4.2.2 Regression Analysis
Table 2 presents the ‘regional’ regression coefficients from LAV regressions of the 10th, 25th, 50th (median), 75th, and 90th percentile of the Real Hourly Wage Distribution for Wage/Salary Workers. In the left panels, the coefficients indicate the average relative difference in Real Hourly Wages for Wage/Salary Workers at various points in the distribution over the entire sample period in each region relative to Auckland. In the right panels, the coefficients indicate the change in the relative difference in Real Hourly Wages for Wage/Salary Workers at various points in the distribution between 1997/98 and 2003/2004 in each comparison area relative to Auckland. The top panels present the “unadjusted” regression coefficients. Again, these are identical to the differences graphed in the previous figures averaged over entire sample period. The bottom panels present the “adjusted” regression coefficients controlling for all covariates. Again, these coefficients allow us to examine whether controlling for differences in the attributes of individuals living in different areas ‘explains’ any of the differences between these areas.
Adjusting for characteristics explains all of the difference in wage/salary wages between Auckland and Wellington at all points in the distribution besides the median, but almost none of the difference in wages between Auckland and the other comparison areas at any point of the distribution below the 90th percentile. Adjusting for characteristics does explain 16% of the difference in 90th percentile wages between Auckland and Christchurch, 30% between Auckland and Other Urban, and 47% between Auckland and Rural. While Auckland has faster unadjusted wage/salary wage growth across the entire wage distribution than all comparison areas, the magnitude of these differences are fairly small and insignificant below the median, and differences in characteristics explain most of the (larger) differences between Auckland and the other areas above the median wage. Overall, adjusted wage/salary wages have grown in Auckland by an additional $0.25-$0.60 at the 25th percentile, $0.40-$0.90 at the median, $0.40-$0.70 at the 75th percentile, and $0.70-$1.55 at the 90th percentile compared to the other areas.
| Mean | 10th Pctile | 25th Pctile | Median | 75th Pctile | 90th Pctile | Mean | 10th Pctile | 25th Pctile | Median | 75th Pctile | 90th Pctile | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| v. Auckland | Unadjusted 1997-2004 | Unadjusted Growth Between 1997/98 and 2003/04 | ||||||||||
| Wellington | 0.84 | 0.45** | 0.65* | 1.03** | 1.86** | 2.47** | -2.40 | -0.19 | -0.84 | -1.05 | -1.85 | -3.65 |
| (0.43) | (0.13) | (0.26) | (0.32) | (0.51) | (0.93) | (1.61) | (0.40) | (0.56) | (0.70) | (1.27) | (2.21) | |
| Christchurch | -2.53** | -0.34** | -0.69** | -1.00** | -1.70** | -3.39** | -1.46 | -0.37 | -0.46 | 0.05 | -0.93 | -2.97 |
| (0.38) | (0.11) | (0.23) | (0.27) | (0.39) | (0.79) | (1.49) | (0.21) | (0.54) | (0.77) | (1.78) | (1.82) | |
