How to Pay for the Next Sovaldi

prescription-bottleImagine a pill that could cure cancer with one course of therapy or reverse an inherited, deadly disease. If it cost $1 million, could you access it?

This was the question asked at a recent panel discussion held by the American Enterprise Institute. The panel discussed a couple of new proposals to finance new medicines that come at a high price. Because these medicines address the needs of only a small number of patients, manufacturers contend that prices need to be high to make the investment worthwhile.

One proposal was put forward by Scott Gottlieb, MD (of the American Enterprise Institute) and Tanisha Carino (of Avalere Health). They put forward a redefinition of spending on specialty drugs as capital investment rather than consumption spending. This is because, for example in the case of Sovaldi, the expensive upfront costs of the drug are more than paid for by dramatically reducing costs over the next twenty or thirty years of a patient’s life, which might end up with a liver transplant. Gottlieb and Tarino’s paper is not technical, and one way to envision the outcome would be a mechanism whereby the patient or insurer would pay the (estimated) $84,000 cost of Sovaldi over twenty years in smaller pieces, rather than all in three months. (Gottlieb and Carino do not actually give an illustration, but I believe my example is an accurate representation of a potential version of what they describe.)

Another proposal was put forward by Professor Tomas J. Philipson and Andrew C. von Eschenbach, both of Precision Health Economics, LLC. Philipson and von Eschenbach are interested in using credit markets to reduce the immediate cost of paying for drugs. Their model suggests that government should incur a significant fraction of such debt, given that these specialty drugs will benefit future patients (so they should bear a share of the burden through an increase in public debt.)

This is a problem that needs solving, and we need “all hands on deck” to find our way there. With respect to Gottlieb’s and Carino’s proposal, what they are describing is a re-insurance market. Re-insurance markets exists, and we would benefit from a greater understanding of obstacles to such a re-insurance market for very expensive, specialty drugs. Philipson and von Eschenbach have a great track record of research on the costs and benefits of prescription drugs, so I understand their idea that future generations should finance innovation. However, the government has a pretty miserable track record in this regard, increasingly preferring to load up debt to finance current consumption rather than investing for the future. Do we really need a “Pharma Mac” or “Meddie Mae” on the federal books?

Our team has been writing about this issue recently (here and here). We’ve got some ideas of our own, which we will roll out in near future.


Comments (21)

Trackback URL | Comments RSS Feed

  1. Barry Carol says:

    I think GILD’s argument about the reduction in future healthcare costs that stem from curing Hepatitis C with Sovaldi is pretty weak. While a liver transplant may cost up to $600,000, we “only” perform a bit over 6,000 of them each year in the U.S. There are some three million people infected with the virus in the U.S. alone and as many as 150 million worldwide. Many have the disease for 20 years or more before symptoms start to appear. How many are well short of needing a liver transplant but die of something unrelated in the meantime? Conversely, for those who take the drug and avoid a liver transplant, how many will be diagnosed with Alzheimer’s a few years later and need many years of expensive nursing home care or cancer? The bottom line is that the lifetime medical costs incurred by any given individual are unknowable until death.

    My other problem with the price of Sovaldi is that GILD is selling it for significantly less in other developed countries including Canada, Germany and the UK. They were willing to sell it for an extremely low price in less developed countries. It seems to me that there is a tailor made opportunity for medical tourism here. Fly to India, stay in a first class hospital for a week or so to ensure that you are tolerating the drug, and fill your prescription there for $25 per pill instead of $1,000 per pill in the U.S.

    That said if patients with advanced Alzheimer’s disease who already need expensive institutional care could be cured by some new drug that would allow them to return home and live independently, that would be worth a lot because the cost of the nursing home care is already ongoing.

    Finally, if drug companies knew that payers would not pay more than, say, $100K per year for a new drug unless they could clearly demonstrate that it saved more than that immediately, perhaps the companies would do more to lower their cost structure by sharply cutting the cost of marketing and advertising. That way, they could sustain their research and development efforts and still ensure satisfactory profitability sufficient to attract capital from investors as needed.

