Pilot Projects are key to an effective UBI

Natural Experiments: In some cases, “control” and “treatment” groups are created organically, by the operation of law or policy. Say that a country’s UBI stipend is distributed only to individuals born after January 1, 1970. Individuals just before the eligibility date can be put into a “control” group while individuals just after it can be put into a “treatment” group. Differences in their outcomes can be attributed to the treatment. To take an example from the patent system, patent rights are territorial, and the same patent application is often filed in multiple patent offices by applicants seeking coverage in different countries. This creates a natural experiment involving “identical twin” applications, by which the impact of different forms of examination on the same application can be compared. “Lab” or Simulated Experiments: Another way to carry out controlled policy pilots is through “lab” or simulated experiments. To test how best to distribute a UBI, for example, one could imagine an agency contracting with a lab to run an experiment, as the Consumer Protection Financial Bureau (“CFPB”) has done.47 The pilot could, for example, vary the way that the UBI is distributed, e.g., through a credit card with all the money up front, or through a regular cash disbursement. Any differences in outcomes pertaining, for example, to the well-being and employability of the participants could then be observed.48 Though doing so wouldn’t generate the same information as a full-fledged field pilot, such a pilot could provide suggestive evidence of the impact of the distribution method, if any. Of relevance to the patent system, laboratory methods have been used to test the impact of changes in the standard of proof required to invalidate a patent49 as well as to test the impact on innovation of patent versus copyright-like incentives.50 The USPTO has also created pilots that applicant can opt into.51 Like lab experiments, voluntary programs simulate but differ in important ways from mandatory programs

Experimenting with Internal Agency Procedures: "Field" experiments vary the experiences of regulated entities. One way of doing so is by experimenting with internal agency procedures. In the case of UBI, for example, one could imagine trying different ways of disbursing stipends to individuals, e.g., through a high-or low-touch process. The outcomes of individuals that receive money one way could then be compared with the outcomes of individuals that receive it another way. In a similar vein, how the Patent Office applies the resource of patent examination can and has been varied across applications, with certain applications getting more scrutiny than others. Rigorously comparing the outcomes of otherwise equivalent applications can provide suggestive evidence the impact of the extra scrutiny.

Experimenting with the Laws and Rules that Apply to Private Parties: Another way to experiment is by varying the rules that apply, not to internal agency procedures, but to the behavior of private citizens. For example, some citizens in the Finnish UBI pilots were entitled to enroll in the partial basic income program while others were not. In related work, I propose a patent pilot program in which certain applicants are entitled to waive or defer the patent eligibility requirement, and some are not. Comparing the outcomes of "treated" applications with the outcomes of "untreated" applications can generate information about the operation of this legal standard, in view of the other standards of patentability. An agency example comes from the Office of the Comptroller of the Currency ("OCC") which, from 1999-2001 extended special lending limits to 277 community banks as part of a pilot program and compared their outcomes with companies not receiving the treatment. It is worth pausing here to note that of these four classes of pilots only the last two involve the mandatory application of different procedures or rules to similarly situated entities.When a pilot is implemented as a temporary program that applies to all, equal protection claims are avoided. Differences in treatment that follow the decision to opt into a pilot are the result of free choice, not government discrimination, the Supreme Court has said. These sorts of pilots do not raise the same fairness considerations that the third and fourth categories do and are likely to be less controversial. Even if they do not vary law or policy, they can still be useful for informing law or policy.