Genealogy Databases and Automation

This week’s interview is with a professor at Brigham Young University named Joseph Price, or Joe. Joe is a professor who graduated from Cornell around the same time that I graduated from the University of Georgia (i.e., 2007 cohort). He’s a labor economist, Fellow at the Wheatley Institution, NBER research associate, Director of BYU’s Linking Lab, co-editor at Economics of Education Review and the author of something like 45 peer reviewed articles in top economics journals like the Quarterly Journal of Economics, Journal of Labor Economics, Journal of Human Resources, Management Science, as well as countless interdisciplinary journals on numerous topics in sports, the family and more.

I will just list two studies that I have for many years found very interesting. One of them with Justin Wolfers made a major splash both in economics on the topic of discrimination as well as the broader public, including the National Basketball Association (NYT link here). who has written on a variety of topics like discrimination in the NBA. “Racial Discrimination Among NBA Referees” with Justin Wolfers appeared in one of the 2010 issues of the Quarterly Journal of Economics and claimed to find evidence of racial discrimination, mostly likely caused by unconscious bias than animus, among NBA referees in calling fouls against players. What impressed me at the time is the same thing that always impresses me: an interesting question, the discovery of some randomness that allows one to plausibly provide some evidence relevant to that question, and the collection of interesting data. Joe and Justin hand collected box scores of every NBA game with specific fouls among other statistics of the players combined with the names (and races) of the officiating referees at each game. While they could not link a referee to a foul called, they used a measure of the percentage of the officiating staff that was White and non-White as a proxy. With random variation in the racial composition, supported by both institutional details and a series of regression analyses, they looked at whether a higher share of White referees “caused” a Black or White player to have a foul called against them more or less often conditional on player fixed effects. You can read the abstract to learn what they find, but given the controversy and antagonism it generated within the NBA, I suppose you can also guess.

But it is another paper of his, a solo authored one, at the Journal of Human Resources, that I have always found to be a truly beautiful piece of economics. “Parent-Child Quality Time: Does Birth Order Matter?” was for a long time my favorite empirical paper I’d ever read. It was a simple idea really. Lower birth order, particularly the first born, typically had better academic and labor market outcomes, despite coming from the same family. Sandy Black and coauthors had written about this in a 2005 QJE using Scandinavian registry data, but the mechanisms were largely speculative. Joe’s paper was not so much conclusive as it was a clever descriptive paper showing that lower birth order children received more high quality time from their parents using the American Time Use Survey, which is a time diary and in my opinion one of America’s more interesting repeated cross sections. The patterns he found fit a rule that was well intentioned but likely led to inequities within the family — first borns received all their parents’ time; second borns received half their parents’ time, third borns received a third of their parents time, and on and on. In other words, equity rules with each stage over quality time, or simply budget constraints themselves, leads to American families to spend less time overall in early years with each new child simply because quality time is a scarce resource. Becker might say that instead of equity, we should aim for optimized time spent with children — spent quality time up to the point where marginal benefit equals marginal cost across all children. But such rules, while sensible economists, are likely unethical because of ironically strong bonds of kinship where parents love their children the same.

These kinds of questions over deviations from optimizing behavior where emotion and quick thinking drives decision making, as opposed to pure economic calculation, was a hallmark of Joe’s work, but more recently he transitioned into a very ambitious project of using Machine Learning and large genealogical databases to link people with other large datasets like the Census, to track them over time and create a large family tree of what he calls the Human Family. This is the Joe Price I have come to know — a deeply curious man, a man with deep endowments in the skills of our professions, a hard worker (you will not find him on social media), a mentor and a man of vision. To say that I hold him in the highest esteem is an understatement. And because of his character, the lack of guile and a positive and egalitarian spirit, he was for a long time an economist my age whose productivity did not cause me insecurity. I was simply honored and amazed by what a good economist he was and tried, as I often have, to see if I could crack the algorithm that made him so successful.

So, it is my pleasure to introduce you to one of my friends who I consider to be a special member of economics of larger story. Special in many ways, but one way being he would likely tell me that we are all special. Thank you again for supporting the podcast. If you like it, consider supporting it by subscribing below. I am Scott Cunningham the host of the Mixtape with Scott!

Scott's Substack is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber.