Markets, Part III: Barriers to entry

Recall that in the first post, I used widgets to show how markets can be self-correcting and result in benefits for everyone involved, without the need for government interference. I then identified five qualities of a market that make it work well.

In the second post, I explained why informational symmetry is a problem when it comes to markets for healthcare.

Now I'm going to briefly discuss barriers to entry. High barriers to entry, if you'll recall from the first post about widgets, mean that new competition has trouble entering the market. This could mean a new competitor (such as Wally, who started Wally's Widget Emporium,) or it could just mean that existing competitors have trouble entering another portion of the market (e.g. geographical.) There are many potential barriers to entry.

Barriers to entry are, generally speaking, bad for competition--as the widget example showed. They lead to inefficiency and create the potential for exploitation. In the worst case, they permit and perpetuate monopolies. Markets for healthcare services are not unusually affected by these barriers, however--which is why this won't be a long involved post.

• Some barriers to entry serve useful purposes. One reason I can't start up my own airline is that there are lots and lots of regulations regarding equipment, training, maintenance, etc. etc. etc. That's kind of a good thing.
• Some barriers to entry we can't do much about because they are inherent to the enterprise. To become a car manufacturer (at least, one that competes with the big makers), one needs tons of expertise, tons of experience, a lot of money, etc., etc.
• Some barriers to entry are erected intentionally (though not uncontroversially). For example, governments usually impose tariffs on imports precisely for the purpose of limiting foreign competition.

The barriers we care most about--the ones we should ameliorate through government--are those that (1) inhibit competition (are 'anti-competitive') and (2) don't give us enough in return. If Microsoft bundles Internet Explorer with Windows, it's very hard for a potential browser-maker to break in--leaving us with an inferior product for much longer than would otherwise be the case! On the other hand, the bundling was, in some senses, an advantage for consumers.

There's probably nothing special about healthcare when it comes to anticompetitive practices, so I won't spend any more time on this section. There are certainly some very large insurers, and some very large hospital groups, that wield 800-lb gorilla-type power in markets. In Massachusetts, there's all sorts of consternation over the one I work in, and the Atty General keeps a close eye on its practices. Any large-scale changes in the delivery system will have to continue to allow policing of the markets.

But my project here is to explain the limitations of unfettered markets when it comes to healthcare. Okay, let me post this, so I can get on to the more interesting parts of the series!

NPR's been reading my blog!

Check out this story on Morning Edition, which proves that my blog, despite being read by only a couple of people, is influencing the healthcare debate!

Chana Joffe-Walt and David Kestenbaum talk about the "information problem" in medicine.

Markets, Part II: Informational symmetry

Okay, it's part two of my series on why markets work well for widgets but not as well for health care.

Recall that in the first post, I used widgets to show how markets can be self-correcting and result in benefits for everyone involved, without the need for government interference. I then identified five qualities of a market that make it work well. I'm about to look at those qualities in a little more detail, to see whether they are qualities that a market for healthcare possesses.

Where I'm not going with all this
Before I address the first quality (informational symmetry,) let me head off a concern that's been raised about where I'm going with all this. Even if I persuade you that markets aren't good at allocating healthcare resources, this will not commit me (or you) to the position that the government must 'run' healthcare. My primary goal, in fact, is much less ambitious: it's simply to expose as naive the commonly-heard claim that 'the way to reform healthcare is to expose it to true competition!' As I'll try to show, market competition can only work for certain limited aspects of healthcare. (What it means for healthcare to 'work' will, I suspect be a bone of contention, but let's leave that alone for now.)

Informational Symmetry (or equality)
(I should've used 'symmetry' instead of 'equality' becaue 'equality' has all sorts of connotations that aren't warranted in this context.)

If I'm in the market for widgets, I can compare the quality of widget brands with reasonable confidence that (1) I'm asking the right questions (how well will this widget work in my wangle?) and (2) the information I'm getting is accurate (because widgets are not too difficult to understand.)

I've fallen and I can't get up!
Now let's say I've sprained my ankle pretty badly. X-rays show it's not broken, and the ER docs said "ice it, elevate it, wrap it, here are some crutches and some pain meds--you should follow up with your PCP." Fine so far--let's stipulate that this is good ER care.

