatile periods when returns are negative around the time
of a market crash, enabling investors to avoid these episodes.
ENDS
Notes for editors: ‘A Three-regime Model of Speculative Behaviour: Modelling the
Evolution of the S&P 500 Composite Index’ by Chris Brooks and Apostolos Katsaris is
published in the July 2005 issue of the Economic Journal.
Professor Chris Brooks and Dr Apostolos Katsaris are at Cass Business School, City
University, 106 Bunhill Row, London EC1Y 8TZ.
For further information: contact Chris Brooks on 020-7040-5168 (email:
c.brooks@city.ac.uk); or RES Media Consultant Romesh Vaitilingam on 0117-983-9770 or
07768-661095 (email: romesh@compuserve.com).