Back when health reform legislation was being debated, there was much talk about its ability to “bend the cost curve” — meaning to slow the rate of health care cost growth. After all, if health care costs are slowly bankrupting us all, something that will slow that process down even further is desirable–second only to stopping it altogether. But even after the bill became a law, opposition has remained, and that opposition continues to look for vulnerabilities in the law. Most recently, the discussion has turned to a revised CBO cost projection that opponents claim demonstrates that reform actually bends the curve upward rather than downward. If you want to read a nice, concise analysis of why that’s not true, you need to read Jonathan Cohn.
What I want to say is that everybody needs to take a breath. CBO scores are projections–they make no bones about that. That means that they take the data they have, and then get creative. They make assumptions. They look at trends and expect those same trends to continue ad infinitum. If something’s never been done before, they basically have nothing to go on at all, and they have to make the best educated guess they can. I would not want to balance my checkbook in 2015 using projections CBO made for me in 2010. They’re just not that precise. Now, that doesn’t mean that they’re not in the ballpark or that their methods aren’t rigorous, they are. It just means that their projections aren’t the end all be all. They are meant to inform decision-making, not predict the future with perfect certainty. After all, if they could do that, they’d be working on Wall Street.
Instead, they’re like the people at Yahoo! who create the projected points for the players on my fantasy football team each week. They have data on a player’s performance, even that player’s performance against the specific team they’ll be playing. But on any given Sunday, anything can happen. Are the projections close to the final score most weeks? Yes. Are they sometimes significantly off in either direction (under or over-estimating)? Sure. But do I use the projections to help me decide which players to start and which to bench? Absolutely.