Estimating present bias and sophistication over effort and money

We design and implement the first real-effort experiment that can jointly estimate present bias (β) and sophistication (bβ), with separate preference parameters for money (βm, bβm) and effort (βe, bβe). In our study, participants chose to (and predicted to) complete 14% (and 10%) fewer tasks on the same day than on a future day, leading to an estimated βe between 0.70 and .79 (and bβe between 0.80 and .88). Participants chose to (and predicted to) complete 2% (and 2%) fewer tasks when the payment happened on the same day than on a future date, implying insignificant present bias or sophistication over money. At the individual-level, estimates of βe and bβe are highly correlated but that of βm and bβm are uncorrelated. We explain and illustrate how assuming identical preferences for money and effort, biases the estimates of key preference parameters.