This is the fundamental insight of the Austrian school of economics. Economists had fixated on finding predictable mathematical relationships between observabl

This is the fundamental insight of the Austrian school of economics. Economists had fixated on finding predictable mathematical relationships between observable macroeconomic variables: Laffer curve, Phillips curve, etc. Mises said that ultimately the economy can not be modeled from data because human behavior is too dynamic. History has proven him correct. The econometric discoveries once heralded as potential "laws" of macroeconomics have all faired poorly against the data over time. His own insights derived from praxeology have held up well. This failure of econometrics is at the root of the current crisis. All the model based derivative valuations and risk management tools worked ... until they didn't. Same thing with the LTCM crisis. You just can't substitute regression analysis for thinking.