Going The Distance: How Travel Behavior Influences Climate-Related Policy Preferences in Washington State

1 Introduction

This research explores the relationship between household travel behavior and voter preferences for climate-related fiscal policy in Washington State. We examine how variations in travel patterns might influence public support for specific climate change change-related fiscal policies such as carbon taxes or “cap and trade” emission trading systems. The analysis builds a harmonized, census tract-scale data set and uses several statistical models to explore the relationship between these variables.

2 Data and Methods

2.1 Data Opertionalization

• Travel Behavior: Represented by average daily vehicle miles traveled (VMT) per household, sourced from the US Department of Transportation’s Local Characteristics for Households dataset. This metric reflects household mobility patterns.
• Voter Preferences for Climate-Related Taxes: Quantified through the results of the Initiative 732 vote in 2016, which proposed a carbon tax aims at reducing greenhouse gas emissions. This serves as a direct measure of voter support for climate-related taxation.
• Political Partisanship: Operationalized using the 2016 presidential election voting results, indicating the political values of voters, which may influence their support for environment- and/or taxation-related policy.

2.2 Model Descriptions

2.2.1 Univariate Linear Model

The univariate Ordinary Least Squares linear model examines the direct relationship between voter support for the climate-related tax and household travel behavior. This model is articulated through the following linear equation:

\[ y_i = \beta_0 + \beta_1x_{1i} + \epsilon_i \]

In this equation:

• \(y_i\) represents the response variable, specifically the share of ‘No’ votes on Initiative 732.
• \(\beta_0\) is the y-axis intercept, indicating the baseline level of opposition to the initiative when average daily VMT per household is zero.
• \(\beta_1x_{1i}\) is the coefficient for the explanatory variable, average daily VMT per household, which quantifies the change in the proportion of ‘No’ votes as VMT varies.
• \(\epsilon_i\) denotes the random error term, accounting for the variation in ‘No’ votes not explained by travel behavior.

Here, \(y_i\) represents the response variable (share of ‘No’ votes on Initiative 732), \(\beta_0\) is the model’s y-axis intercept, \(\beta_1x_{1i}\) is the coefficient of the explanatory variable (average daily VMT per household), and \(\epsilon_i\) represents the random error term.

This model establishes a baseline for identifying potential correlations between travel behavior and voter preferences regarding climate-related taxes, without considering any other confounding factors. It provides a straightforward way to assess the primary effect of travel on voting behavior before introducing more complexity into the analysis.

2.2.2 Multivariate Linear Model

Expanding upon the univariate linear model, the multivariate linear model incorporates an additional explanatory variable: political partisanship. This model is specified through the following equation:

\[ y_i = \beta_0 + \beta_1x_{1i} + \beta_2x_{2i} + \epsilon_i \]

In this equation, \(\beta_2x_{2i}\) is the coefficient of the additional explanatory variable (share of votes for the Republican presidential candidate).

This model evaluates how both travel behavior and political orientation together affect support for climate-related taxes.

2.2.3 Spatial Lag Model

The Spatial Lag Model refines the multivariate Extended Linear Model by including a spatial lag variable that accounts for “spillover effect” (i.e., spatial autocorrelation of model residuals) of the multivariate linear model’s dependent variable.

\[ y_i = \beta_0 + \beta_1x_{1i} + \beta_2x_{2i} + \rho w \cdot y_i + \epsilon_i \]

In this equation:

• \(\rho\) is the spatial-autoregressive coefficient
• \(w\) is a spatial weights matrix

Each model progressively incorporates more complexity to address different hypotheses about the influences on voter preferences regarding climate policy in Washington State. This approach allows for a nuanced analysis, distinguishing direct effects from those mediated by political identity or spatial proximity.

3 Results

3.1 Data

Rows: 1,454
Columns: 5
Rowwise: 
$ geoid           <chr> "53001950200", "53005010600", "53005010902", "53007960…
$ hh_vmt          <dbl> 49.22, 39.50, 36.03, 52.59, 39.31, 31.31, 40.17, 57.66…
$ vote_rep_pct    <dbl> 77.975856, 52.493079, 57.784743, 62.749887, 53.021292,…
$ vote_i0732n_pct <dbl> 78.69265, 65.03399, 71.65084, 72.74983, 65.13168, 60.5…
$ geometry        <GEOMETRY [m]> POLYGON ((-3507230 8109216,..., POLYGON ((-35…

Source: Article Notebook

Data summary

Name	model_data_skim
Number of rows	1431
Number of columns	4
_______________________
Column type frequency:
character	1
numeric	3
________________________
Group variables	None

Variable type: character

skim_variable	n_missing	complete_rate	min	max	empty	n_unique	whitespace
geoid	0	1	11	11	0	1431	0

Variable type: numeric

skim_variable	n_missing	complete_rate	mean	sd	p0	p25	p50	p75	p100	hist
hh_vmt	0	1	41.36	8.28	13.48	35.78	41.36	47.48	61.76	▁▂▇▇▂
vote_rep_pct	0	1	39.22	16.51	2.15	27.66	40.49	50.94	81.60	▃▆▇▅▁
vote_i0732n_pct	0	1	59.71	11.28	23.45	53.13	61.15	67.67	87.60	▁▂▇▇▂

Source: Article Notebook

Figure 1

Source: Article Notebook

Figure 2

Source: Article Notebook

Figure 3

Source: Article Notebook

Figure 4

Source: Article Notebook

3.2 Models

3.2.1 Univariate OLS Linear Model

Univariate model paramaters:

