4th Health Economics Conference: What is Universal Health Coverage ? Keynote

[music] Okay, we're we're gonna we're gonna get started. Uh so please everybody grab your seat. Uh we're delighted to have John Gruber uh here in Tulus from MIT and uh John's work uh we all know as health economist but in addition to being a fantastic economist he's also a delightful person. Uh so thank you John for for coming to Tuloose and for giving uh our second keynote for the conference. Uh John the the floor is yours. >> Thank you. Thank you so much Angie. Um, I have to start by uh recognizing that my wife is here, which is important for two reasons. First of all, it's one of the few times I'm nervous when I'm talking. Uh, second of all, it's important because it helps explain in a way why my voice is gone. Not for the reasons you think. Okay? But, um, because um, my favorite sports team is the New York Knicks. And part of the reason they're my favorite sports team is because when I was a first year grad student at Harvard, very frustrated, I hated the first year. I wanted to drop out. I decided that a a convenient way to end my economics career would be to go to a Boston sports sports bar and root for the New York Knicks and get beaten up. I thought that would be like a way to symbolize the end of my first year. And it was at that trip to the bar that I met my wife. So the New York Knicks are a very important part of my life. And the reason that matters is because I flew home on Saturday to New York for one day to go see the Knicks, win the championship, and that's how my voice is gone. So, I apologize. I apologize especially to conference organizers. I've never lost my voice before. I did not know this was a risk. So, I'm very apologetic uh to the conference organizers for not having a voice. Um, okay. So, I want to talk today about research I'm working on with Mark Shepard at Harvard University. This is going to be a chapter in the handbook of health economics. um we were tasked with the modest question of what is universal coverage uh universal health coverage. So we spent a lot of time thinking about this and we think we have some interesting insights I'd like to share with you today. Let's start with the simple fact which is why do we need health insurance? This is data from five different countries on percentiles of health spending relative to median income. So what you see is in every country the bottom 50% of the population spends virtually nothing on healthcare. Okay? When you get by the time you get to the top 10% in every country the top 10% expenditure on healthcare is larger than the typical median income. By the time you get to the top 1% the top 1% of expenditure is at least three and a half times median income in the US is eight times median income. This is risk. Okay. in a picture. This is what people want to ensure unpredictable enormous risks. So health insurance is necessary. Okay, this kind of distribution of risk is around the world. Obviously, it's wider in countries. This obviously lines up a bit with healthcare spending as a share of GDP. But the bottom line is even in countries with healthcare is not a large share of GDP. There's enormous risk and that's why health insurance is necessary. So what is universal healthcare coverage? We have our own definition which is that no one faces unaffordable costs to get essential healthcare. Now this raises three questions. First, what is no one? Okay, do you literally mean no one? In particular, what about undocumented residents of a country? Second is what is affordable? Neil Mahoney will be will understand this especially affordability is a term with which people are obsessed. Guess what? That's not an economics term, folks. Affordable is not has no meaning. none in economics. It's a term of art. So how do we think about affordable in the concepts of economics? And the third is what is essential? What essential healthcare? Who should decide? [clears throat] And definitions can change over time like they have with pharmaceuticals. So these are the three questions you need to answer to get to universal health coverage. So in practice, what we're going to do in this country, in this paper, I'm sorry, is we're going to focus on in this paper, we're just going to focus on relatively high income countries. It's a sample of countries with incomes of at least $10,000 per capita. Okay? For very poor countries, we feel like there's a separate set of issues about capacity that we're not going to get into. There's a set of countries that have the capacity to have universal healthcare coverage. Okay? Indeed, virtually every OECD nation has universal coverage, but it comes on very different flavors. And we're going to develop a new taxonomy. It's related to the old TR taxonomy, but we think it's a more useful taxonomy for health insurance coverage. Previous taxonomies about about healthcare delivery as well as coverage. For focusing on coverage, we develop a new taxonomy between three types. Universal public coverage which is countries where people are automatically enrolled in birth at birth into a public plan that covers their costs. Okay. Mandatory multipayer systems which are countries this was mentioned yesterday when we discussion of um uh the Netherlands countries that have multipayer systems yet they everyone is insured. And then the last capacity we'll call incomplete and fragmented systems. These are systems which lead many people to fall through the cracks. Okay. What we do is we're going to start by reviewing insurance market failures and arguing importantly between the difference between healthcare a universal healthcare access and universal healthcare coverage. A lot of the market failures we talk about in health care are really about guaranteeing access. But if you really want to guarant if you really want to justify your healthcare coverage, it has to be on grounds of externalities or internalities. The second point we'll make is we'll set up this taxonomy I described. Then we'll review key aspects of health insurance systems. Okay, who is covered? What is covered? How much is paid out of pocket? And we'll conclude with what is necessary and what is not for universal coverage. Okay. Uh so let's start with market failures and insurance. As I showed, healthc care risk is large and unpredictable. There should be and there is large demand for insurance. But why does government need to be involved? After all, private insurance expenditures in the US are in the order of $2 trillion. So why do we need government involved in the production of insurance? Well, this all goes back to Ken Arrow's everything in health economics does. Ken Arrow talked about all the ways that healthcare markets are screwed up. Talked about imperfect competition. Most hospital markets at least in the US I presume around the world are very concentrated. There's imperfect information lots of evidence for ever selection. My colleague Amy Ficklestein and Lonaf have done fantastic work in this area as we all know. The key point is these can justify government regulation and provision of universal healthcare access. But none of these failures justify actually going the way to universal coverage. If you have universal access where you can that will that is enough to address these problems. If you want to actually argue universal coverage you need to go on to the next step which is that requires either externalities or internality justifications. Okay. I in my opinion externalities are not large enough. There's two types of externalities. There's physical externalities. I don't have health insurance. I get sick. Okay. That is not very large in practice. I mean 99.9% of healthcare spending is not on communicable disease. Okay, if we if that was really the main thing we worried about, we