Donehower on the Net Fiscal Effect of Low-Skilled Immigrants

When I emailed the editors of the National Academy of Sciences report on The Economic and Fiscal Impact of Immigration about Jason Richwine’s criticism, they responded swiftly and scrupulously.  Series editor Francine Blau put me in touch with Gretchen Donehower, one of the authors of the section.  Donehower sent me the following response.  Reprinted with her kind permission.

Hi Bryan,

Thanks for reaching out.

Richwine is correct in that piece that he writes, and we actually exchanged a bunch of emails to verify back in February of this year.  The $35k cell (upper left-hand corner of Table 8-12) is the net fiscal impact of someone who comes to the US aged 0-24 whose parents’ average education falls in the <HS category.  The average age in that cell comes from the average age of actual persons age 0-24 who have entered the US in the last 5 years (with 2012 being the base year for the calculations done then).  All of the other age groups are categorized by the person’s own education. The reason that the 0-24 age group must be treated differently from the other age groups is that the vast majority in that cell have not yet completed their own education.  In other words, it wouldn’t make sense to treat a child in 6th grade as a high school drop out, because she hasn’t had the chance to finish high school yet.  What to do then?  We decided to use the taxes and benefits that accrue to kids by parental education group while they are ages 0-24 but starting at age 25, we predict their future taxes and benefits based on a predicted educational distribution.  The prediction is based on educational transmission functions from parent to child in the past (see annex 8.4 for details, it’s on page 488 in my copy).  These result in future trajectories of taxes and benefits based on weighted averages of education groups, meant to represent a predicted path for that child.

So, Richwine is right.  He is also right that in Table 8-13, the estimated impact of a person exactly age 25 at entry with <HS education has a fiscal impact of -186K and this probably implies that it is negative for age 24 also.  However, he is most certainly not right that it is negative for age 5 on average.  Those educational transmission functions come from data.  I did not make them up.  It is more likely that a child of <HS immigrant parents will at least complete HS than that she will not.  That’s not an interpretation, that’s just what the data showed.  If it hadn’t showed that, the $35k number would have been much lower.  So, somewhere between age 0 and age 25 there is an age when the net fiscal impact is probably close to zero.  I don’t have that sitting around, but it would be an interesting thing to dig out if the calculations are ever done again.

I hope that helps.  The whole idea – predicting a person’s future – is pretty complex and if it wasn’t obvious enough in the write up how I did it I do apologize.  If you’d like to talk further, we can keep emailing or set up a phone or zoom sometime.