Is Fee-for-Service Payment the Problem?

Almost everyone says that it is. Conservatives. Liberals. Republicans. Democrats. A vast swath of the health policy community. With near unanimous agreement among everybody who knows anything about health economics, how could we even ask the question?

Trouble is, all these people are wrong.

We pay for most things fee-for-service. Or, more precisely, the payment mechanisms that are predominant in health care are widespread in almost every other market.


I think I’d better think it out again 

By fee-for-service, we mean that units of service are individually priced. The more you consume the more you pay. This is usually contrasted with capitation — under which the fee is independent of the quantity consumed. Under fee-for-service, doctors are said to have an economic self interest in overprovision, since the more the patient consumes, the higher the doctor’s income. Under capitation, by contrast, the doctor gets the same income, regardless of the amount the patient consumes. Under capitation, doctors are said to have a self interest in underprovision.

How do these payment mechanisms work in other markets?

Take the restaurant industry. Most things on a common menu are priced fee-for-service, but not everything. There is typically one price for tea, but unlimited refills at no charge.  Think of this as explicit capitation. Other items (condiments, butter, rolls, water, etc.) are available in unlimited quantities with no charge. Think of this as implicit capitation, since the average per-customer cost is embedded in each customer’s overall bill. Another thing that is implicitly capitated in a restaurant is time. Not your time, but the time of the restaurant staff. You can usually stay at a table indefinitely — with no extra charge.

At a lawyer’s office, these same payment mechanisms are employed, but in different ways. The lawyer’s time is typically sold fee-for-service, while such items as paper, ink, materials, secretarial time, etc. are usually implicitly capitated. Paralegal time might be priced either way.

At the doctor’s office, things are reversed. The doctor’s time is implicitly capitated, while something called a “visit” is sold fee-for-service. But in many ways this is a misnomer. “Doctor visits” are usually discrete capitated events. That is, each visit requires a fixed fee that is fairly independent of the amount of time the doctor actually spends with the patient (although the fixed fee usually varies by what is judged to be the complexity and intensity of the service). Additional services, such as X-rays, vaccinations, etc., are usually priced fee-for-service, while such items as bandages, etc. are explicitly capitated. If we extend from one-visit-one-fee to X-number-of-visits-for-an-annual-fee, you have the typical (capitated) arrangement offered by a concierge doctor.

Which brings us back to square one. If there is nothing particularly special about the payment mechanisms, what is it that makes health care different? There are three things: (1) third-party payment of the bill, (2) rationing by time and not money, and (3) an inability on the part of providers to repackage and reprice their services.

A restaurant can change its payment mechanisms at the drop of a hat. The restaurant can price side dishes ala carte or package them with the entre. It can have an all-you-can-eat (capitated) salad bar or it can sell salads (fee-for-service) plate by plate. It can even employ both pricing schemes in the same restaurant at the same time.  But whatever combination of payment mechanisms it employs, the restaurant is still competing for customers on price and quality. The restaurateur has a self interest in meeting customer needs in the most economical way.

In general this is not true in health care. On the average, we pay only 10 cents out of pocket every time we spend a dollar at a doctor’s office, and 90% of Medicare patients, almost all Medicaid patients and many privately insured do not even pay the 10 cents. That means we predominantly pay for care with our time, not with our money. As a result, doctors don’t compete for patients on the basis of price. And because they don’t compete on price, they don’t compete on quality either.

Even if they tried to do so, health care providers are trapped in a system in which they cannot repackage and reprice their services in patient-pleasing ways. For example, third-party payer bureaucracies prevent doctors from pricing their services the way lawyers do — charging their time out on a fee-for-service basis, but explicitly capitating other services.

In such a world, providers will naturally tend to try to maximize against third-party reimbursement formulas. There is no other way, after all, for them to increase their income. In response, the third parties try to minimize their payout.  As I explained in a previous post, both sides will use their intelligence, creativity and innovative ability in institutionalized warfare that has only tangential relationship to low-cost, high-quality care.

Bottom line: The fundamental problem in health care is not that we are using too much of one payment mechanism and too little of another. The problem is that the person who benefits from the service is not the same as the person who pays the bill.

