Chris Conover has the definitive case for the supposition that the Affordable Care Act is responsible for growth in part-time employment. I highly encourage you to go read it because I consider this a good faith effort to make that case. However, I still have my doubts — not whether the ACA is or will have an any effect at all but by how much. For the sake of brevity I’ll just list my thoughts:
- Anecdotes are arguments against the idea that there are zero people who will or are working fewer hours in part-time jobs because of the employer mandate provision. I haven’t argued that. For the most part the people I read haven’t argued that. Conover specifically links to this post as evidence of someone arguing there is no effect, but Spencer England simply shows that the data doesn’t support the weight some (like Conover) are giving to such anecdotes.
- Conover admirably attempts to parse the static of BLS data to show that the ratio of part-time employment to full-time is quite large for this year so far. Yet he didn’t even have to go that far given that the information doesn’t necessarily require parsing — part-time employment has boomed this year. It’s not a secret. In the industry for which most of the anecdotes are coming from employment growth has been strong for some time. Yet here, as well as elsewhere in the article, I think he misleads readers by assuming that whatever changes have occurred are Obamacare’s fault. Apparently it’s so blatantly obvious that the chart has to have a title saying as much. But what’s the actual evidence? It’s 2013. That’s it. This is a causal inference, which is fine, but it’s most definitely not “proof-positive.”
- He dings the CPS analysis I wrote of yesterday by noting the “lookback” option in regulations that allow employers to choose how far back in 2013 the IRS should go in determining their employees weekly hourly status. Again, as I noted at the end of that post, this doesn’t necessarily disqualify using CPS data but it does bring up the possibility of it being less relevant than the full year. However, to say that due to the lookback option we can’t just look at partial data for 2013 and infer non-effect is to also say we can’t use it for effect. I don’t necessarily agree with this, but if that’s the case then Conover using BLS data for a partial 2013 should also fall under this qualification.
- I’ll say it again; I think it would behoove everyone to recognize that attempts to discount the ACA’s (large) effects on part-time work aren’t being made in a vacuum. They’re largely in response to claims such as these. That is, they’re an argument against prophecies of economic apocalypse, the end of small-biz, etc — NOT an case against zero adversity.
- Be careful when you make ethical appeals. It’s disingenuous to appear aghast at the left by twisting an empirical argument (i.e., this is about a small portion of the labor force) into a moral assumption that the left doesn’t care about a (still) large number of absolute people in question. This is absurd. It’s also hypocritical if your preferred alternative to the incidence of involuntary underemployment is to therefore deny millions of Americans access to affordable health insurance.
Did I miss anything?