Calibration of the difference between actual and hypothetical valuations in a field experiment

We design and implement a field experiment to elicit and calibrate in-sample hypothetical and actual bids given the presence of other goods and intensity of market experience. Using market goods that possess characteristics beyond the norm but yet remain deliverable, bidding behavior was consistent with theory. But we also observe the average calibration factor for hypothetical bids in the auction with other goods to be more severe (0.3) than for the auction without the goods (0.4). The results support the view that the calibration of hypothetical and actual bidding is good- and context-specific.