Working Paper: CEPR ID: DP17346
Authors: Gregory Clark; Neil Cummins; Matthew Curtis
Abstract: Using a new database of 1.7 m marriages in England 1837-1939, and a genealogy of 414,000 peoplein England 1700-2021, we estimate two independent new occupational status indices for England1800-1939. These new indices show that there was much less social mobility 1800-1939 than previousindices, such as HISCAM, imply. The performance of these two new indices, however, illustrates ageneral problem with comparing social mobility across time and place using status indices. All suchindices embody unknown and varying degrees of error. The more error, the more apparent mobility.So in the paper we develop a way of measuring intergenerational occupational status mobility whicheliminates all measurement error. This suggests that intergenerational occupational status persistence in England1800-2021 was always much greater than conventionally measured, and was largely unchanged overtime.
Keywords: intergenerational social mobility; occupational mobility
JEL Codes: No JEL codes provided
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Newly constructed occupational status indices (FOERCI and FOEPCA) (J39) | Intergenerational occupational status persistence (J62) |
| Measurement errors in previous indices (C43) | Inflated perceived mobility rates (J62) |
| Intergenerational correlation declines to approximately 0.78 by the 1980s (D15) | Modest increase in mobility (J62) |
| Comparison of father-son and father-in-law-son correlations (J12) | More accurate estimate of true intergenerational mobility (J62) |
| True correlation in occupational status persistence is closer to 0.9 (J62) | Indicating much lower mobility than previously thought (J62) |