Friday, October 29, 2010

Some Economics of the Employer Mandate

It is exceedingly difficult to figure out from the text of the health bill itself just what are the employer responsibilities under the new health bill. First, which bill does one actually look to? The original House bill, the Senate bill, or the reconciliation bill? I have yet to sort that out...which gives you an idea why people can be justifiably upset about this legislation.

But I trust some lawyers and accountants to sort it out for me. See here for a nice summary piece from the law firm Mintz Levin. I have seen other descriptions that are the same, so I think this is correct.

The basic employer responsibility is two-fold: one, provide "minimum essential coverage" to your employees; and two, make sure it is affordable. Affordable is obviously a key definition. From the Mintz article: "...coverage is deemed "unaffordable" if the premium required to be paid by the employee exceeds 9.5% of the employee's household income." Also, a plan is unaffordable if it covers less than 60% of the total cost of benefits. That 60% rule is actually quite complex, as it involves the actuarial value of the plan with a standard pool of participants. Take a standard pool of participants, simulate them through your health plan -- if the participants pay more than 40% of total costs (through deuctibles and copays) on average, then you do not have an affordable plan.

If the employer does not offer coverage or affordable coverage, it will pay a penalty. The calculation of such penalty -- or is it a tax? -- is itself complicated, depending on the number of employees not being offered coverage and who get a subsidy on insurance they buy on their own (remember there is an individual mandate too). Let us say that it is $3000 per year per employee who does not get offered affordable, minimum essential coverage.

As I write this, I realize just how complex all of this is, and think that perhaps the biggest burden on business is going to be paying the accountants, lawyers and consultants to figure all this out. If the rate of business formation takes a hit, I for one won't be surprised.

But my main focus in this posting are the economic effects of the employer mandate on the labor market.

The basic economics would seem to be the following. Employers must offer employees subsidized insurance in additon to any wage paid. This lowers the demand curve for labor by the cost of the insurance to the employer. Employees get a subsidy, therefore the supply of labor also shifts down by the value of the insurance to the employee. The supply/demand figure shown here illustrates the shifts in the demand and supply for labor.

The key issues are these: (1), Is the cost of the insurance to the employer equal to the value of the insurance to the employee? (2), Can wages freely adjust?

As to point (1), if we took the position that $X of insurance provided by the employer must be valued at $X by the employee, then the way I have the curves depicted would be correct. The demand curve shifts down by $X and the supply curve also shifts down by $X (think of the wage and the insurance all being in annual amounts). Then the new equilibrium is at the same quantity, but at a wage that is reduced by $X. Think of it like this: The government says that employees must be paid 90% in cash and 10% in a dollar-based voucher that can be purchased on a dollar for dollar basis and can be spent on anything. Then the cost of a $1 voucher is $1 and the value to the employee is $1. Clearly this does not affect the equilibrium except that some of the wage is paid in vouchers instead of cold hard dollar bills.

In this case of health care, however, employees do not get vouchers for anything but they get a specific good, health insurance. Economic theory shows that payments in kind are generally worth less than payment in cash. That many employees do not now spend their money on insurance supports this idea. So in this case, the supply curve would shift down less than the demand curve shifts down, and employment will fall.

There is a counterargument, based on the market failure in insurance markets. Suppose that adverse selection is making insurance either unavailable or priced such that some individuals choose not to buy the policies available. Then $X of insurance offered by an employer could not only be valued at $X but even by more than that. In this case, the supply curve of labor would shift down by at least as much as the demand curve falls, and employment could even increase.

Point number (2) from above, can wages freely adjust, is also important to consider. First, wages are notoriously sticky downwards. This is the main reason why labor markets do not clear and we get unemployment. With sticky wages, the adjustments pictured in the graph will not happen, or at least not quickly. Labor demand will drop and labor supply might increase, but the money wage will not fall by the value of the insurance. In this case, labor demand will determine employment, which will be lower than before -- at the point where the old wage intersects the new and lower demand curve. Since more people are willing to work than before, there is observed unemployment.

Another reason for sticky downward wages is the minimum wage. To the extent that the current wage is already above market clearing levels because of government mandate, employment is already being determined by the labor demand curve alone. As that shifts down because of the insurance mandate, employment falls and measured unemployment increases.

The summary of all this? Using my prior beliefs on the differential value of insurance between employer and employed (reasonably higher to the employee) and the very significant downward inflexibility of wages, I predict less employment.

The Democrats' Closing Arguments

Paul Krugman: "So if the elections go as expected next week, here’s my advice: Be afraid. Be very afraid."

Robert Reich:
"Why Business Should Fear the Tea Party"

President Obama: "We're gonna punish our enemies and we're gonna reward our friends who stand with us on issues that are important to us."

