prev next front |1 |2 |3 |4 |5 |6 |7 |8 |9 |review

Age group   Model  Intercept  Median  Median Country R2
                                         Share    Income  Dummy
Working age  US only   848   -21.71        -           -       .33
                    US only    838     -21.80     0.40      -        .34
                 US &CAN    826    -20.92      0.20      -67   .51
       w/dummy                                             

Elderly     US only         5255    -20.58       -          -     .01
            US only           5547     -18.03    -10.50    -     .03
       US &CAN            5490     -14.16   -11.20  -399  .16 w/dummy                                                       

All ages     US only      1110   -15.09         -         -      .13
                    US only   1141  -14.82      -1.10       -      .12
                 US &CAN  1127   -13.84   -1.30     -91    .34
w/dummy                                             

 

 

We then modeled the metropolitan area data. The modeling proceeded incrementally within the different age groups. First we looked at the U.S. states only with the median share of income as the explanatory variable. Then we added median income. We then added in the 53 Canadian metropolitan areas but when we did this we controlled, using a dummy variable, for the longstanding mortality differences between the U.S. and Canada.

The effect on income inequality on mortality is strongest for working age populations - a 1% increase in the share of household income to the poorer half of the income distribution is associated with a decline in the death rate of nearly 22 deaths per 100,000.