could deal with that by just vaccinating everybody and having free primary care visits. Okay, there's financial externalities. These are larger, but they're still not enormous. Once again, in the US, which is, you know, we're talking about $30 billion a year of spending on uncompensated care for uninsured people. Okay, $30 billion is nothing in the context of government policy. Certainly given the enormous battle you would have to have to get un from universal access to universal coverage. It's a trivial amount of money. We think that if you really want to justify universal health coverage, you need to rely on individual choice inconsistencies. And it's clear that individuals are not making rational decisions about health insurance. Okay, we have evidence things like very small premiums or co-ayments matter enormously. Very small premiums matter enormously for health insurance coverage. Taking someone from a zero to $5 has an enormous effect on the decision to purchase health insurance. That's irrational given the value of health insurance. Defaults matter. Defaulting people into health insurance with an opt out has a massive effect on whether they sign up for health insurance. There's the evidence that I've worked on and others have worked on that intensive margin choices are highly inconsistent, which suggests that the extensive margin choices will be inconsistent as well. And basically there's lots of evidence that expanding health insurance coverage even a system of universal access when you expand coverage it in it reduces it improves health cost effectively. So we think that there are that rethink if you want to go from universal access which I'll argue is largely where the US is today. We largely have universal access at least for documented citizens. If you want to justify the next step, jiversal coverage, I think you need to lean on internalities. Here's our coverage system taxonomy. Okay? And really, the only way we tr taxonomy is we collapse his first two categories into one category. The TR taxonomy sort of breaks these first two categories apart based on whether there's public delivery of coverage like in the UK or just public insurance like in Canada. We think that's a false icon when it comes to coverage because the truth is in every country there's public and private providers everywhere. For coverage, what really matters is whether you're automatically enrolled in a public system from birth, okay, as you are um in um and and whether the basic insurance is all publicly provided. There can be supplemental insurance, it's private, but you're automatically enrolled in a public plan. That is the first category and that's most the countries in the OECD. Okay. Um the second category which we find the most fascinating is mandatory multipayer systems. These are systems where you're not automatically enrolled where there is not a single payer but where no one falls through the cracks where everyone is insured despite the fact that they're not enrolled from birth. And we talked a little bit about that yesterday with the with the uh presentation on the Netherlands. We'll talk more about that today. And then finally, we have non-universal systems. They're either disorganized or they're incomplete. These are less regulated. There's much larger role for for-profit. Fragmented systems are things like the USA and China. And incomplete are placed like Mexico, Argentina, and Chile. So, what I want to do now is go through a series of facts categorizing the countries and show you what lines up with these categories and what surprisingly doesn't. So, let's start with um I already talked about this. Let's start with uh coverage. Let's just go right here. Coverage. So what you see here is coverage of um uh uninsurance rates are essentially zero uh in for legal residents. Okay. In um in virt in in in every country that's either universal public system mandatory multipayer. The only countries that have meaningful uninsurance rates are in that last category of incomplete or fragmented for legal residents. But if you actually then add in all residents, there starts to be some uninsurance. So one feature we'll highlight is that in every country there's less than per in most countries there's less than perfect coverage of undocumented individuals. And you see that but the numbers are still very small. Okay. Health spending is a share of GDP here. Basically you don't the pattern is a little bit different than you get. You get some of the incomplete fragmented countries are at the bottom. China, Mexico, and then of course you get the outlier the US and everywhere I've tried to highlight France as well given given where I'm presenting the presenting the talk. France has uh pretty high spending as a share of GDP. Let's go back to France on the uninsured. France on the uninsured is pretty low even when you add in the undocumented 0.4%. Um what's striking about healthcare spending is the vast m is really when I look at this graph my takeaway is basically how flat it is in the middle that bas the vast majority of countries spend you know between 8.8 and 12.3% of GDP on healthcare. There's some outlaws at the bottom there's the US at the top but it's pretty it's pretty flat and I I would say it's sort of not super highly differentiated uh across the categories. Then we get to public versus private financing of healthcare. Here it lines up much more as you'd expect, although not perfectly with the categories. The automatic public systems have a much higher public share. The multipayer systems have a much higher private share. And the non-universal systems have the highest private share. But once again, even in the country with the highest private with priv private share, it's only 55%. Okay? There is no private healthcare in the world. Purely private systems. Okay. Uh a as much as conservatives would like to pretend there is. Um Singapore, which is not in this chart, probably comes the closest and we'll talk about Singapore. That's a particularly odd case. But the bottom line is even in the countries like the Scandinavian countries, still 17th of healthcare spending is privately financed and then it goes down as you go along. Here is to my mind the key insight we want to focus on in this in this paper out of pocket share which completely does not line up with anything with the systems. Look at the two lowest out of pocket share in the world France and the US. Two countries that are totally different in other aspects of this. Okay, you've got countries. It's just you've got Taiwan, which has a public singlepayer system, has 38% out of public pocket. France, which is a public pay single pay is 9% out of pocket. You've got Mexico at 41%, the US at 11%. These are really striking numbers. Once again, none of these numbers are hidden. This is all Commonwealth data plus a little work on our own to get countries the Commonwealth doesn't cover. But this to us is a really striking fact we want to come we want to focus on in this in this research. How do you get this interesting spread in the share of healthcare spendings out of pocket that doesn't seem to line up at all? Okay. Uh financing is varying across different countries. Um universal public systems generally rely on general taxation. Um but there's a broader mix in the man in the multipayer systems. Uh a larger role. So what you see here is general taxation is blue. Payroll and social insurance is a much larger role in the mandatory multi-payer systems. Um uh and then you've got um then you've got a a strong mix in the USA. Okay. Um now now I want to dive into a little more detail on some of these things and talk about them. Let's talk about who's covered and uninsured. So basically, like I said, essentially um you've got universal coverage. You've got essentially zero on