It follows that if an individual has constant absolute risk aversion (CARA), it implies that [r.sub.1] = [r.sub.2], then A < 0 and [[dq.sup.T].sub.H]/dt < 0.

Thus i [greater than or equal to] r under NIARA, with equality holding in the case of constant absolute risk aversion. On the other hand, the sign of p - r is ambiguous, and is the same as that of [g.sub.F]/[g.sub.L][g.sub.LL] - [g.sub.LF].

Indeed, constant absolute risk aversion is sufficient for own-price LeChatelier effects for labor demanded, since in the absence of wealth effects, risk preferences have no direct bearing on factor demands.

Specifically, the constant absolute risk aversion (CARA) utility function is used in this analysis.(3) Each taxpayer's preferences are represented by the von Neumann-Morgenstern utility function u(y) = -[e.sup.-ay], so the Arrow-Pratt measure of absolute risk aversion is A(y) = -u[double prime] (y)/u[prime] (y) = a.

As long as applicants do not exhibit constant absolute risk aversion, the hidden differences in wealth give rise to hidden differences in risk aversion.

Previous experimental studies that analyze (or control for) subjects' risk attitudes employ both constant absolute risk aversion (CARA; for example, Berg, Daley, Dickhaut and O'Brien [1986]) and constant relative risk aversion (CRRA; for example, Cox, Smith, and Walker [1988]) functional forms.