Some years ago, I wrote a paper that showed how housing construction in the United States led the business cycle. The paper included a large number of econometric specifications (detrended levels, first differences, error correction, etc.), and the finding was robust to all specifications: residential investment Granger caused GDP.
For the Jackson Hole conference two years ago, Ed Leamer did a similar exercise, and found that the results held up: indeed the title of his paper was "Housing is the Business Cycle." An interesting question, then, is whether the result holds for other countries.
For at least one other, it appears not to. One student of mine from Wisconsin, Shampa Bhattacharya, performed the exercise for India, and could not find a relationship between housing and GDP. Another student of mine from India School of Business, Katyayini Krishnamoorthy, updated Shampa's work with fresh data and found the same thing.
The difference in results may reflect the fact that Indian data has lower frequency than US data (it is annual instead of quarterly), but it was striking how they got absolutely nothing. Among other things, it means that in India housing is not an effective transmission mechanism for monetary policy. This may have something to do with the fact that housing finance in India is not well developed--a fact that may have kept that economy out of trouble over the past few years. I will be posting more about this soon.
Tuesday, February 09, 2010
Subscribe to: Post Comments (Atom)
Thank you for listing the difference and comparing. Your blog is awesome for giving out infos.
I'm trying to get in touch with the sites admin, can you drop me an e-mail to firstname.lastname@example.org as I'd like to talk through some things.
Thank you for your time, Jamie
Great reviews on your site. I have read out the whole blog with full interest and found it to be both very informative and well-updated. Thanks for sharing... Houses for Sale Manchester.
Your review of the site is great.
Urban economics blog is very important.
Post a Comment