February 15, 2015
As regular readers know, the Washington Post editorial board has problems with economics. They were foremost among the Very Serious People who warned about financial crises and soaring interest rates if we didn’t tame the deficit. They still regularly issue demands for what they consider fiscally responsible policies (e.g. cutting Social Security and Medicare).
Anyhow, today they used their lead editorial to wag their finger at supporters of the Affordable Care Act (ACA) for not acknowledging that it would lead many employers to cut workers’ hours. The issue is a provision in the law (which has yet to be applied) that would require large employers to provide insurance for workers who work more than 30 hours per week or to face a fine.
The editorial noted a directive from Staples to its store managers to restrict part-time workers to less than 25 hours as evidence of this ACA effect. The piece then cited work by Ben Casselmen to support its “Iron Law No 1: Incentives influence behavior”:
“In 2009, 9.7 percent of part-timers worked between 25 hours and 29 hours and 7.7 percent worked between 31 and 34 hours. In about mid-2013, just before the employer mandate’s original implementation date, the gap between those numbers began to widen, hitting 11.1 percent and 6.6 percent, respectively, by year’s end.”
Are you impressed by that iron? Let’s add some more details.
First, the editorial gives us these figures as a share of part-time employment. That’s a strange denominator since most Post readers have no clue how many people are working part-time and this number actually changes quite a bit over this period. If the point was to provide information rather than establishing the veracity of Iron Law No 1, the editorial might have told readers what these numbers were as a share of total employment. (The editorial does later note total employment of 150 million, but never directly puts the part-time numbers in context.)
There were over 28 million people working part-time in 2009. This had fallen to around 26.5 million by the end of 2013. This means that the total number of people putting in between 25 and 29 hours per week went from roughly 2.7 million to 2.9 million over this period. (I realize that Casselmen’s analysis uses less than 40 hours a week as the definition of part-time rather than the standard 35 hours. This would probably raise these numbers by about 10 percent.) As a share of total employment, rather than part-time employment, the increase in the portion of people working between 25-29 hours would have been from roughly 1.95 percent to 2.02 percent.
A second point is that there is a notable drop in this period in the number of people who report working part-time involuntarily. In 2009 roughly 9 million people reported working part-time against their will. This had fallen to less than 7.9 million by the end of 2013, a drop of roughly 12 percent. (This number had fallen further to 6.8 million in the most recent data.) With over 70 percent of the people who work part-time saying that they do so by choice, the odds are that most of the people working between 25-29 hours are doing so because they would rather not work more hours, not because of restrictions on hours imposed by their employer.
This brings us to the final point, the story doesn’t fit with the actual rules on the ACA. The only period in which employers had reason to believe that the sanctions from the ACA would be in effect was the first six months of 2013. In July of that year the Obama administration announced that it would not apply the employer sanctions for 2013 or 2014. As a result, employers had no reason to cut back hours in the second half of 2013 or 2014 based on the ACA.
Yet, almost the entire increase that Casselmen finds in the share of workers putting in between 25-29 hours occurs in the second half of 2013 and into 2014, periods in which employers knew that the sanctions would not apply. That doesn’t sound like a very good job of responding to incentives. (Helene Jorgensen and I analyzed the data for the six months when employers expected the sanctions to apply. We found a small increase in the share of workers putting in 26-29 hours a week, but it was entirely at the expense of a decline in the share working fewer hours. The share working more than 30 hours had increased compared to the same months in 2012.)
This raises another issue. There is an increase in voluntary part-time employment in 2014, which is almost certainly due to the ACA. We found a substantial rise in the number of young parents who report working choosing to work part-time. This was presumably due to the fact that many of these parents are now able to get insurance through the exchanges or Medicaid and therefore no longer have to work at a full-time job to get insurance through their employer. This is likely to lead to an increase in the number of workers putting in 25 to 29 hours a week, however this is not an issue that would be troubling to most people. Most people would probably think it’s a good thing that parents could spend more time with young children.
Anyhow, we wouldn’t want to cause too much trouble for the Post’s iron laws. Undoubtedly some employers have considered the costs imposed by the ACA sanctions in determining hours for their workers. But they have many other considerations in determining hours and the ACA sanctions are not likely to be the most important in most cases. As a result, the impact on hours worked is likely to be small, which is what the data suggest.
Comments