    • John R. Graham says:

      Thank you. I take your point about the small proportion of patients that will eventually undergo a liver transplant.

      However, the logic of your statement that curing someone of one condition means that they will suffer from another condition a few years down the road is that there should be no medical care provided at all.

      When I was a kid, leukemia was a death sentence. Now many young people are saved. Every one of them will get sick and die of something in their lives. Should leukemia not have been treated.

      There is an old witticism that the most expensive drugs are antibiotics, because they are used to cure bacterial pneumonia and save the lives of many elderly patients who otherwise used to die (and peacefully, as I understand it).

      • Barry Carol says:


        I’m all for curing disease so that people can live longer and healthier lives. My problem with GILD is that it’s trying to justify its high price charged for Sovaldi by suggesting that it will lower healthcare costs. It would likely raise lifetime medical costs for the system and it’s not even clear that it will lower costs related to hepatitis C.

        While there are three million people infected with hepatitis C in the U.S., I don’t know how many of them are sick enough to need a liver transplant soon or in the reasonably near future. I also have no idea how much in medical costs they incur each year to manage their disease. The vast majority of people with hepatitis C will never receive a liver transplant either because they won’t get sick enough to need one or will die of something else in the meantime. We also don’t know how many of these patients doctors would consider good candidates to be treated with Sovaldi right now or how many of those insurers would approve assuming they are only approving the drug on a pre-authorization basis.

        The other point worth noting is that the company that developed Sovaldi, which GILD acquired for $11 billion, was assuming in its financial models that it could profitably sell the drug for $36,000 for a course of treatment while GILD chose to price it at $84,000. People don’t buy drugs and other medical services because they want to but because they have to. GILD’s price reflects unmitigated greed which is outrageous on its face in my opinion. BTW, I spent my 40 year career in the money management business and am generally quite sympathetic and supportive of market mechanisms to efficiently allocate resources. This case is different.

  2. bob hertz says:

    Barry is right on as usual.

    In normal bargaining, the buyer is constantly testing the seller to see what is the lowest price the seller will accept.

    The process is plain as day in real estate and labor negotiations. It is a little less blatant but equally present in grocery stores, barber shops, car dealers, hotel rooms, you name it.

    In other countries, the government takes the role of an aggressive buyer who will walk away if the seller is too stubborn.

    It is ludicrous to think that a person whose life is on the line can take that role in the US. For a number of reasons, insurance carriers do not take the role at least with drugs. Medicare and the FDA are too timid insitutionally to take the role.

    I have long advocated for a pharmacy price review board with broad powers. If the drug companies do not like it, where are they going to go? Canada?

    • Frank says:

      Excellent point. I agree that the sellers of drugs have a huge upper hand in any negotiations over the price. This made me interested in how Canada provides prescription drugs for lower prices and it seems they have an board to ensure the ceiling for prescription prices are low.

      • John R. Graham says:

        My research on the Patented Medicine Prices Review Board is dated, but it comes to surprising conclusions. Namely: Price differences are caused by differences in exchange rates and productivity, not the PMPRB. See:

    • John R. Graham says:

      No: They will dramatically reduce their R&D and stop investing in new medicines.

      This is not abstract. Empirical evidence is discussed in scholarly papers by John Vernon, Henry Grabowski, and others.

      I have heard no pressure from American citizens that the federal government should negotiate on their behalf with grocery stores,, et cetera.

      Generic drugs in the U.S. are available for a few dollars for a monthly supply – No government negotiation necessary.

      Patentholders have a period in which they can charge higher prices, but the patents are necessary to induce competition. Reducing the power of the FDA would reduce drug prices without dissuading capital investment. Price controls would not.

  3. Devon Herrick says:

    My complaint isn’t with the price drug makers charge for therapies — any more than I complain about the high prices of BMWs. With BMWs it’s evident that the market will bear the costs or BMW wouldn’t be in business (or would lower their prices). My worry is the financing of specialty drugs. There is not a feedback look allowing consumers themselves to signal how they want resources allocated. For instance, a consumer would likely pay more for a drug that cured a disease than a medication that merely provided marginal benefit. This is more difficult in health care due to the way most medical care is paid for.

