I didn’t want to let this get away without commenting on it and a hat tip to Megan McArdle at Bloomberg for the graph above, the data for which are from the Bureau of Economic Analysis. The graph depicts the quarter on quarter changes in real healthcare spending from 1999 to 2013.
You may have heard that the rate at which healthcare costs have been increasing has been going down. That’s illustrated in this graph by the general downward trend from 2001 to 2012 (when you average the changes over the year for 2011 the overall trend is still slightly downwards despite the slight spike). However, that trend may have ended as illustrated by the spike in the third quarter of 2013, the highest rate of increase since the second quarter of 2004.
I think there are several things to keep in mind.
- As depicted these data are very noisy—there’s lots of up and down motion.
- Since June of 2012 the rate of change in spending has increased fairly sharply.
- This charge does not depict expenditures. It depicts the rate of change in real expenditures.
- Real expenditures are going up and they have been going up for decades.
- No one knows why expenditures have started going up again just as no one knows why they went down. No one knows if the recent increase in spending is a fluke or part of a trend.
- The CBO, etc. found that the slowing in the rate of change in spending could not be attributed to the PPACA.
- Real spending on healthcare is increasing, it’s increasing faster than incomes or non-healthcare spending, and that’s been going on for decades. That’s illustrated by the two quarters over the period of the last 14 years in which spending did not increase. That’s only two quarters out of more than 50.
- I really wish the graph went back another thirty years. If it did, what you’d see is very sharp increases from 1965 to 1980 with the rate of increase moderating considerably after that, roughly to what it is now. You would also see that the pattern is cyclical, i.e. they go up for a while, then they slow, they they go up for a while, and the pattern repeats.
This graph reflects intentional behavior. By that I mean that it’s not like weather patterns, the tides, or the motions of the planets. It’s more like a fencing match. Incentives matter. Different people engage in different behaviors and the net effect is what you see in a graph like the one above. Healthcare providers respond to changes in government policy. Individuals respond to changes in costs. Policymakers respond to political pressure.
Update
Fron the President’s Council of Economic Advisors, also using data from the Bureau of Economic Analysis, here’s a graph illustrating the growth in healthcare spending with a longer time horizon:
The portion at the far right of the graph, roughly from the 2002 spike to the present, illustrates the same data as are in the graph at the top of the page. As you can see typical growth rates in healthcare spending over the period have been at around 6%, with slower growth since about 2006. Remember, both of these graphs depict the rate of increase rather than just plain old expenditures.
Note that this chart completely bears out the thumbnail analysis I made in the body of the post:
very sharp increases from 1965 to 1980 with the rate of increase moderating considerably after that, roughly to what it is now. You would also see that the pattern is cyclical
1) You are correct. You need a graph with a longer time period. We have historically run closer to 2% increases.
2) Quarterly changes are good if you are trying to make a point. They are too noisy to tell us much. Note that we had a spike from June10 to June 11, then it dropped again.
3) CBO (or was it Kaiser?) calculated that about 1/4 of the drop may have been due to the ACA. Since the ACA did not take full effect until this year, I find it hard to ascribe the decrease, or this recent change (if it turns out to be real) to the ACA since it started in 2012.
4) You need a lot more info if you want to speculate. I understand that is beyond the scope of what you are trying to do here. Megan is not very well read on health care issues. She is good at doing “pop” health care and is readable.
Steve
I keep coming back to the fact that I’d much rather have 2014 health care than 1965 health care. It’s gotten more expensive? Nonsense. If you want the 1965 health care I’ll bet you can still get it for less than the 2014 iteration.
People routinely died of stuff in 1965 that they don’t die of now. So, personally, I don’t see why I shouldn’t pay more for life than I would have paid for death. But maybe that’s just me.
I’m not trying to say there aren’t problems with medical, just saying we’re comparing apples to apple pie and becoming outraged that the pie costs more.
