In what is emblematic of every annual Medicare trustee report, the most recent offers a word of warning (which is about as urgent as any warning you can expect from an actuary):

Without unprecedented changes...Medicare prices would be considerably below the current relative level of Medicaid prices, which have already led to access problems for Medicaid enrollees, and far below the levels paid by private health insurance. Well before that point, Congress would have to intervene to prevent the withdrawal of providers from the Medicare market and the severe problems with beneficiary access to care that would result.

Perhaps former CBO director, Douglas Holtz-Eakin summed it up best when he tweeted:

Medicare and Social Security are still broken, going broke...because we've done nothing to fix them.

In a word, yes - this is the message that Medicare's trustees are sending Congress and the President. Neither Democrats (who may otherwise wish to sit on their laurels, confident that the ACA's cuts to Medicare have saved the program), nor Republicans can ignore the fact that Medicare - the health insurance safety net for over 50 million elderly and disabled Americans - is hemorrhaging money and efficiency.

It makes sense then, to take a moment to review how and why, and what can be done about it.

1. When You Assume

All projections - whether they come from the CBO, Medicare's Trustees, the White House, or the private sector - are based on a comprehensive set of assumptions. Assumptions can incorporate the gamut from changes in GDP, productivity, costs and virtually anything else that the authors find is relevant in their modeling efforts.

Of course, as the leading experts in risk-adjustment and financial modeling, actuaries know this well. And while the front-page projections of government accounting are usually based on a "current law" baseline (that law takes effect as written), there are often major differences between what is supposed to occur and what actually does. The latter is known in the policy world as "current policy" - what is expected to happen.

There are several important areas where current policy baselines are important to keep in mind, but one very salient example in the health policy realm is Medicare's infamous "sustainable growth rate" (SGR). The SGR governs the rate at which Medicare's payments to providers changes annually, and for many years the SGR required that physician payment rates be reduced - last year, the reduction would have been close to 30 percent. Because it is widely agreed that reductions in Medicare's payment rate would result in much worse access to care for Medicare's population, Congress has routinely overridden the reductions in its annual "doc fix." So, because it's unlikely that Congress will allow payment rates to be reduced, part of Medicare's "current policy" baseline is to assume that SGR cuts will not happen.

This "alternative scenario" has become so important, that now every report by Medicare's actuaries includes alternatives to current law.

Over time, the difference between current law and the SGR alternative projection will about to about 0.7 percent of GDP (over $100 billion); through the modeling period, the difference in Medicare spending comes to about 11 percent (well over $1 trillion). But that's not all. The top line in the graph assumes that IPAB is relatively ineffective, SGR cuts are overridden, and the ACA's cuts to Medicare will not be sustained past 2020 (a point that was made in the 2011 technical review of Medicare's actuarial reports) - the difference from the current law baseline to the full alternative policy is a massive 3.3 percent. This grows over the modeling period, and while the actuaries don't provide a full estimate, the total difference in Medicare spending between the two scenarios would be massive.

Unfortunately, there is likely no single solution here. The first step (and kudos to President Obama for including this in his FY14 budget) is to eliminate the SGR and immediately develop a new payment formula, replacing the SGR temporarily with a 1 or 2 percent increase in payment rates. The new formula would need to ensure that evaluation and management services are adequately compensated (so as not to skew too far in favor of specialists). This would address part of the issue, eliminating the need for a "doc fix" and moving current policy closer to current law. But ultimately, as long as Congress (and the President) believes in using gimmicks like provider payment reductions to hold down Medicare spending, we will continue to see a disconnect between current law and current policy - this only serves to make Medicare's future less certain, to no one's benefit.

2. Medicare Is (only slightly) More Solvent

During the last election cycle, a popular talking point became the solvency of Medicare's trust fund. For clarification, when we discuss Medicare's trust fund we usually refer to Medicare Part A - Hospital Insurance. This fund pays for inpatient hospital services, Part B (which actually has an increasing fund balance over the modeling period) pays for outpatient services, and Part D (which is funded largely through premiums) covers prescription drugs.

The new ominous deadline for Medicare Part A is now 2026 - a two year improvement from last year's report which projected that the trust fund would be depleted by 2024. This "improvement" is likely due to expenditures last year being lower than expected coupled with more optimistic forecasts about economic growth in later years (which in turn means greater payroll tax revenue).

But there's more to it than that. The 2026 figure comes from "intermediate" assumptions regarding costs - under the worst-case scenario assumptions the trust fund would be depleted 7 years earlier, in 2019.

Perhaps the solution here is simple. The actuaries note several times that there is no provision to pay Medicare Part A benefits with general revenue - so for some, the obvious fix would be to allow general revenue to be used for Part A. However, this would be short-sighted; it misses the entire point that Medicare, as is (even with the ACA's "productivity adjustments"), is unsustainable - both in the short-run and the long-run.

That one federal program alone will eat up at least 6.5 percent of GDP in the future should not be taken lightly by policymakers. Containing Medicare's costs is critical - reworking the SGR (as noted above) and Medicare's payment formula is an important step. Simplifying Medicare, by combining Part A and B deductibles, for instance, can help eliminate some of the program's "hospital favoritism." More drastic measures would forbid MediGap plans from covering deductibles or copays to ensure that beneficiaries still have some skin in the game and make smarter decisions. Going further, Medicare's eligibility age can be raised little by little, which would reduce Medicare spending by $148 billion over ten years.

What policymakers should not do, is rest easy thinking that existing "productivity" adjustments are good enough.

While the Medicare trustees report is rife with scary language (and many other scary charts), the actuaries make one point above all else: Medicare is not sustainable. The program is broken and will not survive without major changes - the earlier the better.

This piece is cross-posted at the Manhattan Institute's state and local issues blog, PublicSectorInc.org

Fiscal instability is not only an issue nationally - driven largely by health care spending - but at the state and local levels as well. A new GAO report illustrates the magnitude of the fiscal challenges facing states, and identifies the (unsurprising) culprit:

The [simulation] show[s] that [state and local] health-related costs will be about 3.8 percent of GDP in 2013 and 7.2 percent of GDP in 2060...[t]he model projects that the [state and local] non-health-related costs will be about 10.5 percent of GDP in 2013 and about 7.7 percent of GDP in 2060.

The ever-growing burden imposed by health care spending means that by 2060, the national state and local fiscal gap will be around 4 percent of GDP - in nominal terms, that's about $5 trillion based on CBO projections. Because health care costs - enshrined in promises to government employees and retirees, as well as Medicaid spending on the poor - will drive this fiscal growth, which is unlikely to slow down (health care spending on current employees and retirees is governed by contracts, which makes it difficult to pare back; Obamacare's Medicaid expansion ensures that in the states that undertake it, many more residents will be covered making it more difficult to slow down its growth) other state and local outlays will fall on the chopping block. This phenomenon of "crowding out" is nothing new; because localities operate with limited funds (revenue must be raised through taxes, bond issuance, or from federal grants), each slice of the pie has to get smaller.

Indeed, the GAO report also acknowledges that wages paid to state and local employees will likely fall as a share of GDP (this phenomenon may ironically increase retirement promises that localities make to employees).

For states that are expanding their Medicaid programs under Obamacare, the expansion may seem like a reasonable way to shift costs to the federal government. At first, the federal government picks up the full cost of new enrollees, while later on, states will only be responsible for 10 percent of the costs. For the majority of states, however, this will still mean an increase in spending.

Reports like this underlie the need to drastically reform government health care spending - Medicare (on the federal side) and Medicaid (on both the state and federal side) are seen, unambiguously, as eating up an ever growing share of revenue going forward. Obamacare expanded Medicaid, shifting some costs to the federal government, while bringing "productivity adjustments" to Medicare in the form of provider cuts. Neither approach represents actual reform, however; analysts in the private sector and at the CBO routinely note that government health care spending is on an unsustainable trajectory.

Worse still, the latest FY14 budget from the White House drew a line around Medicaid, taking it "off the table."

Perhaps the one silver lining for Medicaid is that it remains largely a state-administered program that simply has to operate according to federal guidelines. Because of this, states have become testing grounds for new approaches to offering the poor affordable health care. Managed care penetration for instance, where tight networks of providers are often given financial incentives to provide more efficient, outcome-based care, varies significantly among the states. New York has more than three-quarters of its Medicaid population in managed care contracts; Maine, on the other hand, has less than half. Other innovative approaches have also taken hold - the Healthy Indiana Plan took a "consumer-driven" approach, offering Medicaid beneficiaries an HSA product that was partly funded by the state, to help keep down unnecessary health care spending - the results have largely been positive.

Just because one reform worked in Indiana doesn't make it ideal for other states, however. State medical needs vary greatly, and no amount of central planning will change that - basic demographic differences like income and age distribution can drastically affect a state's health care spending. Fortunately, states can request waivers from the federal government to experiment with new ways of paying for Medicaid services or administering the program.

Whether these waivers will continue to be issues as regularly as they have in the past is unclear - a uniform Medicaid program is essential to Obamacare's coverage expansion, and HHS may choose to approve fewer and fewer waivers as a way of better controlling the program. Other, more drastic approaches are likely only pipedreams for now - Medicaid block grants based on the successful welfare reform of the 90s would create a natural cap on future spending growth, as my colleague Paul Howard has written.

Real reform to Medicaid would take advantage of states' abilities to be "laboratories of innovation"--the current administration could take the first step by assuring states that even with a Medicaid expansion, HHS would continue to consider even the most drastic waivers. 

