One aspect of religious dogma that has entered the medical world is that fee-for-service pricing of medical services is bad and should be replaced by a capitated, or global, arrangement that establishes an annual budget for care for different risk groups of patients.

Like other religious beliefs, this is often offered without rigorous analytic support. Some insurance companies are particularly pleased with this approach because it shifts risk from insurers to providers and makes it easier for the insurers to create budgets and price their products.

Don’t get me wrong. This may be the right way to go, but the topic is worth more time and discussion than it has received.

It may be illustrative to think about other sectors of our economy and see which of them are characterized by global payments. Not many. Sure, there are products like cellular phone service that are sold in monthly fixed dollar amounts. But that is because it is a high fixed-cost product, where the marginal cost of additional phone calls is essentially zero. Fixed prices offer revenue stability to the vendor and a way to recover those fixed costs.

But most other goods and services in our economy are sold on a piece-work basis. Think of groceries, automobiles, electricity, gasoline, televisions, and clothing. Why is fee-for-service pricing appropriate for these? Or, in economists’ terms, why does such pricing lead to a reasonably efficient solution? The answers are pretty straightforward. Other markets are characterized by open entry and exit and by transparent information concerning quality, value, and pricing. Consumers can make more or less knowledgeable choices based on that publicly available information. New firms enter the market when they see an opportunity. Successful firms grow. Other firms fail.

In contrast, medicine is characterized by friction. Doctors are trained as radiologists, pathologists, or other specialists. The only thing they can do for a living (more or less) is sell their services as specialists. As the people at the Dartmouth Atlas have noted, this leads to supply-driven treatment patterns. If there are more radiologists in a given community, the usage of imaging will be greater than in less well staffed communities.

Likewise, hospitals generally do not come and go. They, too, represent huge investment in fixed costs, and they stay in the marketplace for decades.

But in addition to this, the main attribute of the practice of medicine is opacity. You and I as consumers (patients) have no idea what a given service costs because it is covered by insurance, and the actual rates paid to doctors and hospitals by each insurer are confidential. You and I also have no metrics by which to judge the quality of the service being provided. You have every incentive to request or demand more service for your medical problem.

If you are an insurance company holding a hammer, every problem looks like a nail. What is the most direct thing you try to do to influence levels of care that might be excessive? Design a pricing system that shifts risks to providers and is subject to an annual budget.

But, that is not the only solution. In another post, I discuss the path being taken by Harvard Pilgrim Health Care. It is good for Massachusetts that two of the largest insurers are trying different approaches. It establishes the possibility of comparing results across the two populations.

Now, though, let me let you in on a little secret with regard to capitated care. Underneath the global budget, there is still a fee-for-service arrangement establishing the transfer prices among the providers in a network. That GI specialist will still get paid for each colonoscopy. The big thing to work out in this system is the allocation of any surplus or deficit in the annual budget among the various specialists.

Unless that allocation is skewed heavily towards primary care doctors, decisions about the level of care given will not change. But, if the allocation is skewed too heavily towards the PCPs, there is no real income signal for the specialists, leading to a danger that they will not feel invested in the end result. Unless the system is accompanied by intensive, real-time reporting, along with clear penalties for excessive care, it will not work.

Did I say penalties? You bet. Without those, there is no enforcement of the global budget. But with those, global budgets are likely to raises hackles and resentment among specialists. I predict that the biggest issue facing physician groups in the coming years is the perceived interference by the global payment risk unit in the clinical decisions made by specialists.

If we were designing the health care system from scratch, I am guessing that the HPHC approach would be more likely chosen than a global payment approach. It would be accompanied by a shared savings mechanism, where physician groups and hospitals that beat an annual budget target would get a cash reward. It would also have a hefty dose of transparency with regard to clinical outcomes, so that the pricing levels charged by each provider would be accompanied by meaningful medical information that could help consumers make more rational choices. In short, a lot of the opacity of the health care delivery system would be eliminated.

That does not solve the problem of friction with regard to market entry and exit by doctors and hospitals. But global payments are weak on that front, too. Such friction may be an inherent characteristic of this sector for some time to come — unless, as appears likely, overall payment rates for Medicare, Medicaid, and private insurers fail to keep up with the cost of living. In that case, the future will belong to the efficient, hospitals and doctors who implement Lean or other front-line driven process improvement.

Paul Levy is the former President and CEO of Beth Israel Deaconess Medical Center in Boston and blogs at Not Running a Hospital. He is the author of Goal Play!: Leadership Lessons from the Soccer Field and How a Blog Held Off the Most Powerful Union in America.