We use the 2004-'05 wave of the Australian National Health Survey to estimate the impact of private hospital insurance on the propensity for hospitalization as a private patient. We employ instrumental-variable methods to account for the endogeneity of supplementary private hospital insurance purchases. We calculate moral hazard based on a difference-of-means estimator. We decompose the moral hazard estimate into a diversion component that is due to an insurance-induced substitution away from public patient care towards private patient care, and an expansion component that measures a pure insurance-induced increase in the propensity to seek private patient care. We find some evidence of self-selection into insurance but this finding is not robust to alternative specifications. Our results suggest that on average, private hospital insurance causes a sizable and significant increase in the likelihood of hospital admission as a private patient. However, there is little evidence of moral hazard; the treatment effect of private hospital insurance on private patient care is driven almost entirely by the substitution away from public patient care towards private patient care.

We use the 2004-'05 wave of the Australian National Health Survey to estimate the impact of private hospital insurance on the propensity for hospitalization as a private patient. We employ instrumental-variable methods to account for the endogeneity of supplementary private hospital insurance purchases. We calculate moral hazard based on a difference-of-means estimator. We decompose the moral hazard estimate into a diversion component that is due to an insurance-induced substitution away from public patient care towards private patient care, and an expansion component that measures a pure insurance-induced increase in the propensity to seek private patient care. We find some evidence of self-selection into insurance but this finding is not robust to alternative specifications. Our results suggest that on average, private hospital insurance causes a sizable and significant increase in the likelihood of hospital admission as a private patient. However, there is little evidence of moral hazard; the treatment effect of private hospital insurance on private patient care is driven almost entirely by the substitution away from public patient care towards private patient care.

We provide a model in which small and relatively isolated communities can successfully manage local commons informally in circumstances where larger or less isolated communities could not do so. The reason for this is the non-anonymous nature of many interactions between the members of a small and isolated community. Such communities may be able to use these multiple interactions to enforce informal restrictions on the usage of local commons. To the extent that the process of economic development reduces the number of non-anonymous interactions among community members, it will reduce the ability of the community to successfully manage the local commons informally. The resulting need for either explicit regulation or the introduction of private property rights represents a hidden cost of development.

Many service industries, including the medical and legal professions in some countries, display a gated structure. Rather than approaching a …nal producer directly, a consumer will …rst seek a referral from an intermediary. In this paper, we provide one possible explanation for such an industry structure. If the outcome of a transaction depends on producer e¤ort, which is unobservable and unveri…able, then the market may fail to generate a Pareto optimal outcome. This is the standard moral hazard problem. If consumers had a long-run relationship with producers, this type of market failure might be avoided. However, in some industries, consumers will only have a short-run relationship with producers. A gatekeeping intermediary may provide an opportunity for reputation e¤ects to apply in such a setting. By aggregating many potential consumers, gatekeeping intermediaries can create an arti…cial long-run relationship between a consumer and a producer. This long-run relationship reduces the incidence of shirking on the part of the producer.

Many service industries, including the medical and legal professions in some countries, display a gated structure. Rather than approaching a …nal producer directly, a consumer will …rst seek a referral from an intermediary. Such an industry structure might help to alleviate adverse selection problems between parties that interact infrequently. Intermediaries aggregate many short-run transactions between various consumers and a particular producer. As such, they might be able to learn a producer's level of pro…ciency more rapidly than an individual consumer. However, the presence of a positive information externality means that too few consumers will seek a referral. As such, some form of regulation to encourage consumers to seek a referral might be warranted.