Estimating the Intergenerational Elasticity and Rank Association in the United States: Overcoming the Current Limitations of Tax Data ☆

摘要: Abstract Ideal estimates of the intergenerational elasticity (IGE) in income require a large panel data covering entire working lifetimes for two generations. Previous studies have demonstrated that using short panels and only certain portions life cycle can lead to considerable bias. I address these biases by PSID constructing long time averages centered at age 40 both find IGE family United States is likely greater than 0.6 suggesting relatively low rate mobility States. similar sized labor income. These support prior findings Mazumder (2005a, b) are also comparable reported Mitnik et al. (2015). In contrast, recent influential study Chetty, Hendren, Kline, Saez (2014) tax begins 1996 be just 0.344 implying much higher mobility. demonstrate despite seeming advantages extremely samples administrative data, structure, limited dimension used Chetty leads downward bias estimating IGE. further sensitivity checks regarding which children’s measured, length average parent estimate suffer from due limitations. There concerns unlike survey may not adequately reflect all sources Estimates rank–rank slope, al.’s preferred estimator, more robust limitations but biased modestly overstate However, main sizable geographic differences within US rank unlikely affected biases. conclude researchers should continue use rank-based measures depending on their concept It important adequate coverage key consider possible drawbacks data.