But now it's a day later and I'm in the market for proper followup care for my sprained ankle. The first thing to observe is that I don't even know what proper care for my sprained ankle is. Frankly, I'm not even sure exactly what it means to have a sprain.

I see my PCP. He tells me the same things the ER docs told me--ice it, rest it, wrap it, etc. Now, how should I feel about this care? Should I say, as many patients do, "I coulda looked that up on WebMD! What a ripoff!"? Should I say "thanks for the confirmation and the followup"? Should I ask for an MRI? Should I ask for a consultation with an orthopedist?

Dammit, Jim: I'm a neurologist, not an orthopedist!
I don't have any way to know what optimal care is, even for something as simple as a sprained ankle! Mostly this is because I don't know the variety of ways an ankle can be sprained. I don't know the five hundred medical conditions that can predispose someone to sprain their ankle. I don't know what an MRI can and can't show about an ankle. I don't know what the treatment alternatives are, I don't know when to start walking on a sprained ankle.

This is the primary source of informational asymmetry in healthcare--the asymmetry of medical knowledge between the doctor and the layperson. The expert possesses a lot of information that I don't. He cannot possibly impart to me all of the relevant knowledge that goes into making the best decision. At some point, I just have to trust that the doctor is giving me good care.

NB: It's precisely for this reason that professions exposed to large amounts of informational asymmetry (medicine, the law, engineering, etc.) are associated with a professional ethic known as fiduciary duty. The professional has a duty to act strictly in the interests of her patient/client. This duty is enforced mostly by the professionals themselves, but to some extent also by policy/regulation/law.

"Patients don't have any incentive to limit costs, because insurance insulates them from the impact of their decisions."
I've heard this argument, in one form or another, a lot lately--it's sometimes couched in terms of 'moral hazard'. If insurance is picking up the bill, why shouldn't I seek out the highest (perceived) quality care I can, and damn the cost? The argument isn't usually accompanied by empirical evidence--or even analysis--that shows this to be a big driver of cost.

The solution, usually implied, is to have patients share more of the cost of care, which would incentivize them to 'shop around'. Works for widgets, right? Just one of the many problems with this as a large-scale solution is the informational asymmetry. How the hell do I 'shop around' for low-cost-high-quality healthcare, when I don't even know what the hell healthcare I need? This should be absurd on its face. There's a potential solution, though--make the ones with the information (as much as can be had, anyway) make the decisions about cost...

"Doctors don't have any incentive to limit costs (at least when they're paid fee-for-service,) because they get paid without regard to cost."
This argument usually accompanies the previous argument, with the implied solution being that doctors should have direct personal financial incentives to reduce cost. This solution exists currently, in the form of 'capitated' payment plans (the doctor gets $X to care for Y patients for Z months and keeps whatever's left, or bonuses for coming in under-budget.) This would seem to put the cost-management function in the hands of the doctors, who do have the relevant information.

This solution is misguided, though not on grounds of informational asymmetry. It's misguided because it puts physicians under direct, personal pressure to make decisions about each individual patient on the basis of cost. This is the most certain way to erode the physician's sense of fiduciary responsibility to her patient--and is thus a very bad idea.

What's the Solution?
How, then, should we make decisions about cost-effectiveness in the face of informational asymmetry? If patients can't make them (they can't evaluate effectiveness,) and doctors can't make them (we don't want them to consider costs too much when faced with an individual patient,) we're screwed, right? Wrong--we could, if we wished, pool our resources together to insure each other against healthcare risk. We could then ask all the relevant experts to come together and make some assessment of cost-effectiveness. We would, of course, insist that this assessment be transparent and open to debate and revision...

Hmmm. Isn't a version of this the way it works now? We insure each other against health risks. The insurers make cost effectiveness judgments that apply to the risk pool. They cover the more cost-effective things, and don't cover the less cost-effective things.

It's true--the insurers do something like this in an effort to control costs. They compete with each other, and whoever can do it best is more likely to get the business of employers, who buy almost all the healthcare insurance in this country. There are potential hazards inherent in this system (lack of transparency, barriers to entry, switching costs, etc.,) but those are topics for later posts.