insurance rate for legal citizens in all countries in the first two categories, but there is widespread non- coverage of the undocumented. This interesting fact, the bottom line is there's just not that many undocumented in most countries in the US. My estimate is probably if the of the uninsured today between a quarter and a third are undocumented. So a lot of what's going on in the US is not so much a different set of rules. It's a larger share of the undocumented population. So among the documented population, our uninsurance rate is uh is is much much lower. Okay. So still high but lower. Okay. Now, let's talk about the US for a minute. Since we're sort of the exceptional case here, let's talk about the path that we've taken because this will be important for thinking about how hard it is to get to universal coverage. Most of the other countries followed a sort of politically uninteresting path, which is they just passed universal coverage at some point and they were done. Okay. The US has a more interesting path with four key developments. The first was during World War II, we moved to an employer sponsored insurance system, okay, which was really predicated on two developments. One was wage and price controls during World War II, which led employers to uh which led employers to try to avoid them by offering benefits like health insurance. And the second was a decision by the IRS to allow health insurance to be non-t taxable compensation that is a feature in our system which is widely criticized. uh it is um cost us about $300 billion dollars a year in foregone tax revenues. To be clear, other countries in the multipayer system also have this feature to some extent. We're not alone in that uh uh in that in that system. Then in the 1960s, there was the introduction of Medicare and Medicaid. The only thing I called my undergraduates responsible for is understanding the difference between Medicare and Medicaid. Medicare is health insurance for the elderly. Medicaid is health insurance for the poor. Don't get me started on the poor elderly. That's confusing. Uh but there's Medicare and Medicaid. They were introduced in the 1960s. They were then dramatically expanded in the 80s and '90s for two groups for children and for pregnant women to, you know, politically sympathetic groups. And then finally, the Affordable Care Act in 2010 essentially filled the rest of the access gaps, especially when the system was in place under President Biden, as was discussed yesterday in the presentation. Essentially, we had a system where no documented American could not and where every documented American could get health insurance for 8% of their income or less. Full stop. Okay, that's universal access. Now that's been backed off now that the extended subsets are gone for some for some there's still universal access but now with that cliff reintroduced that was discussed yesterday um that 8% number could be more like 25% for some people and it's still universal access okay everyone everyone is guaranteed access to health insurance um although it can be very unaffordable for many I think no matter what we define how we want to define affordability 25%'s unaffordable okay but due to these expansions mentions it's a very fragmented or uncoordinated system. These peace meal expansions led to this very fragment or uncoordinated systems and within this system any attempts to move towards broader coverage have largely failed. Take up is highly incomplete. Take up of health insurance is highly incomplete. Um uh and uh trying to get to mandated coverage is too politically unpopular. So here is kind of how the US health insurance system looks like today. You've got about 50% of people in employer insurance. About 6% are in non-group. That's up a lot with the passage of the Affordable Care Act. That's largely in the ACA exchanges that allow people to buy health insurance. About 20% in Medicaid, health insurance for the poor. There you go. About 14% in Medicare health insurance for the elderly. There you go. Um about 1% military insurance. It's about 9% uninsured. But here's I think a more useful way to think about uh here's a more useful way to think about how the US healthcare system works. If you're 65 plus, you get Medicare. This is singlepayer for the elderly. If you're less than 65, your employer sponsored insurance offered by a job increases with age. This is not a technically sophisticated graph. It's illustrative. Uh if you're less than 18 and poor, you get Medicaid and CHIP, which is an expansion of Medicaid. If you were between 1865 and very poor under the ACA, you get Medicaid. And then there's the individual marketplaces in between. But the bottom line is there's lots of holes. And this is the theme that we emphas we want to emphasize in doing this research, which is the theme we call seamlessness, but I think a better a better decision may be uh is some sort of a theme of filling the cracks. It's not that cracks don't exist around the world in health insurance systems. It's that other countries fill them and we don't. The US doesn't. We are in France. You do. The US doesn't. Okay. That the cracks exist and they're unfilled in the US. These cracks are filled in other countries. Okay. So basically in the automatic public system, it's easy. You're in when you're born. Okay. There's no crack to be filled. Okay. Now there is movement around the country. There is some disruption when you move, but they're very aggressive about making sure those who are moved immediately resigned up for the regional plan. The mandatory multi-payer systems, this is what's interesting. In many ways, they're not that different than the US. They have the same sort of system of but the difference is there's a social acceptance of mandated coverage through a mix of autoenrollment via employers active enrollment and premium collection. And basically the point is they have a default rule to make sure people are insured. So let's talk about how this works. So here's the UK. Here's an example for a typical typical um public typical universal public system. Okay. Typically, there's a public system. Um, if you're an ordinary resident in the U, in the UK, then you're automatically entitled. Okay. When you're born, you get assigned an NHS number. Done. Okay. You get assigned an HS number. Your one practical step is you do have to register with a GP. Okay? But that's free and open to all. And then it's free care at the point of use. Okay? It's the simplest possible system. there's no enrollment to enforce and there's emergency care for everyone. Okay, so here you have um and we'll find this common across all countries including the US, free emergency care for anyone who needs it. So that's simple. Now let's talk about Switzerland. Okay, here's a mandatory multi-payer system. How does Switzerland work? There's a legal mandate to be insured. Okay. Um there is an automatic pathway which is employees are automatically registered by the employer and there's an active registration pathway which is you pay directly. So automatic pathway no no problem. Here if you never here people don't register right that's going to happen if you're not defaulted at birth. It's going to happen. What do you do? First there's reminder notices okay warning you that you'll pay a fee if you're not insured. Second, there's penalties and arars. We penalize you and we keep track of those. They penalize keep track of those penalties. Third, there's restricted coverage. If you haven't registered, then you can't access aspects of the system and they start coming after you with debt collection. Okay? And fourth, there's actually garnishment and seizure of assets. Garnishment of wages, seizure of assets if you're not insured. So basically there's two back stops here. One is retroactive