wavePod

Get Wave AI

How to Bring Down Healthcare Costs - Statecraft | Wave AI Podcast Notes

Back to Statecraft

How to Bring Down Healthcare Costs

Statecraft

Thu Oct 02 2025

Summary

Statecraft Podcast with Santi Ruiz:
“How to Bring Down Healthcare Costs”
Guest: Anoop Malani (Chief Economist, CMS; Professor, University of Chicago)
Episode Date: October 2, 2025

Overview

In this episode, Santi Ruiz sits down with Anoop Malani, the first Chief Economist at the Centers for Medicare and Medicaid Services (CMS), to pick apart the sprawling complexity of the U.S. healthcare system. They dive into why American healthcare spending is so high, why that hasn't bought better life expectancies, the nuts and bolts of tackling Medicare fraud, the shifting economics as America ages, and how policymakers attempt to bend the healthcare cost curve. Practical, candid, and steeped in both academic and real-world experience, Malani walks listeners through the daunting mix of policy, economics, and organizational incentives shaping U.S. healthcare.

Key Discussion Points & Insights

1. Malani’s Role and Transition to CMS

[04:04 – 06:18]

Malani is the inaugural Chief Economist at CMS, an agency overseeing $2 trillion or roughly 23% of the federal budget.
His job blends three key tasks:
- Prospective Decisions: Advising on imminent policy choices through research, quick calculations, and literature review.
- Projects: Running discrete projects aimed at improving economic decision-making across CMS.
- Economic Research: Analyzing past policies and prepping for future issues, even when not tied to concrete decisions.

“There’s no clear definition, so you have to make the job.”
—Anoop Malani [06:36]

Malani stresses the importance of building trust with leadership and colleagues:

“Trust is something you build up by doing a good job… If you want your work to have import, it’s got to be that the decision makers who take it as an input actually value it.” [07:13]
He contrasts academic work—where time and question selection are luxuries—with government, where “almost every day there’s a decision that’s important. Any given week or month, every decision is more important than almost anything in academia.” [09:01]

2. Disconnect Between Academic Economics and Policy-Making

[11:00 – 15:36]

Academics often don’t grasp the constraints real-world policymakers face: legal rules, slow regulatory processes, political pressure, limited human resources.
Real-world implementation is slow; a “decision might take two years to actually take effect.” [11:38]
Academic research should shift to more statecraft- and implementation-focused questions:

“We need to think about the economics of implementation… decision making in the context of political coalition… personnel economics in the context of policy decision-making.” [14:45]

3. Why Does the U.S. Spend So Much Yet Lag in Longevity?

[15:36 – 23:12]

U.S. spends about 20% of GDP on healthcare; that growth has plateaued in recent years.
There’s no economic “natural rate” for healthcare spending—it may just be a normal good.
Life expectancy isn’t explained only by healthcare systems; social factors, demographics, and personal choices play major roles.
Malani references a paper by Alice Chen, Heidi Williams, and Emily Oster showing infant mortality differences arise in months 2–12, i.e., outside direct healthcare intervention—implying community factors are critical. [19:13]
American diversity, geography, habits (risk tolerance, spread-out urban planning, etc.) also explain health outcome differences.

“Health is not the same thing as healthcare. When we want to figure out if our healthcare is doing a bad job, we need to actually look at places where healthcare is having an impact.”
—Anoop Malani [20:28]

4. Demographic & Fiscal Constraints

[23:43 – 28:43]

Low fertility and an aging population mean funding Medicare/Medicaid gets harder:

“As your population ages, your working population declines and it’s your working population that supplies tax revenue.” [25:24]
Simply cutting benefits is generally politically unacceptable. The challenge is to make each dollar go further—to improve efficiency and productivity instead of slashing spending.

5. How CMS Tries to Improve the System

[28:46 – 36:23]

CMS encompasses Medicare, Medicaid, and health insurance exchanges, together more than $2T in spend annually.
Institutional levers to fight inefficiency vary:
- Medicaid: States administer, Feds pay the bills—so perverse incentives abound.
- Medicare:
  - Traditional Medicare: Government as insurer. History has moved from cost-plus reimbursement to prospective payments to Accountable Care Organizations (ACOs), all to check overspending.
  - Medicare Advantage (MA): Private insurance plans paid by government, plans incentivized to game risk adjustment scores.
- Exchanges: Similar structures to MA, with private plans involved.
Common Issue—Fraud:
Historically addressed via slow, post-hoc litigation, CMS is turning to ex-ante data analysis to spot fraud before dollars go out the door.

“Dollars already went out. We want to switch to a program where before the dollars go out the door, we try to capture fraud.”
—Anoop Malani [35:39]

6. Tackling Fraud and Gaming in Medicare Advantage

[36:23 – 52:02]

Fraud Defined:
Fraud occurs when any mismatch among (1) what was promised, (2) what was done, and (3) what was reported occurs.
- Upcoding: Billing for a higher-risk patient than warranted.
- Medically unnecessary care: Doing/providing more than appropriate.
- Substandard care: Billing for more, delivering less.
Risk Adjustment Manipulation:
- Insurers can “game” risk scores by ordering extra diagnostics or exaggerating claims to raise premiums paid by CMS.
- CMS is engaged in a constant game of identifying and closing these loopholes.
- Too little adjustment: leads to “cream skimming” (insurers avoid sicker patients). Too much: overcompensation and waste.

