Numbers are hard. Some people find them boring. It’s much simpler to debate whether illegal immigrants should receive subsidized health insurance and whether there should be a government-run health plan option. Everyone can participate in such a debate without much effort.
But numbers matter in health care reform. I want to highlight some important numbers and economic forces in these bills that are receiving insufficient attention. If these bills become law, these numbers will significantly affect your financial well-being.
The effect on health insurance spending
I agree with the President that the core problem to be solved is the out-of-control growth of health insurance premiums and health spending. Any bill which fails to reduce those costs should not become law.
I am talking about public and private health spending. Too much of the debate is focused on just the government’s balance sheet. The goal is to slow the growth of all health and health insurance spending.
Here’s the super-simple conceptual model of what the pending bills do:
- The bills expand insurance coverage through a combination of subsidies and mandates enforced by tax penalties. Academic research is clear that increases in insurance coverage lead to more usage of medical care (and better health). More usage of medical care increases health spending and health insurance premiums.
- CBO thinks the exchanges will foster competition and reduce insurance premiums for those buying health insurance outside of employment. They also think a public option (which will almost certainly not survive the Senate) would further reduce premiums. I question some of CBO’s views on these points, but they’re the ref.
- The bills shift the cost of health insurance for some onto others, through large tax increases and reductions in projected Medicare spending. Shifting costs certainly does not reduce costs, and probably increases them.
The big question is whether the effect of (1) or (2) is bigger. If (1) is bigger, then these bills will increase health spending and health insurance premiums and fail to achieve the President’s basic goal.
I think (1) is much bigger than (2). I think (3) further increases incentives to spend more on health care and health insurance, but it may be a much smaller factor.
Congress needs to ask CBO, “Would these bills result in an increase or decrease in total health spending, and in total health insurance spending, relative to current law?” CBO analyses have so far looked only at subcomponents of the population. Policymakers need the answer for the country as a whole. They also need to understand the effects on public and private health spending combined, and on public and private health insurance spending combined. Don’t just focus on government spending. That’s too narrow.
I believe CBO will and should conclude that these bills would increase total health insurance spending relative to current law. I therefore believe the bills fail in the core objective defined by the President.
In addition to the above economic point, there’s a simple Washington-based argument that reinforces my conclusion: the industries that generate income from health spending generally support these bills. They know that the government mandates will, on net, increase total spending on health care and health insurance, the opposite of the President’s correctly stated policy goal. If these bills actually reduced health spending relative to current law, the insurers, doctors, hospitals, and other medical providers would oppose them. Remember that the insurance industry champions the individual mandate. How many other industries would like the U.S. government to force you to buy their product, and then prohibit you from buying inexpensive versions of it?
The effect on wages for those with employer-provided health insurance
The so-called “insurance reforms” and mandated minimum actuarial values are restrictions that would increase the cost of employer-provided health insurance. I am not aware of any provision in these bills that would reduce the cost of this insurance.
If I’m correct, then these bills mandate a wage cut for anyone with employer-provided health insurance. I believe this is a silver bullet that by itself can kill a bad bill, if CBO confirms my guess.
Congress needs to ask CBO, “For those who have employer-provided health insurance today and would keep it under the bill, would the bill cause aggregate wages to be higher or lower than under current law?”
They also should ask the same question for each income decile.
One private study suggests that this kind of health law could result in real decreases in wages over time, especially for lower-wage workers.
There is no way Members can vote to cut wages, even if some would be made better off on net through higher government subsidies.
If CBO says a bill cuts wages, that bill dies.
The inequity created by the firewall
In 2016 a family with a worker worth about $48,000 in total annual compensation would get about $9,000 of subsidies for the purchase of health insurance, if their employer does not offer them coverage. If your employer offers you coverage, you are not eligible for these subsidies. The bills create a firewall intended to prevent employers from “dumping” their employees onto the subsidized system. This firewall creates an enormous inequity.
Imagine two families in the year 2016, each with an identical worker whose total compensation is worth about $48,000. Both families are required to buy health insurance.
Family A is offered health insurance through an employer. A “silver plan” will cost about $14,000 in 2016, squeezing out $14,000 of family A’s income and leaving $34,000 in wages.
Family B is not offered health insurance through an employer, and therefore qualifies for about $9,000 in subsidies to buy health insurance. Family B thus has $48,000 in wages plus $9,000 in subsidies, minus $14,000 in health insurance and roughly $4,000 in higher taxes, leaving about $39,000 in wages after buying the same silver plan.
Family B ends up roughly $5,000 better off than Family A, even though the workers are worth the same in total compensation. The family that does not get health insurance through employment is better off because it gets a big subsidy.
This is a rough and oversimplified example. Congress needs to ask CBO and JCT for more precise calculations to better understand the unfairness they are creating between like workers.