Comments (35)

Trackback URL | Comments RSS Feed

  1. Ken says:

    Really nice post. One of your best.

  2. Vicki says:

    How do you think of all these musical pairings? This one is, again, on target.

  3. Joe S. says:

    You are right. Almost everybody else in health policy thinks about this question the wrong way.

  4. Neil H. says:

    What is surprisig is how many right-of-center health policy people say the same thing as left-of-center policy wonks. They are all fixated on fee for service. But as you point out, they are completely missing the boat.

  5. Devon Herrick says:

    There is near universal agreement that the fee-for-service payment system creates perverse incentives for providers. Yet, the people who criticize fee-for-service payments often don’t understand that it’s not the fees that are the problem, but rather the fact that third-parties are the ones making the payments. Pundits and experts talk about the need to move beyond fee-for-service payments; but they are more enthralled than ever with third-party payment.

  6. Matt says:

    You made several good points in this piece. One of the fundamental problems that exists in the private and public healthcare sector is the managed care (PPO)contracts. These contract between a hospital and the insurance carrier are kept under lock and key. A consumer, employer, benefits consultant would have more success uncovering the original Coca Cola formula then they would having access to a hospitals managed care contract. We are misled to believe that the PPO discount is beneficial. Meanwhile as discounts continue increase, healthcare cost continue to rise along with the profits of hospitals.

  7. David R. Henderson says:

    Nicely done, John. There’s nothing I disagree with in this post, but there is something that’s a quandary to me. I agree that doctors don’t seem to compete on quality. But I’m not clear why they don’t. Many doctors are free to charge above the price that the insurer will pay, and then have the patients pay the increment. I myself go to a doctor who prices above what my insurer will pay and I pay the increment. I like his and his colleagues’ attention they give me. So there’s a tiny case where they’re competing on quality. I understand that they can’t do that legally with Medicare, which is a huge part of the market. But why don’t we see them extra-bill for their privately insured patients and then patients who value the increment in quality at more than x are willing to pay at least an increment of x?

  8. Greg Scandlen says:

    A related issue is the oft-stated view that the payment system dictates provider behavior. The Dartmouth people look at localized variations and conclude that providers over-treat because FFS encourages them to do so. Yet the providers in the high-use areas are subject to the exact same payment system as providers in the low-use areas. Ditto with the Mayo Clinic, Cleveland Clinic and all the other models of “efficient delivery.” They ALL exist in the exact same payment system as every other provider.

    The Group Think in health policy circles (in this case, that FFS drives over-use) is a national scandal.

  9. Philip Weintraub says:


    While PPACA introduces a number of references for the use of alternative payment systems, Fee for Service (“FFS”), is not entirely eliminated by this legislation. The intent of PPACA is to seek improvements which lead to improved quality, outcomes and lower costs where possible. As you know, there are many academic studies which show that FFS in healthcare does lead to Supplier induced Demand.

    Health is not a commodity and should no be one. While a buyer generally knows they are thirsty and want a cup of tea, they are unlikely to know the proper treatment protocol for a specific disease.

    Phil Weintraub

  10. Bill Stuart says:

    This piece is just the latest in an unbroken line of essays in which John shows that the way that most pundits, politicians and critics see the health care system and its performance is completely wrong. As always, his simple analogies with other industries expose the myths that prevent a serious discussion about the REAL issues surrounding health insurance. Thank you, John, for your insight. If only those with closed minds would read and think . . .

  11. Francois says:

    I cannot say that you convinced me of anything in this article.

    Eating at a restaurant is not part of an insurance plan. Does the cook tell you what to eat? Does the cook tell you to eat more so that the cook will not be sued for under nurishment? Does the cook also own the meat supply house?

    Poor analogy – way off the mark.

    Payment needs to change; to offer patients the best quality and be the most efficient. Waste can no longer be tolerated in we are to compete in a global economy.

  12. John Goodman says:

    Bill: Thanks. Just for the record, I didn’t pay Bill to say those things.