“It’s a back and forth… You close one loophole, the [insurance companies] find another.”
—Anoop Malani [50:09]

Recent reforms attempt to tighten risk adjustment while also revising overall reimbursement rates—a complex balancing act, given the shifting health and risk profiles of MA’s population.

7. Comparing Medicare Advantage v. Traditional Medicare

[54:03 – 58:44]

MA covers more services, has less patient cost sharing, and allows for allocation of financial risk between insurers and providers.
Metrics are tricky: comparing federal spending per enrollee between MA and Traditional Medicare is an apples-to-oranges scenario due to differences in what’s included and division of costs between government and patients.

“If you just compare the total dollar amounts, it could look bigger in Medicare Advantage simply because the government’s accounting for a higher share.”
—Anoop Malani [57:41]

8. Strategies for Bending the Healthcare Cost Curve

[58:44 – 63:41]

Static cost curve bending: Use existing technologies/therapies more efficiently (ex: “site neutrality,” paying the same for a given procedure no matter the setting).
Sometimes higher payment is justified (e.g., true emergency settings), but often not (e.g., simple procedures in hospitals when outpatient is sufficient).
Broader aim: “For a given level of quality and quantity, are you able to provide that care at a lower cost?” [63:24]

9. The Politics & Economics of Drug Pricing and Innovation

[63:41 – 72:56]

Reducing prescription drug prices is bipartisan but complex.
Counterintuitive complications:
1. Intermediaries (PBMs, wholesalers): They may absorb some price cuts, so not all savings come at the expense of R&D.
2. Global R&D burden-sharing: U.S. often pays more than other countries, bearing a disproportionate share of the R&D burden; changes might just shift cost abroad, not reduce innovation.
To actively spur innovation beyond price-based incentives, policies might use government procurement, prizes, or direct subsidies (e.g., Operation Warp Speed).

10. Unlocking Health Data: The Push for Interoperability

[72:56 – 81:52]

July 2025 White House announcement initiates protocols for patient/provider IDs and encryption.
Goal: Create seamless, privacy-protected pathways for health data to move across providers and systems, enabling AI and analytics to improve care and efficiency.
Malani’s vision: Government as convener, not builder. Getting private sector consensus on standards and protocols is the lynchpin.

“In order to do this digital revolution, you need to have access to data… Our big challenge is making data both protected but available.”
—Anoop Malani [74:02]

Notable Quotes & Memorable Moments

On the challenge of real-world policy:
“You bring a set of tools. For me, it’s economic analysis… that doesn’t give me authority to displace people’s preferences.” [07:57]
On risk adjustment ‘gaming’: “Once you set the risk score function up, the insurance company manipulates; you discover what they’re doing, then you close that loophole, and they find another.” [50:09]
On why U.S. health outcomes lag: “Health is not the same thing as healthcare… differences show up when the healthcare system is not operating.” [20:28]
On fraud detection:
“If a [provider’s] billing practice suggests that the person worked 5,000 hours in a year, probably fraud. We should at least investigate that.” [40:56]
On changing how drug R&D is incentivized:
“When you don’t know what you want, you offer a patent and say: We’ll give you a monopoly if you invent something useful.” [72:12]
On health data interoperability:
“It’s like the equivalent of building rails for railroads. But here, what you’re transporting is data.” [76:16]

Timestamps for Key Segments

04:05 — Malani’s role at CMS; structuring the Chief Economist position
11:00 — Why academic health economics misses the policy mark
15:36 — Why U.S. spends more, but doesn’t live longer
23:43 — Demographic time-bomb: Aging, low fertility, and healthcare finances
28:46 — The levers CMS can use to fight inefficiency, fraud, waste
36:23 — Defining and combating Medicare fraud
43:41 — How risk adjustment gets gamed & CMS’ response
54:03 — Medicare Advantage versus Traditional Medicare: apples to oranges
58:44 — The real levers for bending the cost curve
63:41 — Drug pricing, innovation, and perverse incentives
72:56 — Why data interoperability is critical for future health innovation

Tone & Style

The conversation keeps a collegial, sometimes wry tone, with Malani offering candid personal takes (“never skiing enough,” “too much squash, not enough skiing”), advisory-level clarity, and occasional callbacks to the discipline’s blind spots. Ruiz blends humility as a “remedial econ” student with policy wonk’s curiosity—creating a lively, approachable atmosphere even as technical details abound.

For more in-depth content or episode transcripts, visit www.statecraft.pub.

Loading summary...

Transcript

Anoop Malani (0:00)

Foreign.