enrollment once discovered. So if I find your enroll, we we go we get you and we make you pay pay back the fees you missed with some relief for poor people and there's also but there's also always emergency care. Okay. So the bottom line is Switzerland system not so different from the US. But there is a social consensus that they'll take the steps necessary to make sure people have coverage. Okay. So that's the first question. Who's covered and how do they get there? Okay. So the first thing was when I said the definition of coverage was everyone. Well, this is how everyone's defined. The next step is essential healthcare. What is essential? Well, this is a list of the services covered by countries in our sample. What you see is there's a broad consensus in every country that essentially all care, primary care, hospital care, specialist care, and by and large mental health with the one exception of mixed coverage in Australia should be fully covered with no exceptions. This is striking. There's actually remarkably little variation in what insurance packages cover. A lot of a lot of focus in the US is on skinny packages and how we can make coverage skinnier. That's really a bunch of because the truth is almost all insurance in America covers almost all the exact same stuff. Okay? The skinniest is just about financial cost sharing. The truth is all the plans cover most of this stuff. Where differences start to emerge. Prescription drugs are less well covered, less universally covered, although still mostly covered around the world. dental care and and is much less covered along with vision care and long-term care. Now, long-term care, there's often coverage outside of the health system, but this is within the health system coverage of long-term care. So, that's where differences start to emerge in services coverage, but there's really an incredibly broad consensus across very different countries on what makes health insurance. Okay, the consensus, the most interesting place the consensus breaks down is prescription drugs. Now, how is this determined? Okay. Well, basically, let's focus on pharma because remember when these systems were set up, outpatient pharma was not a big deal. Okay. In in in before the 1970s, outpatient pharma spending was a very small share of medical spending. And it's really in the last couple decades, it's become quite a large share. So, some of these countries were set up not including pharma. Okay. Um but um and what's quite striking is the mandatory multi-payer systems they often covered RX from the start of their systems. They started later and typically carved from the start. But the automatic public nations often built around public hospitals and doctors did not. Pharma was a later add-on um in their systems. Okay. Um uh so in Canada um uh they was covered much later in the UK it was covered from the it was covered early but with cost sharing. Okay. So that is sort of the the interesting difference in pharma. The other big difference on on benefits is how far does choice extend? Okay. In universal public countries there's little or no choice of the basic plan. You get a plan you're signed at birth. you're done. Okay. But in the mandatory multi-payer countries, there's some choice in your basic plan. Okay. Um France and Japan, there's not real choice. Germany has an interesting system where the basic for most people that enrolled in a in a single basic plan, but if you're above a certain income level, you get to choose, which is kind of interesting. Okay. And then in the other [clears throat] in the other multipayer systems, there's real choices. Okay. There's real choice across plans. Individuals get to choose once again with this backs stop that you're mandated to choose something. Now, the US is both similar and different. Most insurers cover a similar set of benefits, but insurers vary a lot more widely. They do in other countries in particular in terms of um but they don't vary in what they cover. This is the key thing. The big difference across plans in the US is not what the plans cover. It's in things like how broad the network is. how large the cost sharing is and how aggressively they manage utilization. Those are the big differences. Okay. Um and the degree of plan choice varies a lot in the US. If you're an employer, you typically have one choice. Not an employer, you may have many. Okay. Now, that is on paper. But let's talk about in practice. On paper, in most countries, most things are covered. In practice there are critical limitations. Okay. So for example there's limitations in quality and convenience like weight times. Okay. Um and there's limitation in qu in the quality of care. Okay. And I want to talk about these limitations for a few minutes. Let me say this is very interesting. We go back to this uh slide here. Sorry it's a long time back. I shouldn't have done this. Okay. We go back to this slide. If you look at the last column, that was based on our own calculations. We then went and looked and we were spot on with the OECD. We're within 0.1%. But we're way off from WH. The WHO rates of universal healthcare coverage, they look nothing like this. Why? Because WHO factors in um factors in all these limitations. WHO says you only have universal coverage if you if you're unlimited. So WHO tries to have an estimate. So you look at WHO lists, they've got they've got countries like uh for one of the Scandinavian countries 20% has 80% coverage where we say it's 100%. Because they're saying it's hard to access some of the care. So the question is when you define universal coverage, we would say it basically comes back to what does it mean to be able to get essential care? Okay. So for example, if you look at waiting times, there's a lot of variation in waiting times. Once again, not super correlated with the healthare system. So you've got Australia and Italy, two different countries with national public systems and dramatically different waiting times. Um, you've got waiting times for hip replacement surgery. The same ones tend to be at the top. The top tends to be Chile, Norway, Australia. the bottom tends to be Italy, Denmark uh tends to be at the bottom, okay, for the waiting times. Okay, so waiting times is one limit. Now, one way countries get around this is they allow an escape valve, which is they allow supplemental coverage. And this is a key feature of universal health care systems. Okay, so France, you have the mutual system. Uh in Canada, there's coverage of excluded drugs. In Australia, um, you can buy supplemental coverage to get, uh, private hospitals, nicer hospital rooms. My favorite is Singapore. In Singapore, if you don't buy supplemental coverage, you don't get an air conditioned hospital room. And, and I've never been to Singapore, but I can't imagine anything worse in life than being an unairconditioned hospital room in Singapore. Okay? So, basically, there's various ways that they basically supplemental coverage basically top up your coverage. Top up your coverage. Okay? But that leads to the fascinating fact which I highlighted before which is this very strange pattern of out-of pocket share of healthcare spending that doesn't seem to line up with anything else. Okay, you've got the US and France have the lowest despite having two different systems. Others have higher and there's no clear relationship to system type. So what's going on here? What's going on here? And if you look at it by service, you see that it's also like all over the place. Brazil is the highest for pharmaceuticals, but it's in the middle for inpatient care. Uh and you know uh um you've got uh Switzerland is uh at the bottom of outpatient care of pharmaceuticals and near the top for outpatient care and back in the middle for dental care. It's all over the place. So