While Republicans and Democrats can't agree on much these days, there is one unimpeachable fact on which the two sides can agree: Medicare (and indeed government health care spending) is on an unsustainable trajectory. Last year, Medicare spent a total of $551 billion - around 3.4% of GDP. This grows to 4.1% of GDP - over $1 trillion - by 2023 in CBO's latest projections. Continuing to implement the "doc fix" every year adds over $100 billion more to this number.

So, yes - Medicare is a huge problem. But what to do about it?

On one side (the right) premium support has been proposed as an option - future retirees would be given a set dollar amount (that grows every year) to purchase private insurance, while retaining the option to remain in fee-for-service (traditional) Medicare. On the other (the left), solutions have mainly revolved around cutting provider payments (which Medicare actuaries have noted will likely have to be undone) and attempting to use Accountable Care Organizations to improve quality and reduce costs. While premium support is off the table for the time being, ACOs are being rapidly implemented through Obamacare. However, because evidence on ACOs is mixed, it is unlikely that accountable care will be a panacea for Medicare.

One other option floated every few administrations (it came up during the Clinton years and under Bush Jr.) would raise the Medicare retirement age to 67 (up from 65 now) to match the full Social Security retirement age. A CBO analysis found that, on net, this would reduce federal spending by $113 billion over 10 years (this accounts for potential increases in Medicaid spending etc.), and would reduce Medicare spending by about 5 percent through 2035. It appears that in recent talks between Republican and Democrat lawmakers, this idea has cropped up once more, albeit with a new flavor - offsetting costs on those without Medicare through a buy-in.

Essentially, anyone 65 or 66 (though some proposals bring the lower end down to 62) would be allowed to "buy-in" to the Medicare program by paying actuarially adjusted premiums. The idea would be to help offset the cost shift to those who are unable to receive Medicare benefits at 65/66, and their employers who may have to pay more in retiree benefits because of this. In 2005, the Urban Institute released a thorough analysis simulating the impacts of raising the Medicare eligibility age coupled with a buy-in program. The findings were somewhat nuanced:

Increasing the Medicare eligibility age to 67 will not solve the program's growing cost crisis. Eliminating 65- and 66-year-olds without disabilities from the Medicare rolls will not save very much money because they do not tend to be particularly heavy users of Medicare services. Moreover, it will be almost impossible to design a cost-neutral buy-in option. The buy-in plan will disproportionately attract people with health problems who are expensive to insure. And to be effective, the option must include some subsidies.

Certainly, as CBO's analysis found, the actual savings to Medicare will be modest, and because a subsidized buy-in is necessary to ensure that the elderly aren't left out in the cold, the cost savings would further be moderated by the increased spending. Also, some of the impact on the low-income population would be moderated by current Medicaid expansions as well as the availability of subsidies on exchanges. But ultimately, the specific parameters of raising the eligibility age along with a buy-in program could vary significantly - for instance, if employers are allowed to subsidize a worker's buy-in into Medicare, the cost to the government would be reduced, and the impact on workers could be moderated.

In its final point, the study notes a powerful dynamic of the Medicare eligibility age - the labor force participation rate - which is arguably more important than Medicare spending in and of itself. With Medicare benefits unavailable at 65, workers may be incentivized to keep working longer - two extra years of working at a median salary of $50,000 helps grow the economy quite a bit more than two years receiving Medicare benefits. The impact of more people delaying retirement and working longer (and paying more in taxes), could theoretically offset any increase in costs for the sicker, low-income population.

So while raising Medicare's eligibility age is far from a panacea for Medicare's spending problems, it can be an important symbolic change spurring more people to delay retirement and contribute more to the economic growth that America needs now more than ever. 

A few days ago, the National Commission on Physician Payment Reform issued a slew of recommendations for reducing national health care spending, based on one core principle:

Our nation cannot control runaway medical spending without fundamentally changing how physicians are paid.

This approach has its merits - physician services make up somewhere around 20 percent of national health spending - around $540 billion annually. But more importantly, the authors of the report make an under-appreciated point: that physicians have significant input on the course of care long after the initial office visit. So if we think that we spend too much on health care, it makes sense to start at the initial point-of-contact with the health care system.

Just as Stephen Brill's recent opus illustrated the tangled web of our health care system's "non-price" approach, this report does a terrific job at identifying the broad, structural problems in our health care payment system.

The report identifies several factors driving growing health care spending: fee for service reimbursement, reliance on technology over care management, a disproportionate reliance on specialists over primary care physicians, and a payment scheme that favors hospitals over other outpatient settings.

While the commission recommends these changes to the health care system as a whole, these reforms are particularly well suited to Medicare reform.

Fee For Service

A fee for service system pays a physician for a procedure/office visit/surgery etc. by volume. In a direct-payer system, without insurance companies and government subsidies, this wouldn't be inherently problematic. When you spend your own money you tend to be frugal, and your physicians - in their own self-interest - have to make sure that the care they give is effective enough to keep you from dying but is affordable enough that they keep you as a patient. It's easy to see how this leads to problems in a third-party payer system - when an insurance company or the government is paying your bills, you become less price-sensitive and physicians have the freedom to charge more and provide more services (some of which may be unnecessary), while discriminating based on who is paying (charging Medicare less than private insurance for instance, because of Medicare's monopsonistic bargaining position).

With fixed budgets and profit-motives, private insurers already have some incentive to insure that medicine is evidence-based and that the cost is worth the expected outcomes. The private sector has already recognized the need to mix some outcome-based measures into standard fee-for-service reimbursements - Wellpoint recently announced an effort to reward evidence-based medicine, and even earlier decided to boost primary care physician fees by 10% if they meet certain outcome metrics.

But on the public side, the lack of a fixed budget and tremendous bargaining power makes the powers that be more resistant to "quality-oriented" payment reforms. The fear is that this may lead to lower quality care for those who truly need it. Quality-focused payments generally consist of attempts at managed care - where care for a particular individual is coordinated through a tight network of providers to best determine the most appropriate, long term treatment route that maximizes quality while minimizing cost. The evidence on managed care, however, is mixed. One the one hand, some studies indicate that Medicare Advantage managed care plans tend to reduce racial and ethnic disparities in primary care quality and reduce preventable hospitalization more than FFS Medicare. Yet other studies studies have found little benefit from Medicaid Managed Care Organizations (MCOs) or pay-for-performance approaches. These disparate results may indicate that Medicaid pays providers too little relative to Medicare; they may also point to differences in the underlying structure of different managed care programs; moreover, there may (and likely are) systematic differences between Medicaid and Medicare populations beyond simply age. The point is, given certain conditions - managed care can work to improve quality while cutting costs.

Technology Over Care Management & Reliance on Specialists

These two points logically go hand-in-hand it would appear. In general, this isn't a problem for the private sector - private insurers must generally be more prudent about what they pay for and how much they pay. The profit motive seems to work well here.

Medicare, on the other hand, skews its payment system to favor high-technology, specialist-focused care over primary and evaluative care. The underlying problem here is the relative value unit (RVU) system that values the work of a specialist at a much higher rate than the work of a general practitioner. While cardiac surgery is certainly a complicated, labor-and-technology-intensive procedure that warrants high compensation, a 15-minute, mostly automated cataract extraction isn't, and doesn't. While it may sound simple, a routine office visit requires the GP to consider a variety of possible conditions, across a very broad spectrum of specialties - not an easy task. Yet the ophthalmologist performing the cataract extraction will receive much higher reimbursement from Medicare because his education, specialization, and time are arbitrarily valued more.

The first recommendation here from the commission is simple - increase payments for evaluation and management services that are provided not only by GPs but also by cardiologists, neurologists, and others. This would provide incentive for more focus on evaluation over expensive, often unnecessary treatment (particularly in Medicare). The second set of changes, again is rather simple (and relatively uncontroversial) - don't just use the RVU (set by the Relative Value Scale Update Committee, which is made up mostly of specialists) to set reimbursement rates; and end the SGR (which can be funded largely by eliminating excessive administrative costs, unnecessary services, and fighting fraud) that annually is slated to cut physician reimbursements. Combined, these three recommendations could drastically change the Medicare landscape - perhaps changing it from a high-risk, high-fraud program into one that is more affordable and genuinely seeks to improve patient outcomes.

Setting-Neutrality

The last point here should really be a no-brainer. It shouldn't matter whether you get an EKG in a hospital or in a lab - the amount paid should be the same. Once again, this is a major issue for Medicare. Part A (hospital insurance), tends to pay more for the same service performed in a hospital than what Part B, outpatient insurance, pays if the service is performed in a doctor's office. The Medicare Payment Advisory Commission, in a report to Congress noted: Medicare pays 80 percent more for a 15-minute office visit in an OPD than in a freestanding physician office.

There is simply no good reason for such a disparity in payments based on location. Doing so encourages consolidation of providers - this drives up prices with little indication that it improves quality of care.

The exact process for mitigating this discrepancy can take several forms: a global inpatient-outpatient reimbursement rate would statutorily close the gap for instance. Switching FFS Medicare (by shifting beneficiaries to Medicare Advantage plans, for instance) would likely also eliminate this problem through capitated payments.

The commission's report should be welcomed by both sides of the health care debate - the left for its encouragement of quality-based care and the right for its acknowledgment that Medicare has serious, systemic flaws. It offers a unique perspective on the American health care system - one that looks at it holistically rather than as a set of silos. And the focus on payment reform is long overdue. With luck, some of the commission's proposals can find bipartisan support before it's too late to fix our health care problem.

