The graph below is from a fascinating new paper, What Happens When We Randomly Assign Children to Families?, by Bruce Sacerdote. Holt's International Children's Services places children, primarily Koreans, with families in the United States. Holt has an interesting proviso to their adoption contract, conditional on being accepted into the program, children are randomly assigned. Sacerdote has collected data from children who were adopted between 1970-1980, and thus who today are in their mid 20's or 30's, and their adoptive parents.
The graph shows how parent income at the time of adoption relates to child income for the adopted and "biological" (non-adopted) children. The income of biological children increases strongly with parental income but the income of adoptive children is flat in parent income. What does this mean?
The graph does not say that adopted children necessarily have low income. On the contrary, some have high and some have low income and the same is true of biological children. What the graph says is that higher parental income predicts higher child income but only for biological children and not for adoptees.
What do parents transmit to their biological children but not to their adopted children? Genes. When we observe, as we do, that low-income parents tend to have low-income children and high-income parents tend to have high-income children we should not bemoan the inequities of nurture but rather the inequities of nature.
Read the paper, for much more of interest. A few additional comments are in the extension.
Some might suggest that parents treat their biological and adopted children differently and this is what accounts for the difference in incomes. The interpretation is very uncharitable to the parents who have volunteered to raise an adopted child and I think it implausible. Moreover, unless every adopted child is treated equally poorly in all families, then we would still expect the income of adoptees to increase with parental income but perhaps starting at a lower level.
The other proviso is that the Holt experiment is only informative for the experimental variation in environment. In other words, we can tell from the Holt experiment that variation in parental income from around 25 thousand to 175 thousand doen’t have much impact on variation in adopted child income but all these children are raised in the United States so culture and other variables are roughly similar. In other words, move a child from a poor country to a rich country and you would expect a much bigger treatment effect than moving a child from a poor family to a rich family.