For now, it's sufficient if I've persuaded you that informational asymmetry is one reason to reject simplistic calls for 'more market competition' in healthcare.

In the next post I'll tackle barriers to entry. I promise, it'll be much shorter, because it's less of a problem for healthcare markets...

Markets for widgets, not for healthcare.

I need some widgets. I can buy from Widgets-R-Us, or WidgetWorld. I visit their websites, and read Consumer Reports' detailed review of quality. Widgets-R-Us has them for $95, and WidgetWorld for $100. Both companies make widgets the same way. There's no difference in quality, which I know because widgets aren't that complicated, and there's plenty of reliable information out there about them. I buy from Widgets-R-Us. The majority of people looking for Widgets can do the same, and Widgets-R-Us ends up getting more business & more profit.

Markets for Widgets Result in Low Costs and Efficiency

WidgetWorld, if it can, should lower its price to at least match Widgets-R-Us. Widgets-R-Us can make a widget for $90, and WidgetWorld for $92 (they're not as efficient). WidgetWorld can lower its price to $93 and still make a profit--so they do, and their market share begins to rise. That is, until Widgets-R-Us drops their price to $91. WidgetWorld must now either get more efficient or go out of business. This is good for everyone...prices are low, and if you can't produce efficiently, you go out of business.

WidgetWorld goes out of business. Widgets-R-Us can now charge whatever they like without fear of being undercut--so they raise their price to $115, and stop worrying so much about efficiency. People with little money go without widgets (even though widgets make wangles work--and you can live without a working wangle, but your life will suck.) They also don't have to treat their widget-making employees quite as well. This is a monopoly. Bad for consumers, bad for employees, although it's friggin' awesome for Widgets-R-Us shareholders and executives.

Happily, Wally, seeing an opportunity, starts a new company--Wally's Widget Emporium. Wally's makes widgets for $90--just like Widgets-R-Us used to. Wally starts to sell for $100 and starts making money hand-over-fist, as consumers flee Widgets-R-Us. But almost overnight the price of widgets drops to $91 as the two companies compete--and employee conditions improve at Widgets-R-Us. Poor folks can afford widgets for their wangles again, and their lives are less bad.

As Wally's and Widgets-R-Us continue to compete, they find more and more efficient ways to make widgets, and prices are continually driven to an equilibrium point just above the cost of production. (If it gets too high above, someone else will come in and make more widgets, driving the price down. If it gets too close to cost of production, the profits drop and the manufacturers will start making something more profitable; widget supply will drop, and prices will rise.)

This is almost miraculous! The widget market, without anyone consciously trying to make it happen, maximizes efficiency and minimizes costs. What features of the market make this possible? For now, I'll just identify some of the features. In subsequent posts, I'll explain why most of these features either do not or cannot exist in markets for healthcare.

Features of an efficient market that needs no government influence:

1. Informational equality.
2. Low barriers to entry.
3. Low switching costs.
4. Elasticity of demand.
5. Few externalities.

Informational equality means that it's possible for me to accurately assess the quality of widgets, and know the price ahead of time. Widgets can be compared side-by-side, so Consumer Reports can competently and accurately assess quality for me.

Low barriers to entry. We were rescued from the widget monopoly when Wally's came in to compete with Widgets-R-Us. This was only possible because Wally could start making widgets without having to be a mega-billionaire (think about starting a car company). He also didn't have to face massive regulatory hurdles (think about starting an airline.)

Low switching costs. The market works well because it costs me nothing to switch from one brand of widget to another. But what if the companies made widgets with different numbers of dongles? If I switched widgets, I'd have to get a new wangle--and I love my wangle! The two companies can raise their prices, knowing that I won't leave them unless there's a BIG difference in price.

Elasticity of demand. If the price of widgets goes too high, I can always just not buy one. I won't like it, but I can live without it, and I can wait around until there's a sale. If enough people stop buying, then the companies will drop their prices until we start buying again.

