Does the accumulation of assets shape voting preferences? Evidence from a longitudinal study in Britain