UNITED STATES: Y Combinator releases proposal for expanded study of basic income

Silicon Valley’s Y Combinator has concluded its pilot in Oakland and released a draft proposal for a large-scale randomized control trial of basic income in the United States.

In January 2016, Silicon Valley entrepreneur Sam Altman announced his intention to spearhead a privately funded trial of unconditional basic income in the United States, hiring social work and political science PhD Elizabeth Rhodes as Research Director later in the year, and eventually assembling a team of expert advisors.

Since this time, Y Combinator has conducted a feasibility study in Oakland, California, and is now working to finalize the design of its full scale experiment. (Contrary to some misconceptions, the Oakland project was not itself an experiment. Its purpose was merely to test and fine-tune the mechanisms for conducting the experiment–such as the selection of participants, disbursement of funds, and collection of data–not to analyze the effects of unconditional cash transfers on recipients. The latter will be the goal of the project described in the new research proposal, which has yet to be launched.)  

Although some details of the experiment remain to be decided, including the precise outcome variables and methods of data collection, Y Combinator has decided to design the experiment as a randomized controlled trial, conducted on a random sample of poor and low-income young adults from two US states (using a stratified sample to ensure adequate representation across race, gender, and income categories).

On the tentative design, the researchers will select a total of 3000 participants, randomly assigning 1000 to the treatment group–who will receive a regular cash payment of 1000 USD per month unconditionally for the duration of the experiment–and the remaining 2000 to the control group. (Individuals in the control group will provide the same type of feedback and data to researchers but receive only a much smaller cash payment, tentatively set at 50 USD per month.) The experiment is planned to continue for three to five years.

Y Combinator expresses an interest in a “holistic approach to understanding the individual-level effects of basic income”, in contrast to past and present experiments which have focused on the labor market impacts of unconditional cash payments, such as Finland’s current basic income experiment and the negative income tax experiments conducted in the United States in the 1970s. Among these individual-level effects, the research group is particularly interested in time use, mental and physical health, subjective well-being, financial health, decision making and attitudes toward risk, as well as  political and social attitudes. Furthermore, although individual-level effects will be the focus of the experiment, researchers also hope to examine spillover effects on recipients’ families, friends, and communities.  

While the research group has not finalized its choice of data sources and collection methods (see its project proposal for a discussion of possibilities currently under discussion), it plans to combine quantitative analysis with regular surveys and interviews (in contrast, for example, to the Finnish experiment, in which researchers have abjured the use of surveys and interviews during the duration of the experiment). Rhodes has explained, however, that participation in surveys and interviews will be voluntary for participants; that is, the payments will continue for the duration of the experiments even if recipients do not respond to requests for data and information.   

The research team acknowledges that the experiment does not, strictly speaking, test a universal basic income. For one, as mentioned, the sample will be limited to young adults (aged 21 to 40) with incomes below the area median. The researchers justify this limitation, however, by noting that “the marginal effect of the additional income on many of the outcomes is expected to be relatively small at higher income levels” and that, under most plans, “the benefit received by higher-income individuals would be paid back in taxes in order to fund the program”.

Additionally, due to the use of a randomized controlled trial, the research will not capture multiplier effects that might result from the implementation of a universal basic income (in contrast, for example, to the saturation study in Dauphin, Manitoba, or GiveDirectly’s recently launched village-level RCT of basic income in Kenya). However, researchers note that “ the intervention is very expensive and our sample size is constrained by the budget. We will not have enough statistical power to detect effects with a geographically saturated study and the increase in sample size required to allow for clustering is financially infeasible.”

To conduct the experiment, Y Combinator has partnered with the Center on Poverty and Inequality (CPI) at Stanford University. The research has been approved by Stanford’s Institutional Review Board for research involving human subjects.

Y Combinator is currently working with state and local governments to coordinate mechanisms for distributing payments without affecting recipients’ future eligibility for existing government benefits, and to obtain the use of registries to collect individual data.

With many details still to be settled, no specific launch date has been set for the experiment (although Rhodes stated at the recent BIEN Congress that the research group hopes to begin the study in “early 2018”), and the states from which subjects will be sampled have not been publicly announced.   

The full research proposal can be read on Y Combinator’s blog (see “Basic Income Research Proposal,” published September 20, 2017).

The organization invites comments and feedback on its project proposal.

Reviewed by Dawn Howard

Photo (Martin Luther King Jr. Regional Shoreline, Oakland) CC BY-NC-ND 2.0 MagicMediaProduction