Political generations and partisanship in the UK, 1964–1997

Summary. Political partisanship is often claimed to be influenced by generational and life-cycle processes, with both being cited as the factor that is responsible for higher levels of Conservative identifications among older voters. Given the existence of over-time change it is difficult to assess the validity of these claims as even with repeated survey data any model is underidentified. This paper uses smoothed additive models to isolate and examine the non-linear component of the generational effect. Some identifying assumptions are presented to try to assess the extent to which linear aging or generational processes are responsible for the increased Conservatism of the elderly. The advantage of the smoothed additive models is their ability to highlight non-linear effects, however, and this paper shows that regardless of linear trends people who entered the electorate during Conservative Parliaments are more likely to be Conservative partisan identifiers many years later. The introduction of a multiplicative term linking age to period effects supports this hypothesis by showing that younger people are more susceptible to the influence of period effects.