The comforting thought of Robert Reich looking out for business interests will keep me smiling all day long.

Wednesday, October 13, 2010

A Parking Proposal

It always strikes me how unwilling many people are to use prices to solve problems of scarce resources. Parking on college campuses is a prime example, and Dartmouth serves well as a case in point.

The parking lot close to the Tuck School fills up by 8:15am; if you arrive after that, you will have to hike a distance of at least 10-15 minutes. Now I know that may not sound like a catastrophe, and even be good for one's health, but the value of people's time is significant. When it is raining or snowing, or just plain cold (as in below zero) a 15 minute walk across campus is quite unappealing. And of course it is on such days that the parking lot fills up by 8:05 am.

The major cost of not having adequate parking is that faculty will simply choose to not come in to work. On days when not teaching, a professor can just as effectively work from home. Unfortunately, they are then not mingling with others on campus and the lifeblood of the university -- collegial, intellectual interaction -- drains away. For junior faculty, if senior folks are not around, this is a very serious problem. This is not a visible effect, but it is there, and college administrators who ignore it are ill-advised.

This effect also occurs during the day, if an employee has to leave for an off-campus meeting. Not knowing if there will be a spot when they return, many will go home and finish the day there.

Other less important effects are the waste of time spent circling for a free spot (and the gas and carbon) and the time spent trying to be the early bird. I really hate seeing people idling their cars in the parking lot waiting for someone to show up and leave, yet that happens regularly. Where are the green police when you need them?

So what is the answer? Raise the price of parking in the lots closest to campus. Raise the prices until the market clears and the excess demand disappears. "The beatings will continue until the whining stops."

Why the vehement objection to something like this? The usual culprit is that lower-paid employees will not be able to afford the increases, or just that such price increases are a larger part of lower-paid employees' income, so they hurt them more. Yes, that is true, and it is a valid complaint. But the complaint is confusing the incentive effects of higher marginal prices with income effects.

The way to get around this objection is as follows. The rate for on-campus parking right now is $30 per month and $22.50 per month in remote lots. First point: these prices are way too low to have any meaningful incentive effects, and the discrepancy between lots is a joke -- $7.50 per month, $90 per year. Ha!

So the rates should be something like $100 per month for on-campus lots and $20 for remote. Now we are starting to get into an incentive-relevant region. I suspect many people would opt for a remote lot if it meant $960 per year in their pocket.

But here is the real kicker. To put just the price effect into play, without the negative aspects of an income effect, the College can give everyone a cash bonus in their paycheck equal to the increase in rates: $70 per month, or $840 per year. So everyone can keep their current parking and their overall financial situation is unchanged.

But! Anyone can also switch to a remote lot and save $80 per month! Note how this is more of a carrot approach than a stick. Instead of just saying, "parking is more expensive" we are saying, "Parking is more expensive, but we are going to give you additional money to spend how you like. For many of you, spending it on parking is probably not the best use."

Yes, to the extent that people switch to the lower cost lot, the College loses some money. I confidently conjecture that the gains from increased presence on campus and time and fuel savings will far outweigh the loss of revenue. But if that is such a big problem, then I propose this: have the cash bonus paid only to lower-income employees, accomplished through a sliding reduction of the payment. Something like this: the payment is the full $840 per year for anyone earning less than $50,000, and then it phases out linearly between $50K and $100K. This will make the proposal revenue neutral or even revenue positive for the College.

There are a few minor complications that would have to be dealt with, such as the fact that lower income employees actually only pay $12 and $9 for on-campus vs. remote parking right now. See how fairness and equity issues get resolved through pricing of resources -- exactly the last thing you want to do!! For crying out loud, a parking spot is a parking spot; its price has to be the same to everyone so that we all face the same cost of using it! Prices of parking will have to be raised significantly to these lower income folks, but again, they can be given a cash bonus to offset the impact. They will in the end be better off -- as evidenced by their willingness to forego expensive parking for more cash but more walking.

I am going to be pushing this one.

Thursday, October 07, 2010

Patient-Based Cost Saving Incentives

There is a lot of talk about new payment schemes for health care providers as possible ways to control health care cost and improve quality. The general principle behind the plans is to create incentives for providers to save cost while maintaining or even increasing quality. Payment systems have to put some risk onto the provider, as through a fixed payment for the care of a population, with residual risk borne by the provider. The theme of "accountable care organizations" includes some form of risk-sharing or savings-sharing, as do payment schemes such as "global payments" or "bundled payments." Of course, Medicare does this to some extent already, by paying providers a fixed fee for a DRG -- diagnostic related group.

What I don't see in any of these discussions is extension of the risk- or savings-sharing to the patient/consumer.