The incentives created by this inequity
The above inequity creates a financial incentive for Family A and their employer to try to mimic Family B. While the legislation purports to ban this dumping, the incentive for firms to work around this firewall would be enormous.
In addition, this inequity is politically unsustainable. The easiest solution would be for Congress to extend the $9,000 subsidy to family A. But there are many more family A’s than family B’s, and so extending the subsidy (in future legislation) would cause a federal spending blowout. Jim Capretta convincingly argues that the true economic cost of the bill, including this incentive effect, is a multiple of the official CBO estimate.
Congress needs CBO and JCT to explain the incentives they are creating for employers to dump, and what creative paths employers might take in pursuit of that goal. Congress also needs to ask CBO what the cost would be if the subsidies targeted by these bills to only the non-employer market were also extended to the employer-based market, the likely political result of creating such an enormous inequity.
Perverse work incentives
When you’re designing a new government subsidy, you have to trade off between using a limit number of taxpayer dollars to help the most sympathetic target audience, and creating perverse incentives against work. The more you focus subsidies on the poor, the more efficiently you use your limited resources. But you also create a huge “cliff” effect, where those who work harder and earn more offset higher income with diminishing government subsidies.
Jim Capretta (again) estimates the effects of the Baucus bill on a family whose gross income climbed from $24,000 to $48,000 in 2016. Health insurance subsidies would decline from about $16,000 to about $9,000, creating a new additional marginal tax rate of about 30%. Add to this the high effective marginal tax rates under current law from phasing out the earned income credit in this range, along with income and payroll taxes, and Jim says:
(T)he effective, implicit tax rate for workers between 100 and 200 percent of the federal poverty line would quickly approach 70 percent – not even counting food stamps and housing vouchers.
Congress needs to ask CBO and JCT to calculate the effective marginal tax rates on modest income workers and families that would result from these bills. I can’t imagine they want to create effective marginal tax rates that high, especially on workers in that wage range.
Young and healthy subsidizing older and less healthy
Mike Leavitt, Al Hubbard and I wrote about this topic in the Wall Street Journal last Friday. The combination of an individual mandate, guaranteed issue, and modified community rating means these bills would force younger and relatively healthier people to pay higher premiums than their expected health losses would justify, so they can cross-subsidize those with higher expected medical costs. In some cases these could result in enormous increases in insurance premiums for healthy workers in their 20s and 30s.
Congress needs to understand the magnitude of these subsidies, and especially the higher costs that would be paid by younger Americans. It’s fun to emphasize the benefits of these policies to those with pre-existing medical conditions, but it’s irresponsible to ignore the higher costs that will be paid by others. This part is a zero-sum game.
The affordability limit
The Finance Committee bill contains a “hardship exemption” to the individual mandate: if health insurance costs more than X percent of your income, you are exempt from the mandate and not required to buy health insurance. The choice of X can have significant effects on the budgets of modest-income workers and their families, and on the cost to the taxpayer.
Congress needs to understand the effects of different values of X:
- If X is low (say, 5%), then many people will be exempt from the mandate, and the universal coverage goal farther away.
- If X is high (say, 8%), then more people will have coverage because they’re forced to buy it, and at a higher out-of-pocket cost. These people will in effect face a regressive tax from the mandate to buy health insurance. They will be insured but will have less disposable income available for other needs.
- The other option is to have X be high, but shift even more of the costs to other taxpayers by increasing the subsidies. This increases the cost of the bill.
When your gross income is $40,000, the difference between a 5% limit and an 8% limit is $1,200. That’s a big deal in such an income range. Congress should know exactly what they’re doing to these workers and families. How many of them will end up insured and uninsured? How many will see their disposable income decline, and by how much?
Delayed implementation
The Finance Committee staff are using timing gimmicks to “game the budget window.” They are slipping implementation dates and new subsidy programs by six months here and there so that a smaller share of new government spending shows up in CBO’s measured 10-year timeframe. This has the effect of allowing them to increase total actual long-term government spending, while holding scored spending constant. The figures being tossed around casually in the press of a $8XX B bill or a $9XX B bill are misleading, because they represent spending over different timeframes. What matters are the long-term cost per year and the growth rate of that cost.
Congress needs a lot more information
In 1994 the Congressional Budget Office produced a comprehensive 104 page analysis of President Clinton’s health proposal. CBO has been doing yeoman’s work in this year’s pell-mell legislative process, but Members are still woefully uninformed. The President and Congressional Leaders are trying to march their Members forward without adequate information. I suspect this is in part because they fear that some well-informed Members would vote no.
Congress should give CBO and JCT the time to do this kind of thorough analysis, so Members can understand what they are on the verge of doing to American society.
These numbers matter.
(Who to watch: Sen. Grassley, CBO, and Jim Capretta’s blog Diagnosis)
(photo credit: Vintage abacuses by H is for Home.)