| Other Urban | -3.22** | -0.57** | -1.08** | -1.53** | -2.45** | -4.33** | -1.54 | -0.12 | -0.14 | -0.05 | -1.45 | -4.42** |
| (0.30) | (0.08) | (0.18) | (0.15) | (0.28) | (0.62) | (1.24) | (0.20) | (0.37) | (0.64) | (0.95) | (1.48) | |
| Rural | -3.74** | -0.78** | -1.38** | -2.23** | -3.08** | -5.06** | -2.13 | -0.16 | -0.32 | -0.66 | -1.73* | -5.30** |
| (0.30) | (0.10) | (0.18) | (0.18) | (0.29) | (0.61) | (1.29) | (0.30) | (0.33) | (0.51) | (0.88) | (1.41) | |
| Observations | 74,395 | 74,395 | 74,395 | 74,395 | 74,395 | 74,395 | 74,395 | 74,395 | 74,395 | 74,395 | 74,395 | 74,395 |
| Pseudo R-SQ | 0.03 | 0.01 | 0.01 | 0.01 | 0.02 | 0.02 | 0.03 | 0.01 | 0.01 | 0.01 | 0.02 | 0.02 |
| v. Auckland | Adjusted For All Covariates 1997-2004 | Adjusted Growth Between 1997/98 and 2003/04 | ||||||||||
| Wellington | -0.42 | -0.02 | -0.10 | -0.31** | -0.26 | -0.45 | -1.36 | -0.28 | -0.62** | -0.87** | -0.72 | -0.70 |
| (0.23) | (0.08) | (0.07) | (0.11) | (0.24) | (0.33) | (0.73) | (0.33) | (0.21) | (0.31) | (0.73) | (0.79) | |
| Christchurch | -2.58** | -0.74** | -0.98** | -1.59** | -2.05** | -2.86** | -1.67* | -0.25 | -0.41* | -0.76** | -0.59 | -1.30 |
| (0.21) | (0.12) | (0.09) | (0.08) | (0.16) | (0.33) | (0.77) | (0.26) | (0.19) | (0.23) | (0.43) | (0.66) | |
| Other Urban | -2.60** | -0.83** | -1.10** | -1.64** | -2.14** | -3.04** | -1.46* | -0.08 | -0.23 | -0.53* | -0.69* | -1.55* |
| (0.16) | (0.09) | (0.07) | (0.07) | (0.12) | (0.23) | (0.66) | (0.15) | (0.16) | (0.22) | (0.30) | (0.63) | |
| Rural | -2.58** | -0.97** | -1.19** | -1.71** | -2.15** | -2.67** | -1.48* | -0.14 | -0.27** | -0.41 | -0.38 | -1.32 |
| (0.16) | (0.07) | (0.07) | (0.08) | (0.13) | (0.23) | (0.72) | (0.19) | (0.10) | (0.25) | (0.45) | (0.68) | |
| Observations | 74,395 | 74,395 | 74,395 | 74,395 | 74,395 | 74,395 | 74,395 | 74,395 | 74,395 | 74,395 | 74,395 | 74,395 |
| Pseudo R-SQ | 0.26 | 0.11 | 0.16 | 0.21 | 0.23 | 0.24 | 0.27 | 0.11 | 0.17 | 0.21 | 0.23 | 0.24 |
Notes: All amounts are in June 2004 Dollars. Bootstrapped standard errors which account for clustering at the PSU level are in parentheses. Coefficients with * are significant at the 5% level and those with ** at the 1% level.
| Mean | 10th Pctile | 25th Pctile | Median | 75th Pctile | 90th Pctile | Mean | 10th Pctile | 25th Pctile | Median | 75th Pctile | 90th Pctile | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| v. Auckland | Unadjusted 1997-2004 (June 2004 Dollars) | Unadjusted Growth Between 1997/98 and 2003/04 (June 2004 Dollars) | ||||||||||
| Wellington | 0.61 | 0.31* | 0.55* | 0.88** | 1.42* | 2.36 | -1.91 | 0.05 | -0.63 | -1.04 | -2.08 | -1.30 |
| (0.51) | (0.13) | (0.22) | (0.32) | (0.59) | (1.23) | (1.90) | (0.30) | (0.49) | (0.79) | (1.76) | (3.22) | |
| Christchurch | -2.74** | -0.49** | -0.87** | -1.23** | -1.95** | -3.66** | -0.77 | 0.41 | 0.47 | 0.35 | -0.50 | -0.62 |
| (0.46) | (0.08) | (0.19) | (0.33) | (0.43) | (1.04) | (1.64) | (0.49) | (0.42) | (0.82) | (1.27) | (2.43) | |