Santi Ruiz (0:05)

Hi, I'm Santi Ruiz and you're listening to Statecraft. Today I'm joined by Anoop Malani. He's a professor of law at the University of Chicago, but he's currently on leave, serving as the first chief economist at the Centers for Medicare and Medicaid Services. What that means is he oversees economic analysis for the agency that manages $2 trillion in annual healthcare spending. That's 23%, more or less, of the entire federal budget. The Centers for Medicare and Medicaid Services, or CMS. Medicare for 70 million elderly Americans, runs Medicaid for low income families and the health insurance exchanges where millions buy coverage. I wanted to talk to Anoop about the transition from academia into the federal government and to hear about what the day to day at CMS looks like. But I also wanted him to teach me a little bit about American healthcare policy. It's a field of policy I just know less about than some other topics.

Santi Ruiz (1:00)

That we cover here.

Santi Ruiz (1:02)

We talked about a couple, maybe pretty basic questions. For instance, the US spends like 20% of our GDP on health care. Why is our life expectancy so bad compared to half of Europe? We also spent a lot of time talking about fraud, specifically Medicare fraud. How do you crack down on it without hurting patients who really need care? And how do you stop incentivizing private insurers to make patients look sicker than they really are? We also talk about some other stuff. Skiing, committing fraud, or matching ACL injuries.

Santi Ruiz (1:31)

It's a good time.

Santi Ruiz (1:32)

As always, you can find a carefully annotated transcript for this conversation and for many others at www.statecraft.pub. if you liked this episode, leave us a rating on your podcast platform of choice.

Santi Ruiz (1:45)

Okay, here we go. Anoop Malani, thank you for joining Statecraft.

Anoop Malani (1:49)

Thank you for having me.

Santi Ruiz (1:51)

It's good to chat. I realized in preparation for this conversation, I think I owe you and my dad an apology because I'm a UChicago alum. You taught there and I mean, you teach there, you're on leave. But I never took an econ class in undergrad at UChicago, despite being at, you know, one of the famous economic institutions. So this is my public acknowledgment to my dad and to you actually taking your class.

Anoop Malani (8:59)

Yeah, so you're exactly right. The main difference between academia and here is speed and the weight of decisions. Almost every day there's a decision that's important, that's being made. And in any given week or month, any decision that's being made is important, more important than almost any decision you'll ever make in academia. And of course, I'm speaking from CMS's perspective where almost all items, when you quantify them, end with a B for billions and not hundreds or thousands or even millions. Even when there are millions, there are three, usually three digits involved. So I think that that's really what's going on now. What does that mean for analysis? When you're an academic, especially if you're an academic economist, you do two things. You pick your questions and then you take your time to answer them to a certain standard of quality, usually whatever the bar is for getting into a very good journal. And in fact you will discard questions when you don't think you can answer them to that level of quality. That's not a privilege you have when you're working in a business or in the government. There you have to make the decision. And given that your goal is to say, what is the best analysis I can do in the time period allotted to make that decision? But you have to do the analysis. So that means sometimes you're going to do back of the envelope calculations, sometimes you're just gonna have to rely on theory. And so the broader point that I would make is you spend a lot more time, more than I anticipated, thinking about theory and letting theory guide you. So basic Econ 101 theory, to some extent, I would call it price theory because I'm from Chicago. And then you learn to do, you know, treat problems as Fermi problems. You just do back the envelope calculations as fast as you can. You pray that the literature's already answered the question. If not, you just gotta get what numbers you can quickly and answer the question before the decision has to be made.

Anoop Malani (11:20)

Yeah, I think you touched on a very important point, which is that there's a disconnect between health economics research and, let's call it American health economics research. I don't want to comment on different countries, but at least the health economics research that I engaged in, that my colleagues and good friends engage in, and the decisions that have to be made at cms. And the main difference is not understanding the constraints that decision makers face here. And there's a myriad of them. That's the first realization I had here. Beyond the complexity of the program. I realized that you can't just toggle switches on and off. You're really constrained. So let me give you a few examples. When you want to change a regulation, there's obviously a whole legal process set up by the Administrative Procedures act, notice and comment, things like that. But even outside that, when you want to make a change, you have internal clearances that you have to go through that add time on top of that, there's prudential concerns about making sure that you send the signal to the private sector so that they can prepare. You can't just pass a regulation and expect compliance right away. There's a lot of work on the private side to be able to comply, and you want to make sure that happens. And so today, if we want to make a decision, and we made the decision today, it might be two years, 18 months to two years before you can actually see the effectuation of that decision. So you got to make decisions way in advance, and that slows down a lot of things. A second issue that really comes up is just there are political constraints. Now, this is not unique to this administration, but you can't just make great policy decisions. You have to think about the coalition that is going to support your decision making and how your decision affects them because you exist in that coalition. That's a second one. A third one is just the resources for implementing things. I think economists, we just think, oh, let's change the premium schedule this way, or provide this little mechanism to provide incentive. And we don't think about, how do you implement that, how many humans do you need, what it do you need, what services do you need to have in place, what kind of education do you have to provide to the regulated party to understand exactly how to comply? That process is complicated and itself either may not exist or takes time to construct. And so I would say those are the three, at least. I can probably list more. But those are the top three categories of things that we have to think about. So you don't have a blank slate on which to rewrite. And I think that's the big constraint. And this, I imagine this affects regardless of which administration you work for, when people come in and they say, like, okay, oh, you should just change things to meet this particular, oh my gosh, you have to understand those constraints and you have to understand when an agency's inability to take an action is due to constraints versus not being proactive, something like that.

