This is an interesting ms. about the mobility between two types of institutions (Historically Black College and Universities, HBCUs, and Predominantly White Institutions, PWIs) in the USA. The focus is on the moving premium and the moving faculty for academics involved in the mobility.

The ms. uses two data sources – Academic Analytics dataset and Scopus dataset – to measure the premiums of mobility in terms of research productivity and research impact. The matching of academics from mobile and immobile groups is used in this quantitative study. Research dynamics before and after moving to and from HBCUs is examined.

The ms. is a very good read: concise, elegant, simple. The Introduction shows the importance of HBCUs for the American Blacks and the role of Black professors in these institutions for Black elites. Research questions are well stated. The mobility to and from the two institutional types is studied tracing the performance of academics involved in mobility. The literature on mobility and performance is rightly shown to indicate mixed results.

The data provided by Academic Analytics are excellent. Mobile professors can be identified (N=139). The construction of the comparison group and the mobile group is well done. The methods are suitable for the analysis and the results are very interesting.

However, there are several issues to be tackled before proceeding further – and these are largely minor issues:

1) Can you say a bit more about Wapman et al.’s SpringRank algorithm? And specifically, why 4 matching professors to each mobile professor were used? What are the limitations of this method?

2) Can you compare briefly this method (and algorithm) to a more traditional method, Difference in Difference (DiD), often used in comparing the treatment and the control groups?

3) Did you have any issues in mapping Scopus classification to Science-Metrix one?

4) Why did you compute journals’ impact in the way it is done in WoS rather than using directly Scopus journal percentile rank (0-100)? Does the WoS construction work any better than Scopus percentiles, yours is an indirect way, is not it?

5) Can you speculate about the distinct motivations driving the two mobility types – in Discussion?

6) Can you speculate why professors in mobility to HBCUs experience challenges with publishing in top journals – in Discussion?

7) I have problems with the simple description that outflow is higher than inflow as the first idea in Discussion. Is it important, why? The paper is about mobility and productivity/impact, not about measuring flows?

8) The section on limitations could be more extended, especially: the limitations of the small numbers of academics; of the two datasets, AA and especially Scopus; time period selected; and the methods used in matching scientists?

I congratulate the authors on an excellent ms., my remarks are minor and are intended to strengthen this research.