Avoid financial meltdown – use a thesaurus
So it’s not just investors who are guilty of moving in a herd-like fashion.
Financial journalists use the same verbs and nouns with greater frequency as stock markets overheat but display more variety in their phraseology after the bubble bursts, a study by Irish computer scientists has shown.
Trawling through nearly 18,000 on-line news articles that mention the Dow Jones, FTSE and Nikkei stock indices between 2006 and 2010, Aaron Gerow of Trinity College Dublin and Mark Keane of University College Dublin found that the language used by the writers had become more similar in the run-up to the global financial crisis.
“Meaningful regularities” in language employed before the crash showed “progressively greater agreement” in “positive perceptions of the market”.
Financial commentaries from The Financial Times, the New York Times and the BBC as well as news wire services such as Reuters, for instance, deployed increasingly similar noun-phrases as the market overheated, possibly reflecting a “narrowing of reporting to a relatively smaller number of key events/companies.”
The verbs “rise”, “fall”, “close” and “gain” were most popular through 2007 but their usage peaked the week of October 12 when the crash begins.
Gerow and Keane argue that this convergence of language can be used to identify stock market bubbles and supplement traditional volatility analyses.
“Current techniques for predicting bubbles rely on the analyses of price movements and volatility but are often explained (away) by claims about new valuation models…In the 2007 crash, the story was the low interest-rate environment, in the dot.com bubble it was the ‘new economy’ story,” they write.
Such language analysis offers an independent measure of “volatile thinking” in the market.
“When you look at the radical shifts towards a common, ‘irrational’ view of the market just before the 2007 crash you see a very strong signal that something is wrong. So the promise of the current work is that it provides a way to assess impending market events by looking at what people are saying (and presumably thinking),” they write.