Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
We propose a method for recovering the compensating variation for a public good based on donation data. People face a take-it-or-leave-it actual donation decision, and the public good is provided if a threshold contribution level is reached. Compensating variation is derived using an expected utility model that accounts for nonparticipation, free riding, and warm glow, and a double hurdle model is used to recover estimates of structural parameters. Monte Carlo simulations suggest that the approach reliably recovers willingness to pay measures with moderate bias, and this bias is reduced when donations are combined with data on the donors’ subjective probabilities that the threshold will be reached. We illustrate the methodology through an empirical application involving the protection of grassland bird habitats.