Sean Crockett

An Experimental Test of the Lucas Asset Pricing Model

We implement a dynamic asset pricing experiment in the spirit of Lucas (1978) with storable assets and non-storable cash. In one treatment we impose diminishing marginal returns to cash to incentivize consumption-smoothing across periods, while in a second treatment there is no induced motive for trade. In the former case subjects smooth consumption, and assets trade at a discount relative to the risk-neutral fundamental price. This under-pricing is a departure from the “bubbles” observed in the experimental asset pricing experiments of Smith et al. (1988). In our second treatment with no induced motive for trade, assets trade at a premium relative to expected value and shareholdings are highly concentrated.We use laboratory experiments to test a consumption-based general equilibrium model of asset pricing, which posits that agents buy and sell assets for the purpose of intertemporally smoothing consumption. Such asset pricing models are widely used by macroeconomists and finance researchers but have not yet been subjected to experimental testing. In the experiments we induced several features which, according to the theory, determine asset prices, such as risk and time preferences and the process for income and dividend payments. Our analysis indicates that intertemporal consumption-smoothing strongly inhibits the formation and magnitude of asset price bubbles, a stark departure from most recent asset pricing experiments. In fact, when subjects are motivated to smooth consumption, assets trade at a discount relative to their expected value and markets are thick; when this condition is eliminated in an otherwise identical economy, assets trade at a premium in thin markets. For useful comments we thank Elena Asparouhova, Peter Bossaerts, Craig Brown, John Geanakoplos, Stephen Spear and seminar participants at Claremont Graduate University Department of Politics and Economics, the University of Utah Department of Finance, the 2009 CARESS-Cowles Conference on General Equilibrium and its Applications, the 2009 Society for the Advancement of Economic Theory Conference, and the 2009 Economic Science Association North American Regional Meeting. Contact: sean.crockett@baruch.cuny.edu Contact: jduffy@pitt.edu

Extreme Walrasian Dynamics: The Gale Example in the Lab

We study David Gales (1963) economy using laboratory markets. Tatonnement theory predicts prices will diverge from an equitable interior equilibrium toward infinity or zero depending only on initial prices. The inequitable equilibria determined by these dynamics give all gains from exchange to one side of the market. We show surprisingly strong support for these predictions. In most sessions one side of the market eventually outgains the other by more than 20 times, leaving the disadvantaged side to trade for mere pennies. We also find preliminary evidence that these dynamics are sticky, resisting exogenous interventions designed to reverse their trajectories. (JEL C92, D50)

Price Dynamics in General Equilibrium Experiments

General equilibrium (GE) theory faces several related challenges. Classical theories of out‐of‐equilibrium adjustment are intuitive but implausibly centralized. Further, standard restrictions on individual preferences place little structure on aggregate excess demand functions. Amongst other issues, this fact implies that economies can be unstable under classical dynamics, and thus instability appears to be an unavoidable feature of GE theory. This paper recounts laboratory studies of GE environments which reveal that classical theories of dynamics do, in fact, organize market activity surprisingly well. These experiments suggest that the Walrasian auctioneer, for example, appears to be a constructive abstraction, and that instability should be accommodated within theory rather than circumscribed. Laboratory applications of GE theory focused on issues beyond stability and dynamics are also briefly surveyed.

Momentum Trading Through Reference Dependent Preferences

The endowment effect is a well-known behavioral regularity in which a person values a good more when he is endowed with it. In their generalization of prospect theory to consumption bundles with multiple attributes, Tversky and Kahneman [1991] imply the endowment effect as a consequence of loss aversion and diminishing sensitivity in gains. It has since frequently been presumed that this form of reference dependent preferences will inhibit trade. However, in this paper it is demonstrated that loss aversion and diminishing sensitivity in gains also imply a dynamic momentum trading effect that increases exchange, so the net effect of such preferences on trading volume is ambiguous. In fact, the momentum trading effect is shown to completely cancel out the endowment effect in an important class of examples.