April 12th, 2011

I don’t spend all of my time blogging.  I teach and I try to write some stuff.  This two pager    summarizes my applied micro economics research on climate change mitigation and adaptation.

Share this post:
  • Twitter
  • StumbleUpon
  • Digg
  • Reddit
  • Facebook

4 Responses to “A Two Page Summary of My Work on the Economics of Climate Change”

  1. paul says:

    Did your analysis of used vehicles exported under NAFTA also look at the carbon footprint of the transportation modes that the exported vehicles replaced?

  2. James Wimberley says:

    From the paper: “In the absence of formal Pigouvian carbon pricing, households have no financial incentive to economize on their production of greenhouse gas emissions…”

    Not quite. Buying a durable energy-consuming good like a house or car involves taking some view on future energy costs and security of supply. It’s a reasonable bet that the price of fossil fuels will rise, quite possibly sharply: either by a carbon tax, or production constraints, or both. Ideology may be strongly correlated with such predictions, but logically they are distinct. Have you tried disentangling them?

    One test would be cars. The worries I mention apply much more strongly to gasoline than to natural gas or coal. To the extent that greens are differentially buying Priuses - saving gasoline - rather than Passivhäuser - saving natural gas and (via electricity) coal -, this would suggest that economic predictions are significant.

    I agree that some green decisions, like installing rooftop solar PV panels now, can’t be explained by economic calculations: these would (in the USA) always lead to waiting till the price comes down more thanks to the Germans and Chinese.

  3. paul says:

    James: I think your caveat loses some of its force because the pricing changes you’re talking about are going to be for energy in general (thanks to market equilibria) rather than for carbon-spending energy in particular. (Then again, in the absence of appropriate regulation, that will be true for a world with carbon taxes as well.)

    Meanwhile, I’m not sure about the disentangling test you propose. In general, it’s easier to ditch a low-mileage car than it is to ditch a high-energy-cost house, so the risk-averse would look more at options with longer-term implications. One the other hands, the market for houses is not that segmentable by energy consumption, so the signal will be noisy; houses can also be retrofitted with insultion or with new energy sources rather more easily than cars can have their engines and powertrains switched out.

  4. James Wimberley says:

    Paul: How can hybrid cars not become much more attractive as energy prices rise? Suppose all electricity were fossil-generated (incidentally removing much of the carbon advantage, but it’s only a thought experiment). Let fossil fuel prices double. The marginal cost per km of electric motoring also doubles - but it’s a quarter of the cost of gasoline. The gap widens as the electricity mix includes more renewables, which is a sure thing over the life of a car. Electric cars and hybrids are only pricey to own because of the battery; an up-front fixed investment cost. The price of batteries must be pretty insensitive to energy costs.

    I take your point about the noisy housing market and the general possibility of incremental improvements. But a true Passivhaus isn’t something you can achieve by retrofitting - for one thing it involves passing all ventilation through heat exchangers. So it’s a lumpy investment like a hybrid. If such houses are rare in Berkeley, and Priuses common, isn’t that evidence for my expectations theory?