Call:
lm(formula = vote_i0732n_pct ~ hh_vmt, data = model_data)

Residuals:
    Min      1Q  Median      3Q     Max 
-28.696  -5.670   1.458   6.567  21.909 

Coefficients:
            Estimate Std. Error t value Pr(>|t|)    
(Intercept) 28.97985    1.27477   22.73   <2e-16 ***
hh_vmt       0.74291    0.03022   24.58   <2e-16 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 9.462 on 1429 degrees of freedom
Multiple R-squared:  0.2972,    Adjusted R-squared:  0.2967 
F-statistic: 604.4 on 1 and 1429 DF,  p-value: < 2.2e-16

Source: Article Notebook

Univariate model assumption checks:

Figure 5

Source: Article Notebook

3.2.2 Multivariate OLS Linear Model

Multivariate model paramaters:

Call:
lm(formula = vote_i0732n_pct ~ hh_vmt + vote_rep_pct, data = model_data)

Residuals:
     Min       1Q   Median       3Q      Max 
-20.4516  -1.6197   0.2296   1.9609  20.1292 

Coefficients:
              Estimate Std. Error t value Pr(>|t|)    
(Intercept)  28.863538   0.416144   69.36   <2e-16 ***
hh_vmt        0.162635   0.011199   14.52   <2e-16 ***
vote_rep_pct  0.614887   0.005617  109.46   <2e-16 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 3.089 on 1428 degrees of freedom
Multiple R-squared:  0.9252,    Adjusted R-squared:  0.9251 
F-statistic:  8826 on 2 and 1428 DF,  p-value: < 2.2e-16

Source: Article Notebook

Multivariate model assumption checks:

Figure 6

Source: Article Notebook

Spatial Autocorrelation check (Moran I test):

    Moran I test under randomisation

data:  residuals(model_lm_multivariate)  
weights: model_spatial_weights  
n reduced by no-neighbour observations  

Moran I statistic standard deviate = 29.982, p-value < 2.2e-16
alternative hypothesis: greater
sample estimates:
Moran I statistic       Expectation          Variance 
     0.4884553310     -0.0007007708      0.0002661758 

Source: Article Notebook

3.2.3 Spatially Lagged Regression

Spatial lag model parameters:

Call:lagsarlm(formula = model_lm_multivariate, data = model_data, 
    listw = model_spatial_weights, zero.policy = TRUE)

Residuals:
     Min       1Q   Median       3Q      Max 
-19.7908  -1.6062   0.1467   1.8119  18.9670 

Type: lag 
Regions with no neighbours included:
 523 1102 1253 
Coefficients: (asymptotic standard errors) 
              Estimate Std. Error z value  Pr(>|z|)
(Intercept)  19.340830   0.681397  28.384 < 2.2e-16
hh_vmt        0.151437   0.010106  14.984 < 2.2e-16
vote_rep_pct  0.454746   0.010966  41.468 < 2.2e-16

Rho: 0.27185, LR test value: 276.53, p-value: < 2.22e-16
Asymptotic standard error: 0.016331
    z-value: 16.646, p-value: < 2.22e-16
Wald statistic: 277.09, p-value: < 2.22e-16

Log likelihood: -3504.669 for lag model
ML residual variance (sigma squared): 7.7375, (sigma: 2.7816)
Nagelkerke pseudo-R-squared: 0.93831 
Number of observations: 1431 
Number of parameters estimated: 5 
AIC: 7019.3, (AIC for lm: 7293.9)
LM test for residual autocorrelation
test value: 405.56, p-value: < 2.22e-16

Source: Article Notebook

Parameter comparison: OLS vs Spatial Lag

Parameter    |  model_lm_univariate | model_lm_multivariate |    model_spatial_lag
----------------------------------------------------------------------------------
(Intercept)  | 28.98 (26.48, 31.48) |  28.86 (28.05, 29.68) | 19.34 (18.01, 20.68)
hh vmt       |  0.74 ( 0.68,  0.80) |   0.16 ( 0.14,  0.18) |  0.15 ( 0.13,  0.17)
vote rep pct |                      |   0.61 ( 0.60,  0.63) |  0.45 ( 0.43,  0.48)
rho          |                      |                       |  0.27 ( 0.24,  0.30)
----------------------------------------------------------------------------------
Observations |                 1431 |                  1431 |                     

Source: Article Notebook

Comparison of Adjusted R²/Pseudo Adjusted R²: OLS vs Spatial Lag

# A tibble: 1 × 3
  univariate multivariate spatial_lag
       <dbl>        <dbl>       <dbl>
1      0.297        0.925       0.938

Source: Article Notebook

Spatially lagged regression model residuals:

Source: Article Notebook

4 Appendix

4.1 Data Sources

• Voting Precinct Shapefiles: https://www.sos.wa.gov/elections/data-research/election-data-and-maps/reports-data-and-statistics/precinct-shapefiles

• Election Results: https://www.sos.wa.gov/elections/data-research/election-data-and-maps/election-results-and-voters-pamphlets

• American Community Survey: https://www.census.gov/programs-surveys/acs/data.html

• 2017 Local Area Transportation Characteristics for Households https://www.bts.gov/latch/latch-data ### Methodology Notes

• Income should not be included in our regression because it is used in the model that estimates household VMT (see LATCH Methodology p. 10)

Figure 1: Figure 5: Figure 6: