Efficiency in the trust game: an experimental study of precommitment

We experimentally test a precommitment mechanism for the trust game. Before the investor's decision, the allocator places an amount into escrow, to be forfeited if he keeps the proceeds of investment for himself. We vary the available escrow amounts-in particular, whether there is a high amount that gives rise to an efficient equilibrium-and whether escrow is voluntary or imposed. We find that when chosen, the high escrow amount does lead to efficient outcomes. We also find substantial investment when the high amount is unavailable or not chosen, though well below that when it is chosen, and declining over time. We find only weak evidence for behavioral theories, such as crowding out and signaling. These results are seen when escrow choices are imposed as well as when they are voluntary.