Knowledge and Freedom
What is the relationship between knowledge about the world and freedom? Under which circumstances is autonomy granted to an individual who may be less or better informed?
The idea that (only) informed choices should be respected has a long history. Vilfredo Pareto wrote in his Manual that economists should restrict attention to preferences that come about from repeated choice. The Talmud and the Bible command us to respect the choices of others, provided they have sufficient knowledge about their condition. Occupational licensing and driver's licenses aim at the protection of the public, but they also protect professionals and drivers from high-stakes errors that result from ignorance. MiFID II requires investors to prove knowledge before they may trade derivatives or other advanced instruments. A recent study showed that 68 percent of American parents fear that their child may regret a tattoo later; in other words, a lack of knowledge about consequences. The present project aids in our understanding of public policy.
In the intellectual history of choice under uncertainty, we take an intermediate position. Learning and experience connect Savage's conceptualization of ambiguity with the full information paradigm of von Neumann and Morgenstern. Reality requires exploration of the unknown. Only after learning can we make informed choices. Suppose that a Choice Architect can overwrite a Chooser's decision between two options: A safe payment and a risky binary lottery, of which only the success probability is unknown. The Choice Architect has asymmetric information about the success probability, and Choosers can draw from the binary lottery a given number of times before making their choice.
Using a simple model of Bayesian learning with von Neumann-Morgenstern expected utilities, we show theoretically that learning can help Choosers make the decision that they would prefer under full information. We demonstrative that a ("utilitarian") Choice Architect without any own normative interest may intervene because Bayesian updating can induce objective errors in Chooser decisions. This effect is weakened given more learning and strengthened if experiences are less representative. However, a ("Leviathanic") Choice Architect who cares only about personally preferred results will always intervene, independent of the amount of knowledge a Chooser possesses. Furthermore, we provide a microfoundation for the false consensus bias observed in studies of paternalism (e.g., Ambuehl, Bernheim & Ockenfels, 2021, AER).
In a series of lab experiments, we vary Choosers' amount of experience to disentangle these motivations. Furthermore, we vary how representative Choosers' experiences are. We can distinguish interventions on the extensive and intensive margins. (I will present results from our lab experiments of spring 2023. At the time of submission, no data had been collected.)
This positive study of freedom highlights one of its essential theoretical underpinnings: Decision-makers are granted freedom of choice only if they are well-informed, and informedness can be enhanced through experience and learning. Furthermore, liberty's central appeal is its ability to grant preference satisfaction to broad classes of individuals with diverse preferences.
The idea that (only) informed choices should be respected has a long history. Vilfredo Pareto wrote in his Manual that economists should restrict attention to preferences that come about from repeated choice. The Talmud and the Bible command us to respect the choices of others, provided they have sufficient knowledge about their condition. Occupational licensing and driver's licenses aim at the protection of the public, but they also protect professionals and drivers from high-stakes errors that result from ignorance. MiFID II requires investors to prove knowledge before they may trade derivatives or other advanced instruments. A recent study showed that 68 percent of American parents fear that their child may regret a tattoo later; in other words, a lack of knowledge about consequences. The present project aids in our understanding of public policy.
In the intellectual history of choice under uncertainty, we take an intermediate position. Learning and experience connect Savage's conceptualization of ambiguity with the full information paradigm of von Neumann and Morgenstern. Reality requires exploration of the unknown. Only after learning can we make informed choices. Suppose that a Choice Architect can overwrite a Chooser's decision between two options: A safe payment and a risky binary lottery, of which only the success probability is unknown. The Choice Architect has asymmetric information about the success probability, and Choosers can draw from the binary lottery a given number of times before making their choice.
Using a simple model of Bayesian learning with von Neumann-Morgenstern expected utilities, we show theoretically that learning can help Choosers make the decision that they would prefer under full information. We demonstrative that a ("utilitarian") Choice Architect without any own normative interest may intervene because Bayesian updating can induce objective errors in Chooser decisions. This effect is weakened given more learning and strengthened if experiences are less representative. However, a ("Leviathanic") Choice Architect who cares only about personally preferred results will always intervene, independent of the amount of knowledge a Chooser possesses. Furthermore, we provide a microfoundation for the false consensus bias observed in studies of paternalism (e.g., Ambuehl, Bernheim & Ockenfels, 2021, AER).
In a series of lab experiments, we vary Choosers' amount of experience to disentangle these motivations. Furthermore, we vary how representative Choosers' experiences are. We can distinguish interventions on the extensive and intensive margins. (I will present results from our lab experiments of spring 2023. At the time of submission, no data had been collected.)
This positive study of freedom highlights one of its essential theoretical underpinnings: Decision-makers are granted freedom of choice only if they are well-informed, and informedness can be enhanced through experience and learning. Furthermore, liberty's central appeal is its ability to grant preference satisfaction to broad classes of individuals with diverse preferences.