When someone claims causality for one thing it might be useful to look at something closely related to understand if they move the same way to make it clear which among the possible choices would explain it.
March 23, 2011, 10:21 AM
The (Nature of) Shock Doctrine
Just a thought on market movements: you often see news stories confidently asserting that factor X was responsible for any given market movement. Serious observers rightly make fun of such stories; after all, how do they know? And it goes beyond funny to downright harmful when the stories assert supposed facts that actually serve some policy agenda. Lately we’ve seen that with regard to interest rates: again and again, upticks in US interest rates are asserted to be the result of debt fears, as evidence that the bond vigilantes have arrived.
But how can you place an interpretation on market movements? What’s the nature of the shock? One way to answer this question would be to ask the traders — although my sense is that really good traders have intuitions that generally go beyond their ability to articulate reasons. Another approach, however, is to look at more than one asset price; if you have a story about what’s driving one price, that story should make sense for the other, too.
Many years ago, when the Fed still claimed to pay attention to targets for monetary aggregates, there was an observable pattern: whenever the money supply came in higher than expected, interest rates rose. But there were two interpretations. Some people said that rates went up because markets believed rising money supplies heralded future inflation; others, that rates rose because people expected the Fed to tighten to get the money supply back on target.
Jeff Frankel cut through this debate by pointing out that we could look at what happened to the exchange rate. In fact, high money numbers were associated with a rise, not a fall, in the dollar — and this meant that fears of Fed tightening, not fears of inflation, were the real story.
via The (Nature of) Shock Doctrine – NYTimes.com.