Thursday, February 26, 2009

Skiing

No blogging this week: I'm going skiing.

Monday, February 23, 2009

Comments/HMOs

A bunch of good comments on my posts lately, and rather than responding in the "comments" area, I thought I'd bring some of them up here. Regarding my post on adverse selection in health insurance, an anonymous reader writes:
So it is apparent that our current costs are known and this points out the flaws in our system. However, a fully socialized health care system can worry people, who think they will not get as good of care. But, wouldn't we paying in what we are now, or roughly the same to a socialized system? And if so and we eliminated the insurance shuffling and paperwork issues (because of this change) wouldn't more of the money we spend go to treating people? What am I missing?

Before I wade into this, let me state that I am not a health-care economist, and I don't know the answer to the question of "what should we do?" But I do know enough to at least see the costs and benefits of various courses of action. This reader is asking about the costs of a single-payer system, so I'll comment on that.

One issue with a socialized, single-payer system is that we're not using the power of competition to keep costs under control.

What happens to automakers who don't contain costs? Consumers stop buying their cars, and they'll be competed out of business (or, perhaps, competed into a massive government bailout... But you get my drift). The key notion is that market competition rewards those firms that figure out ways to keep costs down and still provide goods and services that consumers want.

We'd like the same thing to happen in health care. That is, we'd like someone to be shifting health-care business around to the providers who can offer care that consumers want in a cost minimizing way.

This sounds simple and obvious, but it doesn't work very well in this market. Why not?

One reason is that most of us don't pay the full cost of the health care we receive. Just a quick story about this: I was in Alta's Watson shelter last Friday and walked past the ski shop. The shop is selling helmets and their sign read "How large is your health insurance deductible? Buy a helmet!" They're saying to potential helmet-buyers: "You are on the hook for part of the cost if you hit a tree and have a head injury." Notably, however, you're not on the hook for the whole cost --- once your deductible is paid, any incremental costs are paid by the insurer.

Because consumers don't pay the full cost, they don't shop around aggressively for low-cost care. We don't haggle for medical care the way we do for a car. (And insurance isn't the only reason for this lack of haggling --- Health care is credence good, which means it's hard for consumers to even assess quality.)

So is there any haggling in this market? Any attempt to find low-cost providers? There is, and it happens between your employer (assuming you have employer-provided care) and some HMOs. Think of an employer negotiating with a bunch of insurers. The employer wants to get a good deal, and the HMOs compete by trying to buy low cost care from providers, packaging that care with insurance, and selling it to employers. One way the HMO can get a competitive edge is to negotiate lower rates from doctors --- this is a force that helps control health care costs.

Essentially, the current system uses the profit motive to contain costs in the following way: HMOs are allowed to keep any profits they earn by negotiating for lower-cost care with health-care providers. What keeps those profits from getting too big? Competition from other HMOs. What keeps HMOs from completely ignoring the quality of care? Again, competition from other HMOs. If employers look for the best quality care for employees at the lowest possible cost, then the current system at least has some competition-based mechanisms that pushes things in that direction.

Single-payer relies on the government, not markets, to find the balance between quality and cost.

Now, before you completely flame me for writing something nice about HMOs, note that I'm not saying that the current system is the best we can do, and I'm not saying socialized medicine would be worse than the current system. But I am saying that the big question mark for socialized health care is finding a means for appropriately matching up quality and cost. The current system isn't ideal by any means, but there is at least a profit-motive-based mechanism (an imperfect one, to be sure) to get health-care costs under control.

Read the last couple paragraphs of the Gruber article, for some hints on what the big drivers of health-care inflation have been.

All this is a bit moot anyway, since most Americans are happy with the insurance they have. Read Gruber's discussions of the overinsured and the underinsured in our country on this. As a result, it seems reform will be incremental, along the lines of the Massachusetts plan.

Friday, February 20, 2009

Old Friends and Taxing CEOs

One of the cool things about the internet is that people can find you. I have friends who don't want to be found --- something about a crazy ex-girlfriend? --- but I don't mind it.

Anyway, this week I got two e-mails about this blog from old friends. One was from a Kellogg MBA who took a couple of classes from me a few years back. She's now working for a large global consulting firm. I was thrilled to get her mail --- and also thrilled to know that somebody is actually reading some of what I write here. If there are other Kellogg MBAs out there in cyberspace, come on out of the woodwork. Let me know what you're up to.

She was wondering if I had any thoughts about this article from the NYT.

After the small fortune she paid to Kellogg, I figured I should probably answer one more question from her. (Utah students: You get zero extra questions... jk.)

