ECONOMETRIC ANALYSIS OF LINK BETWEEN BUS ACCESSIBILITY AND EMPLOYMENT
Daniel Johnson, Institute for Transport Studies, Marco Ercolani, University of Birmingham
The modelling work carried out here aims to establish the underlying sensitivity of employment to changes in bus accessibility
The modelling work carried out here aims to establish the underlying sensitivity of employment to changes in bus accessibility. This is a relatively difficult and unexplored area, characterised by complex modelling and onerous data requirements.
An important issue within this analysis is to understand if accessibility and employment are endogenous –which direction is the causality in the relationship between the two?
To discover the appropriate mix of economics and geography, we adopt a multi-pronged approach.
Firstly, we estimate a fixed effects model using panel data. The panel dataset is measured across (324) Local Authority District (LAD) areas and over 2008-2011. Dependent variables were based on the level of employment as published by the ONS. For measures of accessibility, we used DfT derived accessibility indicators for journey times to employment areas calculated by public transport and by car. Other covariates included LAD level measures of population, population density, education and training, gender and ethnic mix, public and elementary occupation mix.
We also estimate a cross sectional model utilising 2011 UK Census data at the Mid level super output area, giving observations on social and labour market measures for England, matched again to bus accessibility data from the DfT. This data set allows us to investigate the relationship between spatial differences in bus accessibility and differences in employment rates, controlling for other localised. In estimating this model we include fixed effects constants for the LADs to capture unobserved heterogeneity between area types. In this model we also investigate the issue of endogeneity through (IV) estimation.
Our results add to the existing literature on labour supply elasticity/employment sensitivity within the spirit of the current WebTAG framework. Across both our datasets, we found a statistically significant and negative relationship between public transport travel time accessibility and employment, which varies in magnitude by urban type and level of car availability. Our models appear plausible in terms of signs and magnitudes for all estimated coefficients. We take the consistency between datasets to be an indication of robustness in the results.
Association for European Transport