January 23, 2011. In about 1976 I entered the graduate program in economics at U.C. Berkeley, where I had previously studied math and physics. Among economists there is talk of physics and the math of physics. The idea seemed to be that math was what made physics a "hard science," and that if you did hard math that would turn you into a hard scientists.

February 5, 2011. Physics math is hard, but that is not what makes physics a hard science. What makes it "hard" is results that describe the real world precisely. Some of those involve hard math and some do not. Measuring the speed of light, requires no hard math, neither did he Michelson–Morley experiment (1887), which showed that the speed of light is the same relative to a moving observer as it is to stationary observe. This led to special relativity.

But physics math does get hard as physics describes the world in ever greater precision. Suspiciously, economics math gets harder as economist abstract ever further from reality. This is because economists are doing math, not science. And they don't seem to know the difference.

In math you can make any set of consistent assumptions you want and then prove theorems about them. For elegant theorems you want elegant systems to analyze, and reality is not relevant. In economics we prove theorems about the perfection of economies with infinitesimal producers, "compete" markets for every kind of financial hedge, and consumers whose preferences cannot be affected by the market. These elegant assumptions make for fun math for those not smart enough to be real mathematicians.

February 5, 2011. Having missed undergrad economics, I was shocked  by graduate economics. Science it was not, and except when modified by "hard" and "soft" I never heard the word science cross the lips of any of these hard, or would-be-hard, "scientists." Never in four years. Not one mention of the scientific method. This is not to say there is no science in economics. There is quite a lot. But at its core, it really is not scientific.

Now that may sound harsh, so it's only fair to give some specifics.

Rationality is a core economic concept. Any scientist would first want it defined and then want to know to what extent economic actors are rational. So if economists were scientists, we should have no trouble finding the definition, and hopefully a set of reproducible and reproduces results showing the extent of rationality.

Astonishingly the word is used extensively with two completely different meanings, and I have never once seen this acknowledged. How can one investigate such a muddle scientifically? The two meanings are (1) consistent preferences, and (2) use of reason to make complex inferences.

A cat may have consistent preferences. Indeed Herbert Gintis insists that virtually all animals do. This definition of rationality is codified in a set of axioms that rigorously define consistency. A good start to a scientific discussion. But the most prominent use of the term is in the phrase "rational expectations." This is applied to a hypothesis (often misrepresented as a theory) that on average the public can calculate the distant implications of government policy for the macro economy, and their best private response to this outcome. In fact it is not clear that even macro economist can do that and it is certain that cats cannot—even if they are perfectly rational by the first definition and notoriously farsighted.

The problems of the economic view of rationality run much deeper, and we will return to them.

Consistent preferences, represented by utility functions, are fundamental to the theory of consumers and the optimality of competitive market outcomes.

China and India have nixed caps. Without these caps, Kyoto fails. What can be done?

Carbonomics explains "wrecking" the economy, "peak oil," caps, carbon taxes, and Kyoto.