So here are a few thoughts:

(1) I'm not a fan of caps on pay. Super smart people have lots of opportunities. They're going to make scads of money one way or another. Do we want them to make their money working as "consultants" to the banking industry, where they can charge whatever fee they like? Or do we want them to be employees, where their fortunes can depend a bit more on how the firm does? I think the latter is better, but the pay caps will push people to the former. (I do think we should be limiting payments to owners --- that is, limiting dividend payments --- for bailout banks.)

(2) Is the Obama administration thinking about this? I am 100% certain they are. How am I so sure? Austan Goolsbee, an economist from the U of Chicago GSB who's on the President's Council of Economic Advisors, is one of the leading experts on taxing CEO pay. Here's a link to a paper of his on this.

(3) What Goolsbee found in his research was that CEO labor supply seems to be pretty inelastic. If you think back to how taxes affect economic activity, a lot depends on how elastic --- that is, how price sensitive --- demand and supply are. If demand and supply are elastic, then taxes can cause buyers and sellers to choose not to transact. The tax increases the price buyers pay and reduces the price sellers get... So if buyers and sellers are sensitive to price, this government interference will lead to reduced buying and selling. If demand and supply are inelastic, then the price doesn't matter much to the buyers and sellers. They're so eager to trade that even the tax doesn't deter them.

When the Republicans talk about how taxes will hurt economic activity, they're essentially arguing that demand and supply are elastic. This is for sure true in many markets, but not all.

Putting this in the CEO labor market context, if sellers (that is, CEOs) are price sensitive and their wages are taxed more heavily, then you might imagine them working less. A CEO might think "Well, I used to keep 70 cents of every dollar I earned, and now I keep 60. So I'll work less." If this were true, you might think that the pre-tax level of pay would go down after a tax increase.

But this isn't what seems to have happened in Goolsbee's data. CEO pay went up in the year before the tax increase and down in the year after. CEOs seem to have found ways to shift a little pay from the future to the present. But then pay went up on pretty much exactly the same trend as before.

So --- I think the advice the President is getting on taxing the rich is this: "Probably it won't cause the rich to work much less."

Let's see in the upcoming State of the Union address if Obama has anything to say about this.

Wednesday, February 18, 2009

Adverse Selection in Health Insurance

Front-page article today in the NYT on young adults who choose not to buy health insurance.

What's wrong with this?

When I teach about adverse selection in Fin 6025, students usually suggest that the big problem is uncompensated care. Uncompensated care happens when an uninsured person gets sick or is injured, and then seeks treatment (usually at an ER), and then can't pay.

But uncompensated care is actually quite a small part of overall health care costs. According to a recent survey by leading health care economist Jon Gruber, uncompensated care is only $30 bn out of the $2 trillion US health care spending. That's 1.5%.

A bigger problem is that when healthy people jump out of the pool of the insured, then insurers figure that you must be pretty sick to want to buy health insurance. This is called adverse selection, because the "selection" of people who choose to buy insurance is "adverse" to the interests of the insurer. And this raises the price of insurance, and causes more of the relatively healthy to jump out of the insured pool.

So we end up with risk-averse people choosing not to buy insurance, because the price of insurance reflects the _average_ cost of insuring those who choose to buy, not the specific cost of insuring that person.

This sort of adverse selection problem can lead to market failure --- that is, it can lead to a situation where everyone's well being could be improved by changing the market outcome.

And this part of why markets don't do as well at allocating health care resources as they do at allocating, say, tin.

And this is why governments (around the world) are so heavily involved in the provision of health care. Health care is no more a "basic human right" than, say, food, but governments play a comparatively small role in the provision of food.

Monday, February 16, 2009

Basketball

Cool article in Sunday's NYT magazine about measuring the performance of basketball players.

It's a harder problem than measuring the performance of baseball players. With the exception of playing defense, baseball isn't really a team sport. It's more a series of one-on-one matchups that can be easily analyzed.

But the Rockets are clearly thinking pretty hard about applying analysis to this performance measurement problem. They've broken down, for example, what actions a player might take that help the player's resume but don't help the team win. And that's a good lesson for any manager --- sports fan or not --- trying to measure employee performance in settings where performance is hard to measure.

Here's the best line of the article:

It turns out there is no statistic that a basketball player accumulates that cannot be amassed selfishly. “We think about this deeply whenever we’re talking about contractual incentives,” he says. “We don’t want to incent a guy to do things that hurt the team” — and the amazing thing about basketball is how easy this is to do. “They all maximize what they think they’re being paid for,” he says. He laughs. “It’s a tough environment for a player now because you have a lot of teams starting to think differently. They’ve got to rethink how they’re getting paid.”

"They all maximize what they think they're being paid for" --- that's a starting place for a course on organizational economics, just like the one I teach.

