Today’s research summary is about a paper co-authored by Angela Duckworth, that is at the intersection of psychology and economics. Though I have been following behavioral economics a bit, I still found the paper a bit challenging to read and comprehend and don’t claim to understand all the attached jargon, functions and mathematical formulations. The fact that the paper is 88 pages long wasn’t of help either 🙂 (the saving grace being that 20 or more pages were filled with references alone), so read the rest of the summary at your own peril!

An illustration of Spearman’s two-factor intelligence theory. Each small oval is a hypothetical mental test. The blue areas show the variance attributed to s, and the purple areas the variance attributed to g. (Photo credit: Wikipedia)

1. The paper aims to throw light on how personality affects (socio)economic outcomes and how concepts from personality psychology can be used in economic equations and modeling.
2. To start with, an important socioeconomic outcome is success in life. IQ or cognitive ability is well established as a predictor of success in life/job,  and slowly but surely, a case is building up for the predictive power of personality traits like conscientiousness to predict success in life/job.
3. Its useful to distinguish cognitive factors like Intelligence/IQ from other ‘non-cognitive’ factors like personality traits and motivation.
4. Perry Preschool study which enriched the environment (an intervention aimed at increasing IQ) of disadvantaged kids with subnormal IQ, found that IQ gains for treatment group (which shot up initially) and control group became equal at age 10 , though the treatment group continued to be much more successful on many socioeconomic outcomes over their life cycle. This can be only explained if we admit that something other than IQ, maybe personality factors, were changed by the intervention.
5. Psychologists use personality, motivation and cognitive factors to explain behavior and success of an agent. Economists however use concepts like preferences, constraints, incentives etc to explain choice/decision/ behavior and ultimately success in life.
6. Cognitive factors are defined as ‘‘ability to understand complex ideas,to adapt effectively to the environment, to learn from experience, to engage in various forms of reasoning, to overcome obstacles by taking thought’’. The various tests like IQ tests that measure cognitive ability have led to identification of a general factor ‘g’ of intelligence. The factor structure of Intelligence is hierarchical;  as per one conceptualization, the second-order factors are ‘fluid intelligence’ and ‘crystallized intelligence“.
7. IQ test are not a pure measure of maximal intellectual performance; for those getting low scores, appropriate incentives can increase their scores. Similarly test anxiety may affect performance; thus IQ measure is affected by factors like motivation and personality.
8. Personality factors also have a hierarchical structure; the most common level contains the Big Five factors, below them are specific facets  and above them two super factors of plasticity and stability.
9. The personality traits of Big Five have been arrived at using factor analysis and are more descriptive in nature, based around clustering of together of traits, adjectives or behaviors. The same can be said of ‘g’ which is again more descriptively arrived at. In contrast, economists prefer measures that have been built based around their predictive power in the real word. MMPI, Hogan personality inventory etc were on the other hand built with the specific aim of predicting real world outcomes.
10. Economists, try to estimate preferences of agents and thus predict/explain their behavior etc. Some of the typical preferences studied are time preference, risk aversion, preference for leisure and altruism/ social preferences. Estimating these preferences help explain and predict behavior that deviates from a purely self-interested rational agent.
11. Time preference is the preference for immediate reward over future reward. This is measured by the phenomenon of time discounting while making decisions.  For example, what would you choose 1 $ today or 2 $ tomorrow? 500 $ this week or 1000 $ next month? based on answers to questions like these (and maybe real world behaviors/ decisions too) economists can infer what is the rate at which you discount future utility for present utility. That function is hyperbolic in nature.
12. Its seems “time preference is tri-dimensional, comprising three separate underlying motives: impulsivity, the tendency to act spontaneously and without planning; compulsivity, the tendency to stick with plans; and inhibition, the ability to override automatic responses to urges or emotions”. Its easy to see how the three components of time preference can be related to personality factors. Also important is to note that a person with low future vision or imagination may be constrained on this time preference dimension.
13. Risk aversion is the phenomenon, whereby a sure or less uncertain outcome is preferred over an uncertain outcome.  For example, what would you choose  1 $ for sure or a 50 % chance of winning 2.2 $? Based on analyzing such decisions, one can again calculate, how risk averse a person is. This paradigm is however prone to framing effects.
14. Those who show little risk aversion, also have poor outcomes like indulging in smoking, stealing and not wearing seat belts. The personality trait of sensation seeking, as developed by Zuckerman, is related to this construct.
15. Preference for leisure is the preference to use time for relaxation etc over indulging in work or economic activity. Some people are driven to work hard and personality traits like Conscientiousness are really relevant here.
16. Social preferences are preferences like inequality aversion where a monkey would not accept cucumber pieces for the same work, if another monkey is getting grapes instead. Doesn’t  make sense rationally, but economists can use social preference to get out of this hole!
17. The big five as well as IQ are predictive of various life outcomes like leadership, grades, longevity etc
18. Most personality traits as well as intelligence measures change with age- they are malleable and follow a pattern. For eg, fluid intelligence decreases while crystallized intelligence increases over the lifespan.
19. Environmental factors like parental investment and social roles can be the mechanisms that lead to changes and stability in these traits.
20. Preference factors, which are studied by economists, however are not clear as to whether they are stable or change with age and more research needs to be done there.
21. The real contribution of this paper is in conceiving psychological traits as constraints under which economic decisions are being made.  For eg. low cognitive ability will constrain a person to figure out and get clarity about the issue at hand and he will be forced to choose in uncertainty and his risk aversion maybe causing him to make sub-optimal decisions.  The intelligent person has a richer choice set and intelligence is a constrain having real world implications; same is true for personality factors.
22. Thus personality traits may be a form of constraints/ preferences and research in either psychology or economics around this shroud inform each other.
23. Overall, despite its challenging economics jargon, I found this really useful; as someone interested in personality psychology, this provided a new perspective.

As always, do check out the original paper here.