what's going on with out of pocket spending? This was the real enduring mystery we faced in doing this. Well, really there's two things that a pocket spending represent. The first is cost sharing in the basic system. But the second we would argue as economists is a market-based representation of the limitations in the system that essentially we have in some sense we think about input and output. Think about input and output regulation right you can regular inputs you regular outputs. One inputs would be you do things like number of doctors per capita, number of hospitals per capita to get access. Output might be things like waiting times. This is the ultimate real preference measure of limitations in the system, which is how much do people spend out of pocket trying to bypass the system. Okay. Now, the problem is decomposing these is hard. And this is what we spent a lot of time on. And it'll be much more detail in the paper, but let me go through two examples of what we've done to try to decompose these. Let's start with Italy. Italy is interesting. Italy is um has a universal healthcare coverage system but pretty high out of pocket. They're at about 22% out of pocket or about more than twice the US. Okay, so what's going on with Italy? Okay, so they have a universal healthcare coverage system with minimal cost sharing. Okay, there's if you look at the system, it's like a couple euros to go see the doctor and yet 22% of their healthcare spending is out of pocket. Basically, essentially people are paying to bypass the basic system. There's two types of things. They're paying to bypass the basic system. They even have something called introa um private doctors practicing in public hospitals. So, you get shorter shorter weight times and faster care. And there's also you're paying for care the basic system excludes like non-essential drugs, vision optical, long-term care. Now, if you consider um basically um uh the question is does OP buy faster care? Basically, here's what's quite interesting. Italy actually has relatively low waiting times. Um I as I said before, so what seems to be happening is that Italians are spending a lot to pay to pay their way to shorter waiting times. Okay. Now, is that a violation of universal health coverage or not? That depends your definition of essential. If essential means immediate access to care, then that's not universal coverage. That's a dub what the WHO would say. If essential means you can get it in and but we don't care about waiting times, then then Italy does have universal health coverage. So, basically, it depends on how you want to define that. Switzerland is a case where there's much higher out-of-pocket spending for the base system. And here's the one place we're actually able to do the math. We collected all the data and did the math. And what we find is that on the order of 40 35 to 40% of out-ofpocket spending in Switzerland is actually co-ayments for the basic system and the rest is buying your way out of the basic system. Okay. So the bottom line is most of this high out-of- pocket number seems to come from people buying it buying their way out of the system. Now here what's interesting is satisfaction with care is much higher in the multipayer countries despite the fact universal care coverage countries it's simple you're enrolled from birth they actually have some of the lowest satisfaction with their health care system. Italy at the bottom. Once again, an interesting case. Italy universal coverage, virtually no co-ayment, incredibly unsatisfied, very high out-of- pocket spending. So, once again, is that universal health coverage? That's in the eye of the beholder. Okay. France has fallen marketkedly in satisfaction on what's going on in France. Uh, used to have a very satisfied population, 81%, now you're down to 60%. I'm not quite sure what's going on with satisfaction with healthcare in France. Um, and the US is in the middle. Okay. So once again another way to think about universal coverage is how satisfied are people with their healthcare coverage. By that measure the US is not doing too badly. By that measure the US is doing better than France. Okay. So once again there are multiple ways to think about this about what is a system a desirable system. Okay. So I want to conclude this part of the talk by saying what is necessary and what is not necessary uh for for universal healthcare coverage. I'm make sure I run out of time. Oh. Um, what's necessary is not universal healthcare coverage. We think three things are necessary for universal healthcare coverage. You need a social consensus for seamlessness. You don't need a singlepayer. I think Bernie Sanders has done an enormous disservice to the left in the US by saying that basically by tying universal coverage to singlepayer. You don't need singlepayer to have universal coverage. Okay? Those are two separate things. What singlepayer is do what singlepayer systems are doing. When you say the term singlepayer, you're really mixing three concepts. Universal coverage, regulated prices. Okay, universal coverage, uh uh regulated prices, and um I forget the third. Oh my god. Who was the guy who forgot that in the debate? I'm like the Texas guy forgot the third point in the debate. Anyway, the bottom line was Yeah. Uh no, whatever. The bottom line is uh he's mixing up universal coverage with oh I'm sorry and one payer. That's the third thing. Okay. Single payers, one payer, universal coverage and regulated prices. Those are three separate concepts. All of which you can have independently and can be mixed in any combination you want. Okay. In France, you have a Bernie Sanders style singlepayer system. You have one payer providing universal access to regulated prices. Okay. But you don't have to have that in in Switzerland. You have universal coverage and regulated prices, but not one payer. The bottom line is these are all independent features. What you need but what you ca what you need to have universal coverage is a social consensus for a seamless system. That's that's the key requirement. The second is empirically, not theoretically, but empirically, you need an escape valve. Every country's universal coverage allows supplemental coverage and there's a large amount of out-of- pocket spending. There's an escape valve. You cannot try to regulate. Everyone gets the same amount of care or at least no country's tried to do that. Canada probably came the closest. For many years, Canada did not allow um people to top up their health insurance, but they've even caved on that now. Okay. And the third thing you need is a large investment in public spending. None of the countries that have universal coverage do not massively invest in their public health systems. Okay, what do you not need? You don't need a single payer. You don't need a particular form of financing. It could be tax financing, employer financing, individual financing. That's not necessary for singlepayer. And quite frankly, in our view, you don't need the highest quality of care for everyone. Most countries, people would agree, have universal coverage, have a lot of people buying their way out. We have these sort of market based indicators that people aren't satisfied. Okay, so that's the first thing I want to conclude on that paper. But I want to talk about looking forward two big issues that all the countries face regardless of uh universal coverage. The first is um the first is um there's is population aging. Okay, we know these facts. So, this is these facts are all from a book I wrote with Kathleen McGary called Long-Term Care Around the World, where we put together a team of experts from 10 countries to review their long-term care systems. We know the population is getting older and we know that the share the share 65 is getting