By the way, how does that second graph in any way prove that health care costs are in some upward spiral? Am I missing something? I’m seeing 8 and 9% rates versus 2%. And it sure looks like the long-term trend is downward not upward, and definitely not some wild upward spiral. The current rate is lower than at any earlier point on the graph.
I don’t personally think it’s got anything to do with O-Care, what with that policy only now taking effect and time’s arrow being so pigheaded about only moving in the one direction. I don’t know what caused it to be so low now. But what this clearly does not show is that O-Care is responsible for spike.
That’s because you don’t understand what it means when I say the graph illustrates the rate of increase. You cannot ask that question and have even a vague understanding of what that means because that’s exactly what the graph illustrates.
Let me try to explain it. Let’s say you have a stack of ten pennies. If you add one penny to the stack per hour, at the end of each hour the stack has more pennies in it than it did at the start of the previous hour. If you add a penny each minute, the stack will have increased in size sixty times as much as it would have if you’d added one penny per hour.
How fast you add pennies is that rate of increase. Sometimes the rate of increase has gone up. Sometimes it has gone down. But it almost never goes down from year to year. And, importantly, the total number of pennies in the stack always goes up.
That’s unlike practically everything else in the economy. And, particularly, it’s unlike incomes. Here’s a graph of real median household incomes (in 2011 dollars). In 1965 it was about $42,000. Today it’s about $50,000. If real median household incomes had increased at the same rate as healthcare costs have from 1965 to 2011, the real median income today would be around $279,000.
Our little jog at the end is unlike most prior increases. Slope is much smaller.
Basically, steve, I think it’s too early to tell. It might be just ordinary variation. It might be the start of another major surge. It might be a smaller second derivative than the second derivative in previous increase cycles. We just won’t know for a while.
A big difference between now and 1965 is that healthcare is a significantly larger part of the economy than it used to be. That has employment implications.
But Dave, the graph shows the rates of increase from year to year. So although it’s always going up (more pennies) it’s either adding pennies more quickly or more slowly. I grant you that I’m a math illiterate, but adding pennies at a rate of 8 a year is worse than adding pennies at a rate of 2 a year, right? So if we used to add pennies at a rate of 8 a year and we’re now adding them at a rate of 2 a year that’s an improvement, right? Or would it be better to go back to 8%?
Michael, you’re asking if increasing more slowly is better than increasing faster. Obviously, increasing more slowly is better. What Megan McArdle (and others) have pointed out is that in the last quarter of 2013 we added 3 pennies to the stack rather than 2 and that’s more pennies added to the stack than in any quarter since 2004.
As I replied to steve above, it’s too early to tell what that means. It’s not to early to be concerned but it’s to early to be confident that we know.
One of the persistent themes of this blog has been that healthcare needs to start behaving like other sectors of the economy. That would mean that in some quarters it would add pennies to the pile and in other quarters it would remove pennies from the pile. We can’t afford a healthcare system that’s always adding pennies to its pile. We don’t have enough pennies.
Okay, but that’s not some escalating upward spiral, that’s a slowed upward spiral. And as you and Steve both agree there’s no lesson to be drawn yet from one quarter.
I don’t see why health care should necessarily behave like other sectors of the economy. Obviously it would be great if it did, but if I’m getting better product each year (debatable) then why should there be no increase in cost? I understand that’s what happened with computers (way better product, way cheaper) but there are other areas that behave differently. TV for example. We used to get TV for free, now we pay for it, but we get more and better TV. Or phones. I spend way more on my phone now – but it does a million more things. Or coffee. 3.75 for a latte versus ten cents for a cup of Joe back in the day. But the coffee’s better and you have a lot more choices.
I do realize that spiraling coffee expenditures don’t threaten the solvency of government in quite the way health care does, but I’m unsure why we’d expect health care to behave like computers and not like phones or coffee – better product, more expensive. As I said in the other comment above, if you compared 1965 health care to the exact procedures and drugs today, would they be significantly more expensive? I don’t know where to find it, but I wonder if the cost of plain old vanilla penicillin has risen since 1965.
