This page pertains to the case of Lathern v. NationsBank, a putative class action brought against NationsBank Corporation in the United States District Court for the District of Columbia in 1995. The case had the potential to result in a recovery of many hundreds of millions of dollars. But ultimately, due to the failure of the plaintiffs to timely move for class certification, a class was not certified. The complaint is available here.
The case was the subject of my January 1, 1996 Legal Times article styled “When Statistics Lie.” It was also the subject of the two unpublished works – The Confusion over Credit Discrimination” (1997) and “The NationsBank Case and the Misunderstanding of Statistical Proof of Discrimination” (1995), both of which discuss studies underlying the case. The 1995 unpublished item does so at considerable length and provides a useful guide to the substantial shortcomings of lending discrimination studies, including but not limited to the implications of the failure to recognize the pattern whereby the rarer an outcome the greater tends to be the relative difference in experiencing it and the smaller tends to be the relative in avoiding it (which pattern is addressed on the Scanlan’s Rule page and many other pages and sub-pages of this and was most recently explained in the credit context in “The Lending Industry’s Conundrum,” National Law Journal, Apr. 2010).
In the years since 1995, I have the addressed more fully some of the matters treated in the 1995 item. The material in note 3 of the 1995 item regarding the adjustment procedure in the Lawyers’ Committee study underlying the Lathern case is treated further in Section 5 of the Adjustment Issues sub-page of the Vignettes page. Section 3 of the Adjustment Issues sub-page and Nov 2009 Comment on Lynch JECH 2006 also pertain to matters mentioned in that note.The subject of note 6 is treated at much greater length in the Times Higher sub-page of the Vignettes page. The points about odds ratios at pages 21-23 (which are now also addressed in Section 6 of the main Lending Disparities page) have been illustrated with income data in “Can We Actually Measure Health Disparities?,” Chance 2006, and with income data in various conference presentations between 2006 and 2011 (e.g., in Figure 4 of the 2009 Royal Statistical Society presentation). The cases discussed in the final section of the 1995 article are also the subjects of individual articles collected in Section A of the Employment Discrimination page.