Kessler, Judd B. and Alvin E. Roth, Organ Allocation Policy and the Decision to Donate," June 2010. (As an added bonus, Judd is on the jobmarket this year.)
Here's the abstract: "Organ donations from deceased donors (cadavers) provide the majority of transplanted organs in the United States, and one deceased donor can save numerous lives by providing multiple organs. Nevertheless, most Americans are not registered organ donors despite the relative ease of becoming one. We study in the laboratory an experimental game modeled on the decision to register as an organ donor, and investigate a variety of strategies for increasing the donation rate. We find that an organ allocation policy giving priority on waiting lists to those who previously registered as donors has a significant positive impact on registration.
And this is from the concluding section
"Before further considering the benefits of the priority rule, it is worth noting that there are other ways to change policy that could positively affect the number of registered organ donors. For example, one proposal that has received a good deal of attention would change the current “opt in” registration method used in the United States to an “opt out” system in which everyone is presumed to be a donor unless he or she actively indicates otherwiseAnother proposal, “mandated choice” would require everyone (e.g. who applies for a driver’s license) to specifically indicate whether they wished to be a donor or not. We want to briefly argue here that the priority rule that we consider may create a more direct link between registration as an organ donor and subsequent successful organ recovery and transplantation than policies that change the procedure by which individuals register as organ donors.
Attempts to increase organ donation rates by changing the default organ registration status (and adopting an “opt out” policy) would surely generate more registered organ donors since those who do not take any explicit action would automatically be registered as donors (see Johnson and Goldstein 2003, 2004, who find direct evidence that registration rates are higher with an “opt out” system).[2] However, such a policy may weaken the link between the registration decision and the legal clarity of the potential donors’ last wishes. Under current United States gift law, changing the default status is likely to have legal consequences that could be detrimental to organ retrieval.
Since the Uniform Anatomical Gift Act of 1968 (UAGA), an individual can make his or her own legally binding decision to be an organ donor after death, which does not require the consent of next of kin (Glazier 2009). However, a donor symbol on a driver’s license has not been considered sufficient evidence of the deceased’s intent to donate in order to proceed without permission from the next of kin. Aside from the fact that the driver’s license is often not available in a timely way, the law allowed that a registered donor could have changed his or her mind about donation subsequent to the issuance of the driver’s license (Glazier 2006).
In recent years, computer registries have allowed for fast checks of organ registration status. They also provide individuals with a way to easily change their organ donor status online, which allows the presence in the registry to be interpreted as intent to donate. The legal status of the anatomical gift has meant doctors can recover donated organs without receiving explicit permission the next of kin (see Glazier 2006). In contrast, a donor registration that does not reflect a positive decision to donate (as under an “opt out” policy) may not be taken as evidence of the deceased’s intent in the legally compelling way that registration does currently. Under an opt out policy, approval from next of kin might again become necessary for an organ to be transplanted.[3]
A "mandated choice" system would also change the way in which individuals became registered donors (see Thaler and Sunstein 2008 and Thaler 2009). Under “mandated choice,” every individual who registered for a driver’s license (or potentially other state or federal documentation) would be required to indicate that he will be an organ donor or that he will not. While there is evidence that a “mandated choice” policy would (like “opt out”) generate more registration of organ donors (Johnson and Goldstein 2003, 2004), similar concerns arise about whether a change to mandated choice would lead to more donated organs and transplants. While the UAGA makes registering to be a donor legally binding under an “opt in” policy, failing to register as an organ donor is not a legally binding decision, whereas registering as a person who declines to donate would likely be legally binding on the next of kin. Discussions with the staff at the New England Organ Bank suggests that they are able to recover organs from about half of all non-registered potential donors in New England by approaching next of kin. This means that more than half of the people who are not currently registered under “opt in” would need to choose “yes” in mandated choice to increase the recovery rate.[4] Consequently, it remains an empirical question whether a change to “mandated choice” would generate more organ transplants.
Even though registration under “opt out” and “mandated choice” systems may raise legal concerns about the intent of registrants under the UAGA, changing the procedure by which individuals register as donors may still be a fruitful avenue to pursue to increase organ donation and recovery.[5] Gift laws can also potentially be changed to address any legal concerns that might arise from new policies. We simply see these legal issues as additional hurdles to monitor and overcome in successfully implementing a change in registration policy. One attraction of the priority rule is that it seems to avoid these additional hurdles since it preserves the current donor registration process as is (and thus is consistent with current United States law regarding donor intent at time of death).
While comparing the different mechanisms in our experiment, the priority rule, rebate, and discount all generate an incentive to donate that offsets the costs of donation. But the priority rule has two advantages over the rebate and discount both inside and outside of the laboratory. First, the simplicity and elegance of the priority rule (as suggested by its outperformance of the rebate and discount at the start of the game) suggests that its benefits are particularly clear and salient. Second, and more importantly, the priority rule is feasible to implement and can be implemented without any additional costs to the system. In contrast, decreasing the costs of registering to be an organ donor is difficult (it is difficult to both understand the costs and to decrease them) and providing a rebate through the form of monetary incentives is not currently allowed by the U.S. National Organ Transplant Act and by similar legislation in many countries.
...
The priority rule used in Singapore appears to be a powerful policy tool. Results from this experiment suggest that it performs as well as or better than discounts and rebates that are of a similar magnitude to the benefits of priority. It is a plausible mechanism to increase rates of registration, and policy makers should consider allocation policy along with other policies to generate more organ transplants."