That’s where it gets more math-y. Here’s a good picture of the difference between exponential growth (the kind that healthcare spending is seeing) and linear growth. It doesn’t look like much of a difference at first but over time the difference is enormous. We’ve seen this sort of growth for almost 50 years.
Yes, they’re enormously more expensive. Uwe Reinhardt has written on this subject extensively. Here’s one place to start.
Frankly, I doubt it. Thirty years ago a cellphone cost $4,000. And that’s in 1983 dollars. Remember, phones now are by and large a loss leader for phone service. Forty years ago a ten minute phone call to London would probably have cost as much in real terms as you spend on your phone service in a month.
Because the other sectors pay for the healthcare sector (two-thirds or more comes from tax dollars) and the number of jobs created per dollar of spending is much smaller in healthcare than in other sectors.
I didn’t see anything there that compared standard 1965 care to what the identical care would cost now. But I just skimmed.
I’ve seen you point out that healthcare costs are rising all around the world, though starting from lower numbers than we have. If costs are rising in France and Switzerland (they are, I checked) then obviously whatever we’re doing is not responsible. Something other than US policy is causing Germany and Japan’s costs to escalate (they are). So one might guess that there’s some common cause other than our government’s policy.
What might that be? Maybe that more people are being cared for. Or that populations are aging. Or that the quality of the care is better. Or that the way we calculate what is and what is not care has changed. I don’t know. I’m just pretty sure O-Care is not the reason Australia’s health care costs have doubled since 1995.
Mostly I think it’s the effect of expectations. We used to expect pain and suffering and death. Now we don’t. Now we expect to be cured of anything and to feel no unpleasantness while doing it. We expect to live to be 100. We expect all of our children to survive. We demand a lot more health care, and we get it, good and hard. If we changed our attitude as people, if we took a less self-cosseting way, the amount we spend would go down. So long as we all expect the maximum quality and quantity of health care (triple soy milk caramel latte rather than coffee, black) we’re going to drive costs up.
I think that U. S. policy is an important component in the increases in the cost of healthcare worldwide, particularly primary care. There’s a worldwide market in primary care physicians and, as you would expect, the wages of GPs and internists are reasonably close in all OECD countries.
However, in most countries other than the U. S. the wages of specialists are much lower relative to what they are here. That’s because it’s much harder for a foreign-trained physician to become a specialist than a GP. There’s not as much of a worldwide market for, say, pediatric oncologists as there is for GPs.
Whatever the reason for the cost increases state and local governments simply can’t afford them. In Illinois healthcare expenses are already the state’s largest budget item and that’s only going to go up. It’s crowding out other spending, like education and infrastructure maintenance.
This isn’t a partisan thing. Democrats have had complete control of state government here for the last 15 years. It’s effectively a one party system. The problem is that the state can’t raise taxes fast enough to keep up with healthcare spending.
Something Michael Reynolds points out — populations are getting older around the world. We’ve a larger number of elderly people in absolute terms; we’ve a larger percentage of elderly people; the ages those elderly folk live to keep increasing; the medical treatments for those elderly become more extensive each year; the number of perscriptions those elderly receive (and the number of ailments treatable by pharmaceuticals) keep increasing. Moreover, those dying and those with incurable ailments (spinal bifida, for instance) are commonly shifted into high-maintenance hospital beds and hospices and long-term care facilities, rather than being left at home for care by relatives.
All of which push up medical costs.
On the flip side, ny dad had a badly infected leg as a teenager in the 1930’s. They finally nailed his leg up so it would never bend again and scraped hell out of the bone, and even then it got re-infected a few more times, requiring months long hospital stays. These days, I suspect his condition might be treatable with a couple office visits and 50 bucks of penicillin. Granted, office visitis cost a whole lot more in 2013 than in 1936.
