Is Economics Advancing? - Considering Expertise

I’ve been thinking a lot about the problem of expertise. Alvin Goldman, a philosopher at Rutgers, has written some interesting papers on the interaction between novices and experts. Next week I’m going to be leading a discussion on the topic.

I restarted my research on this topic after participating in some other recent discussions about economics and the reactions to the recent recession. How can economists propose such dramatically different explanations and remedies to the current crisis? The result is Paul Krugman asking How did economists get it so wrong?

Chris Dillow posts some interesting thoughts on whether economics is more certain about it’s conclusions now than it was 80 years ago when John Maynard Keynes was writing his General Theory.

To answer the question Dillow looks at two recent papers by economists about the rationality of markets. In comparison to Keynes the contemporary papers are more advanced, referencing more empirical data and having greater theoretical clarity. But they reach opposite conclusions by studying the same data, one finding evidence of rationality and the other finding evidence of irrationality.

Keynes wrote without any reference to empirical data. So which case is better? Someone of prestigious economic intelligence writing without any empirical data or two contemporary economists analyzing the data but reaching different conclusions.

This suggests that, if it is firm beliefs you want, economics regresses. Reading Keynes, you’d infer clearly that markets were somehow not rational. Reading the later papers you wouldn’t know whether they were or not. 70 years of advances in economics has merely generated doubt.

This kind of situation puts the novice economic observer into a major pickle. The novice must rely on the experts to analyze the information and data because the novice lacks the knowledge to even begin the analysis. But after the experts finish the conclusions are contradictory. Is there a way for the novice to resolve this problem?