Anoop Malani (18:48)

Your second question is why do we get so little? And I would argue that that too is too quick a statement. Not that it's wrong, but it's too quick a statement when put that succinctly. And the reason why is twofold. The first is you have to understand the inputs into health in the United States versus other places. And if the inputs into health, meaning the underlying health, human capital of the population, is lower to start with, then, you know, it might not be that we're getting a bad result. It could just be that we're doing pretty good with a worse base. And if you don't mind a quick detour, I'M going to talk about a really nice paper that I think kind of addresses this. Sure. To me, kind of. This is my usual response when I talk to folks, when they say, hey, why, why does the US do so poorly on longevity measures? So this is a paper by Alice Chen, Heidi Williams and Emily Oster, my colleague Heidi Williams. Yeah, they're fantastic. And this is one of my favorite papers. It's actually kind of a descriptive paper, but super insightful. So what they look at, I'm going to try to be as precise as possible, but succinct. They look at why one year child mortality is higher in the US than in Europe. And it is higher, it's higher in the US than it is in Europe. And when they do their breakdown, a really important result that comes out is that the gaps that emerge in mortality rates don't emerge in the first month, they emerge in months two through 12. Why is that important? If you think about when the healthcare system where we're spending our money, when you think about where it's going to have the biggest impact, you think the first month and the first month, you don't see the differences. The differences show up when the healthcare system is not operating. It's in months two through 12. It's what's going on in the community back at home. And what that tells you is again, we may be starting with different baselines that our population might not have the same longevity in the abstract or they may engage in other behaviors that lead to outcomes being worse, longevity being worse. And so I think we want to take that into account. That's great paper. You should read the paper also. All three are great researchers, always fun to read. But that's an important takeaway is that health is not the same thing as healthcare. And when we want to figure out if our healthcare is doing a bad job, we need to actually look at places where healthcare is having an impact and not necessarily where just the community or families or activity differs.

Anoop Malani (29:58)

Okay, that is a very complicated question. And I think it's helpful to break it down by type of program. And really you need to think about three programs. Medicaid, Medicare and exchanges. Sure. And each of them have their own distinct challenges. And then there's going to be some challenges that are common across all. So if we think about Medicaid, one of the problems with Medicaid is that it's a program with peculiar incentives. The federal government pays for a majority of Medicaid, but states control how Medicaid is operated. Okay. And one implication of that is that states don't fully internalize the costs of their decisions. It's like going to a dinner and, you know, your friend tells you they're going to pay 60% of the check. You might order a little bit more than you need to. Right. Or you might not be careful in making sure that you're getting the best deal when you're choosing the restaurant. Right. So I think some of those issues arise just by the structural features of Medicaid. And Medicaid's an important program, don't get me wrong. But I think we could do a little bit better in making sure that there aren't efforts to either increase the federal share or to engage in inefficient payments, like state directed payments. So that's one way. And I think that has to do with the structural program. I'm going to turn to Medicare and then I'll do exchanges, because Medicare and Medicaid are the big ticket items, close to a trillion dollars of spending each year on each of those. So in Medicare, the real challenge is that Medicare is actually two programs that are combined. There's something called traditional Medicare where the government is the insurer, and they cover hospital care and they cover physician care. And after 2006, drugs, you get through Medicare Part D plans. But still, that's traditional Medicare. And then there's something called Medicare Advantage, which, you know, if you're old like I am, you'd call that Part C, but the young kids call it Medicare Advantage. And this is where the government is not the insurer. Instead, the government pays premiums to insurers, private insurers, who then provide, you know, hospital and physician and drug benefits, among other benefits. Okay, so it's really two programs. It's a public insurance program and a. You know, the government provides premiums for private insurance, and people can choose which of those to go into. And they're run differently. Meaning in traditional Medicare, where the government is the insurer, we just kind of set a whole range of prices, and it's fee for service, so providers just decide what to provide. And the history of traditional Medicare has been the government realizing that leads to too much expenditure. And so we try to pull back. So we go from cost plus pricing to prospective pay pricing to more recently, you have, under the Affordable Care act, you have something called Accountable Care organizations. And the first one was, we just pay whatever your costs are, plus. Then we say, oh, my gosh, that's too much. Are you charging too much? So we're going to pay per sickness, roughly, and we'll pay you a certain amount for sickness, and then you've got to manage. And that was still not controlling costs as well as we'd like. So then we switched to something called Accountable Care organizations. And there we said, the federal government's going to shift the risk associated with extra costs from the government to providers. Okay, Kind of like capitation, but not exactly. If you remember back in the 1990s, for those listeners that follow health insurance, that'll make sense. That'll have resonance. So that's, that's the issue there. The government as insurers trying to get the providers to behave efficiently in there. You know, you could imagine that the biggest gains that we could engage in are expanding accountable Care organizations and trying to get providers to be a little bit more careful with their spending. And there, that's the margin for improving cost or efficiency. I would say providing better care for customers without overcharging the government or without charging government for things that. That is not necessary or not a high return, but doing in a way that empowers the physicians. Right. Like, you also have to think about making sure that people participate in the program, but that's the balance that we're trying to do. Okay. Now, Medicare Advantage, which is the private plan thing there, we're not dealing with the providers directly, we're dealing with the plans, private insurance companies. And we have to incentivize them both to take on patients that are costly, but then to treat them well. Okay, Right. And there we want to make sure that we're getting a good assessment of how costly different patients are to make sure beneficiaries, I should say, to make sure that the insurance companies take them on. And there we're doing risk adjustment with the insurance. We're setting insurance prices and determining risk, which is different than what we're doing with the providers. So we can do better with risk adjustment in that regard and to keep down costs, again, without harming patients, in fact, maybe even making the patients better off, because we try to make those insurance companies a little better aligned with what the patients actually want, what the beneficiaries actually want. Exchanges look a lot like Medicare Advantage. We're paying a premium support. It's slightly different than in Medicare Advantage, but we're paying for people to get insurance through a private insurance company. And again, we want the insurance company to take on even high risk patients. And so we have to think about exactly how we do risk adjustment. Okay, Right. So those are the big issues. But then you can step back and say, okay, but is there anything that's common across all of these? Now, one thing that's common across all these is fraud, waste and abuse. Let's just look at the fraud component of that. That's the easiest when you charge for something that you shouldn't have charged for. Legally, that can happen in all these programs and we just need to be a little bit more proactive about it. You know, traditionally, I apologize for the wrong answer, but we've been thinking about this retrospectively. Usually what happens is, you know, Dr. Smith, not to pick on Dr. Smith in particular, but Dr. Smith, sorry, Dr. Smith builds for something that he or she didn't do, and that's fraud. Right. And we, what we do is if those billings became big enough that we'd notice, we then start a suit. You know, False Claims act suit, prosecuted by DOJ would take a few years. Maybe there'd be a settlement, maybe there'd be a payment, and maybe Dr. Smith be kicked out of the program. But that's a slow process. It doesn't capture everything. And importantly, the dollars already went up. Okay. And what we want to do is we want to switch to a program where we're doing it ex hand look at the data, we see patterns that suggest fraud. And before the dollars go out the door, we try to capture some of that and just ask, hey, Dr. Smith, did you actually do this? Things like that?