Another thought: When I was a tad, every kid in America came down with measles (several kinds), mumps, chicken pox, and other goodies. And every kid in America, more or less, stayed at home and was nursed through the misery by a stay-at-home mom. And what did this cost in terms of GNP? Why, amazingly enough, absolutely NOTHING! Because they were stay-at-home moms. receiving the sort of wages SAHMs got back then — zilch. Today, the mothers mostly work, and there are vaccinations against childhood diseases, and there’s some cost to being vaccinated, so I suppose “the cost of childhood illness” has risen dramatically since the 1950’s. But not really.
I’d really like to see a chart of “Inflation-Adjusted Medical Care Cost-Per-Capita” for the past century. I’d really like to see a chart of “1950-Standard Medical Costs For Common Illnesses In Current Dollars” for the past century.
“Yes, they’re enormously more expensive. Uwe Reinhardt has written on this subject extensively. ”
Hard to say. (Uwe mostly writes on our prices compared with other countries. Or how costs for same procedures have gone up recently.) You need to compare the exact same procedures and drugs, which is hard to do since we dont have many that are exactly the same. Drugs that were patent protected and used in 1965 and used now cost much less. All ten of them.
For procedures, it is hard to ferret out true costs. A cataract operation in the 60s required a week long stay in the hospital afterwards. Those were often billed separately and hard to figure out how they were included in procedure costs. (This was before DRGs) Same with a gallbladder or most other surgeries. (What I think we can safely say is that if we did those same surgeries the same way now they would cost a LOT more.) This also ignores total costs, what we really want to know. How soon can people go back to work and how productively? Maternal mortality in 1965 was about 30/100,000 (over 100 if black) and about 10 now. How do you factor in the costs from deaths? Negative appendectomy rates used to run around 20%, now less than 5%.
The real increase in spending is in new procedures, new drugs and new tech. We werent doing coronary stents in the 60s. Premies under 36 weeks or so just died. We had nothing to offer them. (Remember the Kennedy baby?) No MRIs. No CT scans. No total joint replacements. No surgery to stent aneurysms. Heck, most hospitals didnt even have ICUs in the 60s.
Steve
Dave- I have not seen a real, live GP in years. Do they still exist? To the best of my knowledge, FMGs still need to do residencies here in primary care specialties like family practice and peds. I dont think it is quite so easy to explain why specialists in the US make so much. Part of it is that specialists dominate the Medicare board that sets pay, but that doesnt explain the wide variation among specialists (MGMA median compensation for ortho docs doing spine surgery is over $1 million) and why private insurers pay so much. Heck, plastic surgeons make a lot and a lot of that income does not come from any insurance.
We should have reciprocal agreements with other countries that have quality training. We also need to increase our residency slots. Congress froze funding for those in the 90s. We are increasing the number of medical students, but not residencies.
Steve
Part of the reason specialists cost more: it’s like cheap wine – it takes a strong mind to believe that a $10 dollar bottle is as good as a $50 bottle despite the fact that they often are. High cost “proves” they’re the best.
So we are comparing apples to apple pie. It’s like phones to cell phones. Or broadcast TV to cable TV. Or diner coffee to Starbucks. It’s a different animal. “Go home and die” became “We can save your life for $100,000.”
Sometimes things get cheaper (computers) and sometimes they don’t. The fact that the economy “needs” medical to be less expensive is interesting but doesn’t change anything. I “need” gravity to stop pulling at me.
I think what we have here is a case of inaccurate labeling. We’re slapping the label “health care costs” on the 1965 “go home and die” and the 2014 “we can save you for a small fortune.” But they are different creatures. Like “Here’s three channels for free” vs. “Here’s 500 channels but it’ll cost you.”
michael-I think that is largely true, but there are real costs that have increased. Salaries for docs, especially specialists have increased faster than the CPI. (There are exceptions, but in general this is the case.) Medical device costs increase faster, even when they are updates of established technology, like with transplant hardware.
