Table 6

Estimated coefficients of the ordinary least squares (OLS), instrumental variable (IV) and fixed effects (FE) regressions of relative income price (RIP) of cigarettes in Bangladesh, 2009–2015

OLSIVFE
CoefficientRobust SEP valuesCoefficientRobust SEP valuesCoefficientSEP values
Tax0.430.030.000.560.090.000.340.020.00
Age−0.030.010.00−0.030.010.00
Female−1.370.910.14−1.150.920.21
Years of education
 1–8−0.390.380.30−0.160.410.70−0.810.580.16
 9 or more−2.260.500.00−1.960.540.00−1.350.760.08
Occupation
 Self-employed in non-farm agriculture−1.100.840.19−1.550.860.07−0.040.910.97
 Self-employed in non-agricultural activity−0.770.380.04−1.550.400.00−0.500.510.33
 Farm wage labourer1.540.500.002.110.530.000.800.610.19
 Non-farm agricultural wage labourer0.130.740.86−0.790.700.260.440.940.64
 Non-agricultural wage labourer1.430.490.000.740.480.121.310.540.02
 Professional−1.730.980.08−1.331.140.24−0.931.200.44
 Managerial, administrative or clerking−1.260.680.06−2.540.760.00−0.190.760.81
 Student2.541.370.062.371.560.132.661.120.02
 Unemployed1.270.680.060.430.730.560.660.700.35
 Housewife/housekeeper/household
 manager
1.370.980.160.961.060.370.991.100.37
 Others0.410.430.34−0.080.430.860.000.480.99
Socioeconomic status
 Moderate−0.590.370.110.030.390.95
 High−2.670.360.00−2.090.380.00
Rural area of residence2.392.100.262.760.370.00
RIP
Cigarette brands
 Medium−0.990.530.06−3.251.750.060.170.470.72
 High−2.311.290.07−9.414.240.033.420.880.00
 Premium−10.052.340.00−24.507.940.001.501.510.32
Trend
 Wave 2−1.290.340.00−1.140.350.00−1.520.270.00
 Wave 3−2.480.420.00−3.390.840.00−2.880.350.00
 Wave 4−7.110.460.00−8.901.290.00−7.220.390.00
Constant12.122.690.007.970.770.007.690.700.00
 Number of observations588258825882
 R2 0.540.470.35
  • Reference categories include male, persons with 0 year of education, owner/tenant farmers, low socioeconomic status, urban area of residence, low cigarette brands and Wave 1 (2009).

  • The SEs of OLS estimates are adjusted for autocorrelation of error terms of multiple observations on the same individual. The village effects in OLS regression are suppressed for the brevity of presentation.

  • The village variable is used as an instrument for tax in the IV estimation. The test statistics for exogeneity of the tax variable using the Durbin-Wu-Hausman test is statistically insignificant (robust regression F(1, 2552)=2.18, p=0.1403) suggesting exogeneity. The adjusted R2 from the first-stage regression of tax is 0.9021, suggesting strong predictability of variations in tax using village-level variation.

  • The age variable drops out of the FE regression due to perfect collinearity with the wave variables. Gender, socioeconomic status and area of residence variables are time invariant and hence drop out of the FE regression as well.