    David: You are not alone. Some doctors have concierge like services (telephone, email, etc.) tacked on to their insurer’s plan in return for extra fees. But not all insurers allow exta billing for covered services and most people feel they have been promised the service for free or for a small copayment and resent extra fees. When demand exceeds supply and there is rationing by waiting, quality changes do not affect the physician’s income. More explanation is at the link I gave.

    Phil: You wrote: “The intent of PPACA is to seek improvements which lead to improved quality, outcomes and lower costs where possible.” Surely you can’t believe that.

  13. Donald Palmisano says:

    Good article! I just tweeted it:

  14. Darrell L. Dean, D.O., M.P.H. says:

    The rationale, used in the past atleast, was that the reason for not treating healthcare like other industries was that it did not respond economically in the same ways. I have not heard that argument in a long while. The current system tries to see that everyone has equal access to all healthcare that they need and want, then tries to figure out how to pay for it. That “system” is unaffordable. Your analyses are thought-provoking.

  15. Dr. Jay Gregory says:


    According to a recent FOX report, a hospital charged $1,000.00 for a toothbrush, $130.00 for Tylenol, and $53.00 for a 23-cents-pair of disposable gloves. It is easy to see where the dollars flow. Just look at what the public aviation costs are today! I don’t own a plane, but parts and maintenance are high to cover liability.

    Hospitals charge ridiculous prices because they can, but the prices that have to pay for equipment are grossly overpriced to cover the liability. Why do physicians have to carry liability? When you go to FedEx and ship a parcel, you are asked “Do you want to purchase insurance?” to replace the parcel should something happen. Remember when you could purchase flight insurance from a vending machine in the airport?

    The physicians are not the problem! We are just easy targets as we have very poor lobbyists.


  16. Robert Suter says:


    Nice post.

    The problem is, when (thanks to government meddling) there is no free market, incremental free-market tools can’t solve the problems. You have to push the reset button. How does that happen politically today?

    Thanks for keeping people thinking…

  17. Tom P says:


    Thank you for giving better structure to an argument I’ve been trying to make for a while. Maybe most of the folks in MA and other places who are desperately hopeful that capitation will save their bacon just don’t recall how it has worked in the past.

    In addition to your #1 – consumers aren’t spending their own money, I’d add #2 – when they do, we won’t tell them how much. That is, most big insurers refuse to disclose how much something costs most of the time.

    #3 is that even if a price were to be disclosed to a consumer, it is usually pretty nonsensical. This is driven in large part by Medicare and how it reimburses for services. Those FFS elements that are set by central planners and not the market are often curious. Think of what would happen to your restaurant if some people in Baltimore dictated the relative pricing of the items — or worse, ingredients of the items — on the menu. This may have little to do with the restaurant’s cost structure or the demand of the customers and, since it represents half of the customer dollars spent, restaurants would likely have to resign themselves to pricing the rest of their menu this way as well to keep from complete administrative meltdown. The result: even those growing numbers of consumers who pay more out of their own pockets are doing so in an environment where the pricing signals are not market-driven. Everything is pegged to a percentage of Medicare, not market.

    Thankfully, the first two are easily fixable and some in the market are focusing on them. The third will be more difficult to complete but is at least underway.

  18. Greg Scandlen says:

    Phil Weintraub wrote —

    “As you know, there are many academic studies which show that FFS in healthcare does lead to Supplier induced Demand.”

    Wow, I haven’t heard Roemer’s Law (“a built bed is a filled bed) invoked in a long time. There may be “academic studies” galore, but that is just evidence of what idiots most academics are. It is simply not true. In fact, occupancy rates have been all over the place over the years. Yet this little bumper sticker slogan has driven decades of failed public policy from health planning and CON to capitation. None of it works because all of it is based on a false premise.

  19. Mike Bond says:

    I agree that the problem is that the patient is not paying most of the bill. We also need to recognize that they aren’t choosing who actually does pay the bill since the boss usually picks our insurance. Some more recent evidence suggests that a big driver of health costs come from a very small number of people who have NUMEROUS health issues (as opposed to somebody having a bypass and being back on the golf course in a few months)and don’t die. I’m thinking some kind of integrated plan is needed for them. These, of course, will never develop with community rated premiums. Any thoughts on fee for service vs. prepayment for that kind of plan? By the way, Greg, I have worked in universities for over thirty years. Idiot is such a nasty term. I think depressed, angry socialists with other peoples money and hard core capitalists with their own is a better description!