In the past month, analysts, pundits, bloggers, journalists, economists, politicians and everyone in between has had a chance to voice their opinions on the CBO's latest budget outlook. For wonks, these updates are like scripture - but of course, as all other mere mortals, even policy wonks fall prey to confirmation bias. That means that we often choose to ignore the parts of these reports that make our preferred policies seem less appealing or we try to rationalize them away.

And it turns out that one critical part of the budget outlook has been ignored by those on the left.

A favored argument against deficit reduction is that we can't cut our way to growth - indeed, President Obama said as much in his State of the Union address. Some on the left have taken this argument further - Paul Krugman, the Nobel prize-winning economist often launches into diatribes against the conservatives' push for austerity and in favor of more government spending - as long as we have near zero percent interest rates, the argument goes, we should be borrowing like mad. (Almost) free money is a tempting mistress.

On the surface, this would appear to be a solid argument. But what does the CBO say about it?

At current projections, net interest costs will rise from 1.5 percent of GDP in 2014, to 3.3 percent of GDP by 2023. This assumes that CBO's current projected interest rates remain on track - 4 percent for 3-month t-bills and 5.2 percent for 10-year notes. This alone should be a cause for concern - rising interest rates keep domestic investment down as more borrowing happens from abroad and more of the federal budget is devoted to paying down interest. A vicious cycle ensues as interest rates go up because of growing federal debt. Under normal economic circumstances, this may not be a problem - when interest rates are low, inflation and GDP grows; interest rates are then driven up in order to keep the growth stable and moderate the eventual downturn. But we are now at a point with near-zero interest rates and near-zero inflation - far from ideal, which means that interest rates are certainly something to be worried about.

But the CBO offers one piece of advice that seemingly, no one on the left chooses to heed:

If interest rates on all types of Treasury securities were 1 percentage point higher or lower each year from 2014 through 2023...for the 10-year period would be about $1.1 trillion  higher or lower (excluding the additional costs of servicing the federal debt).

This means that if interest rates on 3-month bills hit 5 percent while hitting 6.2 percent for 10-year notes, non-interest spending would be over a trillion dollars greater over 10 years - an extra $100 billion in spending annually! 

Krugman's "don't worry, be happy" attitude towards interest rates might be defensible if current U.S. borrowing was being used to fund innovation or growth. But the money we're borrowing now isn't being invested in areas that are likely to spur higher productivity or GDP growth later - it's being invested in health care, a sector notoriously resistant to productivity gains.   In fact,  cumulative spending on government health care programs by 2023 will be the single largest source of federal spending.

 Source: CBO 2013 Budget and Economic Outlook; Table 1-3

Even when accounting for offsetting receipts in health care programs (like Medicare premiums), the cumulative total comes to $12.2 trillion.

Health care programs are largely what drive the national debt - this has been acknowledged by CBO and other analysts time after time.

If the cost of borrowing rises even a smidge above CBO's expectations (1 percentage point) - these programs can wreak even more havoc with the federal budget. Stemming the growth of major health care programs is even more critical now, when interest rates are low, and national health care spending has slowed somewhat. Addressing the cost of the "doc fix" for instance, now stands at $138 billion plus $29 billion in debt service. An increase in the interest rate of one percentage point above CBO's projections (between a 19 and 25 percent increase in borrowing costs) would increase the debt service portion to between $34.5 billion and $36.25 billion over 10 years - and this is just for one relatively inexpensive fix to Medicare.   

Capping federal health spending would also free up more money on the kinds of investments that Professor Krugman likes - and that Republicans might find more palatable. 

In the meantime, President Obama and supporters want to shield health programs from additional cuts.  It's a $1 trillion gamble.  Who would take that bet?

But when it comes to the question of health care costs overall, Medicare is the solution. Its vast bargaining clout lets it get much better prices than any private insurer, and we should be relying on it more to pay our bills, not less.

It makes sense to start with one of Yglesias' first comparisons - the U.S. versus Canada. He notes that while American doctors get paid quite a bit more than Canadian doctors, Canada has 25% more doctors' consultations per capita than America. He also briefly mentions "overprices" for medical equipment and pharmaceuticals as other cleavages between us and Canada, but not surprisingly, leaves it at that.

If there's a bell going off in your head right about now, there's a good reason. Canada has one of the longest waiting times for elective surgery in the OECD - in 2010 25 percent of people that received elective surgery had to wait more than four months for it! In the U.S., this is a mere 7 percent. The comparison also holds when looking at waiting times to see specialists. This is an important tradeoff - if you want more government sponsored health care, or at least more intervention, you will likely see reduced access.

But this doesn't change the fact that American doctors do get paid a lot more than those in other countries - this alone, however, isn't terribly problematic. Wage variations among countries exist - American office clerks make about 20 percent more than British office clerks; meanwhile, British firefighters make about 13 percent more than French firefighters. Does this mean we should "bargain" down the salaries of British firefighters or American office clerks? There are literally hundreds of reasons that certain jobs have wage variations among similar economies - unionization, differences in work weeks, or the amount of training required - these factors can and do affect wage differences. The last point is especially salient in explaining the American variation - American doctors face the largest barriers to entry, in terms of the amount of education required (minimum of 11 years including undergrad for a general practitioner), licensing exams, and the steep cost of a medical education. Given that we are already facing a shortage of GPs, it seems unwise to focus on restricting their salaries. (Although, part of this shortage would certainly be explained by Medicare's hugely specialist-skewed reimbursement rates).

The other point that should be skewered is that Medicare is "cheaper" than private insurance. Part of this argument stems from the statistic that Medicare's administrative costs are a mere two percent. The first problem is that this only uses one measure of Medicare's administrative costs - the one published by Medicare's trustees. In reality, there are two different measures - one from the National Health Expenditure Accounts and the other from the trustees. The former shows Medicare's admin costs growing to about 6 percent in 2010; the latter shows them at less than 2 percent in 2010 - quite a large variation. But even with these two divergent measures of admin costs - one important source of costs is excluded: fraud and waste. The GAO has routinely found that Medicare and Medicaid (with Medicare making up more than half) made about $70 billion in improper payments in 2010 - close to 10 percent of the two programs. Other estimates by the RAND Corporation, however find that fraud and abuse may be as high as $98 billion in the program. The second problem with thinking that Medicare is cheaper than private insurance is simply that it isn't - the latest Health Expenditure Accounts show that benefits provided by Medicare cost more than twice the same benefits provided by private insurance. The overall point is that not counting the fraud rate or these per-enrollee costs paints an incomplete picture of Medicare - to no one's benefit.  

Yglesias comes from the perspective that Medicare's reimbursement of physicians is more in-line with their actual costs and reducing all-around reimbursement to those levels would help shave down health care spending - but this looks at Medicare in a vacuum. The literature has found that hospitals respond to lower Medicare reimbursement by shifting costs to private insurers (in a competitive market they focus on cutting costs, but other literature has indicated that hospitals often operate in more concentrated markets). Indeed, what Yglesias seems to miss is that Medicare may very well drive part of the growth in health care costs by shifting them.

To his credit, Yglesias touches on some important physician-related reforms - he acknowledges that medical school is extremely expensive and that addressing malpractice reform can encourage more people to become doctors. But he stops short of real reforms - reducing the requirements to become a GP; fixing Medicare's terrible specialist-skewed payment model; and aligning Medicare away from the ACO model which encourages consolidation (and drives up costs). These reforms would begin to address both the supply (barriers to entry and consolidation) and demand (Medicare's payment model) side of the equation in a meaningful way.

In an issue brief for the left-leaning Centre for Economic and Policy Research, economist Dean Baker resurrects the idea that Medicare should "negotiate" (read set) prices for drugs. After all, if other federal health insurance programs require mandatory "rebates" for prescription drugs - the Veterans Administration and Medicaid, for instance - why shouldn't Medicare?

Baker's analysis largely focuses on cross-country comparisons with several countries that set prices for prescription drugs - Canada, Denmark, Japan, the Netherlands, and the UK - all of whom spend significantly less per-capita on prescription drugs than the U.S. He concludes that that through "negotiation" with drug companies (cleverly, he never uses the less palatable "price controls") we could save anywhere between $309 and $726 billion over 10-years - savings which would accrue to the federal government, states, and to individuals.

To his credit, Baker at least tries to rebut the pharmaceutical industry's argument that price controls would stifle innovation. According to Baker, strong patent protections (which give companies monopoly pricing power over their products) lead to a more corrupt drug industry, which in turn leads to dangerous drugs being approved (like the recent Vioxx scandal). He offers a proposal by Nobel Laureate Joseph Stiglitz as an alternative - that clinical testing of drugs should be financed entirely by the government, which he thinks would nullify the pharmaceutical industry's argument that patents are required to recoup the high costs and enormous risks required to bring new medicines to market.

(This is a bad idea layered on top of another bad idea. A bad idea sandwich. This would put the government in the position of both paying for new medicines and funding the clinical trials to test them, creating a massive conflict of interest, since approving fewer new medicines would also lower government expenditures. And the process of choosing which medicines to take through clinical trials - ED drugs, HIV, cancer, diabetes - would become hopelessly politicized by Congress.)

Still, Baker's love of price controls is clearly popular with the White House. The president voiced support in his SOTU address for health care cuts similar to Simpson-Bowles (which also included Medicare Part D price controls), and claimed that paying drug companies market prices through Medicare Part D is tantamount to a subsidy.

