n on having online access. If theonly people with online access in a technologically “backward” locations (that also tend to havelow tax rates) are the diehard Internet users who frequently buy, this will bias the coefficientdownward.To deal with the unobserved city characteristics, I first present (columns 3 and 4) the sameprobit and Tobit regressions but restrict the sample to include only the top 30 metropolitan areaswith the idea that these may be more comparable locations and will help eliminate any bias causedfrom comparison to rural and small town locations. This restriction reduces the sample by moreഊ10than 40%. The coefficients on the sales tax, however, are not significantly different than in thefull sample and the point estimates are larger. They still show that taxes appear to have asignificant effect on Internet purchases. The elasticities here are 2.9 on buying and 3.4 onspending. Obviously this does not rule out the spurious correlation, but does suggest thatwhatever the correlation between technological sophistication and tax rates, it must be just as trueamong the top 30 cities as it is between large metropolitan areas and more rural areas.Next, in table 3, I use the variation in tax rates across geographic areas to further narrowthe comparison groups. Column 1 and 2 include region dummies and asks whether individualswith the same observable characteristics and living in the same region are more likely to buyonline in a city with a higher tax rate. In both the probit and Tobit regressions, taxes have asignificant impact within regions. The elasticities at the mean are 2.4 and 3.5. Column 3 and 4include state dummies and ask whether people with the same observables living in the same stateare more likely to buy online where the sales tax is higher. Here the coefficients are particularlylarge and significant suggesting that taxes matter within a given state. The elasticities at the meanare 8.4 and 11.7.M...
