Working Papers INTERGENERATIONAL MOBILITY IN BRITAIN: NEW EVIDENCE FROM THE BHPS

This paper provides the first full account of intergenerational mobility in Britain using the British Household Panel Survey (BHPS) for the period 1991-1999. We analyse two samples. The first is on all adults who provide information on their parents’ occupation when they were aged 14. We label this as the General Sample (G-Sample). The second is a special sample in which young adults (aged 16 or above) are matched to one or both of their biological or adoptive parents interviewed in at least one of the nine years of the panel survey. This sample is labelled M(atched)-Sample. The estimates obtained from the G-Sample point to an intergenerational elasticity – measured in terms of Hope-Goldthorpe scores of occupational prestige – of the order of 0.15 to 0.30. After adjusting for possible measurement error, the estimated intergenerational correlations range from 0.45 to 0.75 for father-child pairs and from 0.30 to 0.50 for mother-child pairs, after being adjusted for possible measurement error. We also find strong nonlinear effects (typically, the intergenerational elasticity increases with parental status), and a substantial decline in cross-generation correlations in occupational prestige over time. The analysis of the M-Sample uses more conventional measures of economic status (i.e., monthly earnings and annual income) for both children and parents. If the OLS estimates of the intergenerational correlation, which are at most between 0.05 and 0.07, are biased downward, the IV estimates, which are at most around 0.20, are likely to be an upper bound of the true elasticity. Either way, they are clearly lower than the ‘consensus’ estimates of around 0.4 found in earlier British studies. We conjecture that the very high economic mobility for this sample is due to life cycle considerations. This is also an area that deserves more attention and is left for future research. JEL Classification: J62, I2, D31