is, well, conventional. Professional economists, not wanting to be subject to ridicule by their peers, will look at existing trends and come to the conventional conclusion that these trends will continue. What else could they do?

Economies have as many variables and unknowns as the weather. I'd like to know how many weather forecasters predicted the existence and track of hurricane Katrina a month before it struck Mississippi and Louisiana.

What we're left with is SWAG -- scientific, wild-ass guesses.

Your comparison of economic forecasting with weather forecasting hits the nail on the head. Both the economy and the weather are very complex dynamic systems described by nonlinear equations. The nonlinearity of these systems gives rise to highly destabilizing positive feedback loops.

In weather systems this is called the "butterfly effect". A random breeze off the coast of west Africa can give rise to a hurricane. In economies the most destabilizing positive feedback is introduced by the interaction of human psychology with free markets.

In classical free market theory, price equilibrium (and economic stability) is achieved through the dynamic interaction of supply and demand. The real world is not that simple. Human greed inflates economic bubbles and human fear deflates those bubbles.

So who could have predicted the outbreak of tulip mania in 17th century Holland? <link> Or that "sophisticated" Wall Street investors would fall in love with the idea of buying junk mortgages? The fallacy in using economic models for forecasting boils down to the assumption that humans in general are as rational as the builders of the economic models. Well, it simply ain't so.

Right on! You would think that many people would have been smart enough to learn from the tulip mania fiasco, but no.

Revrick: Economies have as many variables and unknowns as the weather.

LeRoy: Rather more, I would say. Many, many, many more, to be a bit more 'precise'. After all, weather is one of the variables that affect economies....so, in order to make accurate economic forecasts, economists would also (or first) have to be able to make accurate weather forecasts! They would also have to be able to predict political revolutions, military coups, assassinations, industrial accidents and oil spills, scientific and/or technological breakthroughs, non-weather related natural disasters (such as earthquakes, volcanos, tsunamis)...and be able to fit all of that into a model, a mathematical model, that would incorporate all these variables. Whew! Good luck with that!!

[All of this, by the way, is why posters like Sovereign are being beyond ridiculous when they demand that economists tell them exactly how much the housing implosion is going to affect the rest of the economy...]