Bargaining/Arbitration June 2023

Regressions comparing police to other city workers (2000-2017 pt change)

Regressions

Cut One: Police/Fire vs. Select Set (City FE, City-Clustered SE)

Set = Transit, Solid Waste Management, Sewage, Parks & Rec, libraries, and housing/community development

## 
## Call:
##    felm(formula = pct_change_salaries ~ police_fire | named_id |      0 | named_id, data = reg_1_data) 
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -0.8107 -0.0563  0.0000  0.0563  0.8107 
## 
## Coefficients:
##             Estimate Cluster s.e. t value Pr(>|t|)    
## police_fire  0.06017      0.00774    7.78  1.7e-14 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.185 on 1137 degrees of freedom
## Multiple R-squared(full model): 0.76   Adjusted R-squared: 0.457 
## Multiple R-squared(proj model): 0.0505   Adjusted R-squared: -1.15 
## F-statistic(full model, *iid*):2.51 on 1436 and 1137 DF, p-value: <2e-16 
## F-statistic(proj model): 60.5 on 1 and 1137 DF, p-value: 1.66e-14

Cut Two: Police/Fire vs. All but Police/Fire (City FE, City-Clustered SE)

## 
## Call:
##    felm(formula = pct_change_salaries ~ police_fire | named_id |      0 | named_id, data = reg_3_data) 
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -0.9171 -0.0492  0.0000  0.0492  0.9171 
## 
## Coefficients:
##             Estimate Cluster s.e. t value Pr(>|t|)    
## police_fire  0.04628      0.00624    7.41  2.3e-13 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.158 on 1287 degrees of freedom
## Multiple R-squared(full model): 0.822   Adjusted R-squared: 0.592 
## Multiple R-squared(proj model): 0.0409   Adjusted R-squared: -1.2 
## F-statistic(full model, *iid*):3.58 on 1661 and 1287 DF, p-value: <2e-16 
## F-statistic(proj model): 54.9 on 1 and 1287 DF, p-value: 2.27e-13

Regressions interacting with “implicit” arbitration (police right to strike prohibited vs. not)

Just Police/Fire Salaries - No City FE

## 
## Call:
##    felm(formula = pct_change_salaries ~ police_strike_2014 | 0 |      0 | 0, data = reg_1_data) 
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -0.9361 -0.1377 -0.0266  0.1231  1.5942 
## 
## Coefficients:
##                    Estimate Std. Error t value Pr(>|t|)    
## (Intercept)         0.19948    0.00702   28.42   <2e-16 ***
## police_strike_2014 -0.03837    0.01974   -1.94    0.052 .  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.236 on 1287 degrees of freedom
##   (5 observations deleted due to missingness)
## Multiple R-squared(full model): 0.00293   Adjusted R-squared: 0.00215 
## Multiple R-squared(proj model): 0.00293   Adjusted R-squared: 0.00215 
## F-statistic(full model):3.78 on 1 and 1287 DF, p-value: 0.0521 
## F-statistic(proj model): 3.78 on 1 and 1287 DF, p-value: 0.0521

Interaction

Cut One: Police/Fire vs. Select Set (City FE, City-Clustered SE)

Set = Transit, Solid Waste Management, Sewage, Parks & Rec, libraries, and housing/community development

## 
## Call:
##    felm(formula = pct_change_salaries ~ police_fire * police_strike_2014 |      named_id | 0 | named_id, data = reg_1_data) 
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -0.8116 -0.0569  0.0000  0.0569  0.8116 
## 
## Coefficients:
##                                Estimate Cluster s.e. t value Pr(>|t|)    
## police_fire                      0.0583       0.0084    6.94  6.4e-12 ***
## police_strike_2014                  NaN       0.0000     NaN      NaN    
## police_fire:police_strike_2014   0.0159       0.0216    0.73     0.46    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.185 on 1131 degrees of freedom
##   (12 observations deleted due to missingness)
## Multiple R-squared(full model): 0.76   Adjusted R-squared: 0.457 
## Multiple R-squared(proj model): 0.051   Adjusted R-squared: -1.15 
## F-statistic(full model, *iid*):2.51 on 1430 and 1131 DF, p-value: <2e-16 
## F-statistic(proj model): 20.7 on 3 and 1131 DF, p-value: 4.74e-13

Cut Two: Police/Fire vs. All but Police/Fire (City FE, City-Clustered SE)

## 
## Call:
##    felm(formula = pct_change_salaries ~ police_fire * police_strike_2014 |      named_id | 0 | named_id, data = reg_3_data) 
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -0.9169 -0.0495  0.0000  0.0495  0.9169 
## 
## Coefficients:
##                                Estimate Cluster s.e. t value Pr(>|t|)    
## police_fire                     0.04650      0.00675    6.89  8.9e-12 ***
## police_strike_2014                  NaN      0.00000     NaN      NaN    
## police_fire:police_strike_2014 -0.00107      0.01809   -0.06     0.95    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.159 on 1281 degrees of freedom
##   (12 observations deleted due to missingness)
## Multiple R-squared(full model): 0.822   Adjusted R-squared: 0.591 
## Multiple R-squared(proj model): 0.041   Adjusted R-squared: -1.2 
## F-statistic(full model, *iid*):3.57 on 1655 and 1281 DF, p-value: <2e-16 
## F-statistic(proj model): 18.3 on 3 and 1281 DF, p-value: 1.34e-11