For what it's worth, Baker's paper is correct (or at least strictly tautological) in its assessment: if the United States were Canada, or Denmark, or Japan, we would pay for drugs like Canada, Denmark, or Japan - and probably pay lower prices. But that's not the case.

The U.S. is different in more than just how we structure our health care system. The U.S. is different in its demographics, per-capita income, social attitudes, and income distribution. All of these are areas where the U.S. varies tremendously with its OECD competitors, and these are salient factors when it comes to evaluating at U.S. health spending.

With that in mind, it is worthwhile to address some of the faults in Baker's assessment.

Comparison of Per-Capita Spending

To compare other countries' spending with that of the United States', Baker looks at PPP (purchasing power parity) adjusted per-capita spending levels on prescription drugs. Certainly, this offers an interesting comparison of prescription drug spending across countries while holding constant purchasing power (essentially what one American dollar buys elsewhere - hint: it buys less health care but more of other goods). But this ignores potential differences elsewhere in the health system - for instance, it may be that higher prescription drug spending reduces spending in other categories (as the CBO has seemingly, though not explicitly, taken into account in their modeling efforts).

For instance, it turns out that the U.S. tends to spend significantly less than Denmark, Canada, or the Netherlands (Baker's OECD sample choice) on long-term care services for the elderly.

Source: OECD StatExtracts http://stats.oecd.org/Index.aspx?DataSetCode=SHA# 2010;
Note: latest data for Japan was 2009; UK Data was unavailable

Does this necessarily mean that these other countries should work to reduce their spending on long-term care? Not really. Demographic needs and provision of healthcare should (and do) reflect country specific policy and political choices - not rote conformity with economic peers. One size really doesn't fit all.

But this argument probably won't assuage those concerned about pharmaceutical price-gouging - so let's look at the issue from a different perspective. We already know that the U.S. spends more on healthcare as a percent of GDP and per-capita than any other OECD country. But what about the share of total health spending represented by prescription drugs?

Source: OECD StatExtracts http://stats.oecd.org/Index.aspx?DataSetCode=SHA# 2010;
Note: latest data for Japan was 2009; Netherlands & UK data on drug spending was unavailable

Here, America's spending on prescription drugs as a share of total health spending is lower than both Canada's (by 40%) and Japan's (by 80%). It's also about two-thirds higher than Denmark's. Again, per-capita costs only provide a crude comparison for health system comparison that doesn't take into account total spending, and the share of prescription drug costs as a burden on the health system as a whole. We might spend more on prescription drugs, but it's also far from the real sources or drivers of U.S. spending considered as a whole - well worth taking into account if you're really concerned about the drivers of U.S. health care spending (as opposed to just singling out a politically convenient industry).

The last aspect of these cross-country comparisons that Baker fails to address is the makeup of prescription drug spending. Who ultimately pays for the drugs? (Baker takes this into account when projecting U.S. savings, but assumes that the makeup would remain the same - probably an unrealistic assumption).

Source: OECD StatExtracts http://stats.oecd.org/Index.aspx?DataSetCode=SHA# 2010;
Note: latest data for Japan was 2009; Netherlands and UK data on drug spending was unavailable;
U.S. data from CMS's National Health Expenditure Tables;
Japan's data does not sum to 100% due to rounding

In fact, when looking at who ultimately pays for prescription drugs, the relative share of spending varies significantly by payer. Likely, to imitate a country like Denmark, the U.S. government's share of spending on drugs would have to almost totally displace private insurance for prescription drugs. (President Obama or Baker might like this, but it's obviously a non-starter in Congress or for the American people.)

Without this shift, lowering prices for Medicare would just lead to a cost shift from public to private payers as drug makers tried to maintain their margins. In other words, employers and patients outside of Medicare would have to pay more, so that the government could pay less. There is no such thing as a free lunch, and so Baker's plan would be to leave someone else to pick up the check.    

Savings are Likely Exaggerated

Baker at least concedes that inflicting massive price cuts on drugs would hurt pharmaceutical innovation, and acknowledges that something would have to be done to offset the impact of the lost industry revenue. Implementing Joseph Stiglitz's proposal (aside from the bad features we mentioned earlier) would cost quite a bit of money and as such, that cost - at least tens of billions of dollars annually - should be deducted from any potential savings.

More on this point: A working paper by health economist Austin Frakt looks at potential savings from restricting Medicare's drug formulary to that of the Veteran's Administration (which is about 40 percent less generous, according to the authors). The results did find savings of about $14 billion if accounting for all Part D enrollees. But there is a significant loss of "consumer surplus" (the difference between what a person is willing to pay and what they do pay) to Medicare enrollees from reduced drug access - and because the authors assume that all drugs are valued equally between VA patients and Medicare beneficiaries (an important simplification), and the net savings are relatively small compared to the gross, these estimates are highly sensitive. In any case, Frakt et al's estimates are significantly less than those that Baker arrives at even over a decade. Certainly, the proposals are different - Frakt offers an operationally feasible proposal which would likely be a more realistic implementation of Baker's.

Finally, it is useful to look at the effects of price controls and restricted formularies on innovation (something that Frakt's paper doesn't explicitly address since it is cross-sectional in nature). In a 2005 report for the Manhattan Institute's Center for Medical Progress, Frank Lichtenberg, a Columbia University Economist, noted that after the VA tightened their drug formulary, VA patients' life expectancy may have declined because of reduced access to newer drugs for VA patients.

All in all, Baker's conclusion - that European-level price controls are compatible with saving money and sustaining innovation - is wildly optimistic.

Concluding Thoughts

Price controls have been shown to stifle innovation, and shifting to a European-based system would require numerous other changes to our health care system, and indeed, our economy as a whole, that would be innovation dampening.

Broadly speaking, European price controls are sustainable (at least in the short term) because the U.S. still allows something close to market pricing. As in the case of global defense spending, Europe can afford to pay much, much less, because the U.S. spends more - proving a global security umbrella.

 If the U.S. were to sharply reduce spending on prescription drugs, Europeans would have to pay much more, or global pharmaceutical innovation would decline sharply. If nothing else, access to new drugs would suffer tremendously as a 2011 essay in the Annals of Health Law notes:

The U.S. pharmaceutical industry has historically been characterized as the market-driven pharmaceutical system of the world...companies in the U.K. have endured profit controls...[this] has led to vast differences in the advancement of pharmaceutical innovation and to significant disparities in patient access to medicines.

Baker indulges in a favorite habit of the left - assuming that free lunches really are possible. In every other sector where governments have imposed price controls (food, housing, automobiles), supply, quality, and innovation dwindle sharply.

Pharmaceuticals may be an exception for the moment, because capital can still flow to jurisdictions (like the U.S.) where market-friendly rules still apply. Ironically, European price controls may benefit the U.S. by making America a destination for risk capital, and pharmaceutical R&D investment - along with the millions of jobs and the health benefits that come with enhanced innovation. Americans undoubtedly do pay more for drugs because Europeans pay less (economists call this third degree price discrimination) but the solution is to ask them to contribute more to global medical innovation, especially through bilateral trade agreements. The Europeans can free ride to some extent, but global medical innovation is lower than it would be if they were paying prices commensurate with their wealth level - and these losses are felt just as acutely by European patients suffering from Alzheimer's, cancer and other diseases.

America does need to rein in health care spending. The best idea in this regard may come from the bipartisan Simpson-Bowles Commission, which suggested capping federal health care spending growth at GDP +1%. (It also recommended Medicaid-level drug rebates for Medicare, but no plan is perfect). This would have the benefit of forcing providers and patients to shift towards the best mix of products and services. This might entail more use of pharmaceuticals, and thus more, not less drug spending but would also be welfare enhancing. Obamacare actually attempts something like this through its ACO model, and its selective productivity cuts to some Medicare providers. The problem is that policymakers insist on keeping their hands on the till, shifting the market to and fro based on the political winds of the moment.

It's a recipe for shipwreck, albeit a time honored one on the left. We can only hope that Washington has the foresight to recognize Baker's idea for what it is - a gross oversimplification that will only produce unwanted consequences. 

Paul Howard and Yevgeniy Feyman

For policy wonks, the words "Simpson-Bowles" hold a special place in their hearts. The commission, headed by former Senator Alan Simpson and former Clinton Chief of Staff Erskine Bowles, released a massive report at the end of 2010 that sought to completely eliminate the federal budget deficit by 2035 with a 2 to 1 spending/revenue ratio. The plan wasn't perfect, and ultimately didn't pass muster at the Senate, but many of the ideas it brought up were based on solid evidence with a real chance of achieving the broader budgetary goals.

One of the less palatable ideas proposed in the plan was to impose Medicaid drug rebates (price controls, in reality) for Medicare Part D for the dual-eligible population (those who are eligible for Medicaid and Medicare coverage). While it's been estimated that this plan could save $49 billion in Medicare spending, the impact on innovation could be significant - cutting prices for drugs dampens the incentive for drug companies to innovate. The dual-eligible population in particular often suffers chronic conditions for which innovative medicines are in short supply - these rebates wouldn't be of much help.

But bad ideas are like a boomerang - they always seem to return.

In his State of the Union address, President Obama characterized the price Medicare pays for drugs as subsidy to the pharmaceutical industry:

On Medicare, I'm prepared to enact reforms that will achieve the same amount of health care savings by the beginning of the next decade as the reforms proposed by the bipartisan Simpson-Bowles commission...We'll reduce taxpayer subsidies to prescription drug companies

To put it simply, paying the market price for a drug isn't a subsidy - it's business.

