DO SUNK COSTS MATTER?

That sunk costs are not relevant to rational decision making is often presented as one of the basic principles of economics. When people are influenced by sunk costs in their decision making, they are said to be committing the “sunk cost fallacy.” Contrary to conventional wisdom, we argue that in a broad range of situations, it is rational for people to condition behavior on sunk costs because of informational content, reputational concerns, or financial and time constraints. Once all the elements of the decision‐making environment are taken into account, reacting to sunk costs can often be understood as rational behavior. (JEL D0, D01, D8, D81, D83, D9, D90)     

Control Variates for the Metropolis–Hastings Algorithm

Abstract.  We propose new control variates for variance reduction in estimation of mean values using the Metropolis–Hastings algorithm. Traditionally, states that are rejected in the Metropolis–Hastings algorithm are simply ignored, which intuitively seems to be a waste of information. We present a setting for construction of zero mean control variates for general target and proposal distributions and develop ideas for the standard Metropolis–Hastings and reversible jump algorithms. We give results for three simulation examples. We get best results for variates that are functions of the current state x and the proposal y , but we also consider variates that in addition are functions of the Metropolis–Hastings acceptance/rejection decision. The variance reduction achieved varies depending on the target distribution and proposal mechanisms used. In simulation experiments, we typically achieve relative variance reductions between 15% and 35%.     

THE EFFECTS OF AUTOMOBILE RECALLS ON THE SEVERITY OF ACCIDENTS

The number of automobile recalls in the United States has substantially increased over the last two decades, and after a record of over 30 million cars recalled in 2004, in the last few years it has consistently reached between 15 and 17 million, and in 2010 alone 20 million cars were recalled. Toyota's recall crisis in 2010 illustrates how recalls can affect a large number of American drivers and the defects connected to them can result in loss of life and serious accidents. However, in spite of the increase in public concern over recalls and the loss of property and life attached to them, there is no empirical evidence of the effect of vehicle recalls on safety. This paper investigates whether vehicle recalls reduce accidental harm measured by the severity of injuries in vehicle accidents. The results of our analysis show that if a recall for a new‐year model is issued, then the severity of injuries of accidents continuously diminishes during the first year after the recall, something we do not find among cars not subject to recalls. This is because defects are repaired over time but also because drivers react by driving more carefully until the defects are fixed. To minimize the losses attached to having dangerously defective cars on our roads, both quick and timely recall issuance are needed and more detailed information on defects should be delivered to owners of defective vehicles. The latter can be made possible through simple but important policy changes by the U.S. government regarding recall information sharing with drivers and insurance companies. (JEL L51, L62)