Anoop Malani (38:14)

Okay, so let me address that in two parts. First is let me give you a simple definition of fraud. And then the second thing I'm going to do is talk about concrete changes you can make and address both fraud, but at the same time, keep in mind patient outcomes because you don't want to cut necessary care. So let's think about that definition. You know, one of the challenges with fraud is it's a whole litany of both federal and state rules, so statutes that govern fraud. It's very hard to teach because the rules do different things. And so to understand fraud, I think it'd be helpful to have a simpler definition that captures what all the different statutes are trying to do. And Jetson leader Louise and I have this paper that we wrote just before I started this job where we try to provide that. And here, here's our attempt at providing a simple definition. So fraud can be simply defined as when the following three items do not match. That is what the provider should have done for the patient given the contract with the payer, second, what the provider did for the patient, and third, what the provider told the payer the provider did. When any of those three things differ from the other two, that's fraud. Sure. So let me give you three simple or very common categories of fraud. Take upcoding. Upcoding is simply, you may have done the right thing for the patient according to contract, but you told the insurance company you did a little bit more. Okay, that's upcoding. Let's take the second most common class of fraud, which is medically unnecessary care. That's a situation where you told the payer that you were going to do what was appropriate, but you actually provided more care than is required and billed for that extra care, so you did too much. The third is the opposite of that, which is substandard care, where you're supposed to provide a certain high quality of care, and you did less than that, even though the contract required to provide the higher quality care, and you still billed for higher care even though you provided less care. And very often that happens when, you know, a doctor has somebody that's not qualified or a nurse or just a tech do the procedure that a doctor. That would be a classic example. And that's how you define fraud. Okay. And I think that that really helps when you use those three categories, because that captures almost everything. The only exception is something called kickbacks, which we can talk about if you'd like. Now, the second part of your question was, you know, what's going to be different? And I already kind of answered that question to some extent. Again, in the past, in general, we've been doing fraud enforcement after the. After it's occurred, after the money's out the door, and we usually use litigation tools to go after it. I think that an important thing that we are doing now is we're shifting towards a little bit more ex ante. Let's use the data that we get in the claims data to identify patterns that suggest fraud. For example, if there's a procedure that is supposed to take 30 minutes. And a doctor or a hospital or a specific provider is billing for that. And the billing practice suggests that the person worked 5,000 hours in a year, probably fraud. We should at least investigate that.