Without bringing the patient to bear on the equation, I fear that we will be trying to make the proverbial horse drink from the stream. We will encounter the problems that HMOs (health management organizations) encountered some years back, when consumers paying good money for health insurance ran up against providers who had incentives to reduce care. Yes, I understand that quality is in the forefront today, but I still think there is a basic conflict with consumers who face a marginal price of zero and a provider who wants to do less. That is a recipe for trouble. We have to somehow reduce the moral hazard problem of consumers demanding care to the point where marginal value is zero (which is typically the marginal price they pay).

One idea one of my colleagues has is for insurance companies to give consumers a lump sum when they are sick -- like your car insurer does when you have a crash. Broken leg? OK, that usually costs $10,00, so here is a check for that amount, do what you want.

The big problem with this is it puts all the risk onto the patient, who is even less able to bear financial risk from cost uncertainty than the provider. There is also the problem that some folks won't get the leg fixed -- we will be a nation of limpers.

But there is another way to do it. How about the insurer says: OK, broken leg, that usually costs $10,000 in our pool. If you can get your leg fixed for less than that, you get to keep 33% of the difference.

Voila!! No risk to the consumer, just upside potential. The insurer will have to price policies a bit higher on average, since they bear all the downside risk and share the upside. But that could be priced easily. As consumers started shopping around and asking providers to cut costs, the whole distribution of cost would shift down. This would unleash tremendous forces to cut cost while keeping quality high. And consumers would not be complaining, for they would be getting paid to save!

I like this idea a lot. It creates tremendously powerful incentives on the patient side, without the problems that other incentive mechanisms have. For example, high deductible policies have the (possible) risk of inducing patients to not take enough preventive care, and both deductibles and copays have to run out at some point if the consumer is not going to bear a tremendous amount of risk. In fact, this is such a good idea that it must be out there somewhere already.

Tuesday, October 05, 2010

More Fuel for the "Administration is Anti-Business" Fire

The Obama Administration's Justice Department announced an antitrust suit against American Express, after Amex failed to agree to changes to its contracts with retailers. See here.

I have three thoughts on this case. First, I see it as an example of the "economic engineering" philosophy of the Obama administration economics policy. If they see something that they don't think is right, like credit card fees to retailers that seem too high (or health insurance prices), they look for some regulatory scheme to fix it. In this case, the regulatory scheme of choice is antitrust law (which is meant to prevent inefficient exercise of market power, not to simply push down prices that seem too high).

We had a seminar by Ed Leamer of UCLA last week, and in his paper he quoted Frederic Bastiat (1848) as follows: "There is only one difference between a bad economist and a good one: The bad economist confines himself to the visible effect; the good economist takes into account both the effect that can be seen and those effects that must be foreseen."

Economic engineering of the sort we are seeing is bad economics. It tries to regulate the visible and ignores the unpleasant fact that economic forces will cause adjustments and outcomes that are even worse. Regulating credit card fees sounds great for consumers, but what if it causes less competition in the credit market, or causes companies like American Express to change their very successful and consumer-friendly business model as a result? Or what if threatening insurance companies causes them to stop issuing policies?

My second thought is on why this is essentially an anti-business policy. At best, this policy is a misguided attempt to help "consumers" without consideration of the impact on companies and their owners (also consumers, but in the form of shareholders). That is antibusiness. Even worse, the policy smacks of pitting large business -- banks and payment networks -- against "small" business -- the retailers (is Gap really a small business though)? Even scarier is the hint that just like Secretary Sebelius in threatening insurers, this case is the follow-through of a threat by Justice against Amex: either change your behavior or we will bring you to court. While such threats are OK in many instances, I get the feeling that this Administration likes to flex its muscles a bit too much, and the flexing is usually aimed at getting prices to change from free market levels.

Third thought is on the antitrust case per se. Amex has about a 25% share of the card payments business, a level that is reasonably high but, I believe, below thresholds normally used in such cases. The overall market is somewhat concentrated, at least on some measures (not at the issuing bank level, but on the network level). These facts I agree make the case interesting. However, Amex has a very good efficiency argument for its policy of not permitting retailers to offer consumers discounts for using non-Amex cards: such behavior is free-riding off the investment that Amex has made in its brand name and what "American Express Accepted Here" means. Consumers are attracted to stores that display the Amex logo, but once in the store, the retailer has incentive to induce them to use other payment means. But Amex only collects revenue if the consumer who was brought into the store by the Amex logo then uses the Amex card. Go back and read the classic article, Howard P. Marvel, Exclusive Dealing, 25 J. L. & Econ. 1 (1982).

If the retailer does not think that the Amex logo on its door is worth the restriction, it is perfectly free to drop Amex as a card and no longer display the logo.

That is a pro-business attitude: freedom of contract.