Few externalities. Externalities are effects of a transaction that are not measured by the parties to the transaction. Externalities can be positive or negative. When I buy a widget, it appears to affect only me and the widget-maker. There's no reason for the government to interfere, right? Right. But what if widget manufacturing creates a lot of pollution? Well, my one little widget doesn't contribute much to that problem--and I don't personally live near the plant--so I don't care. And the widget-maker, in making one more widget for me, reasons similarly. What's it to him if there's a little more pollution? Pollution is a common example of an externality. By definition, market actors (buyers and sellers) don't care about externalities.

We sometimes want government to step in and account for the externality. For example, the government should regulate the widget-makers' pollution. In doing so, the government will increase the cost of production and thus the price. But if government does it properly, then all it's doing is making sure that widget-buyers are paying the true cost of widget production (the direct cost, plus the cost of keeping the environment clean.) Externalities are key to understanding when government regulation or control is justified.

------------------------

In later installments, I'll show why these features, which are required for market efficiency, minimizing cost, and measuring the true cost of goods, apply poorly or not at all to healthcare.

Kindle 2 Review

I have had one request for a Kindle 2 review. And never being one to disappoint my audience (thanks, Neil!) here it is.

IT'S AWESOME! No one should be without one.

Let's get the negatives out of the way--they're minor and quick.

1. If you're going to the beach or the pool or outdoors, bring a plastic ziploc to protect the device from the elements. (No biggie.)
2. If you are just running around doing errands, you can't really just leave your book on the seat in the car--your run-around paperback is now worth $300. And don't leave it on the table at the restaurant, or the counter at the store, etc...
3. The only medical journal available so far is the New England Journal of Medicine. I'm sure this will change, though. Journal reading would be the PERFECT use of this device. WHY aren't publishers all over this?!
4. Related: why, why, why aren't .pdfs better supported? They're not bad, but it costs money if you want Amazon to convert the .pdf and send it wirelessly to your device. You can convert it free, but to avoid the delivery charge (admittedly nominal) you have to upload it to the device via USB cable.
5. Detailed graphics (e.g. charts, medical images) are hard to view. If I really needed to read a chart from a NEJM article I'd have to go online.

Now for just some of the positives:

1. Instant access to something I actually want to read (rather than whatever happens to be laying around--which occasionally is nothing, and frequently is some crap I don't really care about.)
2. Related: I haven't found anyplace that didn't have connection to the (Whispernet) wireless provider.
3. The books are cheap--NYT bestsellers are $10, classic books are free (I'm reading the Adventures of Sherlock Holmes.) However: BEWARE OF IMPULSE BUYING! The library is still a better value.
4. Related: you can buy Kindle stuff from the website, and it gets sent to your device wherever it is. (I bought a medical ethics text the other day from a computer at work, and when I got home, there it was on my Kindle!)
5. Read in any position--turning pages is never fumbly, the book never accidentally shuts.
6. Easy on the eyes--font sizes scale across a big range of sizes. Text is as crisp & clear as you could wish (see caveat above about graphics.)
7. Battery life is insanely long--the thing draws NO POWER except when you change the page (assuming wireless is off.) I took it to Disney and left the power cord home. I read quite a bit there, came home with more than half the battery left.

Do you really need more? Well, I'm sleepy and want to read my Kindle before I nod off. So tough noogies.

Health Care Reform

Okay, I admit it. I have been subjecting anyone who will listen to a rant about the quality of the debate over health care reform. And on reflection, I realized I was uninformed. For this, and for the generally crappy quality of the debate, I fault the Obama administration and the legislature.

They are not explaining things well--even to a sympathizer like me.

Here is a first step toward improving my understanding of just what the hell is being proposed.

Click here to look at the OpenCongress Website, where you can find a pretty well-designed way to review a proposed bill such as HR3200--America's Affordable Health Choices Act of 2009, and make comments.

If I understand the process correctly--and I don't--this is just one of several versions in the House. There are others in the House and the Senate.

Let's all have a debate!

Back from the Dead

All right folks, The Cochranes' Blog is back. All zero of you followers listen up!

I'm bringing it back to life because of a recent 40-post Facebook thread that I subjected people to. (Apologies to Craig Mazin and his FB friends, who had to see that unfold.)

I needed a place to conduct a discussion...