SEARCH

SEARCH BY CITATION

Paik M., Black B. S., Hyman D. A., and Silver C. (2012) ‘Will Tort Reform Bend the Cost Curve? Evidence from Texas’, Journal of Empirical Legal Studies, 9, 173216.

Table 7. DiD Results: Top- Versus Bottom-Risk Counties
Dependent VariableLn(relative spending)
Geographic AreaCounty
(1)(2)(3)(4)(5)
Medicare Spending AreaTotalPart APart BMedical and SurgicalImaging and Lab
  1. Note: Difference-in-differences regressions, with county fixed effects, of ln(relative Medicare spending) over 1999–2009 on postreform dummy (= 1 for 2004 and later), high-risk dummy (= 1 for 25 percent of Texas counties with highest long-term med mal risk), interaction between these variables, year-and-after dummies, ln(population), and constant term. Variables are defined in Table 2. Control group is 65 “bottom-risk” counties with zero paid claims over 1999–2003. Middle-risk counties are dropped. Amounts in 2007$. t statistics, with standard errors clustered on county, in parentheses. *, **, and *** indicate significance at the 10 percent, 5 percent, and 1 percent levels. Significant results (at 5 percent or better) are in boldface.

Postreform dummy * High-risk dummy0.01460.02020.00860.0133−0.0120
[0.98][0.91][0.53][0.53][0.50]
95% confidence interval[−0.015, −0.044][−0.023, −0.064][−0.023, −0.041][−0.037, −0.063][−0.060, −0.036]
Year-or-after dummies, Ln(population), constantyesyesyesyesyes
Observations1,4301,4301,4301,1381,138
No. of counties130130130127127
R20.59470.41910.70000.72380.7417

In Table 7 above, the values in columns 1 to 5 in the row called “95% confidence interval” are erroneous.

The values have been corrected online on 13th September 2013 and shown below:

Table 7. DiD Results: Top- Versus Bottom-Risk Counties
Dependent VariableLn(relative spending)
Geographic AreaCounty
(1)(2)(3)(4)(5)
Medicare Spending AreaTotalPart APart BMedical and SurgicalImaging and Lab
  1. Note: Difference-in-differences regressions, with county fixed effects, of ln(relative Medicare spending) over 1999–2009 on postreform dummy (= 1 for 2004 and later), high-risk dummy (= 1 for 25 percent of Texas counties with highest long-term med mal risk), interaction between these variables, year-and-after dummies, ln(population), and constant term. Variables are defined in Table 2. Control group is 65 “bottom-risk” counties with zero paid claims over 1999–2003. Middle-risk counties are dropped. Amounts in 2007$. t statistics, with standard errors clustered on county, in parentheses. *, **, and *** indicate significance at the 10 percent, 5 percent, and 1 percent levels. Significant results (at 5 percent or better) are in boldface.

Postreform dummy * High-risk dummy0.01460.02020.00860.0133−0.0120
[0.98][0.91][0.53][0.53][0.50]
95% confidence interval[−0.0158, 0.044][−0.024, 0.064][−0.023, 0.041][−0.037, 0.063][−0.060, 0.036]
Year-or-after dummies, Ln(population), constantyesyesyesyesyes
Observations1,4301,4301,4301,1381,138
No. of counties130130130127127
R20.59470.41910.70000.72380.7417

We regret any inconvenience caused by these errors.