More importantly, however, by pursuing what are effectively price controls, the president is wasting a potential watershed moment for dealing with the dual-eligible population - a group of beneficiaries that account for 15 percent of Medicaid's population and 21 percent  of Medicare's population, but account for 39 and 36 percent of the programs' spending, respectively. Rather than focus on poorly conceived price controls (which will dampen the incentive to innovate and will likely shift costs to non-dual eligible) for a narrow population, the president should have instead announced a greater focus on care coordination and elimination of duplicative services for dual eligible.

Hiding behind the feel-good moniker of "bipartisanship" may help approval ratings, but it may leave dual-eligibles without the innovative drugs they need.

In a move sure to excite his base, President Obama recently declared that cuts to Medicaid are off the table, while simultaneously ruling out savings of over $100 billion to Medicare achieved through raising the eligibility age.

No doubt, in the State of the Union tonight, the president will say that he's still open to making "balanced" spending cuts.  The problem is that, as a matter of pure accounting, the president's position is deeply disingenuous. And a recent brief by the CBO illustrates precisely why.

The above chart, from the first page of the CBO report, illustrates the growth in means tested programs since 1972, as a portion of the economy. While direct cash assistance and nutrition, housing and education programs have grown from 0.9 percent of GDP to 2.1 percent, means-tested healthcare programs (primarily the federal portion of Medicaid) have on their own have reached almost 2 percent of GDP. If you factor in state spending on Medicaid, that number is close to 2.5 percent.

But that's not all. Future growth is projected to increase at an increasing rate!

This growth in health care spending will crowd out spending from the other two categories - spending on Nutrition and Education programs will actually fall, while cash assistance programs will grow at relatively slower rates.

The Obama administration operates on the idea that we might have a spending problem - if you squint your eyes hard enough you might be able to see some inefficient spending, and some places where we could shave off a few billion.  But the President's basic stance is that the largest and fastest growing parts of federal spending are the holy-of-holies, miraculously efficient and effective. 

Still, if you don't have a spending problem (but you have to acknowledge that there is some problem) the only thing left is a revenue problem! The president's continued insistence that "the rich pay their fair share" certainly echoes this mentality.

Unfortunately for the president, reality rears its ugly head again.

Spending & Revenues % of GDP

Source: Bureau of Economic Analysis / Federal Reserve Bank of St. Louis Data (FRED)

As the table above shows, current U.S. spending is a full 5.3 percentage points above the 65-year average; meanwhile, revenues are only 1.3 percentage points below. As the economy continues to recover, revenues will catch up to the long terms trend fairly quickly.  But it's far from clear that the economy can grow fast enough to drag spending down to trend.

Recall that total Medicaid spending represents about 2.5 percentage points of GDP, but by 2021 is projected to hit nearly 4 percent. If we are to abide by the president's preference and leave Medicaid alone, our options for deficit and debt reduction become excruciatingly constrained. And not dealing with the deficit - as the CBO recently noted can lead to a long-term 0.5 percent drop in GDP over an 11-year horizon.

So as you watch the president's State of the Union address tonight, watch for the president to tell you that we've got to protect Medicare and Medicaid, and have a balanced approach to deficit reduction. 

And remember that if you do the first, you can't do the second. 

In the Washington Post on Monday, Louisiana Governor Bobby Jindal issued an invitation to the President to meet with Republican governors to discuss Medicaid reform:

Medicaid operates under a 1960s model of medicine, with inflexible, one-size-fits-all benefits and little consumer engagement and responsibility. Expanding the entitlement program as it stands would further cement a separate and unequal tier of health coverage. Without fundamental reform, Medicaid will continue to deliver what it has for decades: limited access, poor quality and budget deficits.

A number of Republican governors have asked to meet with President Obama to discuss their solutions, but the White House has ignored these requests. The president claims that he wants to work across party lines to get things done for the American people, so perhaps he could start by meeting with Republican governors who want to solve our nation's health-care problems.

Jindal knows what he's talking about. The governor inherited a Medicaid program that cost $7 billion and covered nearly a third of the state's population (27%), but the Louisiana still ranked nearly last in major health rankings (49th), performed badly on pre-term births and infant mortality measures, with high rates of chronic disease and low use of preventive services.

In other words, like a lot of states, Louisiana was paying a lot of money for Medicaid fee-for-service care and not getting a lot of value for it. In January 2008, Jindal launched a Medicaid Reform Advisory Group to look at improving coordinated care for Medicaid enrollees. Today, the state's Medicaid program (Bayou Health) is focused on prevention, coordination, and actively managing patients with chronic diseases through patient-centered medical homes.

The state is also collecting Medicaid data on nearly 40 health measures, and can penalize private Medicaid plans that don't meet quality benchmarks. If plans meet their health benchmarks, they can also benefit from "shared savings" with the state. Medicaid enrollees can choose from a menu of private health plan offerings, but if they don't, they are auto-assigned to the best plans. Currently, Bayou Health is projected to save $135 million thanks to Governor Jindal's reforms.

You can't call these reforms liberal or conservative - they're just focused on improving value, and states from Louisiana and New York are implementing similar programs. In fact, the Obama Administration - which doesn't hesitate to vilify private insurers when it's politically convenient - is running a demonstration project whereby about 40 percent of all Medicaid/Medicare "dual eligibles" nation-wide are being moved into private managed care plans.

What Governor Jindal does want is increased flexibility to design his state's Medicaid program to meet the needs of different populations - along with the ability to vary Medicaid co-pays, benefits, and plan designs. There is no "one size fits all" Medicaid reform that will work for all populations and all 50 states, and greater state flexibility (along with better transparency for state outcomes) would just recognize this reality.

Today, the only way for states to get broad program waivers is through a lengthy 1115 waiver process; and every new administrations (Democrat or Republican) also have their own policy preferences, which limit states' ability to experiment and the federal government's ability to consistently learn from what works. Here, Jindal has a simple idea that should appeal to governors of both parties:

The process by which states negotiate for flexibility, called "waivers," is broken. Federal officials should have greater accountability for timely review of waiver applications. In particular, waiver applications based on those already approved in other states should be fast-tracked. HHS should allow states to opt in to a more flexible long-term-funding arrangement, allowing them to design programs that best meet residents' needs, rather than requiring the same package of services for every individual. At the same time, federal and state officials could agree to greater accountability for improvements in health outcomes, not just processes.

Jindal's proposal would go a long way to giving Republican governor's what they want (flexibility) while assuring provider and Medicaid advocacy groups that states will still be held responsible for improving outcomes for Medicaid enrollees. In other words, it's a win-win.

A broad based conversation about Medicaid reform should have happened long before Obamacare was passed, but instead the governors were largely locked out of the discussion - a big reason many governor's are balking at Obamacare's massive Medicaid expansion today.

It's not, however, too late to go back to the drawing board and fix a program that millions of Americans depend on for safety net care.

The president keeps saying that he's open to any good ideas, whether they come from Republicans or Democrats. Governor Jindal is taking him at his word.

Mr. President, are you willing to meet him halfway?

Imagine that your car has been overheating and the starter motor is quickly failing. The wheel alignment is also askew. You take your car to a shop. The mechanic at the shop suggests installing a new battery, reupholstering the seats, and installing some oversized off-road tires. You protest that these proposed solutions won't address the key issues and will perhaps make them worse. After all, bigger tires will tax the car's engine, leading to more overheating. Instead of explaining how these fixes work help, your mechanic belittles you for being resistant to change and for not trusting his expert knowledge. He says that his approach is the only way to fix the problems your car faces. Reluctantly, you agree to the repairs.

After you get your car back from the garage, and the bill was double what you were expecting, the starting and overheating problems return in earnest. The alignment is worse than ever. When you contact your mechanic to alert him to the unfixed problems, he then tells you in a solemn voice that your car has an overheating and starting problem. Oh, and the alignment is off.

That's what you brought the car in for initially!!! That's what you wanted him to fix!

Unfortunately, this analogy describes President Obama's misdiagnosis and improper "fixes" to the nation's health care system. Obama described his flavor of health care reform as "one of the best ways--in fact maybe the only way--to reduce those long-term costs." Further, he called the Affordable Care Act "one of the biggest deficit reduction measures in history." He added: "Everybody who's looked at it says that every single good idea to bend the cost curve and start actually reducing health care costs are in this bill." But that didn't stop Obama from saying last week that, "We don't have a spending problem. We have a health care problem."

When the Affordable Care Act passed, Obama said that "From this day forward, all of the cynics, all the naysayers--they're going to have to confront the reality of what this reform is and what it isn't." Okay, if Obamacare doesn't really help reduce health care costs and doesn't really reduce the federal budget deficit, then what does it do, Mr. President?

While 11^th hour negotiations prevented the country from toppling over the fiscal cliff, we're still teetering on the edge. Most provisions of the sequester were simply pushed back by two months (to March 1^st) and the deal included no reform to entitlements - which play a huge role in driving federal spending. This means, effectively, that lawmakers have simply kicked the can down the road - it is a matter of fact that some savings will have to be squeezed out of entitlement programs; a new report by United Health Group, one of the largest insurers in the country offers some ideas on how to do so without cutting benefits or reimbursements to providers.

Through a combination of managed care innovations, better use of fraud-fighting technology, and various care coordination reforms, the report argues that some $542 billion in savings can accrue to the federal government over 10 years, and an additional $69 billion can accrue to states.