I sometimes compare it to computers. The newest, bestest always costs a lot more than the ones that have been around for a year or two. The problem is that, just like computers, the difference is actually pretty small sometimes and we are probably way overpaying sometimes. Look at all of the “me too” drugs. I remain convinced that the constant turnover of new total joint hardware is just churning to keep prices up. Just a couple of examples.
Steve
Steve
That’s basically untrue. The top of the line IBM PS/2, a personal computer, cost $10,000 in 1990. That’s in 1990 dollars. Now even the “newest, bestest” personal computer is a fraction of that cost. In the technology sector costs are falling and they have been for years. A few years ago the basic business computer cost $1,000. Today the basic business computer costs half that (much of that is the Microsoft tax).
If technology were the most significant part of the increase in healthcare costs, it would show up in balance sheets as an increase in capital budgets or an increase in consumable medical appliance and pharmaceuticals costs. Since, as I’ve previously shown here, neither of those is increasing at the rate that healthcare costs are, generally, the real answer is payrolls or, as the head of Mayo put it a couple of years ago “too many people making too much money”.
However, we’re getting far afield. The issue is not whether healthcosts are increasing (they are) or even why. “Why?” is a question for another day.
The question is how are we going to afford it? It won’t be from savings. For that to happen the real savings rate would need to rise faster than healthcare spending and that isn’t happening (it’s about half).
That leaves either from taxes or from extending government credit. If you think it should be from taxes, you’re effectively saying that all other sectors of the economy (the ones that actually employ people) should shrink while the healthcare sector should grow. You also need to identify a way to increase the effective rate of taxation. Regardless of the marginal tax rates that has proven remarkably stubborn over the years.
Doing it by extending credit is an extremely unwise practice in the case of an ordinary operating expense. Not only will it cause every other sector of the economy to contract, it will ultimately result in a collapse of the currency.
Two stories straight off the NYT front page this morning:
http://www.nytimes.com/2014/04/07/business/breast-cancer-drug-shows-groundbreaking-results.html?hp
And, http://www.nytimes.com/2014/04/06/health/even-small-medical-advances-can-mean-big-jumps-in-bills.html?hp&_r=0
(My emphases)
Expensive new drug, expensive new devices.
How might this relate to higher salaries? How much will you pay the guy who’ll keep you from dying of breast cancer, as opposed to the guy who sends you home with aspirin and sympathy? If the profession is objectively worth more to the patient, why wouldn’t you pay them more?
I’ll offer an anecdote from a related field, dentistry. I had my first root canal 30 years ago. It was exceedingly unpleasant. The dentist was new, he kept having to march me back and forth to an x-ray machine in a different room and when I’d been tortured for two long days, they finally told me the tooth was done for and it was all a waste.
I had another root canal a few years ago. It wasn’t my favorite way to spend an afternoon, but the x-ray was digital so the endodontist had instant updates, the pain management was infinitely better, and the whole thing was sort of a big “meh.”
How much more will I pay for the second option, the well-trained, well-tooled modern endodontist? Quite a lot.
Maybe doctors are paid more because they’re delivering a service they didn’t used to deliver. Maybe they’re better paid because they’re better.
What if they’re better paid not because they’re smarter, harder working or more capable but because what they’re delivering is better and competition is prohibited by law?
Look, my view is pretty simple. I think that docs are by and large smart, hard-working, well-intentioned guys who are paid too much. I think that higher wages for docs pull up wages throughout the sector. I think that healthcare has reached the point of bureaucratic displacement and increases in output in the sector (however measured) are completely disproportionate to increases in cost. Since the majority of spending is by the government, the rate of compensation needs to be regulated. Our politicians refuse to do that. Increasing costs in healthcare are damaging the rest of the economy.
You’d think that at least we could institute a five year freeze, or a rule requiring government compensation of doctors to equal the rate of inflation.
Crazy idea out of left field: Medicare/Medicaid could cover the costs of flying (or driving) some patients out of the country when the difference in costs was significant enough. If we knew Mexican hip replacements were as good as US hip replacements but cost 10 grand less, we could throw some of the old folks on a bus, spend a grand on hotel and meals, bring them home and be up 9 grand.
