Discussing the logsum as an accessibility indicator

In recent years the so called Logsum as an accessibility measure increasingly received attention. The advantages of the logsum measure (LM) are discussed in literature, but to the best of our knowledge no systematic discussions on limitations exist. The fist aim of this paper is to fill this gap. Our second aim is to explicitly compare the LM with its main competitor: the gravity measure, or potential accessibility measure. Limitations of the logsum include, amongst others, the ignorance of option values and the related love for variation in consumption, limitations related to the willingness to pay, the fact that the utility at the choice moment may differ from the utility of the choice option itself, limitations related to the marginal utility of income, the fact that the logsum relates to changes in accessibility, not the absolute level of accessibility, the fact that the logsum is a measure for the valuation of accessibility but not for accessibility itself, the fact that the indicator is difficult to communicate. Comparing the LM and gravity based measures we conclude that to go from the LM’ measure (that is very close to the gravity based GM measure) to a welfare measure, the logarithmic transformation has to be used in combination with the (1/α) standardization. This reflects the difference between accessibility and the utility of accessibility: an absolute change in accessibility measured according to the LM’ definition has a welfare impact that depends on the initial level of accessibility.