Saturday, February 14, 2009

Friedman and Hayek

A former student wrote this week to ask some questions about an old lecture. I directed him to what I think is one of the very most important ideas in economics: The notion that a fundamental role of prices is to convey information that guides individuals' choices in a socially productive direction.

I can't improve on what Nobel-laureate Friedrich Hayek wrote about this in 1945, so I'll just link to it.

Focus on H.21 through H.24, in particular. (H.24 might be my favorite paragraph of economics ever. It totally rocks!)

Thomas Friedman, the NYT columnist, will be speaking in SLC next month, and a frequent theme of his columns is that we need to get the price signals right so that people will make socially efficient investments in energy technology. He supports higher taxes on carbon-based fuels.

This notion of getting the price signals right is straight out of Hayek. Now, there's nothing in Hayek about taxes and further Hayek's work is often held up as an intellectual justification for free markets not government taxation. How do we get from Hayek to "tax carbon"?

In Hayek's conception, prices come from perfectly competitive markets with no externalities. His discussion of tin, for example, fits this. Perfectly competitive markets without externalities "work", in the sense that they achieve a socially efficient allocation.

My interpretation of Hayek is that when markets "work" --- which they very frequently do --- we should be really, really careful about messing with them. Because government interference can mess up the price signals, which will mess up individuals' choices. Government subsidies for housing loans, for example, can mess up the prices (that is, interest rates) for home loans, and cause people to make poor choices. (Sound familiar?)

But, economists have learned a lot about when markets "work" and when they don't. Markets can fail to achieve the best outcome when externalities are present. In cases like this, prices will send the wrong signals, and we'll get a not-the-best outcome. Friedman's point is that we need to use the power of markets (with a boost from tax policy) to send the right signals.

So, go listen to Friedman (and look for me there), and read Hayek.

Friday, February 13, 2009

Whaling

More winter reading: I recently finished Nathaniel Philbrick's "In the Heart of the Sea." It's about the 1820 wreck of the Nantucket -based whaling ship Essex. This true event is believed to have been the catalyst for Melville's fictional Moby Dick.

I learned a lot about the industrial organization of whaling --- Very interesting industry! Seems to have been a lot of labor-market-related knowledge spillovers that allowed the little island of Nantucket to dominate this industry. But those advantages were eventually outweighed by the fact that Nantucket's harbor isn't really that good. Any really big ship had to be unloaded far from shore, and whale oil brought in to the town by small boat. This is expensive, and eventually the industry moved.

Oh, and plus there was a horrible disaster on the Essex when an angry sperm whale attacked. For some reason Philbrick spends most of his pages on that rather than the economics of whaling...

The most relevant passage in the book to today's economy was this:

Making (Nantucket's) level of profitability all the more remarkable was the state of the world's economy in 1819. As Nantucket continued to add ship after ship to her fleet, mainland businesses were collapsing by the hundreds. Claiming that the "days of our fictitious affluence is (sic) past," a Baltimore newspaper reported that spring on "dishonored credits, deserted dwellings, inactive streets, declining commerce, and exhausted coffers."
Sounds familiar. And serves as reminder that economic times have been bad before. They'll get good again.

Thursday, February 12, 2009

Chaffetz: Wrong on Stimulus

Our 3rd District Congressman Jason Chaffetz voted against the stimulus package and recounts his reasoning on cnn.com.

Most economists, myself included, think he's wrong on this issue.

The big immediate problem we face in the economy --- the problem that's leading to the half-million jobs lost each month --- is a kind of coordination problem.

To understand it, think about a firm making decisions about how much output to produce over the next six months and, by extension, how many people to employ. If our firm expects demand for its product to be strong, then it will plan to produce a lot. It will plan to employ a lot of people. If it expects demand to be weak, our firm will cut back on output, and we'll have to lay some of our workers off.

Now, what determines demand for our firm's product? One factor that determines demand is whether our firm's customers are likely to have jobs. So, if our firm expects other firms to have layoffs, then our firm will expect demand for its product to be weak. Expectations of other firms layoffs can trigger layoffs at our firm.

The reason I refer to this as a "coordination problem" is that our firm's best strategy --- layoff or no layoff --- depends on what other firms are doing. One outcome of this game is for no one to have a layoff. Another outcome is for everyone to have a layoff. We're on the "everyone" track as of now.

So how can we stop this cycle?

Well, this isn't the sort of thing that any single firm can undo. One firm deciding not to lay off its employees won't have a big enough impact. A group of firms --- a very large group --- could undo this by all simultaneously deciding to maintain or increase employment. But think of the problems of getting all those CEOs on board.

