Faulty Health Cost Accounting
What is the cost of a Medicare prescription drug benefit? According to official estimates cited in the Washington Post today, “[t]he drug benefits, to start in three years, are the most popular and expensive element of the plan, which is predicted to cost $400 billion over its first decade.”
Currently the “cost” of such programs and proposals is measured and reported, misleadingly, as the accounting cost to the federal government – that is, the amount of additional outlays needed to finance the program. However, this is a poor measure of the true cost of the program – to get a better measure, we need to also measure the value of the associated benefit.
To illustrate, suppose that every day you consume about a gallon of water. Suppose the water costs about $1 a day – so over the course of a year you spend about $365. Suppose an average American drinks about the same as you, so total water consumption would be somewhere around $110 billion per year.
Now, let’s say that the government is debating a water benefit – all citizens will receive allotments of 1 gallon of water per day. What is the cost of this program? According to federal government accounting – the cost would be $110 billion per year. But in reality, the total cost to everyone would be close to zero, since the extra government expenditure would be offset by the reduction in individuals’ expenditures. What does it matter if you pay the $1 to the water company or to the federal government in taxes?
If the program were financed by an increase in general revenues, there would be some winners and losers from the program as a whole, but on average people will see a net cost of zero.
For prescription drug care, this means that the cost of a prescription drug benefit would be offset by the lower out-of-pocket expense of medication purchases.
Of course, some choose not to drink ordinary tap water, but rather to spend a little extra for fancy bottled water or soda. These people then face the decision of whether or not to “top-up” their government allotment with additional purchases of beverages – and they would not be receiving the kind of liquid they would otherwise desire. In addition, some people might not drink that gallon of water provided by the government and are then (inefficiently) receiving too much. In both these cases the government provision of the “water” benefit may contain some inefficiency.
In the case of health insurance more broadly, however, most people do not typically actively choose how much, or what quality, insurance they desire. They typically get what they need, or what their employer offers, and then pay what they have to. (There is often some choice, but the price usually ranges from expensive to very expensive.) The point is that if you need open-heart surgery, you will likely get an operation – so there is little choice about the bulk of health care expenditure. The inefficiencies that arise because the government is the provider are thus likely to be small.
So, by having, say, universal health insurance financed by the government, you no longer need to pay for health insurance yourself – (or your employer wouldn’t have to pay on your behalf and could raise your salary.) The government accounting cost needs to be offset by the cost savings at the individual level when deciding if it is a good policy.
In addition, universal health insurance is very likely to save money, since those without health insurance typically incur health care at a greater cost to society by, for example using emergency room service more often or not receiving preventative care. These costs are born by the system as a whole; and just because they are not accounted for in the government’s books, doesn’t mean they don’t exist.
Ok, so the water example is a bit silly, since there is not a very strong case to be made that the federal government should be in the water business – at least not in a modern US economy (although we can always talk network externalities, increasing returns to scale, and/or health benefits… perhaps another time).
However, there are two strong reasons why government needs to be in the market for health care.
First, the private market doesn’t work. Because of what economists call “information asymmetries,” the market is likely to break down. Since only the sickest are likely to buy really good insurance, the price for good insurance will be very high – and most of us have settle for cheaper, but inadequate insurance coverage. By segmenting consumers of insurance, we lose the real benefits of insurance – the spreading of risk across individuals – and we end up having to pay the full cost of care. Try buying insurance directly from a company and you’ll see what I mean.
The solution is to group consumers together for purposes of purchasing insurance – this is one reason you get health care offered by employers. But the same problem arises; the information problem is just shifted to the group level. The best option is to create a single group, which contains everyone – universal health insurance. Allowing people to opt out and purchase private plans just returns us to the same problems as the private market.
Second, the government needs to be involved because there are significant “externalities” involved in the health care system. No one likes to see someone have to forgo needed medical care just because they can’t afford to a doctor. When someone can’t get an operation, can’t afford needed medication, or can’t take a child with asthma to a doctor, we all suffer. In these cases there is a role for government to intervene in the private market and to provide what the market is ill equipped to handle.
Would a real prescription benefit have an accounting cost more than $400 million? Yes. But prescription drugs are being purchased anyway, the government should provide this benefit; and, importantly, we should be willing to pay for it through a fair tax system and not put costs off to future generations.
And just to be clear – I oppose the bill that just passed the congress; it would have been simple to just add a real benefit for prescription drugs, without loading the bill with all kinds of funky semi-privatization schemes, a ban on drug price negotiation, billions in subsidies to HMO’s and other businesses, and 1,100 pages of other who-knows-what. If it’s so great, why wait for three years to have it take effect?
Filed under: Economics
John Irons's Blog 