There were two interesting developments around universal basic income schemes in north America this week. (I’ve previously covered some other stories in this space here.) First, the government of Ontario confirmed that it plans to launch a UBI pilot project in early 2017. Notably, the pilot was designed by a conservative political strategist, and follows on the earlier success of a social pension program for the elderly.
[Strategist Hugh] Segal’s interest in the idea was sparked in the mid-1970s. A series of news stories documenting high rates of seniors living in poverty in the province – including one report of some resorting to eating pet food to get protein in their diets – had ramped up pressure on lawmakers to address the issue. The response was a basic income policy for seniors in the province.
The policy sent poverty rates among seniors in Ontario downwards, from the low 30s to 5%, and sparked a slew of ripple effects. “Food security went up, longevity went up, independence of the healthcare system in terms of needing long-term care, all those indicators went up,” said Segal. The program soon spread across the country.
Second, Marginal Revolution pointed to a new paper by David Price and Jae Song evaluating the Seattle-Denver Income Maintenance Experiment (SIME / DIME), which gave several thousand US families up to $25,900 in annual grants over 3 – 5 years in the 1970s. Because of some program rules involving taxation of outside income, the average increase in an individual’s income during the treatment period was about $2400 annually. The authors used data from the Social Security Administration to track the earnings of the experiment participants over the next 40 years. Their findings are somewhat counterintuitive:
We find that treatment caused adults to earn an average of $1,800 less per year after the experiment ended. Most of this effect on earned income is concentrated between ages 50 and 60, suggesting that it is related to retirement. Treated adults were also 6.3 percentage points more likely to apply for disability benefits, but were not significantly more likely to receive them, or to have died. These effects on parents, however, do not appear to be passed down to their children: children in treated families experienced no significant effects in any of the main variables studied.
They find that treated adults were slightly less likely to work whilst receiving the grants, but no less likely in the period immediately after the program ended. There’s some circumstantial evidence that the reduced lifetime earnings might have been caused by earlier retirement, although it’s also possible that time out of the labor force during the SIME / DIME program years led to lower wages afterwards.
For me, the headline point about this experiment was that it actively discouraged work effort during the program period by taxing non-grant income at marginal rates of 50% – 80%. This is a bad design choice if you care about continued engagement with the labor market, and it’s not clear to me why the program’s designers made it. At any rate, this is very different to most modern proposals for a universal basic income, which are intentionally not conditioned on recipients’ employment status. If the drop in lifetime earnings was driven by time out of the labor market during the SIME / DIME program period, it should be clear that these results aren’t generalizable to UBI proposals which are employment-agnostic. Of course, if the program lowered lifetime earnings due to early retirement instead, the concerns raised here may still be relevant. I’m hard pressed to understand how receiving receiving an extra $2400 for 3 – 5 years in the 1970s would affect one’s retirement decisions potentially decades later, however.