One of the hottest trends in health policy — wait, where are you going? — is for insurers to pay doctors and hospitals based on the value of their services. I know: how novel. But as it happens, in the United States we generally pay providers based on volume, rather than value. The new trend of value-based purchasing goes by the name “pay-for-performance” or P4P.

P4P can be tricky, since patients vary in terms of their preferences and how they respond to even high-quality care. So if you’re going to have insurers define and reward value, it makes sense to let patients choose their health plan. That way, patients can avoid a P4P system that might (ironically) leave them with lower-quality care.

Unfortunately, the government doesn’t see it that way. The federal Medicare program — which covers 45 million seniors and disabled Americans, making it the nation’s largest purchaser of medical services — is developing a P4P scheme that enrollees will find unavoidable, for two reasons. First, many Medicare enrollees do not have the option of switching insurers. Second, even if they do have that option, private insurers will simply follow the P4P scheme developed by Medicare. Why spend your own money doing something when a big government bureaucracy will do it for you?

The reactions to a study in this week’s Journal of the American Medical Association illustrates another concern with government-administered P4P. According to that study:

Safety-net hospitals tended to have smaller gains in quality performance measures over 3 years and were less likely to be high-performing over time than non–safety-net hospitals. An incentive system based on these measures has the potential to increase disparities among hospitals.

This is not necessarily a problem: a P4P scheme should penalize low-quality care and spur those hospitals to improve quality.

The problem is that when government administers the P4P scheme, the low-quality providers and their advocates will lobby their elected officials for more money in advance of improving quality. In some cases, such as safety-net hospitals, providers may indeed require more resources to improve the quality care. But because the P4P scheme is politically controlled, even providers who don’t will demand more resources — and get them. Moreover, what will those providers say if the “pre-improvement” subsidies fail to deliver any improvement? I’m willing to bet my hat that they’ll ask for even more subsidies.

As I argue here, better that we just let patients choose their health plan, and leave the P4P-ing to the private sector, where the low-quality providers won’t be able to lobby their way out of making some badly needed behavioral changes.