Sometimes politicians appear to `take care of their own'; at other times, loyal supporters seem to be taken for granted, while redistributive benefits are aimed at groups of `swing voters'. In Discussion Paper No. 1054, Research Associate Avinash Dixit and John Londregan explore politicians' decisions to use redistributive benefits to bid for the support of swing voters. The construct a model of locational majority voting where competing parties offer special favours to interest groups and where each group's membership is heterogeneous in its affinities for the two parties. Individuals face a trade-off between party affinity and their own transfer receipts. Each party has its core groups of constituents whom it understands well, and this greater understanding translates into greater efficiency in the allocation of particular benefits. A party's core constituencies need not prefer its policy position. It is the party's advantage over its competitors at swaying voters in a group with offers of particular benefits that makes a group core.

The model is sufficiently general to yield the two competing theories as special cases. If the parties are equally effective in delivering transfers to any group, the outcome of the process conforms to the `swing voter'. If groups have party affiliations and each party is more effective in delivering favours to its own supporters, the result is a `machine politics' outcome, where each party dispenses favour to its core support group. The model also offers insights into the changing treatment of various special interest groups over time.

The Determinants of Success of Special Interests in Redistributive Politics
Avinash Dixit and John Londregan