Local Job Accessibility Measurement: When the Model Makes the Results. Methodological Contribution and Empirical Benchmarking on the Paris Region

This paper focuses on local job accessibility measurement. We propose an original model that uses national exhaustive micro data and allows for i) a full estimation of job availability according to an extensive set of individual characteristics, ii) a full appraisal of job competition on the labour market and iii) a full control of frontier effects. By matching several exhaustive micro data sources on the Paris region municipalities, we compare the results produced by this benchmark model to a representative set of alternative models, we show that the model may indeed make the results as far as local job accessibility is concerned. Significant empirical differences do stem from the use of different Local Job Accessibility measures. Moreover, these differences are spatially differentiated across the Paris region municipalities. In particular, we show that failing to use a model where job availability is fully estimated according to individual characteristics may lead to the over-estimation of the job accessibility levels of notably under-privileged municipalities.