older in all these countries and the share of that group that's over 85 is also getting older. We are about 10 years from peak demand for long-term care. Okay, if you look at the baby boom hitting 85, about 10 years away from that. Um, the key feature of why people need long-term care. Okay, now we talk about population aging, we talk a lot of pressures on pension systems, but I want to focus in this book focuses with Kathleen on pressure on long-term care systems. And that pressure comes from the fact that disabilities are highly concentrated with age. So listen this is in the US 65 year olds among 65 plus threequarters of them have no limitations um uh ADLs are activation or activity of daily living that's unable to toilet bathe get dressed feed yourself IEDs are instrumental activity daily living that's inability to drive to balance a checkbook these are sort of less severe uh barriers but still barriers threequarters of those 65 plus have none of those problems However, only 42% of those 85 plus have none of those problems. And you can see among those with who are 85 plus, four, 17% have four or more limitations. They basically can't function uh at all, can't bathe, can't feed themselves, can't toilet. So, as we get older, these problems are going to get more severe. Okay, here's what's interesting. There's there's a we talk about we talk a lot about the variation in healthcare spending around the world. We don't talk enough about the variation in long-term care spending around the world. So basically in 2019 there's a lot of variation in how much people spend in long-term care. Much like healthcare, there's sort of a broad there's a broad middle of around 2% of GDP, but there's variation with countries like Spain, uh, like the US, I'm sorry, spend only 1.3% of GDP on on long-term care. Spain at.9%. The Netherlands, 4.1%. The Netherlands has a 10% payroll tax just to finance long-term care. Okay? They spend a lot of money on long-term care in the Netherlands. Okay? And you see it's growing everywhere but the ch but the rate of change differs a lot as well. So long-term care is a growing share everywhere with variation that we can study. Um the financing of one commonality is long-term care financing is mostly public everywhere. Even in the US at 71% the lowest is Singapore 51%. The vast majority of financing long-term care is mostly public and ins private insurance is negligible everywhere. There's almost nowhere has really created a successful private insurance market for long-term care. It's mostly public and out of pocket. But countries vary a decent amount in how they deliver long-term care. So this shows spending on on nursing home care and care in home. What you see is there's a lot of variation this share at home. You've got some countries uh like Spain where uh this is by the way this is formal care spending. This is excluding informal care. I'll come to that next. Spain among those used formal care, the vast majority are in nursing homes. On the other hand, if you look at um I'm having trouble reading the rows here. Uh if you look at Japan, the vast majority are the majority are at home. So there's a lot of variation in where people are getting their home their formal home care delivered both in terms of nursing home uh and at home. But what's really interesting I don't have this here I should have had this there's also a huge amount of variation how much care is formal. There's a lot of variation how much informal care. So while Spain has most of their formal care is delivered in nursing homes a huge share of their care is informal of the healthcare delivered in Spain is informal. And indeed if you go back to these percents of GDP that I had a minute ago um and you what we did in our paper with in our volume with McGary is we then added in a valuation of time providing informal care. we find that there is that that sort of equalizes these numbers considerably that a lot of the countries that spend little on formal long-term care are devoting a lot of human resources to informal long-term care and we're going to need this is going to be this is a growing issue worldwide that countries are going to face is how are we going to finance this growing need for long-term care and to what extent will we continue to rely on informal care which is costly is detering the ability of people to devote their labor supply to their to where it may be best best targeted. So that's one issue I think we have to face in all these systems. The second issue we have to face is radical healthcare innovation. Okay. Cell and gene therapies for rare disease are one example. My favorite example here is Zulma. Zulgensma is a drug for uh spinal muscular atrophy which is a terrible SMA is a terrible disease. Kids are dead by two. Well, you can imagine nothing sadder. Okay. Well, we've cured it. There's a cellular therapy which is injected into babies which can actually cure them um of this disease and keep them alive. And it costs $2.1 million. Now, $2.1 million is actually a pretty good deal for saving a baby's life, okay? But it's a big number. So what do you do? This is a challenge. You know, we talked about the, you know, we talked yesterday about drug price regulation, MFN and other things. I think the challenge is what Europe, I think, has done very well relative to the US is technology assessment and thinking about setting prices as a function of value. And a lot of the reason drugs are cheaper here is because quite frankly, they're not nearly as valuable as the prices that are set in the US. The problem is these drugs are as valuable. Zulensma was reviewed by NICE was priced at $2.1 million in the US was reviewed by NICE in England and they priced it at $2.1 million. That almost never happens. Okay? Because it's worth it. Okay? So what are we going to do when we have part of the way that Europe has kept its costs down is through I think appropriately assessing that we're overpricing things. What are we going to do when there's incredible sets of things that are actually worth it? The current regulatory frameworks are not going to be enough to keep healthcare costs down. Okay. We're also going to have AIEL discovery of new therapies. Ideally, there's a lot of interesting work on how AI is speeding up the discovery of drugs, speeding up science use in general. And what is going to happen as we cure more and more disease that's more and more valuable through AIEL therapies. Once again, the existing regulatory framework does not solve the problem. Okay, this is something the place Europe has to lead is going to break down when in fact we have a bunch of new therapies that are expensive but worth it instead of what we have now which are expensive but not worth it. Okay. Um, GLP1, huge debate. I think we heard yesterday they're just being covered now in France. Um, uh, in our review of countries, in most countries, GLP1s have not been covered. Okay. GLP-1s are a miracle drug. Okay. They are incredibly successful, uh, at those, especially those with pre-diabetes and diabetes. Um, uh, you know, there's still some debate. There's no real evidence that they're necessarily lowering a lot of costs in the near term. But if you look at, for example, evaluations, they're valued by Iser in the US, they're a valuable drug. They are improving health. They are having effects we may not even realize in terms of dealing with other addictions. I think the mental health benefits of GLP1s are understated, um, etc. But they're expensive. And how are we going to deal how are we going to deal with that? uh uh and once we let them in to place like Europe, that's going to be a big expense. How do we deal with that? So basically, this raises a fundamental question for universal healthcare countries, which is can you be universal if you're