    Another problem is the regulatory bottleneck getting new drugs to market. In a normal market, consumer could decide on the level of risk (or risk avoidance) they are willing to take on. Not so in health care.

    • Frank says:

      Isn’t there already some level of risk when deciding to take a drug. Sure, there are ways to reform the FDA to make sure that drugs quickly get to the people they need to, but to revert to a system where we don’t have an agency ensuring that the claims of a drug are true or that it is not overwhelmingly harmful is dangerous. No one would want to be the early adopter of a new drug. This may end up hurting the pharmaceutical companies that produce the drugs in the first place.

  4. Frank says:

    I’m not sure if incurring public debt because future generations will benefit is the most convincing argument, especially considering the ever changing landscape in health care. Who knows what new technologies and services can be provided that make treatments of today irrelevant.

    • Steve says:

      That’s true. And why in the world should we load our children with ever-increasing debt? We are in a huge fiscal mess, with previous generations racking up a tab that we cannot possibly repay…and yet some people think that pharma ought to also extend its hand for another taxpayer handout?

  5. Bob Hertz says:

    Thanks Steve.

    Of all the reasons for incurring public debt, rewarding pharma companies for savage price gouging is about the worst reason for public debt I have even heard.

    We sometimes incur debt to pay for Middle Eastern oil because the oil fields are a long way away, and it would take much blood and treasure to sieze them and go back to the 1950’s.

    There is no excuse however for rolling over to corporations in Connecticut and Switzerland who would never shoot back if we regulated them.

  6. Bob Hertz says:

    John, your article about drug prices for Fraser makes a good point that America does OK on many drugs, but America is an outlier on patented drugs.

    My mother in law needs gleenex for leukemia. The retail price in the USA is about $96,000 a year, versus about $8,000 in Canada. That is not caused by exchange rates or productivity, eh?

    Dean Baker has had this one right for several years. We should nationalize drug research, and quit paying for valuable innovations by extorting from helpless patients. It is like making the astronauts pay for the space program.

    • Barry Carol says:


      I think nationalizing drug research would be a total disaster. The reason is that many R&D projects fail, lead up blind alleys, etc. Government R&D dominated largely by political considerations and political power wouldn’t know when to pull the plug on failing projects as well as the private sector does. The NIH does mainly basic research and is not focused on commercialization or potential market opportunity.

      Instead, I would attack high drug prices through strategies like more restrictive formularies, reference pricing, pre-authorization and, yes, QALY metrics. You’re correct that many and probably most brand name drugs are significantly cheaper in other developed countries. Government price negotiation is not the issue per se. It’s whether the drug makes it on to the national formulary or not. If there is no meeting of the minds on price, the drug company is free to not sell the drug in that country. If CMS had the authority to negotiate drug prices but refused to keep any drugs that won FDA approval off the formulary, it would not be effective.

      • John R. Graham says:

        The Bayh-Dole Act was passed, to encourage commercialization of taxpayer-funded research, because the research conducted by NIH and others was sitting on the shelf and not getting developed and dispensed to patients.

        Without the profit motive, there would be good research, but few new medicines.

    • John R. Graham says:

      Is your mother-in-law paying herself? Most Canadian provinces have drug plans for seniors, financed by taxpayers.

  7. Bob Hertz says:

    Barry, could any government agency survive if it kept a vastly overpriced drug off the formulary, and some poor patient died because the drug was not available to them??

    I may have this wrong and welcome your response. My fear is that the American habit of considering death as almost a crime will prevent us from excluding any drugs. That is why I advocate upfront price controls. Let a drug company charge the actual cost of production plus 25%.

    A law like mine would keep private drug companies from developing exotic medicines that save the lives of 2,000 people.

    Well, nowhere else in the world do private companies spend ten years working on products for 2000 buyers, unless the 2000 buyers are very rich as in Lamborghinis.

