The methodology proposed (DfT, 2005) is based on the observed correlation between density of employment and productivity.
where:
WB1 are the agglomeration benefits of the option (the ‘do-something’ situation (S)) compared with the do-minimum situation (M), to be calculated; i is a zone for which agglomeration benefits are being calculated - all of the modelled zones are included in the summation; diS, diM are the effective densities of zone i in the do-something situation S and do-minimum situation M respectively, calculated as shown below; diB is the effective density of zone i in the base year (all other values are for the forecast year), likewise calculated using the formula shown below; e is the elasticity of productivity with respect to effective density (Graham, 2005); h is GDP per worker in i (see 9.3.5.1 below); and EiS is employment (in the do-something case).
Effective density is a measure of the accessibility of zone i to jobs in all zones. The formula for effective density, of zone i, in situation X (X is base, Do-minimum or Do-Something) is given by:
This notion of generalised cost is given by a weighted average over:
The weights used in these steps are the numbers of trips (persons or goods vehicles) by mode and purpose in the base case. These weights are based on the numbers of trips between the pair of zones considered in each calculation.
A detailed transport model, such as Transport Model for Scotland (TMfS), can be used to produce a value of agglomeration benefits using the methodology outlined above.