• There are still so many questions to answer

    If you’re following me on Twitter, you already know I’m a little annoyed at how the Supreme Court watch has turned into a sideshow. I’m frustrated every time this turns into a game. If the law is overturned, we are going to be back where we started. If it’s kept in place, there is still so much to do. Spending our time treating this like a circus gets us nowhere.

    Here’s an example. What are we going to do about the issue of people whose salaries come close to the edges of the subsidies? I’ve covered this before, but evidently the Kaiser Family Foundation has updated their Health Reform Subsidy Calculator in the last few days, allowing me to make this point more explicitly.

    Scenario one: It’s 2014. You are 35 years old, married, and have two children. You live in a area with medium regional cost factor, and you can’t get employer based coverage. Your income is $93,699, which is just under 400% of the poverty line.

    In this scenario, your required premium contribution is $8901, as your government subsidy of $2202 covers the rest.

    Scenario two: It’s 2014. You are 35 years old, married, and have two children. You live in a area with medium regional cost factor, and you can’t get employer based coverage. Your income is $93,700.

    In this scenario, your required premium contribution is $11,104, as you get no subsidy.

    Anyone else see the problem? There’s an incentive for you to make less, as making $1 more (before taxes) costs you more than $2200 in after tax dollars. That’s crazy. That has to be fixed.

    It’s even worse at the other end.

    Scenario three: It’s 2014. You are 35 years old, married, and have two children. You live in a area with medium regional cost factor, and you can’t get employer based coverage. Your income is $31,155, which is 133% of the poverty line.

    In this scenario, you get Medicaid. You owe nothing in premiums. Insurance is essentially free.

    Scenario four: It’s 2014. You are 35 years old, married, and have two children. You live in a area with medium regional cost factor, and you can’t get employer based coverage. Your income is $31,156.

    In this scenario, your required premium contribution is $960, as your government subsidy of $10,144 covers the rest.

    Hey, lucky guy at 133% of the poverty line. We’re going to give you a $1 raise in your pre-tax income! But it’s going to cost you more than 3% of your income in post-tax dollars. What? You don’t want that raise? Of course you don’t.

    Go try out the calculator for yourself. We’ve still got a lot of things to fix. I wish we could get started.


    Comments closed
    • This is a problem, but it’s not unique to the healthcare law. There are NY tax rate cliffs, AMT issues, other aid to the poor issues, my wife’s employer health care plan has subsidy cliffs, etc., etc.

      If we’re going to fix healthcare issues, I’d rather start with the government’s restrictions on competition, such as patent monopolies and restrictions on who can practice medicine (licensing and immigration restrictions). These increase cost by hundreds of billions a year.

      • Actually, with taxes, usually the higher rates kick in on money made above certain points, so this kind of thing doesn’t happen. When you move into the 28% bracket, you’re not taxed on all your money at that rate, just the money above the new level.

        But I take your point on there being other things that do this. Few on this scale, though, and few that will hit so many people at once this much.

      • “restrictions on who can practice medicine (licensing and immigration restrictions). These increase cost by hundreds of billions a year.”

        In the real world — as opposed to isolated economic theory — increases in the number of doctors cause increased cost, more waste, and generally more ineffective care. This idea has been beaten to death in study after study, but keeps rising from the floor in true zombie economics fashion.

        I would accept arguments that using more nurse practitioners and physicians’ assistants and broadening their privileges to practice might well decrease costs if they were used effectively, but increasing the numbers of doctors just makes things worse.

        Health care does not respond to classic supply and demand, because — as Adam Smith is the first to tell us — severe asymmetry of information leads to market failure. The invisible hand trips over health care, falls on its face, and breaks its teeth.

        Patents do drive costs up, on everything from medications to iPhones. Generally the idea of patents is fairly popular, but the notion of ending patent and copyright rights is an accelerating trend.

    • On example 1, the $2200 marginal effect is too large, but presumably there’s some non-zero administrative cost (to the insurer, the government and the individual) so there will obviously be some non-zero optimal penalty at the margins. No one really wants a $10 subsidy if there’s any sort of regulatory burden attached to it. I don’t know the specifics of health subsidies work, but if it’s anything like (I presume it will be somewhat more complicated than declaring interest income on bank accounts while filing taxes which has been obnoxious for the past few years with near-zero rates on both checking and savings accounts). I’m also not so worried about this because almost anyone making $94k in adjusted income whose sophisticated enough to recognize the penalty has enough flexibility in itemizing deductions and capital gains to avoid the problematic income level. In fact, someone in a stable job situation who receives a raise could defer income to unfairly receive a subsidy for some number of years (unless there’s also a marginal rate increase near the subsidy cap).

      On example 2, I’d be all for tying premium support to a Medicaid expansion (which is the only way that you could fix this, I think) if it weren’t such a rhetorical gift to people who want poor people to pay more as well. A more practical way to go about this would be funding Medicaid through a flat payroll tax that applies to everyone but is refundable to some degree for people within a few grand of the benefit cut-off. Let’s say $M is the value of the Medicaid benefit, $N is the income cutoff. Introduce a tax at a rate $R where $R*$N = $M (so that people at the cut-off are paying for the benefit entirely; obviously implies a Medicaid expansion). For anyone not receiving the Medicaid benefit, the tax rate on every dollar of income below $N is zero. The rate on income above $N increases gradually to whatever rate is consistent with paying for the program (possibly with caps on contributions).