Anoop Malani (44:11)

Yeah. So I'm going to answer the spirit of the question rather than that specific company, is what I'm saying. The second thing I want to say is I want to be a little bit careful and I think you, by the way, hit with this question on important topic, which is United as an insurance company, and it's not necessarily, although it has some vertically integrated segments that includes providers. So sometimes it's contracting with providers arm's length and sometimes it's actually employer of the providers. And obviously when it's an employer, it can actually tell, maybe has a little bit more power to tell the provider exactly what to do. An ARM select contract, it has to do indirect incentives, things like that. But what you want to ask all the time is why United or a company like United, okay, so let's just call it. An insurance company might have an incentive to get a doctor to report more or less. Right. If the insurance company is acting on its own, at least in hindsight, within a plan year, it doesn't necessarily want the provider to exaggerate exactly what the claims are because it affects the bottom line. But if the insurer is being paid by the government, let's say the federal government, to provide insurance either through premium supports or an ma, the Medicare Advantage, the premium itself, it might, but the reason is because that premium might itself depend on how much care the individual is getting. So now we have to go Back and discuss risk adjustment and then think about how we answer that, how we make risk adjustment better. What is risk adjustment? Risk adjustment says, if you decided in Medicare the federal government was going to pay the exact same premium for every individual that chose to participate in Medicare Advantage. Insurance companies would have an incentive to try to find individuals who are low cost and healthy and cover those guys, but not cover those individuals who are high cost and not healthy. Especially if they couldn't charge any more than what the government was paying in premiums. That's called cream skimming.

Anoop Malani (52:02)

Yeah, great question. Also gets at the complexity of the program. So let's answer this in parts. How do you address risk adjustment gaming even as you're increasing spending? And the way to kind of square that circle is to understand that it is possible that the composition of risk in the program in MA is increasing. @ the same time there's also inaccuracy in estimation of that risk. So imagine if, and I'm not saying this is true, I'm just going to give you a kind of a proof by counter example. But that illustrates how this could work. Okay, let's suppose that you're spending a dollar this year on average per person. And you know that the composition of the population is going to change such that the actual cost of that population is going to be $1.10. Okay? New people coming into the program from traditional Medicare. It could be new people coming into the program because they're aging into the program. It could be just underlying changes in the population, whatever it happens to be. Let's suppose that's the truth, okay? And now that's, on average, some people are going to cost a dollar, some people are going to cost $1.20. Some people are going to be in between. And the problem we face is not merely that the cost is rising. That's something that we just have to acknowledge and we have to pay for. But at the same time, it could be the case that some individuals who might cost A$5, their costs are exaggerated to $1.10 by the insurance company, okay? Through manipulation of claims so as to exaggerate their risk adjustment score. So what we want to do is we want to say, look, on the one hand, we definitely want to account for changes in actual costs for the program, but at the same time, we want to cut back on fake costs, exaggeration of risk scores. And so you can have both things happen at the same time. You just have to be careful about it. And I think that's what we're trying to do now. Can I use this as an opportunity to provide another lesson that's important?

Anoop Malani (54:04)

So we already talked about one margin just a second ago, which is the program is not static. The population in the program is not static. That's changes. And one place where it changes is people going from traditional Medicare to Medicare Advantage. Medicare Advantage has a number of interesting properties, useful properties that make it an attractive program. So one thing that makes a very attractive program is that it allows for allocation of financial risk to both insurance companies and to providers, and to do that in a way that makes sure that both insurance companies and providers are giving patients the best care at appropriate prices. So in traditional Medicare, you do that through what's called accountable care organizations. I mentioned that earlier. You shift the risk just to providers, but you can imagine that there's going to be some times where you want to shift the risk to the insurer because the insurer knows a little bit more information. And sometimes you want to shift it to the provider. You can't do that in traditional Medicare, but you can do that in Medicare Advantage. You can shift it to the private insurance company who can then take a portion of that risk and shift it to provider and optimally allocate it in such a way to get the best deal for the patient and for the federal government. Okay, so that's one reason why you might prefer Medicare Advantage. Another reason that patients really like is that Medicare Advantage actually provides more coverage than traditional Medicare. So, for example, traditional Medicare is hospitals, physicians, drugs. But maybe you're going to get some additional care through a MA plan. Additional care benefits like vision and dental and things like that that patients value. Okay, so that might be another reason why you like Medicare Advantage. Anyway, there are a range of regions. One of the things that you might want to do is pull people into the MA plan if you think it's better for patients. Better quality, lower cost, both. So that's one thing that's important. And to keep in mind when we think about that cost estimate, that's an economic point. But here's an accounting point that's also very important. When you hear this number. So when you hear a number, something like we spend, we, meaning the federal government, spends a dollar ten. I'm going to make up that number because there's that dispute about what the right number are. But let's suppose that somebody asserts the government spends a. For every dollar spent on a traditional Medicare patient, the government spends A$10 on a Medicare Advantage. I'm not going to deny that that's actually true, but I want to put that in context so you understand that that doesn't necessarily imply that MA is inefficient or that it's not providing quality. And the reason is, I was going.