United Health Group Savings Estimates ($ billions)

We already know that Medicaid Managed Care programs have resulted in savings for states, without reducing access to services, (in a managed care plan, beneficiaries are generally limited to a tighter network of providers, but this leads to better coordination of care and lower out-of-pocket spending for beneficiaries) so it shouldn't be a surprise that applying a similar framework would lead to savings in Medicare as well. Indeed, the report advocates adopting a "Medicare Administrative Service Organization" program, which my colleague Avik Roy has written about extensively.

One of the more interesting highlights of the report is the enormous savings potential from reforms to the treatment of dual-eligibles - a full $189 billion over 10 years

Dual-eligibles are arguably the most vulnerable, and difficult to treat population - they are those who qualify for both Medicare and Medicaid - generally extremely low-income seniors or the disabled. The nature of treatments that they require, often involving complex chronic diseases, makes coordinating care between Medicare (which covers doctor's visits and prescription drugs) and Medicaid (which would cover long-term care if it is needed) extremely difficult. The lack of care coordination is an important contributing factor to estimates that total spending on dual-eligibles may reach one-third of all spending on Medicare and Medicaid over the next 10 years.

Through better care coordination and high levels of participation by dual-eligibles, the study argues, these significant savings can become a reality. The Centers for Medicare and Medicaid Services are currently testing two models of care coordination - the first is a capitated payment model where states and the federal government to contract with a third-party to provide integrated Medicare and Medicaid benefits, while the second allows states to enroll dual-eligibles into a managed care program. Both models have large savings potentials, and the study doesn't come down in favor of one model over another, explaining that both models have important cost-saving benefits.

Under the capitated model, for instance, insurance plans or other intermediaries could be assured of two steady, but capitated, payment streams from the states and the federal government in exchange for coordinating and integrating care for dual-eligible populations. The managed care approach would probably require some sort of capitation, but would put more control of the programs in the hands of states, which are arguably better positioned to know unique the needs of their own populations.

In any case, the results from CMS testing these models will have to be monitored to see which performs best; one thing is certain, however - for once, we may be able to have our cake and eat it too.

An editorial in the LA Times has received some backlash from physicians for expressing support for reforming Medicare's payment system from fee-for-service to a bundled payment.

One Physician from Santa Clarita perfectly sums up the concerns:

Putting financial burdens on doctors for better results ignores social factors, including personal compliance.

Some skin in the game, however, is exactly what we need to demand of physicians.

Under its current structure, Medicare - as with private insurance - reimburses providers based on the complexity (determined somewhat arbitrarily through the Resource Based Relative Value Scale) and volume of their procedures. Predictably, as with any volume-based payment system, this encourages overuse of the system and contributes to fraud. While private insurance, not reliant on taxpayer money, has significant incentive to reduce waste and fraud resulting in higher overhead, Medicare instead has an incentive to keep such "overhead" costs low, resulting in unrealistically low administrative expenses (if Medicare were to combat fraud at the same level as private insurance, their administrative expenses would likely be similar). These dynamics mean that fee-for-service reimbursements may work with private insurance (which tries to reduce waste and fraud) but may not be appropriate for a government program with less incentive to do so.

Under a bundled payment structure, care providers would receive a sum of money to provide for the healthcare needs of their patients while still ensuring quality. By tying providers' income to the health of their patients rather than the volume of care provided, two problems are addressed: first, the perverse incentive to needlessly increase the volume of high paid specialist services is minimized; but secondly, it eliminates the need to constantly return to bickering over Medicare's Sustainable Growth Rate, since growth would be more manageable under pre-determined budgets.   

Indeed, the original LA Times article cites a study showing a 10 percent reduction in costs under a bundled payment approach, with the same level of quality of care.

But it's important to also realize the difficulties that physicians would face if Medicare adopted bundled payments. The same physician concerned about taking on financial burdens also worries about tort reform:

There is also a chance that should I neglect to order a test, I'll get sued by an avaricious malpractice lawyer.

For real Medicare reform to take place, physicians have to be held accountable not only for a patient's health, but also for unnecessary costs incurred by the publicly-funded system. At the same time, other healthcare reforms will be necessary to align the rest of the system with the new reimbursement schemes, and this will no doubt include tort reform. Bundled payments offer a good start to help cut the overuse of Medicare, while ensuring that beneficiaries still receive the same quality of care. 

Paul Howard and Yevgeniy Feyman

The recent bill passed by both the House and the Senate effectively kicked the fiscal cliff can two months down the road and was just approved with the President's signature. Among the provisions in the bill is one that also postpones, for the next year, a 26.5 percent cut to physician reimbursements from Medicare. Commonly known as the 'doc-fix', this measure has been used for many years to avert reimbursement cuts required by Medicare's Sustainable Growth Rate. While physicians can rest easy for the next year, part of the cost of averting the payment cut is being funded by cuts to hospital reimbursements.

So what are we left with? The underlying problem with how Medicare's SGR is calculated remains unattended - come 2014, Medicare providers will once again be at the mercy of congressional maneuvering. Moreover, the hospitals facing the cuts are those that primarily treat poor populations. (Disproportionate Share Hospitals).

The broader problem, which isn't addressed, is how Medicare's payments are calculated - the Resource Based Relative Value Scale. Developed decades ago, the RBRVS guides how Medicare reimbursements are structured based on four categories: mental effort, physical effort, skill, and time. Seemingly uncontroversial, the RBRVS has steadily grown to favor specialists over primary care doctors - reimbursements for specialist services have grown tremendously (even as many procedures - like a cataract extraction - have become more routine and automated) while primary care physicians have seen their reimbursements remain static. Certainly, specialists perform often complicated procedures that require years of training to perform properly - and they deserve to be compensated fairly for their work. But primary care is similarly demanding, and patients rely on their physicians to help diagnose one out of possible dozens of ailments and refer them to the appropriate specialist - no small feat with an ever growing number and variety of chronic diseases.  We're also asking primary care physicians to shoulder more of the burdens of chronic care management, in effect asking them to become health care's version of air traffic controllers.

Of course, it's possible to avoid dealing with the RBRVS entirely by simply changing how the SGR fee update is calculated with a method to always insure a positive increase.  But is this the right way to approach the question?

Congress should be agnostic about who performs a service, as long as the service is delivered effectively and efficiently.  Congress should also set up a system that encourages innovators to replace expensive labor (services) with much less expensive diagnostics.  By basing the RBRVS on the "mental skill" required to perform services, the system implicitly biases the increased utilization of labor rather than diagnostics. 

Or, to put it another way, IBM's Watson could eventually deliver routine and complex analysis of a patient's health through a low-cost tablet app offering supercomputing services to a physicians' assistant, nurse, or primary care physician. This reality isn't that far away - The Tricorder X-Prize, offered by manufacturer Qualcomm, seeks to reward the first company to "put healthcare in the palm of your hand" by essentially creating the ubiquitous Star Trek gadget. Mark Mills, senior fellow at the Manhattan Institute, writes:

The ultimate Tricorder idea is to access the wealth of (voluntary) data about what you've been doing, eating, how your own biological machinery has been operating, and marry it with a rich stream of highly precise and real-time physiologically-specific information about what's going in your body right now - wherever you are - and link this wirelessly to the analytic computing power that resides in the Cloud.

The new world of "Big Data" makes this possible - and with the exabytes of health data out there, it will help put healthcare decisions into the hands of patients.

The RBRVS and the SGR lock American health care into labor arrangements that are swiftly being overtaken by technologies that have the potential to radically change the cost and quality of American health care.  But their use will be constrained as long as pricing signals are based on assumptions about the value of labor that are woefully outdated.

A better solution would be to get out of the business of pricing services entirely, through a premium support mechanism that encouraged robust competition among many different networks of competing health care providers. Pricing competition will encourage insurers and providers to seek out the most cost effective and innovative mix of pricing and services.    

Then we won't have to worry about the SGR or the RBRVS ever again.  And that would be a priceless gain for American health care.   

Paul Howard & Yevgeniy Feyman

Democrats sold -and continue to sell the ACA - as a way to cover the "millions of people" with pre-existing conditions who can't get affordable insurance. For instance, in defending their recently released guaranteed issue regulations, HHS claimed that 129 million Americans have pre-existing conditions.

This is a huge bait and switch. The vast majority of Americans with "pre-existing conditions" already have insurance. Why? Age is strongly correlated with developing a chronic illness - and seniors are covered by Medicare. If you're disabled and poor, and can't work, you're eligible for Medicare and Medicaid. The low-income poor (healthy or not) are already eligible for Medicaid. In between, the majority of Americans have employer-provided insurance, and are also already protected from pre-existing insurance exclusions or rate hikes due to illness, through HIPAA.
Who's left then? Not that many people.

In fact, preliminary results from the Center for Disease Control and Prevention's (CDC) National Health Interview Survey indicate that even among the uninsured, only 1.7 percent considered themselves to be in poor health, compared to 6.8 percent of those in Medicaid, and just .6 percent of those with private insurance.

A Medical Expenditure Panel Survey report from 2007-08 also estimated that only 16 percent of the uninsured had two or more chronic conditions - compared to one-third of those with private insurance and 50 percent for public (Medicare and Medicaid).

In a 2010 National Affairs article, James Capretta and Tom Miller estimate that only 2-4 million uninsured Americans with pre-existing conditions need additional financial help accessing insurance, preferably through high risk pools.