  20. Linda Gorman says:

    Note that what hospitals charge for various services is not what people actually pay for those services. The big problem is that bill charges mean nothing.

  21. artk says:

    Fee for service; “concierge” charges for decent access; referrals to physician owned facilitates; these innovations have nothing to do with patient outcomes, their only purpose is to allow physicians to view patients as their own special ATMs. If money is such a strong motivator, have fees keyed to outcomes. If your diabetes patients end up in the hospital with complications, you get paid less. If their diabetes is controlled, you get paid more. When you hire an investment manager, do you want to pay him on the number of trades he makes or on the performance of your portfolio?

  22. steve says:

    Insurance is a big part of the problem. Yet, I dont see how you can have a medical system without it. No one else does, in first world countries. The big ticket items, where the real spending occurs, just cost too much. How would young couples pay for premie care? How would we pay for trauma care? AS long as you have insurance, people will find a way to make it pay.


  23. Chris Ewin, MD says:

    Greg…the payment system dictates provider behavior.

    It sure does…If patients (the customer) pay physicians (restaurants) directly (cash) and give quality service (a good steak), then physicians will be on their best behavior (time, access & a reasonable price) to keep them as patients.

    That said, FFS encourages seeing more patients in less time to improve income.

    You may want to ask actual patients (customers) if they like the service (steak). They decide if it’s worth it. Like the 85 yo with insulin dependent diabetes that went into diabetic ketoacidosis on a cruise ship in the Indian Ocean that I worked with a Norwegian surgeon (ship’s doc) and started an IV insulin drip…By the time they got to Mumbai, he was fine….
    For $167/month, the family and the patient were pleased and want to return for another steak….

  24. artk says:

    So, Dr Ewin, you’re telling me that if that patient’s primary care physician wasn’t a “concierge” physician, he wouldn’t have taken the call? Or when he took the call, he would have told the doctor at the other end “sorry, I don’t get paid for this call, you’re on your own” The question is are you charging for steak but delivering hamburger?

  25. Cotton M. Lindsay says:

    Great post, John. The issue of capitation vs. fee-for-service has little to do with quality or over-underservicing. As long as patients themselves are footing the bill, competition among providers should insure that the service provided satisfies the customer’s expectation regarding quality and quantity.

    A more perplexing question is, given that third party paying is the established norm in this market, what system of reimbursement best influences providers to supply something close to the mix of quality and quantity supplied under patient-pay. The present system of fee-for-service overlaid by managed care, regulation and second guessing by third party payers is possibly inferior on efficiency grounds to some form of capitation.

    I am unaware of any serious attempt to quantify the relative costs and achievements of these alternative reimbursement mechanisms. However, for private patients this issue resides with the third-party payers. They must pay the bills and sell their plans to patients. One imagines that competition would lead these insurers to push for the mechanism that produced the best results at lowest costs. So far as I can see, fee-for-service with managed care seems to be the preferred mechanism.

  26. David McKalip, M.D. says:


    I really appreciate your attention to this issue. It is a common theme for third party payers to attack fee for service. They would far prefer that I be put on a budget as a doctor and be punished financially or professionally if I spend too much of “their” money on a patient. This is defacto rationing. Patients are better served when they hold the purse strings and pay me for my time –- the time I spend evaluating them and advising them and providing treatments. Then they can force me to see them quickly, spend time talking to them, helping them understand their choices and convincing them that the tests I am ordering are worth their time and money –- as would be the surgery I would perform. They can then be backed up by their high deductible plan for rare medical expensive that would surpass their out of pocket contribution.

    This is true patient power and is despised by government and insurance companies alike.


  27. Robert Kramer, M.D. says:


    The loyal opposition thinks you are wrong. Until we get some metrics on quality of care, fee for service won’t work. Why? Because there is no way to differentiate within a speciality.

