st: Simultaneous equation with no endogeneity?

[Natascha Nisic <natascha.nisic@soz.unibe.ch>]

After testing for endogeneity in a simultanous equation framework, I found
no endogeneity in the data in both equations. But what does this mean 
theoretically?

I examined the endogeneous relationship of income and commuting to the 
workplace, supposing that longer distances of commuting lead to higher 
income (because of bigger access to labour markets), whereas higher 
income makes commuting affordable. So I had two structural equations: 
one for income and one for commuting distance. When  estimating these 
two equations separately by OLS I get a highly significant positive 
effect of income on commuting and of commuting on income. Then I used 
IV-Regressions to estimate the simultaneous equation model. But when 
testing for endogeneity the results of the test show that there is no 
endogeneity in both equations. Moreover neither of the two instrumented 
variables has an effect anymore in the IV-Regression.

I dont't know how to interpret these results, as it seems that 
empirically there is no endogeneous relationship, but theoretically 
there is? (Of course my instrumental variables might be bad, but 
supposed they are not, what do these results mean?)

Thanks a lot.
Natasha



*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/