High risk pools allow people with serious pre-existing conditions get affordable coverage without increasing insurance costs for young and healthy uninsured. Yet this is where Obamacare has also failed, despite a modest effort. Under the law, federal high-risk pools were established to provide access to healthcare for patients without insurance, and with pre-existing conditions. A recent evaluation has found that only about 45,000 people signed up for these pools; a fraction of the 375,000 that CMS expected. Reasons proposed for the failure of the pools include low funding (only about $5 billion) and high costs for signing up. Regardless, for the last four years Obamacare has failed to expand healthcare to those with pre-existing conditions who really needed it.

Ironically, Obamacare also attacks consumer driven health plans - which a recent Mercer report credits with helping to hold down health insurance inflation to a 15-year low - threatening to drive up insurance costs just as we're identifying the tools to keep them in check . Various requirements such as the Minimum Loss Ratio (that insurers must spend at least 80 percent of premiums on benefits) and minimum actuarial value (that plans must cover a minimum of 60 percent of expected healthcare costs) make consumer driven health plans - which often have low premiums with high deductibles - less viable.

Ultimately, the biggest flaw with the ACA's insurance market reform is that it enforces expensive insurance regulations on the entire small group and individual insurance markets, increasing the cost of getting insurance for the vast majority of uninsured who are basically in good health. It also scales those subsidies up to 400% of the poverty level, to people who could easily afford to purchase it on their own.

Obamacare's failure at what should have been its primary goals leaves the door open for conservatives to start pushing for reform. The House could pass legislation repealing Obamacare's community rating and guaranteed issue regulations (as our colleague Avik Roy has suggested), and fixing Obamacare's flawed high risk pools. Paring back the subsidies (from 400% to 200% or 300%) would also lower Obamacare's price tag while still helping people who need it the most.

Governors of states that refuse to establish Obamacare's health exchanges (or expand Medicaid coverage) could also push for legislation to allow Medicaid funds to be used to help purchase private insurance for that vast majority of non-disabled or elderly Medicaid enrollees. This would provide high quality private coverage, and prevent people from shuffling between Medicaid and private insurance as their income changed. True state flexibility in Medicaid program design might also convince many governors to re-think their opposition to Obamacare's Medicaid expansion.

The debate on fixing or fighting Obamacare is likely to continue to for years to come. In the meantime, moderates and conservatives should point out that Obamacare's biggest shortcomings are self-inflicted - they didn't have to happen in the first place but can (and should) be remedied.

While states mull over decisions on health insurance exchanges, another decision is looming on the horizon - whether or not to expand Medicaid as the Affordable Care Act (ACA) calls for.

Many governors have held off until after the election to make their decisions, with a number of states like Texas and Florida deciding not to, but the majority still remains undecided. Though the ACA calls for the federal government to foot 90 percent of the cost, the remaining 10 percent still represents an increase in states' costs in the midst of a dire budgetary crisis plaguing much of the country.

Budgets, however, are not the only decision rule states should be using to determine whether or not to expand their Medicaid rolls. Medicaid's track record on access to care and health outcomes is generally poor when compared to private insurance, and these measures often end up being only slightly better than for those who are uninsured, and are often statistically insignificant.

A study released today in the Journal of the American Medical Association (JAMA) looked at postoperative morbidity rates for patients that received surgeries for brain tumors, controlling for the insurance status of the patient. The results? At first glance, unsurprising. After fully adjusting for patient characteristics and comorbidity, uninsured patients were 2.62 times as likely (a hazard ratio of 2.62) to die after surgery as those with private insurance.

While in the adjusted analysis, the results for Medicaid recipients were not statistically significant (the results found Medicaid patients 2.03 times as likely to die relative to those with private insurance, but were not significant at the 5 percent confidence level [analogous to a 95 percent confidence interval]), the original, unadjusted  analysis (which doesn't take into account differences in patient characteristics) shows no statistically significant difference between Medicaid patient survival rate and the uninsured (2.6 percent and 2.3 percent, respectively).

This is yet another study that sheds light on the relatively poor outcomes of patients on Medicaid, which may not be significantly better than those of the uninsured. The relatively poor outcomes should come as no surprise, as Medicaid's lower reimbursement rates for specialists tends to cause significant access to care problems. And poor access to care will likely make Medicaid patients on the whole less healthy than those with private insurance, increasing the chances that they will have other complications that may lead to higher mortality.

A critique in JAMA brings up important issues that need to be taken into account. Of relevance is one in particular - the number of uninsured patients who enrolled in Medicaid during their hospital stay; these patients would be better categorized as being uninsured rather than having Medicaid. Nonetheless, the results echo what many have been arguing for some time - that Medicaid delivers poor outcomes that are cost-inefficient for states.

In a report that is sure to be misused and abused by those championing states' Medicaid expansions, the Government Accountability Office (GAO) announced that based on their survey of the states, and an analysis of national expenditure data, Medicaid beneficiaries have no more trouble accessing medical care than those on private insurance. If true, this finding would dispel the notion that cutting provider payments creates access problems. The problem isn't with GAO's report or methodology - it's with people that will only read the top-line heading and run with it.

The report found that among Medicaid beneficiaries who had full-year coverage, differences in those reporting problems with access to care were not statistically significant. Specifically: 3 percent of those with private insurance had difficulty with access to medical care, compared to 3.7 percent of those with Medicaid; 2.4 percent of those with private insurance had difficulty with access to prescription medicine, compared to 2.7 percent of those with Medicaid; and 3.7 percent of those with private insurance had difficulty with access to dental care compared to 5.4 percent of those with Medicaid.

According to GAO's methodology, the confidence interval for these estimates is about ±1.5, which makes only differences in dental care access problems statistically significant for those with a full-year of insurance.

If this was all that GAO reported, it would be fair to say that Medicaid has no access problem compared to private insurance.

It isn't.

First off, GAO notes that across the years they analyzed, more states increased payment rates for providers than the number that reduced them - this would mitigate the access problem to an extent.

Note: some states increased payment for some providers, cutting payments for others; totals don't add up to 50.

Where Medicaid's access problems show up, however, is in GAO's analysis looking at working-age adults and whether coverage was maintained for a full-year or less.

The difference between those with private insurance and Medicaid is astounding. While only 3.3 percent of those with full-year private insurance had access problems, a full 7.8 percent of those with full-year Medicaid coverage had access problems. Even the uninsured who had coverage for part of the year fared better than those with partial-year Medicaid coverage; the difference between them and those who had Medicaid coverage for a full-year was statistically insignificant. (The confidence interval is ±1.7 for this sample.)

Finally, waiting times were a significant problem for those with Medicaid (9.4 percent versus 4.2 percent of those with private insurance versus 6.8 percent of those who were uninsured).

The reality is that the GAO report merely reinforces what should be common sense about Medicaid - if you pay providers less than private insurance, you will encounter access problems. But it also indicates that there is something systemic about younger (under 18) and older (over 65) patients that causes them to have fewer access problems than working-age adults, even with Medicaid. The aggregate numbers presented earlier in this post include children covered under Medicaid and Medicare/Medicaid dual-eligibles (seniors who qualify for Medicaid and Medicare). Children are likely will be healthier and will likely be seeing a care provider for more routine causes like a checkup or a flu shot. Because these are more routine, less expensive procedures, it will be easier to find a provider. For dual-eligibles, Medicare would cover many routine procedures; because of its higher reimbursement rate compared to Medicaid, access problems would be mitigated. As a report from MedPAC in 2004 confirms, dual-eligibles generally have good access to care, and those with supplemental insurance, even better.

GAO's findings should make states question whether to expand their Medicaid programs as they are - particularly when outcomes, relative to private insurance and even no insurance, are questionable. Avik Roy, at the Apothecary has written extensively about this. To be fair, Austin Frakt at The Incidental Economist has pointed out repeatedly that outcome-based measures in Medicaid don't necessarily imply causation and could very well be a product of self-selection into the program.

My good friend Steve Parente, professor of finance at the University of Minnesota, had a great op-ed in Politico on a "grand bargain" on deficit reduction in July 2011 . It is well worth re-reading before the President and Republicans meet on Friday to try and avert the "fiscal cliff" and, just possibly, find a serious way to address the spending challenges facing the nation.

First, a word about taxes. The president's proposal rests on the premise that fixing our fiscal woes cannot be accomplished without repealing the Bush tax cuts for Americans making more than $200,000 for individuals, and $250,000 for families. Veronique de Rugy, from the Mercatus Center, puts the president's proposal in context - how much he proposes to raise, and how much he proposes to spend over the next 10 years.

To put it another way, the president's proposal will raise about $80 billion annually in new revenues from "the rich" - when the U.S. is running annual deficits of about $1 trillion, or more than 10x that amount.

So where is he proposing to get the other 90% in spending cuts or revenues to balance the budget and pay down the debt, currently pegged at over $16 trillion?

(Cue the crickets.)

Taxing the rich is nice campaign slogan, but as a policy proposal it is utterly vacuous.

What would a serious policy proposal look like? That's where Steve comes in. His proposal would
generate $4 trillion in deficit reduction over the next 10 years, and over twice that further out. How?

1. Eliminate the employer tax deduction for health insurance and replace it with a much smaller individual tax credit. The current deduction costs the Treasury over $250 billion a year today, and helps drive up health care inflation. But wait - wouldn't the president object to this? Yes, but he'd be breathtakingly cynical about it. Obamacare already contains a 40% excise tax on Cadillac plans that is a back door way of attacking the employer tax deduction. Conveniently, it doesn't take effect until years after president Obama leaves office. Move it up, and make it serious.

