ABSTRACT
We examine the 2016 Survey of Consumer Finance, using weighted logit regressions and employing the repeated-imputation inference method. Loan denial differences across race disappear once credit history and loan types are controlled for as well as other financial and demographic variables used in the literature.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1 Since Lee and Hanna (Citation2012) showed many differences between Hispanic and African Americans, we categorized these two races separately.
2 Original SCF item is revised so that higher EstateImp means stronger belief in importance.
3 NumberHH equals the number of people in the PEU plus the number of people in the non-PEU.
4 Some control variables in logit are not reported in Table 2. Full table is available upon request.