Saturday, January 29, 2011

Durable Housing and Overall Trends in the Economy's Energy Efficiency

Now that Universities such as Texas A&M are ranking professors with respect to their "profitability", I expect that UCLA will embrace similar performance standards.  Since I believe that "incentives matter",  I turned off my television and I wrote a new paper with Dora L. Costa.  Here is the paper link.  I'm now considering launching an IPO of myself the day before Facebook goes public.  A free copy of this paper is available here.

In our new paper, we document a cool fact.  A fact that I didn't know and I doubt that even you knew it.

In the United States, homes consume about 30% of overall electricity generated.   At any point in time on your street, there will be some new homes, some homes built 5 years ago, 15 years ago, 35 years ago and even some built 90 years ago or longer. Housing is a long lived durable good. It lives longer than computers or cars.

So what? From a green perspective, if we build a "bad cohort" of homes say 1956 motel architecture, this cohort will be with us for a long time.

As a believer in "progress", I had assumed that homes built more recently always consume less energy than older homes. It is true that newer homes are larger and have more "modern" features but you can control for many of these factors using simple statistical techniques.

Dora Costa and I first document based on a large sample of California single family homes in calendar year 2000 that homes built in the 1970s have higher electricity consumption than homes built in the same geographic communities built more recently and homes built in the 1960s.

In the more interesting second step of the project,  we have invested the effort to track down the electricity price in the home's vicinity at the time it was built.  So, a home built in Westwood in 1968 would have faced a different "birth price" than a home built in Westwood in 1982 because LADWP changed the price of power. Similarly, a home built in Westwood in 1968 would face a different "birth price" for power than a home in Berkeley also built in 1968 because LADWP  and PGE were charging different prices for power at the same point in time.

Dora and I exploit these two types of pricing variation to document that homes built when power is cheap  are less energy efficient today!   Note that "today" can be 30 years after the home was built.  Dora was interested in this topic because of her work on cohort studies.  Housing at any point in time represents a set of birth cohorts (1 year old homes, 55 year old homes) all on the street. The average home can become more energy efficient if new homes are more energy efficient and  a growing share of the total (think of China today) or if we scrap the older inefficient vintages.

Our work has a similar flavor as the induced innovation literature.  During times of high prices, it is rational to substitute away from the scarce input (this lowers your costs).  When prices are low, why conserve?

Given that homes live longer than cars, it is more important to document this effect for homes because past mistakes will live on the longest.