      This only works if paired with an insurance mandate, though, and it’s a lot of trouble unless there’s any evidence that a lot of people are changing their behavior in otherwise dumb ways to avoid the penalty.

    • Yeah, this is much more pronounced with things like Medicaid and food stamps where there’s an all-or-nothing cutoff point. They could have done much, much worse than a sliding scale up to 400% FPL – the EITC was designed to reward work at low income levels, and it seems to have been a success in this goal, but it phases out around 300%.

      Also, how is this to be fixed as long as we keep it a means-tested program? Ending the cutoffs and making the decline as gentle as possible would result in subsidies being available to an even larger population, which is politically difficult.

      Very important to medical access, I think, is the need to work out some system so that people who bounce above and below 133% FPL (which is very common) keep their same PCP and provider network.

    • Under Scenarios 3 & 4, states do have the option of adopting the Basic Health Plan option under the ACA, which was designed to deal with this “churning” issue around 133% FPL between Medicaid and private insurance. In short, a BHP would use most of the funding for ACA tax credits for people between 133% and 200% FPL to contract with a Medicaid managed care organization for those in that income segment.

      Urban and other think tanks have analyzed this issue, but I generally agree that incentives to earn more, vis-a-vis the ACA subsidies, are not ideal.


    • What I find astonishing is that someone who makes over $90,000 a year gets a $2,200 government subsidy.

      • Why? That’s probably not far off from the tax deduction someone would get if their employer bought them insurance.

        I got a number of emails on this, so I made it a follow-up post.


      • Like Dr Carroll said on the other post, if this person was privately insured, s/he is already getting a subsidy through the tax system.

        But look it at another way. This person has to purchase an $11,000 policy. S/he has to cough up $8,900 of that cost, or 81 percent of the cost. The government is subsidizing 19 percent of this family’s insurance premiums. That’s not a lot of money in the grand scheme of things.

        And for reference, the Milliman Medical Index depicts they typical, total cost of healthcare for 2 adults and 2 kids as $20,700 in 2011. The average employee contribution to premiums was $5,100, plus $3,500 in copays/deductibles. Put another way, employers normally cough up the majority of the cost of healthcare and of insurance. Over 50 percent, at least. This reinforces the point above: the subsidy above is not that much.

      • Also important to note that a “person” making $90,000 (or even $70,000) wouldn’t get a subsidy at all.. those subsidies only apply when that amount is the HHI for a family of 4 (as the federal poverty levels are based on household size).

    • Why should we expect the people who designed such an obviously flawed program to adequately fix it? This was spotted pretty much immediately by critics of the bill, it’s unbelievable that someone could even write this and not recognize the huge implicit marginal tax rates. What other hidden problems are there we haven’t even discovered yet?

      This is the issue to me, not any specific criticisms of parts of the bill, but that we’d entrust the people who made such egregious errors (this subsidy issue, CLASS, exchange eligibility depending on the cost of your employer premium for a single person rather than family coverage, etc) to begin with to implement and run these programs.

      If I brought my car in to a mechanic and when I came back he’d attached one of the wheels to roof, “oh whoops, don’t worry I can fix that” would not be good enough for me. I wouldn’t let the guy who did it touch my car anymore.

      • What do you suggest as an alternative? We should do nothing at all until we manage to elect a completely new set of government folks?

        Seems to me we need to continue trying to solve our problems while also working towards and fighting for better governance.

        • An alternative? How about we reduce the amount of control that they have over the health care system, rather than ceding even more control to them? “Continue trying to solve our problems” does not have to mean Congress micromanaging the system to the degree they are with PPACA, since they’ve demonstrated time and again that they’ll fail to solve problems, create new ones, fail to anticipate obvious unintended consequences, etc.

          Why would you entrust them with more control when they write a bill that anyone with a bit of experience with health care can find obvious flaws in on the first reading? The income/taxation cliffs created by the subsidies they designed were obvious to anyone.

    • There’s another marginal cliff in the ACA, this time from the employer’s side. The law requires all businesses with 50 or more employees to provide a comprehensive healthcare package, but only for full-time employees. Thus, for an employee to go from 29 hours a week to 30 hours a week will cost the company thousands of dollars, and this will create an incentive for employers to keep as many employees at a part-time rate as possible. Of course this won’t really be a factor for people with high-end jobs, say a software engineer or a marketing executive, but for someone with a low-end, relatively unskilled job this could be a very serious problem.

      • 50 full-time equivalents

        I have been through this more times then I care to count. Not only do we have to count part timers but also seasonal.

        see examples here

        I have already had discussions with employers but freezing hiring, changing seasonal pratices and starting multiple companies. Not the type of problems our job creators should be wasting their time on.

        We already see this as a result of HIPAA, groups halt hiring to stay guarantee issued or occasionally hire to get out of small group if they are healthy and want rated on their own.