03
Sep
2014
Health Insurance Coverage in the U.S., 1999-2012
I had known that coverage fell sharply when the recession struck, but I didn’t realize how much it had been falling the entire decade, even amidst the prime housing bubble years:
Depending on one’s worldview, the above chart could be spun either way. If you like ObamaCare, you could say, “See? The U.S. health care system was truly broken.” If you hate ObamaCare, you could say, “See? The massive government interventions in health care drove price hikes that made insurance more and more unaffordable. ObamaCare is just going to mess things up even more, except that the damage may show up in the form of shortages and delays, if prices are suppressed.”
What is the methodology of the chart? Who are they counting and not counting as “covered” in the US. Trust me, common sense might not dictate what the methodology is.
Census uses CPS, SIPP, and ACS data all of which ask simply whether you have public or private health insurance coverage – some are more detailed about what kind. So it’s self-reported in the same surveys that produce the unemployment rate.
Until 2013, the census bureau only tracked whether a person was uninsured at any point during the year, counting that entire number as the total uninsured population. That’s how you get to such high numbers. So technically when I switched jobs a few years ago, and was uninsured for 30 days (even though still within the 60 day COBRA election period), I would have been counted in the total number uninsured that year.
(Using this methodology consistently may or may not impact the trend observed in the chart. For example: it’s possible that the trend is that more people are uninsured for 1-2 months per year, while fewer people are uinsured for the entire year. But to my knowledge no one tracked the data this way).
In other words “never reason from a quantity change”. Wow you ARE diametrically opposed to Scott Sumner aren’t you? (of course he’d agree with that… but then it wouldn’t be as funny).
Interesting that the graph comparison is % to # instead of # to #.
That red and blue are the same, just flipped.
I don’t understand what the last 3 of you are getting at. The point of this post is to show information which I was not fully aware of. The percentage and total # of )un)insured are not the same thing. E.g. even if the percentage were constant–a flat line from left to right–the total number would be rising, since the population grew.
And Kuehn I’m offering two different ways people would fit that graph into their prior worldview. I’m not sure why that made you snap.
No not a snap! I feel like lately I try to be jokey and you think I’m angry!
You think we cannot – just from a movement in quantities – derive anything from this. I agree. Made me think of Sumner’s “never reason from a price change”. Even when you guys are disputing bad logic you do it in opposite ways.
As I noted “opposite” isn’t really correct… he’d agree never reason from a quantity change as readily as you’d agree to never reason from a price change.
Ah well – next time hopefully I’ll hit the mark.
OK sorry. I get it now. Never reason from a CAPS word in a comment.
WHY WON’T YOU LET THAT GO!!!!
^ also joke
“Depending on one’s worldview, the above chart could be spun either way. ”
Exactly. Ergo, bad idea to reason from a quantity change.
Just to be clear DK. You are not arguing that one should never reason from a quantity change, right? You are merely asserting Murphy and Sumner both say that.
Any reasoning from a price change or a quantity change ought to bring other information to the table.
There is an implicit “only”.
I’m with Sumner and Murphy and all learned folk on this.
I counted the patient’s heartbeats. Sunday before the shot, 105000. Monday, 0.
I draw no conclusion. I am a cautious diagnostician.
I assume you’re just having fun, but I think “heartbeats imply life” counts as “other information”.
You could also argue that the chart shows Obama care has had little effect on the numbers, which have increased in the last few years as the economy has recovered a bit from the recession and we’re now back to the (declining) trend line of the years before 2007. The uptick in the numbers covered seem to be too early to corresponding to ACA , right ?
Transformer try this post which has 2014 data. I make your point, contra Krugman.
Yup, it is useful to show graphs like this other FRED graph showing the total number of covered people in the US:
https://research.stlouisfed.org/fred2/graph/?id=USHICCOVER
which makes it appear likely the rate of growth slowed during the recession then went back to trend.
Okay, so I see the chart as something a little different.
Map it against Labor Participation rate. I know several families that when they lost their jobs stop having healthcare coverage and rather than spend money on healthcare decided instead to go without until the Father and Mother picked up better jobs again and took on coverage. It had to do with an economic choice and examination as to when and what was best for them.
Another chart I would suggest is the rate of illegal immigration. Not saying it is bad but I wonder if there is a correlation.
Finally. Who cares if you have insurance or not? The only people who really care about insurance rates are insurance companies…
Their opinion matters a lot, after all, they are the ones drafting the law…
If people can argue that the ramp up in house prices is just an efficient market at work, then why not say the ramp down in insurance is also an efficient market?
The market is expressing a preference towards housing and away from medical care. Show me proof this is not efficient.
Bob,
Map that against undocumented population. As this population increased steadily in the 2000’s (peaking, some say, during the bottom of the recession), the percentage uninsured in the total population also goes up. Numbers by state with radically different undocumented immigration experiences?