| Other Urban | -3.41** | -0.69** | -1.14** | -1.72** | -2.90** | -4.82** | -1.49 | 0.26 | 0.18 | 0.08 | -1.04 | -2.54 |
| (0.34) | (0.09) | (0.16) | (0.17) | (0.36) | (0.77) | (1.42) | (0.31) | (0.46) | (0.58) | (1.01) | (1.75) | |
| Rural | -2.95** | -1.60** | -1.83** | -2.32** | -3.01** | -3.14** | -0.39 | 0.71* | 0.34 | -0.22 | -0.92 | -0.83 |
| (0.35) | (0.06) | (0.15) | (0.19) | (0.39) | (0.76) | (1.42) | (0.31) | (0.44) | (0.58) | (1.20) | (2.07) | |
| Observations | 84,875 | 84,875 | 84,875 | 84,875 | 84,875 | 84,875 | 84,875 | 84,875 | 84,875 | 84,875 | 84,875 | 84,875 |
| Pseudo R-SQ | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 | 0.01 |
| Adjusted For All Covariates 1997-2004 (June 2004 Dollars) | Adjusted Growth Between 1997/98 and 2003/04 (June 2004 Dollars) | |||||||||||
| Wellington | -0.38 | 0.01 | -0.24* | -0.34* | -0.46* | -0.70 | -1.17 | -0.34 | -0.59* | -0.74** | -0.40 | 0.27 |
| (0.28) | (0.06) | (0.11) | (0.17) | (0.20) | (0.40) | (1.03) | (0.24) | (0.30) | (0.27) | (0.86) | (0.76) | |
| Christchurch | -2.50** | -0.93** | -1.16** | -1.73** | -2.17** | -3.29** | -1.17 | 0.20 | -0.23 | -0.28 | 0.31 | 0.24 |
| (0.27) | (0.15) | (0.10) | (0.10) | (0.27) | (0.45) | (0.99) | (0.31) | (0.24) | (0.24) | (0.56) | (1.49) | |
| Other Urban | -2.60** | -0.94** | -1.25** | -1.80** | -2.26** | -3.34** | -1.36 | 0.14 | -0.13 | -0.18 | -0.36 | -0.76 |
| (0.19) | (0.11) | (0.08) | (0.05) | (0.17) | (0.06) | (0.85) | (0.23) | (0.15) | (0.34) | (0.26) | (0.69) | |
| Rural | -2.39** | -1.12** | -1.39** | -1.85** | -2.25** | -2.90** | 0.28 | 0.25 | -0.05 | -0.11 | 0.08 | 0.10 |
| (0.22) | (0.07) | (0.12) | (0.09) | (0.26) | (0.25) | (0.85) | (0.29) | (0.17) | (0.21) | (0.44) | (0.93) | |
| Observations | 84,875 | 84,875 | 84,875 | 84,875 | 84,875 | 84,875 | 84,875 | 84,875 | 84,875 | 84,875 | 84,875 | 84,875 |
| Pseudo R-SQ | 0.14 | 0.08 | 0.10 | 0.13 | 0.14 | 0.17 | 0.14 | 0.08 | 0.10 | 0.13 | 0.14 | 0.17 |
Notes: All amounts are in June 2004 Dollars. Bootstrapped standard errors which account for clustering at the PSU level are in parentheses. Coefficients with * are significant at the 5% level and those with ** at the 1% level.
Table 3 presents the ‘regional’ regression coefficients from LAV regressions of the 10th, 25th, 50th (median), 75th, and 90th percentile of the Real Hourly Wage Distribution for All Workers. The layout of this table is identical to the previous table. Once again adjusting for characteristics, the results are very similar to those found using only wage/salary wages, and controlling for characteristics explains almost all of this difference between Auckland and Rural at the bottom and top of the all-workers versus wage/salary worker wage distribution. Including the self-employed, Auckland has similar unadjusted wage growth as the comparison areas at and below the median, and faster, but insignificantly different, wage growth at the upper-end of the distribution. Adjusting for characteristics also explains most of difference found between wage growth in Auckland and the comparison areas.