    As a pediatrician, no one takes into consideration one’s education, training, knowledge, experience, commitment, sensitivity, availability, etc etc. You talk about the restaurant business; you can get a hamburger at the French Room at the Adolphus, and one at McDonalds. There is a marked difference in price. All the years I was in practice, I was considered to be the French Room of pediatrics, yet fee for service did not allow me to charge accordingly. A visit to the pediatrician for an office visit was the same regardless of the complexity of the complaint or the nature of the illness.

    I remember that years back, I saw a patient, who as soon as I walked in the examining room I knew was critically ill. In less than a minute I diagnosed bacterial meningitis, did a spinal tap, called the ambulance and had the child at Children’s Hospital in a matter of minutes. When I sent a bill for my services, the family complained, because I charged them for a full visit when I spent such a short time with the child. I could have well saved the kid’s life because of my training, knowledge and experience, but could get no larger fee than if the same kid came in for a simple cold.

    I had my pediatric training at Yale, and for two of my three years, I earned $25/month. I could have gone to a lesser institution and made lots more money, but I wanted the best training I could get to become a better physician. At that time Harvard, Yale and Johns Hopkins were the three most highly rated programs in the U.S. Yet this did not let me charge accordingly, when a visit to the doctor, regardless of his training, was worth one set fee. And now that the practice of medicine is so skewed by fear and greed, the fees for the primary care providers will be determined by the insurance companies. And docs will upcode and order lots of unnecessary test or treatments to increase his income.

    Sorry it got so long, but once I get my dander up, I become more verbose and maybe a bit irrational. But this time I think I am right.


  28. artk says:

    Dr Kramer: Risk adjusted outcomes are well established. How would you like to get paid on outcomes? Better outcomes you get a bonus. Worse outcomes you get paid less.

  29. John Goodman says:

    artk, I’m sympathetic with paying for results, but risk adjustment is no where near able to do that adequately at the physician level. Medicare (which has a lot more data than a private insurer) does an adequate job of risk adjusting among insurers who enroll thousands of patients. The misses on any one individual patient tend to get cancelled out when combined with other misses over thousands of enrollees.

    But at the individual practice level (i.e., Dr. Kramer) we can’t even come close.

  30. Chris Ewin, MD says:

    So, Dr Ewin, you’re telling me that if that patient’s primary care physician wasn’t a “concierge” physician, he wouldn’t have taken the call? Or when he took the call, he would have told the doctor at the other end “sorry, I don’t get paid for this call, you’re on your own” The question is are you charging for steak but delivering hamburger?

    artk…it’s a fair question? The answer to the first…sometimes..In my previous practice, on call weekends involved covering for 5 docs and 25-30,000 patients. Yes, someone was there to take the call. But, the majority of the time we don’t know the patient and many times we send them to the ER and don’t treat them over the phone..Especially all the calls for refills over the weekends (I’m out of my insulin)…We do sorry, you have to go to the ER or a walk in clinic because the risk is too high to practice over the phone with someone you don’t know, not necessarily working for free.
    Do you think physicians should jeopardize losing their licenses and not get paid either…
    Further, you decide if you think the service is a good steak and not a hamburger….

    Bob..u hit the nail on the head. well-said.

    Cotton “I am unaware of any serious attempt to quantify the relative costs and achievements of these alternative reimbursement mechanisms”.
    MDVIP’s numbers show that their “concierge” physicians decrease hospitalizations by 60%.

    “So far as I can see, fee-for-service with managed care seems to be the preferred mechanism”. Respectfully, I disagree. Anytime TP’s are involved, costs increase and they dictate our value/payments..See Bob K.’s points.

    David..I agree with you for specialties. For PCP’s, overhead is too costly for individual practices to deal with more claims than the fewer claims most specialists have to deal with b/c of procedures.

    artk..I understand your point about outcomes, but how do you document them unless physicians spend more time documenting then doing patient care. How would anyone know what a fine job B K did with literally saving this child’s life…how much of a bonus would he get. Similarly, as a pediatric intern (Charity hospital, New Orleans), we had 2 patients with meningitis…one died within 6 hours..the other survived with long term complications b/c we were all over him…You just can’t document this and have some nurse tell you that you deserve a bonus.
    That’s why we get board certified (we take personal responsibility to be life-long students) and we have a state medical board thta tries their best to wean out those not giving good care.

