For the last week or so, I’ve been preparing to speak against a Universal Basic Income (UBI) motion at a debate club.  Here’s the Motion:

The Motion: This House Supports a Basic Income for All US Residents

Motion Summary:  Basic income recipients would include children and adults; the employed and unemployed; and citizens, permanent residents, and all other residents who could prove a residency duration of at least three years. The amount given would start at $1,000 per person per month and be pegged to GDP growth going forward. No programs in the existing social safety would be replaced by this policy.

The last five posts addressed the above UBI’s cost, tax implications, and effects on labor market participation. In the most recent post, I estimated how many US citizens and residents would be tempted to stop working, reduce their work hours, take longer between-job breaks, or plan gap years from work should each member of their household, child and adult, receive $1000 a month. Long story short: 66 million, give or take. Check the post for more details.

Here we're looking at how government benefits influence the decision to work, especially when the benefits are not means-tested or not means-tested until a significant income threshold is reached. UBI advocates often argue that government benefits only disincentivize work when they’re means-tested or stopped if the recipient gets a job. They argue that if you eliminated this “work-penalty”, there’d be no work disincentive.  For proof, they point to Alaska, where the state issues a yearly “Permanent Fund Dividend” to most residents. For example, my UBI debate opponents provided a link to a Quartz article with the headline  “When you give Alaskans a universal basic income, they still keep working”, in which we are told:

“Since 1982, the fund has sent a dividend check to every Alaskan resident. In recent years, it’s been up to $2,072 per person, or $8,288 for a family of four”…[Researchers] found that full-time employment did not change at all, and the share of Alaskans who worked part-time jobs increased by 17%.”

 A few points:

The $2072 Alaska dividend was in 2015 and that was the largest dividend Alaskans have received, before or since. In 2017, the dividend was $1100.

The Alaska dividend is received just once a year. So it's like a tax refund.  But most large expenses are incurred on a monthly basis (think rent, utilities, mortgage, car payments, credit card bills) and the annual Alaska dividend looks less impressive when divided by twelve. The 2017 dividend averaged less than $100 a month per resident. The most generous dividend of them all, in 2015, averaged just $173 a month per resident.

According to the US Census Bureau, the median household income in Alaska is $74,444. That’s much higher than the lower-48, because it’s expensive to live in Alaska. For instance:

Median home value is 37% higher than the U.S. average.

The average food budget for a family with two parents and two children is $10,627. 

The Quartz article is based on a recent paper by Damon Jones and Ioana Marinescu, in which the authors breezily dismiss the possibility that the size of a basic income matters - that is, that the effects might be different depending on how big the UBI is. For support, they cite one not-yet-published paper (Cesarini et al) on the labor market effects of the Swedish lottery. Jones and Marinescu also neglect to seriously consider alternative explanations for the rise of part-time employment in Alaska, which they attribute to the Alaska dividend (aka a "basic income" in all but name).  Rather than look at real-world changes in employer demand for part-time workers, they rely on "synthetic-controls" to reject alternative explanations for the rise in part-time workers. Finally, Jones and Marinescu make it very clear from the first paragraph of their paper that they support a UBI. That their conclusions match their preferences comes as no surprise.

Bottom line: Alaska’s “basic income” wouldn't even pay for one year’s food bill, much less housing or transportation expenses. It really isn't comparable to the type of UBI schemes being proposed for developed countries like the US. There is a world of difference between $100/month per person and $1000/month, Swedish lotteries be damned.

References:

Cesarini, D., E. Lindqvist, M. J. Notowidigdo, and R. Ostling (Forthcoming): “The effect of wealth on individual and household labor supply: evidence from Swedish lotteries,” American Economic Review.

Damon Jones and Ioana Marinescu The Labor Market Impacts of Universal and Permanent Cash Transfers: Evidence from the Alaska Permanent Fund.  NBER Working Paper No. 24312. Issued in February 2018. http://www.nber.org/papers/w24312