2. Turn Medicare into a defined contribution plan with growth pegged at inflation plus one-half of the three years' previous growth in U.S. productivity. This would allow seniors to shop for plans that met their basic needs and focus lawmakers on making the U.S. economy as productive as possible, growing the wealth needed to sustain our entitlement programs.

3. Cap subsidies under Obamacare at 300% of the federal poverty level (down from 400%), repeal the medical device tax, and cap Medicaid growth and allow states to draw down their capped allotment from the federal treasury - this could be modeled after the current Rhode Island global Medicaid waiver.

The mix of revenue raising strategies, entitlement reforms that spur competition in health care markets, and Medicaid reforms would produce enormous savings and help rationalize federal spending.

As Steve notes,

This health bargain is likely to yield a savings of $4 trillion over 10 years, extrapolating from existing Congressional Budget Office estimates, and more than twice that amount over 20 years. Because these policies are based on existing CBO estimates, legislation can be gift-wrapped and delivered to the president's desk.

The likelihood of such a grand bargain seems slim at this point, since the President has given no serious signals that he's willing to consider reforms to existing entitlements, let alone the Affordable Care Act.

But it would be worth pushing for. The President could be calling for raising taxes on the "wealthy" as cover on his left flank for fundamental reforms and a grand compromise with Republicans. After all, at this point he doesn't need to worry about his base for his next re-election.

And if conservatives and moderates push for a grand bargain and are rebuffed it will at least clarify who is really serious about fixing the house of cards that is the U.S. budget.

At the last interim meeting of the American Medical Association (AMA), which concluded on November 13, delegates from around the country voted to officially support transitioning Medicare from a defined benefit entitlement, to defined contribution - this is essentially what Mitt Romney proposed during the past election.

Currently, Medicare operates under a defined benefit scheme - a set of benefits are covered under Parts A and B (inpatient and outpatient care, respectively) with the option to purchase Part D (prescription drug) coverage as well, for relatively low premiums. Because neither premiums nor the Medicare tax revenue come close to covering the costs of the coverage, Medicare tries to shift costs around by using an outdated formula called the resource based relative value scale (RBRVS), and ends up paying doctors around 80 percent of what private insurance does.

Even with Medicare's creative fiscal magic, Medicare's actuaries routinely note that the pace of spending is unsustainable - neither in the short-run nor the long-run, and its solvency is at best projected to last through 2024.

A proposal routinely offered as an alternative to Medicare's unsustainable trajectory is competitive bidding with premium support. Under this type of reform, private insurance plans would bid for Medicare beneficiaries, with a benchmark plan (under the Romney plan, the second-cheapest plan would be the benchmark) setting the level of support from the government (in the form of subsidies to beneficiaries), which would be scaled based on income. Most proposals also keep traditional Medicare intact for seniors who choose to remain in it.

In essence, this is what the AMA endorsed as a means of protecting seniors' access to affordable health insurance, while trying to fix Medicare's financial woes.

Critics are often quick to point out that Medicare Part C (also known as Medicare Advantage), which allows beneficiaries to purchase private insurance, costs about eight percent more than traditional Medicare. While this may be technically true, these critics address only one set of numbers to draw their conclusions - the amount that the government pays these plans. These amounts are set by a complicated benchmark formula, which private insurers routinely underbid. However, rather than paying private insurers the amount that they do bid (which is about $717 less expensive than traditional Medicare), Medicare pays them the benchmark amount, which is greater than traditional Medicare.

If the benchmark were to be set through the bidding process (based on what the second-cheapest plan bids), rather than government administration, there is more than enough reason to believe that the plans could deliver the same coverage as traditional Medicare at less cost.

While premium support may be off the table for the time being, the AMA's backing may lead Senate Democrats to take such proposals more seriously in the future.

As Congress teeters on the fiscal cliff, expect liberal advocates to resurrect the idea of having HHS "negotiate" directly with drugmakers for medicines covered by Part D, in addition to inflicting automatic Medicaid rebates onto medicines for dual eligible seniors (eligible for both Medicare and Medicaid) as well as seniors who are eligible for low-income subsidies.

This is a bad idea. First, CBO has said that there is no evidence that HHS could get any better deals than the Part D Pharmacy Benefit Managers (PBMs) are getting, at least not without creating a restrictive drug formulary that would ration seniors' access to new medicines. Since Medicare accounts for about 25% of U.S. prescription drug sales, this would represent a tremendous blow to innovation, allowing some savings today in return for fewer medicines tomorrow - along with more unalleviated suffering and death.

The President is always talking about the need to invest in the nation's future "infrastructure," and Part D is a market-based investment in America's ability to continue to produce cutting edge medical advances. It's a program with a proven track record of high satisfaction among seniors (about 90%), fiscal savings for taxpayers (about 40% below initial estimates), and promotes a much needed focus on competition and value in health care markets. Rather than figuring out how to gut Part D, we should be modeling the rest of Medicare around it.

Markets also show signs of developing new paradigms for cost control in areas that haven't seen much evidence of it in the past, like oncology. Take the recent decision by Memorial Sloan Kettering to not use a high priced colon cancer drug from Sanofi that doesn't (at least according to MSK) offer any advantage compared to existing colon cancer therapies like Avastin.

Individual cancer centers haven't normally entered into the cost debate, because they are directly reimbursed by Medicare at cost-plus, a 6% mark-up based on the average price of the drug. Kudos to MSK for pushing back, and getting a price concession from Sanofi.

This is the way markets are supposed to work, with providers, insurers, and manufacturers negotiating in an open market. Insurers can then compete on price and quality for consumers' business - something that doesn't happen much in Medicare outside of Part D and Medicare Advantage plans. Instead, most seniors opt for fee-for-service health care provided by any willing provider, no matter how inefficient or expensive. Until seniors have better incentives to actually shop for value, Medicare won't be able to consistently deliver high quality, cost effective care.

For all of the bad press that drug companies get, pricing for pharmaceuticals is actually much more competitive and transparent than in many other parts of the health care sector (like surgery, for instance), because drug companies submit reams of data on product safety and effectiveness when they submit New Drug Applications to the FDA. This data can then be mined by insurers and PBMs to create drug formularies and tiered pricing arrangements for consumers. Also, when drugs lose patent protection, they become cheap generics, generating massive cost savings for the health care system while still delivering enormous benefits.

But this all depends on companies being able to recoup their massive costs in drug research and development during the relatively brief time their medicines are still protected by patents. Price controls for Medicare Part D would undermine incentives for innovation - without encouraging providers to create better bundles of care that would also include drugs, tests, physician services, etc.

In other words, we shouldn't be setting prices in any discrete health care silos (for drugs, hospital care, etc.) if we want to generate maximum value for seniors and taxpayers.

Medicare's history of price setting has unleashed a tsunami of unintended consequences across the health care system, since it is the nation's single largest health insurer. As the HHS inspector general noted recently, providers are in the habit of gaming Medicare reimbursement costs ("upcoding") to maximize reimbursements.

As a result, under Medicare's administered pricing scheme we get both too little of some services (that are poorly reimbursed) and too many of other services (that are overcompensated). Medicare can never find the "right" price, because providers have much more fine grained information about the costs of their products and services, and their own patient mix, than Medicare does. The result is a game of Medicare price whack-a-mole that never really controls costs or offers consistently high quality care across Medicare's vast universe of doctors and physicians.

Since premium support is likely off the table for the time being, there are still many other things that Medicare can do to improve care coordination and value. We should bundle Medicare services by putting Parts A&B together, with one premium for seniors, which would encourage providers to better coordinate care. We should allow administrative services organizations (ASOs), widely used by large private employers, to set up networks of preferred providers in Medicare, and offer seniors incentives - through reduced co-pays or enhanced benefits - to utilize low-cost, high quality providers.

ASOs could also represent an appealing ideological mid-point between premium support, traditional Medicare FFS, and Medicare Advantage plans. The key would be to bundle payments and have all providers "go naked" on their outcomes data so we have some correlation between the money spent and actual performance. Additional, web-based tools could then help seniors find the providers who offered the best care at the lowest cost.

Indeed, this approach is already being tested by United Healthcare at a number of oncology centers around the country. In an effort to control costs of cancer treatment, the insurer will provide up-front payments for a typical 6 to 12 month course of treatment, and allow the oncologist to determine the specifics, rather than paying by volume of care.

An earlier study published in the Journal of Oncology Practice found evidence to support this type of approach, identifying some $9,000 in savings for patients on evidence-based pathways in the treatment of lung cancer, with little change in 12 months survival rate. Studies like this can provide a benchmark for weighing how different treatment strategies and practice designs affect the cost of care and health outcomes and - most importantly - inform patient choice in the oncology setting.

ACOs, traditional insurers, physicians groups, etc. could all bid to create these types of networks, available on the same type of exchange mechanism used in Medicare Part D. The upside of this approach is that providers will finally have better incentives to offer the best bundle of care for seniors, rather than perennially lobbying Medicare and Congress for this or that billing code or reimbursement tweak (think SGR). For more information on how this might work, see my colleague Avik Roy's discussion of ASOs on his Apothecary blog.

There are very few true pricing signals available to seniors in Medicare today. Without those signals, Medicare will inevitably drag the rest of the economy over a fiscal cliff, no matter how many price controls Congress mandates for the program.