As if on cue, an article published in the American Banker today really drives home my point about how difficult it is to predict the future.  There’s so much uncertainty down the road that trying to base an earnings-at-risk projection off of a two-year forecast is no better than throwing a dart.

The article is about how banks are now in the awkward position of discouraging clients from making new deposits (emphasis in passage below is mine):

In Cash Glut, Banks Try to Discourage New Deposits

With attractive lending opportunities hard to come by, bankers are finding themselves doing what would have been unthinkable just two years ago: discouraging deposits.

American Banker  |  Tuesday, July 27, 2010

They said it, not me – “unthinkable two years ago”.  See my prior post for a list of other “unthinkable” things that have happened in the last two years.  I hope you can see my point that a two-year forecast time horizon for earnings-at-risk is just too long.  We’re not that good at forecasting.  We also already have a measurement for looking at long-term interest rate risk: Economic Value of Equity (EVE) at risk.

Let me put the question to you this way.  If a two-year forecast is so good, then why not a three-year or five-year forecast?  The only realistic answer I’ve heard is that three or five years is too long because “the future is so unpredictable”…but we’re better at predicting two-years ahead?  Unthinkable.