Anoop Malani (56:33)

Great. Yeah. Because, you know, it connects to this point that it looks like the government is spending more on ma. So when we think about how much is spent on the patient, we should think about all possible spending, meaning you should think about total costs of treating the patient in traditional Medicare and the total cost of treating the patient in Medicare Advantage. For simple comparisons. The problem is that the coverage of those costs varies. In traditional Medicare versus Medicare Advantage. In traditional Medicare, the government pays for part of the insurance. But there's a substantial amount of cost sharing that's involved in traditional Medicare. In Medicare Advantage, there's very little cost sharing. And it kind of maps onto, if you study health insurance, it kind of maps on this idea that, like when you have a fee for service program where you're going to pay for anything you want to, you could go to any physician, et cetera, usually to avoid excessive spending, you have a high coinsurance high deductible, high co pays, etc. To make sure that the patient at least sees part of that cost. And when you do some sort of managed care, whether it's a PPO or an HMO like what you might see in Medicare Advantage, you don't have very much cost sharing. Instead you have the provider that determines whether you can go or the plan determine can you go out of network, or you have the provider that has to get approval for, you know, give approval to go to a specialized physician, things like that. So those are the two different ways to control costs. But what that means is that in traditional Medicare, if you take total cost of a dollar, a higher percentage of it is paid by the patient and a smaller percentage is paid by the government. When you go to Medicare Advantage, higher percentage is paid by the federal government, a smaller percentage is paid by the patient. And so if you just compare the total dollar amounts, it could look bigger in Medicare Advantage simply because the government's accounting for a higher share. And that's why it's a bit of an apples to oranges comparison. You really need to think about which of these two programs are appropriate for each patient and then determine first whether or not we're spending that cost appropriately overall, regardless of who's actually paying the bill, and then second, filter that down to the actual dollars being sent out of the treasury to providers.

Anoop Malani (65:45)

Right. But let me take a step back for a second. Okay. Because I think that, you know, a lot of time, if I were a listener, I would look at this and say, why is the system so complicated? Like, why can't we do things simple? And the usual answer is, as it turns out, each of these changes addressed an old problem that we've now forgotten about. And so if we just come along without knowing that old problem, it looks like the new system is just like this, you know, ball of. Of yarn that is so knotted up you can't possibly understand why it ever got here. So let's just pause for a second to talk about PBMs, and then talk about the problems with PBMs, and then I'll connect it back to exactly what you said. Okay, so imagine you're a payer and you have to. You. You're going to COVID patients at local hospitals. That's perfectly fine. You negotiate with those local hospitals, your deals, but you also have to cover drugs. Now, the problem is that the drug companies are located around the world. There's a lot of drug companies, and you need to negotiate with them to make sure that your patients, your beneficiaries, get access to a portfolio of drugs that are useful. Right. That a health insurance plan would cover. You can negotiate with them, but, you know, as it turns out, there's a lot of them. In addition, it's the case that other insurance companies also have to negotiate with the same people, same drug Companies to kind of get contracts on coverage. So these entities called pharmacy benefit managers emerge and they both negotiate on behalf of insurers with the many drug companies. But they also then help the drug companies a little bit further and help them figure out, here's the portfolio or formulary of drugs that you ought to cover. And that helps the insurance companies. Okay. By the way, this is not the only type of intermediary that exists like this. Hospitals have these, they're called GPOs. These are organizations, the purchasing organizations that help the hospital find gurneys and sutures and band aids and things like that. Right. They're just intermediaries that buy a bundle of things that you need. So this is a phenomenon that exists in a lot of places.

Anoop Malani (67:39)

Problem is that over time, PBMs, you don't need a lot of them to do the negotiation because the duplicate of negotiations, they ended up being a few small, very large PBMs. And as often happens in that sort of situation, there's a concern that those PBMs generate profits. Okay. Now, I'm not going to make a contention that there's excessive profits or things like that, but there's this concern that exists. And the important thing, for our purposes, that means that those PBMs which had a purpose may now be adding to the overall cost of drugs. And so that if you reduce the spending on drugs, you take a dollar out of that total spending, some of that dollar is going to come from the PBM which doesn't engage in R and D, instead of coming out of the pocket of the drug company which might engage in R and D. And that's the argument made for why it is that you don't. Whenever you cut prices, that doesn't translate immediately into a cut in R and D. That's the first complication that I want to highlight, which is why the simple Econ 101 model isn't exactly right. Good intuition, but a little bit different than reality. And let me provide another one even more important, I think so. What's really interesting with drugs and with many products is that once you do the innovation, you figure out, oh, this small molecule or this large molecule has some medical benefits and a tolerable safety profile. We should use it. You don't need to do that in a bunch of different countries. Once you've done the discovery, you can go sell in a bunch of different countries, sell in the U.S. sell in Europe, sell in Japan. You pick your area. And the prices that those countries pay in some sense is compensating. You for your R and D. So R and D is done in one place, but compensated by all these different countries. Okay, sure. And you could have a situation where the United States pays a very high rate of compensation, say 280% of what's paid elsewhere. And what that means is that the US Is shouldering a much higher share of the total R and D than other countries are other developed countries.

