Table 2

Program Effects on Per Capita Solid Waste Generation in New Hampshire

OLS (Robust Errors)Conventional Treatment Effects Modela2SLSa2SLSb
Constant0.777 (0.589)1.429* (0.813)1.442* (0.834)1.447 (0.832)
PAYT- 0.566*** (0.140)- 1.132** (0.459)- 1.143** (0.486)- 1.137** (0.498)
CurbCycl0.133 (0.111)0.140(0.121)0.141 (0.121)0.164 (0.460)
CurbPkUp0.013 (0.129)0.029 (0.140)0.029 (0.141)0.019 (0.244)
MandRec- 0.188* (0.100)- 0.209* (0.111)- 0.210* (0.111)- 0.212* (0.118)
AvgHSize- 0.722*** (0.234)- 0.908*** (0.279)- 0.912*** (0.284)- 0.915*** (0.286)
College- 0.004 (0.005)- 0.002 (0.005)- 0.002 (0.006)- 0.002 (0.006)
PCInc0.018 (0.018)0.009 (0.017)0.008 (0.018)0.008 (0.017)
X_PAYT0.348 (0.266)
R20.1270.135
Estimated % f in43.7770.9871.6771.66
PCMSW with
PAYT
  • Note: Standard errors are in parentheses. The estimator proposed by Kennedy (1981) is employed to compute the estimated percent change in PCMSW due to the discrete change in PAYT from 0 to 1: 100*(eEmbedded Image-0.5*Embedded ImageEmbedded Image - 1), where Embedded Image is the estimated coefficient of PAYT and Embedded Image(Embedded Image) is the estimated variance. Dependent variable: log(PCMSW). OLS, ordinary least squares; 2SLS, two-stage least squares.

  • a Endogenous PAYT.

  • b Endogenous PAYTand CurbCycl.

  • * Significant at the 0.1 level;

  • ** significant at the 0.05 level;

  • *** significant at the 0.01 level.