Three Microeconomic Applications Using Administrative Records

David Hedengren

Advisor: Thomas Stratmann, PhD, Department of Economics

Committee Members: Bryan Caplan, John Nye, Alexander Tabarrok

Carow Hall, Lecture Hall
April 16, 2013, 10:00 AM to 07:00 AM


This dissertation provides three examples of how using naturally occurring datasets, data collected independent of the consideration of researchers, can answer important research questions. These types of data are called “administrative records” or “organic data” and include sources as diverse as W2 tax filings, stock prices, and Google searches (see Groves, 2011).

In the first example, The Dog that Didn't Bark: What Item Nonresponse Shows about Cognitive and Non-Cognitive Ability, I show that what survey respondents choose not to answer (item nonresponse) provides a useful task based measure of cognitive ability (e.g., IQ) and non-cognitive ability (e.g., Conscientiousness). Using the German Socio-Economic Panel (SOEP) and the National Longitudinal Survey of Youth 1997 (NLSY97), I find consistent correlation between item nonresponse and traditional measures of IQ and Conscientiousness. I also find that item nonresponse is more strongly correlated with earnings in the SOEP than traditional measures of either IQ or Conscientiousness. I also use the Survey of Income and Program Participation (SIPP) Gold Standard, which has no explicit measure of either cognitive or non-cognitive ability, to show that item nonresponse predicts earnings from self-reported and administrative sources. Consistent with previous work showing that Conscientiousness and IQ are positively associated with longevity, I document that item nonresponse is associated with decreased mortality risk. Our findings suggest that item nonresponse provides an important measure of cognitive and non-cognitive ability that is contained on every survey.

In the second example, Adverse vs. Advantageous Selection in Life Insurance Markets, I assess the predictive power of the theories of adverse and advantageous selection using administrative mortality records. Adverse selection theory predicts that people with a high death risk are more likely to purchase life insurance. The advantageous selection hypothesis predicts the opposite. Using a unique dataset merging administrative and survey records I find support for the advantageous selection hypothesis. I exploit variation in risk aversion among consumers and between pricing structures in individual and group life insurance markets to study and identify the relative strength of advantageous and adverse selection forces. I find that the overall negative relationship between risk and coverage is likely due to advantageous selection swamping adverse selection forces.

In the third example, Totally Tuber-ular? A price based analysis of the impact of the potato in England 1724-1913, I use organic data from the 17th and 18th centuries to investigate the importance of the potato in England’s economic growth. The potato is credited with speeding urbanization, promoting the industrial revolution, and even ending European infanticide. Though many believe the potato was the super-food which supported England’s rapid development from 1724 to 1913, this paper questions that received knowledge using three separate metrics: a price per calorie comparison to other staples, an overview of the rate of culinary adoption, and an evaluation of the potato’s share of caloric output relative to other crops. Each analysis shows the potato to be a valuable addition to the diet of the English, but hardly the super tuber its proponents describe. The robustness of these conclusions is tested using data from America.