As you probably know, the Congressional Budget Office (HBO), a respected, non-partisan research arm of the U.S. Congress, recently issued an economic projection related to the effects of the Affordable Care Act (ACA); a.k.a Obamacare. Here is the complete study. http://www.cbo.gov/publication/45010. Stories about the study were hardly buried; but accurate accounts, with context, have been scarce
The CBO has been dragged into “scoring” ACA right from the start. I don’t envy their position. Their work has angered and elated parties on both sides of the health reform issues, suggesting they’ve probably been as fair as can be trying to score a policy change with many moving parts and unpredictable outcomes. This latest effort by the CBO, however, appears gratuitous, and stretches economic forecasting almost to the point of silliness. The results were also poorly communicated by CBO, and then mangled further by the mainstream press. Of course the results were molested by partisan propagandists.
Even in an enterprise inherently prone to large errors – forecasting long term fiscal and economic impacts of new policies — the CBO could have handled this a lot better (or touched it more lightly). They are now in recovery and walk back mode. That said, organizations like CBO have enormous value. If they didn’t do the scoring, the job would go by default to partisan propagandists, or to competent and well meaning “private” or academic economists who, alas, often don’t understand the policies, the government data, and the budget environment very well. Without the CBO, our information would be a lot worse.
Here is (a perhaps dangerous) effort at a concise capture of the CBO report as it applied to the ACA; it won’t do it full justice. But we have to start somewhere. The report says that, by 2024, the equivalent of 2.5 million workers in the U.S. will choose not to work, i.e. withhold their labor, because it makes more sense for them to quit working so they can qualify for subsidized health insurance. The predicted labor force change number is a net figure, already reflecting unemployed workers who would fill some of those relinquished jobs. In other words, according to the CBO, there won’t be enough able and willing replacements for people leaving the job market. Thus, there is a net loss equal to about 2.5 million workers. (The reasons for the lack of replacements is not clear from the report. That’s what the CBO model produces).
After a few days of vetting and debate, it is now better understood that there is a very big difference between people choosing not to work, and people being laid off because their company’s revenue and profits are down. But the problems with the CBO report go further than this (vital) distinction.
I read parts of the report, but not its entirety, most of which has absolutely nothing to do with the ACA. That in itself is noteworthy, but was hardly reported anywhere. The CBO is charged with periodically providing long term U.S. economic and fiscal projections, reflecting numerous factors and policy changes. The ACA was just one part of the stew this time. My overall impression is that the CBO really blew it in the way it handled the ACA as part of its much broader economic and fiscal outlook. Here is why.
I used to direct a small (about 25 staff) group located in the Washington State Budget Office which had responsibilities very similar to those of the CBO, though it was in the executive branch, and not quite as insulated from partisan pressures as the CBO. We also produced long term budget, economic and fiscal projections. Here are a few things I (painfully) learned about this type or work:
1) Its down-right silly to try to be too precise. This is commonly called false precision. And it makes economists or demographers look bad. (I did insist that our 30 year population projections have zeros in the ones, tens, and hundredths columns).
2) Stick with a few really big picture factors which are known to have material impacts on the economy, and where trends are clear, or departures from trend already underway and discernible.
For example: the age composition of the workforce, large population movements (in and out migration), changes in the cost of energy, availability of water resources; trends in productivity from technology; personal savings behavior; the amount of government and household debt. The CBO of course totally understand this, and uses all these perspectives, but it’s often pushed (sometimes chooses) to get more refined.
3) Stay away, as best you can, from attributing economic growth or decline to any highly specific, lightning rod policies, unless you are on very firm ground. I know, that’s easy to say.
4) Due to pressures from the people you work for, you often can’t avoid that. So, if you have to get in the middle of the fight, state the results in a way that partisan propagandists (of both sides) have to sweat a little to distort your findings. Let the toxic opposition researchers earn their keep.
5) When you’re forced to touch radioactive issues, provide as much context as possible. Today’s media is not equipped to do that.
For example, in this case: How large is a 2.5 million swing in the labor force? Is it bigger than a bread box? What other features of the ACA, which can have a material effect on the economy, have you not examined? At least mention them. What are the labor force participation rates in countries which, for decades, have had truer socialized medicine, and larger welfare safety nets, than the U.S.? Is the CBO projecting even lower labor force participation for the U.S. That should at least raise a red flag.
So, how does the CBO report. as it involves the AC. fare on some of these grounds?
By 2024, the U.S. labor force, according to the Bureau of Labor Statistics (BLS), will number about 164 million. Today it’s about 155 million. Thus, the CBO’s projected drop in the labor force “due to ACA work disincentives” would be about 1.5% of the total workforce in 2024. To borrow an old phrase made (more) popular recently by John Boehner: “Are You Kidding Me!” Are CBO models really able to discern a 1.5% change in the U.S. workforce 10 years out, and pin it on the ACA? Can anything that small on a base that large, ten years out, other than perhaps predictable deaths (and the futility of the Seattle Mariners), be forecast with an acceptable expected error?
What about numerous (equally unfathomable, but equally plausible) impacts in the other direction? One possibly big one: The gains in labor productivity when workers are not forced to stay in jobs they hate or where their skills are being underutilized? Almost everyone agrees ACA will have some of that effect. The amount is not calculable. But I do know it is a positive number; and potentially very large. Even modest productivity gains due to increased labor mobility would have a material impact on the size of the U.S. economy.
Is there any illuminating context in the report for the predicted drop in labor force participation? Perhaps it is not the CBOs job to range very far in offering useful context? But it is the job of the mainstream media. Yet, no where in the news did I see a chart, for example, showing the labor force participation rates (LFPR) in countries that have, for decades, had a truer form of socialized medicine than ACA.
The CBO’s projected LFPR for the U.S. in 2024, after adjustment for the putative effects of the ACA, is about 60%. Here are the LFPRs for the “socialized medicine/welfare state” poster children in 2012 after many years of social-democrat policies, based on World Bank data. http://data.worldbank.org/indicator/SL.TLF.CACT.ZS
United Kingdom, 62%
With the exception of France and Japan, all of these countries with more socialized health care systems have labor force participation rates at or above what CBO projects for the U.S. in 2024 under ACA. Now, I of course know there are many factors affecting LFPR. But it should be of some interest that most countries which provide people with large (not always wise) incentives to withhold their labor, don;t seem to have lots of people whiling away days at the beach. That;s not what the CBO said; but they sure made it (unnecessarily) easy for others to draw that conclusion,