“You’d think that at least we could institute a five year freeze, or a rule requiring government compensation of doctors to equal the rate of inflation.”
That was tried and every year the “doc fix”…er…fixes that idea. It’s really a political problem – the US population is not yet ready to shove that kind of reform down the throats of providers.
” If we knew Mexican hip replacements were as good as US hip replacements but cost 10 grand less, we could throw some of the old folks on a bus, spend a grand on hotel and meals, bring them home and be up 9 grand.”
We’re already doing that, to a limited extent, for people who can pay for their own procedures. That’s not many people. If insurance covered medical tourism, that could be a game-changer… It appears, based on some recent news reports, that insurance companies are seriously looking at providing coverage for medical tourism. I’m sure the AMA will try to strangle that idea in the cradle. I can hire that surgeon from Cameroon with the perfect Angie’s List grade for my penis enlargement surgery (which, I’m hoping, will soon be a required benefit under Obamacare).
And really, the American Way ™ to reduce costs is offshoring – why not do that with health care? We can offshore much of it like we did with manufacturing. Patients may not even have to leave the US – probably within my lifetime, remote surgery will be common.
Lets consider the effects of this, and other, trends. Soon, the US could be comparable to a high-security upper-class suburb – a place for people to live, eat at chain restaurants, shop in chain retail stores, etc. Like those neighborhoods there won’t be much else except for the massive homeowners association (ie. US government). Most other economic activity will occur elsewhere, but for much better prices! That will leave more money for McMansions which will, of course, be constructed from modular pieces shipped in from India.
What a grand vision for our future….perfect for the Gen X and Millenial malingerers…..
Andy:
Few things worry me more than discovering that actual grown-ups are considering an idea like one I just pulled out of my ass.
I think the legal barriers to telemedicine are pretty hard to justify, especially when you don’t impose similar barriers on imported cars, food, or clothing. It gives you the suspicion that they’re picking winners and losers.
Telemedicine doesn’t even need to be at that sophisticated a level to have a major impact. Presumably, that’s why the barriers exist. It doesn’t even need to cross international borders to have impact. The barriers to telemedicine within the United States are being relaxed but it’s happening painfully slowly.
“That’s unlike practically everything else in the economy. And, particularly, it’s unlike incomes. Here’s a graph of real median household incomes (in 2011 dollars). In 1965 it was about $42,000. Today it’s about $50,000. If real median household incomes had increased at the same rate as healthcare costs have from 1965 to 2011, the real median income today would be around $279,000.”
This is the money line in the whole thread. All the speculation, right or wrong, is just idle musing. $8K in real median income increases is supposed to finance the increase in health care? If health care expense was just 10% of income, or $4K, then, it has gone up some 7-ish x. That leaves “Median Guy” some $20K in the hole.
How was it financed? Borrowing and transfer payments by the government, taxing your rich uncle and neighbors, Mr. Median reallocating consumption expenditures to health care.
This is not sustainable. ObamaCare and the excuses in the thread won’t change that.
1) High end gaming computers still set you back $10,000 Dave. That doesnt even include a monitor.
2) Medical tourism has some problems. First, we have found that we can’t get patients to travel an extra ten miles to save money. You will need major cultural changes to get them to leave the country. There are lots of practical problems like finding someone to watch the dog. Then, follow up becomes a big problem. Having to fly back to Mexico for every follow up visit will chew up savings.
3) Telemedicine sounds good, just like lasers, which have not amounted to much in medicine. We really havent figured out how to make it useful for much of what we do. It might eventually work to save some costs, but it is not very useful right now. Has its best potential for imaging (where it already works pretty well).
4) A freeze on salaries would be a good way to go, especially for specialists. The problem with the doc fix is that after not employing it for so many years it imposes cuts. Much harder to take.
Steve
Well, if it’s “unsustainable” I guess we can relax, since that which cannot be sustained, will not be sustained.