not paying for things which can cure the uncurable. Okay, that's a fundamental question you're going to face. Okay, so basically we need new financing models in the US. I think we need to move towards a singlepayer model for cell and gene therapy. Now I realize singlepayer I just a few minutes ago talked about what a nasty term singlepayer was in the US but in the US does have a singlepayer system. We have singlepayer system for those who have endstage renal disease. If you have kidney failure you are covered on a singlepayer system under Medicare. Why can't we do the same thing for those born with rare genetic diseases? Those my view is those born with rare genetic diseases. There's no moral hazard problem. Okay, there's no I think this is a perfect case for singlepayer. We should move towards Europe in that case. But that doesn't solve the problem you all face that it's expensive. So the other way to think about this is to start think about advanced purchase commitment models. Why are these drugs so expensive? These drugs so expensive because the this typical case of high fixed costs, low marginal costs, but the marginal costs are not spreading among many people. So the problem is you have to charge the average price becomes enormous because the marginal costs are not spread enough people. This is a classic case for an advanced purchase commitment model where we come together and pay a fixed amount ideally as a prize for curing the unccurable in the famous Michael Kramer framework. We say look if you develop a drug that cures this you will win a prize of a billion dollars but then you're going to price it at marginal cost or marginal cost plus 15%. Okay, we I think we need to start thinking about that. That is very very hard. Okay. Uh to do it's hard to do in two senses. One is how big does the price have to be? Well, that is really determined by financial markets. We have to ask how much how big does the price have to be before folks are going to take a risk on developing a new drug. Okay, that's point one. Point two is if you then want to price it at a marginal cost. No one's ever really measured marginal cost for pharmaceutical development before. That's really hard, right? It's made in a big plant that's making lots of other things, etc. So, we have two enormous challenges, which is how do we set an APC? Question. Sorry. >> Coordinate. >> Uh, ideally you coordinate across countries because then the market would be larger. Ideally, you'd have a it's a great point. I believe you'd have a worldwide APC uh where basically you would uh you would because you're right once that's an excellent point. Once one country is paid for it's discovered all the other countries free ride on it. I didn't even thought of that. You're right. You really need a worldwide APC to make this work. It's a great point. Um and then um and then basically the idea would be you would essentially have the government because another important expense is delivery of these gene therapies in particular. So government could contract for delivery and they financed by general revenues. Okay. Now can this work? Here's one positive example which is what happened with Hepsi in Louisiana. So funny story about this. Um I got a call from a reporter uh from the Wall Street Journal saying I've been talking to this is when so in the 2013 um uh Gilly Camp was Savaldi which cured hepatitis C. It was a miracle. Okay. Hepatitis C is a deadly disease. 40,000 Americans dying a year, many more around the world. Um, they came with a drug that cured it. There was $84,000. So, I got a call from a reporter who said, "I've been speaking to the head of HHS in Louisiana and she's complaining about this and I said, tell her she's an idiot. That basically $84,000 a bargain. She should give everyone some." So, I got a call back later. She said, "You're from the woman who was Secretary of Health and Human Services. She said, "You're the idiot." Okay. Do you realize that to cure heepsi in Louisiana would cost more than we spend on our entire education system? three billion dollars a year because we have so many people with Pepsi. But I'm working on a new solution. So I myself and Reena Ki and others worked on setting up a new system in Louisiana which was this called subscription Netflix model which is essentially we had Gilead and NAV bid against each other for the for a fixed commitment to eradicate he in Louisiana. So they bid and basically Gilead agreed for no more money than the state was currently spending on a limited set of people to provide an unlimited amount of uh savaldi to cure heepsi. And so far we've uh the rate of of heepsi eradication has dramatically increased by far the highest in the country. Uh estimates are probably we've eradicated about a third of the population in Louisiana. Um this was really set back by co right before co but it's picked back up and I have a recent paper with uh with Reena Ki and Jacob Wallace and um Kevin Kison which shows that it more than paid itself back in terms of saved medical spending. So these alternative models can work um whoops these alternative models can work and I think that the next step for Europe comes to this interesting debate we had yesterday about country versus Europewide. Your question ra comes to the debate we had yesterday about countries versus Europewide innovation and how do we think about setting up systems and the here's the other piece I would emphasize which is I guess there's a fourth thing that makes this difficult which is the right price depends on the qualies that are gained by curing these diseases. The problem is those qualies rely on very outdated data. So for example, if you look at the ISER report on Zulansma, ISER is the is the sort of shadow equivalent of NICE uh in the US. It was a nonprofit organization that sort of considers drug valuation. If you look at their valuation of Zulanma, 200page report, incredibly well done, but like it relies on incredibly weak data. Like for example, you have to ask how valuable is it that a kid can hold their head up versus not being able to hold their head up? Well, it's worth something. How much? Well, they had one study of three people in 1977. Okay. To base this on, we need to invest now in saying what are the diseases we want to cure. Let's work on measuring now the damage they're doing so that we can actually put out an appropriate valuation for curing them. If we want to say what's it worth to cure this disease, we need to first measure what's the damage it's doing. And that's research that needs to be publicly financed and publicly funded ahead of the game. We should really developing sort of a database, an international database of what's the value of curing these diseases so that when those drugs come along, we're ready to pay the right price. Anyway, that's I'll I've gone on too long. I'll stop there. Thank you very much. >> Well, thank you John for the very thoughtprovoking uh and fascinating talk. Uh, we have some time for questions. I'm sure there are tons of questions. So, all right. >> Just quickly, I do a lot of work on Alzheimer's and dementia and a lot of the costs are outside of the healthare system. I'm curious if any approaches that you've seen around the world have been able to um not perhaps combine but kind of to consider the implications of healthcare um investment or lack of investment into other say fiscal um domains. partly where informal care or productivity impacts are so huge that perhaps treating rare disease or even obesity would actually pay off but the healthare system is not willing to do it efficiently. You know this is an it's a great question enormous challenge that government scorers so I've worked a lot with the congressional budget office in the US that government scorers face all the time um and