    • Barry Carol says:

      Bob – I think, or at least I hope, that the American people have enough common sense to understand that resources are finite and we can’t just let drug companies arbitrarily charge as much as they think they can get away with. That’s why I like QALY metrics if all else fails. We have to set some rational metric that we can consistently apply which will allow us to sometimes just say no, we won’t cover this unless the price comes down to a more reasonable level.

      I don’t think price controls would work especially for new therapies. They work better when there are several alternative drugs within a therapeutic class but the approach often amounts to reference pricing where the least costly therapy is paid for in full and others are reimbursed at that price or sometimes a bit more if it’s more convenient to take or has fewer potential side effects. By the way, the cost to actually manufacture drugs in pill form is often literally pennies per pill. I’ve read that the manufacturing cost of Sovaldi in $1.50-$2.00 per pill. The big cost for drug companies is R&D including the need to cover the cost of all the failures. Gilead Sciences acquired a company called Pharmasatt which developed Sovaldi. It paid $11 billion for the acquisition.

      I generally support the Orphan Drug Act of 1983 which gives drug companies an incentive to develop therapies for conditions that affect relatively few people which is defined as fewer than 200,000 people in the U.S. If there’s a drug that cost $200K per year that will allow people suffering from the condition a chance to lead a normal or near normal life that would die without it, that’s fine by me. I would certainly be fine with it if the person suffering from the condition was me or a member of my family. While the price per person is high, the cost to the system is manageable because there are relatively few patients who need it and can benefit from it.

      Sovaldi treats a condition, Hepatitis C, which affects 3 million people in the U.S. and 150 million worldwide. Even though it’s an actual cure which is a good thing, it’s priced way too high in my opinion. Pharmasatt, in its financial models before it was acquired by GILD, expected to price Sovaldi at $36,000 for a course of treatment. I think that’s adequate.

  8. Bob Hertz says:

    Good comments but I am not going to let go of this issue quite yet.

    1. The $11 billion price tag that the drug company paid for the formulas would shrink drastically in about 2 minutes, if the ultimate selling price was controlled to German or Canadian levels.

    2 I know this may be apples and oranges, but are there not many products in the computer world that have millions of dollars in research behind them, and yet the product sells for $100 or $200?

    They sell for low prices because they have to. They are competing for consumer dollars directly, without a government or insurers covering for bloated prices.

    3 See articles by Dr Donald Light, especially “Demytholgizing the High Cost of Research”

    • Barry Carol says:


      The electronics industry is an entirely different animal. First, chip technology has evolved to allow every more capability and features to be built into a chip. Second, the learning curve allows manufacturers to steadily lower their cost per chip as growing cumulative manufacturing volume leads to higher yields or more good chips per wafer. Third, people can live without these products. They buy them because they want to, assuming they can afford to, and not because they have to.

      In the case of Sovaldi, I don’t know where prices will settle out in Germany or Canada or any other developed country. We’ll see but the initial pricing calls for $66,000 for a course of treatment in Germany and, I think, $57,000 in Canada vs. $84,000 in the U.S. The price in Egypt is $840.

      We often hear that our VA negotiates drug prices with considerable success. What people generally don’t hear is that it has a highly restrictive formulary. I’ve read that the number of approved drugs on its formulary is something like one-third as many as are on a typical commercial insurance formulary. Drug companies often give the VA a voluntary 24% discount to start with because the number of beneficiaries is both quite limited and veterans are a highly sympathetic group. A similar mentality is not extended to any other group as far as I know. I have no idea whether Sovaldi is on the VA formulary yet or not and, if it is, at what price.

      • John R. Graham says:

        Quite right. The most equitable way to finance this is to allow the drug-marker to charge different prices in different jurisdictions, which will mostly be driven by differences in national incomes.

        If this is forbidden, and governments try to enforce a law of one price, that price will be the high one and the drug will not be available in lower income countries. The response will be to break the patent, and that woudl be terrible for future investment.