  31. Chris Ewin, MD says:

    the other survived WITHOUT long term complications b/c we were all over him…

  32. J L McGee says:

    I take issue with your post on two fronts.
    The argument about fee for service could be reduced to a matter of semantics.
    What is the service?
    My staff and I administer an employee benefit plan for approximately 10,000 active and retired public employees. We are very much a part of the mess and the maze that is American health care.
    We do not charge those who come to our office to enroll dependents, change coverage or dispute an Explanation of Benefits. No, we are what John Goodman would call “explicit capitation”. We are expected to serve our customers within our budget. There are very busy days and less busy days, but that has no bearing on our compensation.
    Now imagine that we functioned like a doctor’s office.
    We would have to have some sort of coding mechanism that could parse all the different services we provide into tiny little units. How much to add a dependent, or multiple dependents, to copy documentation, to offer guidance or interpret plan provisions or resolve a dispute with a carrier. There would need to be a back office that could complete the necessary forms that would need to be submitted so we can get reimbursement.
    We might bill the individuals directly. Would that deter individuals from seeking our help? What would be the unforeseen long-term consequences to those individuals who might delay or avoid seeking our help in order to save money in the short term? Certainly we would have to factor in bad debt to our pricing.
    Or imagine further, that the reimbursement claims were submitted to a third party payer. Now that back office staff needs to be further expanded to discern and decipher the variety of rules and provisions for each of those third party payers. Staff would need to be educated about these varying provisions so that they did not spend time providing services to an individual for which they might not get reimbursed by that individual’s third party payer.
    And let’s not overlook the bureaucracy on the side of the third party payers. They will develop lengthy contracts that detail in exquisite minutia the services that will pay for, and when and to whom and the services they will not pay for.
    They will have customer service departments that will answer questions from their paying customers and they will have provider relations departments to answer and respond to issues from suppliers like us.
    My point is a simple one. On its face, leaving aside issues of incentives and issues of quality, you cannot have the system of piecework reimbursement that characterizes American health care, without creating a bureaucracy to accommodate it. That, in and of itself, adds to the cost of health care.
    Instead, I suggest a different model – one that characterizes the way our office delivers its services, the way the Veterans Administration delivers its health care services, the way the Indian Health System delivers services to its constituents – a system of global budgeting.
    That really means redefining the “service”. The service is or should be – provide for the health and well being of this defined population of people.
    Fee for service may or may not distort incentives for quality or quantity, but it most certainly obscures the incentives.
    I also take issue with the final point or, as you call it, your fundamental point. Implicit in that point is that the one receiving the service is the only one who benefits.
    To expand on that would take far too much space.

  33. D.S.Wells says:

    Brilliant article to jump start some lively debate. However, having said that, fee-for-service is not in it’s design the problem with the economics of health care. In every country the disease is transparency of cost. The system of reimbursing providers and facilities on a fee for service basis is necessary to influence consumer behavior if and ONLY if providers and facilities can provide the cost of their services. In the health care marketplace today the end cost of identical services depends on the insurance carrier’s contract with each provider or facility along with the health insurance plan of the consumer.

    UHC, for instance, negotiates aggressively and ‘enjoys’ some of the lowest costs in the industry for their health plans. However even within the UHC family of coverage there are dozens of reimbursement contracts. When I was developing a discount card program the issue was the inability for provider’s offices to generate fee invoices at the time of service. Combine that with providers and facilities gaming the system by diagnosis exaggeration and we have the issue of unnecessary cost inflation.

    Of course these issues are when providers and facilities will accept consumers as patients. Witness the issues coming about because of reduced reimbursement rates for Medicare with providers and facilities no longer willing to provide care. So is Fee-for-Service the problem? In my experience, and personal opinion, No.

  34. William Wehrly says:

    Thanks, John, for finally questioning the conventional wisdom. For the last few years I’ve bridled at hearing experts on healthcare panels here in DC assert that Medicare is made more expensive because of fee for service–with no explanation.