Controversial, but worth considering. I believe societies have different capacities for assimilation (changing immigrants) and appropriation (changing themselves), with the hallmark of any era's great societies being their ability to maximise both.
That said, the evidence is mixed [1], with fairness and economic inequality [2][3] seeming to matter more than racial homogeneity. (Lots of tiny, racially-homogenous societies–high trust or not–bordering each other also have a one-way historical track record.)
A very often ignored fact is the cultural homogeneity. I do not thing racial homogeneity is of any benefit whatsoever, but I do believe that cultural is.
When someone raised in a culture where cheating to win by any means is acceptable (most of India) or where bartering, persuading and microfrauding in trade (most of Middle east and sup-sahara Africa) is not frowned upon, it is not a stretch to imagine that the introduction of such cultural elements will lead to dilution of the overall interpersonal trust in let's say, Swedish society.
Putnam indeed reported a correlation between the mean herfindahl index of ethnic homogeneity and trust in societies (both own-race trust, other race trust & neighbour trust).
If you had actually read the paper (which I have), you would realise that the relationship between ethnic diversity and social trust is inverse.
That said, the evidence is mixed [1], with fairness and economic inequality [2][3] seeming to matter more than racial homogeneity. (Lots of tiny, racially-homogenous societies–high trust or not–bordering each other also have a one-way historical track record.)
[1] https://journals.sagepub.com/doi/abs/10.1177/000169931772161...
[2] https://www.jstor.org/stable/23324182
[3] https://pmc.ncbi.nlm.nih.gov/articles/PMC7454994/