A final possibility is to take the single biggest player in the US economy --- the federal government --- and have them spend money. And this is what the stimulus package is for. The idea is that the government puts people directly to work --- and this will support demand and induce the firm we considered a few paragraphs above to not do layoffs. This will then support demand at other firms, and we can limit the spiraling job loss we are currently seeing.

This is what the president means when he says that only the federal government is large enough to have an impact on this crisis.

Don't take my support for the stimulus bill to be support for each and every provision in it. My view is that it would be better if we saw more immediate spending and less in the way of tax cuts for anyone with high earnings (unless those cuts are in rates that could somehow be made permanent, which I don't think is feasible.)

But when Congressman Chaffetz votes no because he "spoke with a guy who employs 12 people in his small trucking company, (who) sees a trillion dollars in new deficit spending and yet nothing that will help him," I just have to respond.

Ask the trucking company owner whether his business has been affected by the general economic slowdown. Ask him whether he'd prefer Utah's unemployment rate to rise to 6-7 percent (which it is likely to do even with the stimulus package), or whether he'd prefer Utah's unemployment rate to hit 8-9-10 percent. Most economists think the stimulus bill will mean lower unemployment. And this is likely to be good for the demand faced by local trucking companies.

And since I'm criticizing a Republican here, let me praise some as well. How should governments manage their fiscal policy, if they want to limit the impact of the business cycle on residents? They should do pretty much exactly what I'm guessing the State of Utah will do. Run surpluses and eliminate debt in good times. Bond --- that is, borrow --- and use the Rainy Day fund to support spending in bad times. This allows the state to work against the business-cycle coordination problem outlined above, without running massive long-term deficits.

Monday, February 9, 2009

The Middle Class

Article in the D-News yesterday on "The Middle Class".

Reporter Lois Collins did a nice job summarizing the facts, so I won't repeat them here. But I will add a little about some theories on why this might be happening.

The leading theory among economists is something called "skill-biased technical change." The idea is that production technologies --- that is, the way we make good and services --- have changed. And they've changed in such a way as to make the skills of very educated people (think of college graduates and those with graduate degrees) more valuable relative to the skills of somewhat educated people (think high school grads).

If your skills become more valuable, then employers will be willing to pay higher wages to try to hire you. If your skills become less valuable, then employers won't be as eager to hire you, so your wages won't grow.

Why might this have happened? Computerization is one possible answer. A computer isn't a good substitute for a truck driver. But a good piece of accounting software is a good substitute for a bookkeeper. So, maybe in 1980, a CFO needed a staff of ten bookkeepers to track accounting numbers in a large firm. By 2005, though, a similar CFO could do the same work with only five bookkeepers. Computers substitute for bookkeepers, and therefore make bookkeeping skills less valuable.

At the same time, the computers are allowing the CFO to do more with less. This might make it more important to have a really good CFO. If computers complement CFO skills, then good CFOs will be even more highly sought after, and their wages will rise.

David Autor is an MIT economist and a friend who works on these issues. He's written recently on the polarization of the labor market.

I want to emphasize that skill-biased technical change isn't the only potential explanation for what's been happening in our labor market. This is not a settled question in our field --- others think globalization or tax policy might be driving changes in the labor market.

Friday, February 6, 2009

Present Value Math, Mitt Romney, and Tax Cuts

Mitt Romney says he wants permanent tax cuts rather than direct government spending.

One problem with tax cuts in the current environment is that it's hard to commit to making them permanent. Even John McCain's economists think that overall tax rates have to go up --- due to the large entitlement spending that our government has to do over the next 30-40 years. And if businesses expect tax rates to go back up after the current crisis is over, then they won't respond (as much) to a current tax cut.

Here's some present value math on that:

Businesses decide whether to invest by following the net present value rule. Consider a project that costs $7.5 this year, but will pay $1 in revenue annually starting next year. If the tax rate on those revenues is zero and the interest rate is 10%, then this project has a positive net present value.

The present value of a $1 perpetuity is $10, and so the present value of $10 exceeds the present cost of $7.5.

If the tax rate is 30%, then this is a negative NPV investment --- the after-tax NPV is $7, which is less than the cost.

If we permanently cut the tax rate 20%, then the the after-tax NPV rises to $8, and the project is positive NPV. The firm will invest today, which will create jobs and stimulate the economy.

But what if the owner of the firm doesn't think the tax cut will be permanent? In particular, what if the business owner thinks the tax rate will be 20% next year and the year after, but will return to 30% after that?

Then the after-tax NPV of the investment is $7.17. So no investment!

What this means is that expectations of future tax rates matter a lot for the stimulative effect of tax cuts.