Efficiency and Equity of Posted Price Markets for Irrigation Water: Theory and Experiment

Abstract

We study the performance of thinly traded posted price markets under two competing trading rules, recursive and non-recursive trading. Recursive markets allow sellers and buyers to trade multiple times over a growing season so as to increase the chances of market clearing. By contrast, non-recursive markets allow for one-shot trading. Our analysis shows that recursive trading, in line with intuition, raises efficiency when heterogeneity among traders is limited. Under low heterogeneity, and relative to its non-recursive counterpart, recursiveness alleviates congestion frictions (which take place when more than one buyer tries to purchase from the same capacity-constrained seller, forcing rationing) and worsens mismatch frictions (which take place when low-valuation consumers and/or high-cost sellers are matched, crowding out other, more remunerative trading matches), but the first force dominates, thereby inducing higher efficiency. Conversely, recursive trading can reduce efficiency when agents are highly heterogeneous. In this setting, recursive trading worsens mismatch frictions much more than it lowers congestion frictions, thereby reducing market efficiency. But these insights rest on the assumption that agents play a sub-game perfect Nash equilibrium. This assumption can be strict, considering bounded rationality or other forces that may induce behavioral deviations from equilibrium. In this study, we use a laboratory experiment to test the extent to which subjects play the predicted equilibrium and its implications for the relative performance of competing market designs. We find that subjects deviate from the Nash prediction and that such deviations reverse our theoretical predictions regarding the relative performance of competing designs. Most prominently, a recursive design always increases market efficiency and decreases distributional equity (i.e., the extent to which both sellers and buyers of water rights benefit from trade), regardless of the agents’ degree of heterogeneity in water valuations.