their approach is very conservative and I think probably right which is in the absence of convincing evidence they assume it doesn't exist so if you ask anybody who advocates for any treatment of anything to support it the first thing they'll argue is offsets they'll say oh well if you do this you'll save the the folks do believe in dental insurance. That may or may not be right, but the first thing they say is it stops heart disease. Not like it makes your teeth healthier, but it stops heart disease because everyone wants their wants it to be a free lunch. Now, that may or may not be true, but there's no great evidence for it. So, I think that may be right, but I'm worried about that argument as an economist. I'm worried about the slippery slope of that argument, absent evidence. So I think if we really want to argue so you take for example I think a fascinating research topic is the effect of health improvements on productivity. We have almost no evidence on that. Think about that. What could be more important than thinking about the effect of health and productivity? We really have almost no evidence on how making people healthier actually makes them more productive. Okay. If we had evidence on that if if we had even a small effect of that that could massly pay back. But right now, if you go to the Congressional Budget Office, I don't know if the equivalent is in France for scoring legislation and you say company with health insurance, they'll give you zero credit for improving productivity. Okay. So, I think that you raise a great point and you raise and there's sort of there's two sides to your point. One point is the effect of improving health in terms of other social benefits. Here the best data is probably the evidence that Amanda Kowolski and others have done showing that when you expanded Medicaid, 30 years later, people earning more income. Okay. Um but there's also the other side of that which is non-health investments in terms of improving health, housing, food, those what we call social determinance of health. Another great area where there's very little evidence that they actually improve health. So I think these are things where we need a lot more research. >> Yeah. Uh thanks so much. Excellent talk. Really great. Um I was wondering John how how do you see the prospect of the United States to move toward a universal coverage system in let's say the next 10 20 years and and like the followup is like how in that context how important do you see the >> the issue of regulated prices and because we have a high share of underinsured people in the United States which is probably part related at least to the high prices and not not a lot of utilization. So, so do you think like I would just would love to hear your thoughts on that. It's to me it feels like we're stuck somewhere after the ACA and that >> yeah, >> we at equilibrium and I'm not sure if we find a way out there. >> Yeah, I mean we have this fascinating pattern of almost exactly every 18 years we take up healthcare reform. So, we're coming due uh on the next round. Um, I think that basically there's not a lot more to be done on access other than fighting on the margins about the ACA. Do the Biden subsidies come back in? What are we doing about enrollment? I think the access fight is kind of largely over. I mean, the big part of the access fight is the undocumented, but that's I don't see any way we're we're taking that on anytime soon. Um, so I think the access fight on the undocumented is largely over. I think the question is, do you take the next step of the coverage fight? There is an obvious step we can take which is roughly this is a few years old but as of a few years ago certain under the Biden subsidies twothirds of the uninsured we're entitled to free health insurance free so why not just default them in why not just say if your tax return indicates that your income and situation is such that you're entitled to free health insurance then you're defaulted in we get a letter saying unless you want to opt out here's your health insurance. Okay, first of all, we showed in Massachusetts when we autoenrolled people in health insurance that can cause a terrific effect on competitive bidding across plans to be the default plan can really bring the bids down. This is kind of a bit like the conversation yesterday we had about about default plans. If you get to be the default plan brings so it's very strong competitive bidding effect. Um and you could get you could solve a lot of the uninsured problem. Now, how's that different than a mandate? I would argue obviously it's a nudge instead of a mandate because people can always opt out but we know most won't. Okay. Um so that would be I think once again nothing's easy but if you think about potentially feasible steps towards universal coverage I think that's the next potentially feasible step towards universal coverage that doesn't solve like you said the so-called under insurance problem. I don't like that term but the orf affordability like either but the high out of pocket cost problem. Um there I think once again the ACA has in many ways solved that problem which is we have income we need income related out-of- pocket costs. That's the right way to do it and that solves that problem. Okay. The ACA has done that in exchange income related out-of- pocket costs. We need to sort of figure out a way to extend that have more income relation in uh in in the out- of- pocket costs. I thought you know the paper yesterday which showed that you know a small deductible uh didn't really seem to matter that much I thought was interesting. There's a big debate, you know, Amy and Lauron have a book arguing we should have no out-of- pocket cost for basic services. Um, I think there's a debate about that. Um, but I think uh that that's sort of a separate issue. And then the biggest issue is regulating prices. Look, I have completely shifted in my career uh to thinking that the only answer to cost control in America is regulating prices. I think there's no other answer. We've tried other things. Doesn't work. That what a nightmarish debate. I don't know. You know, we started a little bit. You know, I always said I always said I didn't know if we'd get to price regulation in the US, but I knew we'd come for drugs first. And the reason is twofold. One is the prices are transparent, and the two is the ads. Okay? Now, my kid was of the age where the first ads for Viagra started showing. Okay? And my kid was eight-year-old would say to me, "What's Viagra? Why does that man keep throwing that football through the tire swing?" Okay? And then that led to one of my favorite alltime stories. We have a friend who's a pediatrician, Lex in Massachusetts. I swear it's true story. Andrew heard it. Okay. And he said that a kid came to his office, an 8-year-old, and said like, "Talk to you about Seialis." Seialis is a erectile dysfunction drug. For those who don't know, the pediatrician said to the child, "Why do you want to talk about Seialis?" Said, "Oh, well, I was watching football and said, talk to your doctor about Seialis." So, like, talk to about Sealis. So, I think there's reasons why drugs came first. They sort of led with their chin. But I think that what's ironic about that is drugs in many ways is the last place you want to regulate prices because the most innovative part of the healthcare sector. Hospitals aren't innovative. We should regulate hospital prices, right? Hospitals aren't innovative at all. Drug companies are innovative. I'm not saying we shouldn't regulate drug prices, but it's sort of ironic that we went there first given that that's probably the most innovative sector we have in healthcare. Um, so I think that uh but it's just it's just politically just such a challenge. [music] >> [music]