RE: Ignore the flawed CBO
The accounting literature differentiates between projections and forecasts.
A projection is a "what if" analysis. If A, B, and C happen, then the result is likely to be D. You can postulate any assumption. You can have multiple projections with different assumptions.
A forecast is somewhat different. You do additional work to determine which sets of As, Bs, and Cs are most likely to occur, and then base your calculations on those most likely assumptions. You can have as many different projections as you have assumptions, but you can only have one forecast.
The problem with CBO estimates is that they tend to be treated as forecasts, when all they are is projections. And those projections are based upon assumptions that are fed to them by congress, not developed independently by CBO. What CBO does is basically add up the numbers and make sure that 2 plus 2 does in fact equal 4. They do a very good job of what they do, it's just that what the actually do is vastly different from what it is represented to be.
So in pricing out a bill, like the farm bill or Obamacare, congress may tell CBO to calculate the effect if 100,000 people take advantage of the program, and the average person gets $10,000. That's a fairly easy computation, the cost is $1 billion. But what if in actuality 150,000 people take advantage and the average per capita is $20,000. Now the cost is $3 billion. If CBO had the power to say, "Wait a minute, it looks like the number of people taking advantage is likely to be much closer to 150,000 than 100,000, and the average payout is likely to be closer to $20,000 than $10,000," then CBO's estimates would come a lot closer to reality. But that's not the way the system is set up.
If I'm sponsoring a bill, and I want it to pass, I'm going to give CBO scoring assumptions that understate the cost. And when CBO calculates the number, I'm going to say, "See, CBO says it will only cost $1 billion." Now CBO, and most likely I also, probably know that the cost is likely to be much closer to $3 billion than $1 billion, but the way the game is played nobody can mention that.
CBO came pretty close to telling the truth in its projections on Obamacare. If you look past the summary tables, and dig down into the detailed discussion of methodology, you will find that way back in the back they pretty much say that the assumptions on which the estimates are based can be achieved only if one of the following happens: 1) significantly reduced administrative costs (and we have seen that Obamacare requires significant additional administration), 2) significantly reduced access (take a number and get into the queue), or 3) significantly reduced quality (which will happen of its own accord if payments to providers are significantly reduced). But of course, nobody picked that up and ran with it. Democrats didn't want the truth, and republicans were too stupid to read the whole report.
As I understand the process, any senator or congressman can ask for a report based upon any set of assumptions. What republicans should have done with Obamacare is ask CBO to prepare a different projection based upon differing assumptions. But they were too stupid to do that.
CBO does generally employ Keynesian methodology, which tends to be very week on the supply side, and does generally employ static rather than dynamic analysis. This means that if, for example, paying docs less would likely result in fewer docs, they don't consider that in their methodology.
It's not that the CBO is incompetent or does a bad job of what they do. They actually do a decent job of what they do. You just have to understand the limitations of what they do. It's not intended to forecast a most likely scenario.
(This post was last modified: 07-24-2017 11:56 AM by Owl 69/70/75.)
|