Anoop Malani (73:34)

Yeah, certainly healthcare offers a number of opportunities to work on tongue twisters. Look, I wasn't responsible what happened at the White House, but it is one of the reasons why I'm excited to work with this team at cms. So let's step back. We want to bend the cost curve. We want to do this dynamically. We want to affect future technologies that might provide care same quality, but at lower cost. One way to do that is to ride on the kind of digital revolution that we've seen in the last 25 years and then accelerate in the last five years with AI. You know, if I was an economist and speaking to other economists, I would actually say what we want to do is increase capital labor ratios. That would be the term that we would use with the specific form of capital we think about as software and data. But the key thing is, in lay terms, it's, you know, we want, we want to ride this digital revolution to improve care. And that ranges from everything like finding patients the right answers to medical questions without even seeing a doctor, to finding them the right doctor, to helping doctors do diagnoses better, et cetera. None of this displaces the doctor. The goal is to actually empower the doctor to be more productive and to empower the patient to find the treatment that's best for them. And to do that by either raising quality, holding cost constant, or holding quality constant. Lowering cost. Okay, but data is the new oil. In order to do this digital revolution, you need to actually have access to data. Okay, we need access to both claims data and EHR data that patients have. But at the same time, we need to respect two things in getting that data. We need to make sure the patient's privacy protected. That's very important. And the second thing to do is we need to protect ip. Sometimes that data actually contains intellectual property beyond just the information on the patient. We need to protect that. So our big challenge is how do we make data both protected but available so that we can have these innovations that are going to create new gen AI healthcare tools? Well, there's a lot that you can do there. And what we want to think about is broadly thinking about a healthcare data marketplace that protects that underlying data, but also allows it to be utilized to improve technology. And that usually involves four things. And this is, you can study other markets. Agricultural markets, equity derivative markets, water markets, energy markets. There's all these markets. Basically, we do Four things.

Anoop Malani (76:47)

Rails for the railroads, building a set of protocols for patient IDs, for provider IDs, and for encryption that ensure that patient data can flow. So let me give you a very concrete example. My experience with going to doctor's office is that you spend the first 20 minutes filling out a paper on a clipboard, filling out some paper, information that, like, I don't understand why they don't already have. And part of that is because there's a lot of friction to the flow of information both within and across organizations. And what we want to do is we want to make it so that happens, but at the same time, information doesn't get leaked so that your privacy is protected. So if we have a protocol for that, that means that whenever you go to the doctor, you know, ideally you. They would just know you, oh, Santi showed up. Okay. And then you go to the dentist, and Santi showed up. No paper forms to fill out. Maybe they'll ask you questions if there's something new, but not your old stuff. And so your experience with the physician should be much better, but it'll transcend not just this, but a whole bunch of different interactions and encounters you have with providers and patients. So these are the sorts of things we want to reduce those frictions. And this is just the tip of the iceberg. There's a whole bunch of other things that you can do to reduce those frictions, those transaction costs. But the next thing we want to do is try to figure out, well, how can we encourage this market to work more efficiently and grow that market? How can we encourage physicians and providers to gather data with greater fidelity? Meaning, like, make sure you take proper patient records without increasing the time that doctors have to do spend on it. So for example, AI scribes, can that make the doctor's life better and also give us better records? At the same time, can we encourage hospitals to cooperate with each other so that when patients go from one hospital to another, their data is shared efficiently and so there's no medical errors? So those are supply incentives. And on the demand side, can we encourage providers to really start leveraging data to improve the care that they provide to patients? So that's, this is kind of a multi pronged approach to build this infrastructure that we think is important to make sure that we really can apply AI to improve healthcare. Lower cost, maintain quality or improve quality, maintain cost, sure.

Anoop Malani (80:03)

So I mean, if you think a little bit more like a data scientist, less like an economist, you kind of say, you know what would make my life easier to basically manage a database. So you want your fields to be very similar. You want the data to be in a machine readable format and that kind of uses a common format. You want to have field attributes that tell you how to deal with it and protect those data and what data you can transmit and what data you have to very much protect. That's the way that, that I would view this problem when it comes to thinking about how to build this. Now, if you want to think about how you go from that ideal database where you're protecting the data but using it to the best of its ability, you also need to think about, how am I going to get from here to there? Now, in an ideal world, payers and hospitals would all get together, they'd agree on a protocol and shake hands and maybe have a beer after work to celebrate. But that's kind of hard. It's hard because there are at different levels, some people have already gathered data, they've got their own databases. Other people are not gathering data, but they've got this great patient population that's really useful. So there are different stages, development, and they might not all agree on protocols, even though they would all benefit from agreeing on protocols. And part of the role that the government can play is just to say, hey guys, let's all sit down in a room and figure this out and figure out some rules. Obviously, the benefits might not be evenly distributed, and we have to address that in any kind of discussion like this, but the government could play a role in just making sure that the private sector does come together to have this kind of agreement to build this marketplace. And I think that's the, the role that we would want to play. Knowing that it's not about the government building a database, that's not something we want. What we want is the private sector to go build this database, to build this data infrastructure, and then use that to improve the care that they provide. Sure.