Steven A. Camarota from the Center for Immigration Studies has published a reply to my critique of the report he co-authored on refugee resettlement’s fiscal cost. I pointed out in my piece that the report used arbitrary cutoffs based on the Annual Survey of Refugee’s years of schooling question to effectively downgrade refugees’ educational attainment before assessing their likely fiscal impact. Camarota objects:
[W]e never “downgraded” anyone. The whole point of combining the ambiguous degree responses with years of education was to figure out who had which types of degrees, not to alter the status of refugees with known degrees. (In fact, when the ASR degree category was unambiguous — e.g., “secondary” clearly means a high school diploma — we did not utilize years of schooling at all.)
This is unambiguously false. Camarota’s report explicitly assigns refugees with unambiguous medical degrees but less than 17 years of education to “some college” or “BA” categories. This is a clear case of using years of schooling to downgrade the education of medical school graduates when the ASR degree category has no ambiguity. While there can be reasonable disagreement about the extent to which the report downgrades education, it’s plainly untrue to say that there is no downgrading.
On the question of the extent of downgrading, any assumption will produce some errors, downgrading education for some refugees and upgrading it for others. The relevant question is about reducing errors, not eliminating them. I maintain that the assumption I used in my adjusted estimates—namely that refugees with university degrees have bachelor’s degrees—will produce fewer errors than Camarota’s assumption that no refugee with a university degree with less than 16 years of education has a bachelor’s degree. For some suggestive evidence, the data show more than twice as many degree-holding refugees have 15 years of education as have 14 years of education, when one would expect more 14-year degrees than 15-year degrees if Camarota’s assumption were true. In any case, we can be confident that the report downgraded education on net, even if there’s uncertainty about precisely by how much it did so.
One last observation is that of the two major criticisms I leveled at his report—educational downgrading and ignoring the effects of refugees’ descendants—Camarota’s objections have been limited to the former (which is the smaller of the two). For example, in the most optimistic scenario, educational downgrading is responsible for inflating the cost of refugees by “only” about $11,000. That alone is enough to switch the sign on the net impact–but the effect of excluding descendants is over an order of magnitude higher, inflating the cost by over $136,000.