Jane the Actuary General Interest

What if Trump fixed the healthcare system – and no one knew it?

So let’s start with an article in yesterday’s Chicago Tribune, a wire report from AP, that took me by surprise:

“Employers start sending workers shopping for health coverage.”

The article described a new form of employer healthcare provision, the Individual Coverage Health Reimbursement Arrangement or ICHRA.  This approach takes the well-established Healthcare Reimbursement Account, which combines a high-deductible healthcare plan and a reimbursement account, money provided by the employer which can be used to offset some healthcare costs while meeting the deductible, and adds a twist:  the employer can provide money which individuals can use to purchase health insurance — and they can do so with the same tax advantages as if they were providing the insurance coverage directly.

To be sure, there are limitations that mean that the government’s forecast is that only 11 million employees will benefit:  the employer must not offer group health insurance already (that is, employees can’t select this as an alternative to group health insurance), or can only offer the ICHRA to categories of employees to whom it doesn’t otherwise offer insurance (like part-time employees).  Also, an employee buying insurance through the Obamacare “exchange” can’t “stack” the employer benefit and the premium credits, and can’t pay for the additional costs in a pre-tax manner, but that is possible if an employee purchases insurance outside of the ACA exchange.  And, of course, a “regular” employer-sponsored healthcare plan is still preferable for employees when they benefit from group rates and don’t have to wade through a potentially overwhelming number of plan choices.

(For lots of detail, see “New Final Rule Lets Employees Use HRAs to Buy Health Insurance” at SHRM.)

Now, it turns out, this isn’t new.  This was a rule issued by the departments of HHS, Labor, and the Treasury issued the rule enabling this on June 13 of 2019.  And, yes, this is a “rule” — an administration interpretation/implementation of existing legislation, so in principle Biden could simply issue a new rule which overrules this (with the applicable comment period and other bureaucracy).  It seems likely that two other Trump “rules” — one allowing “association health plans” and the other allowing low-cost short-term insurance — will be sent to the circular file, but I have a hard time imagining that Biden will oppose this one (though perhaps my imagination is faulty).

But I do believe that this small regulation, over time, could have a very outsized impact on the healthcare system.

Bear with me for a minute here:

Remember the staff model HMO?

That was supposed to fix our healthcare system.  Rather than the existing expectation of “consumer-driven healthcare plans” that we healthcare customers will work with our healthcare providers to ensure that our medications are the lowest-cost options possible, that no unnecessary procedures and tests are performed, and that such tests and procedures as are necessary, are done by the most cost-effective provider (e.g., through look-up tools at insurer websites), the staff model HMO’s providers did all that as professionals.

And back in the day — well, not only is my own family’s current health plan a high-deductible one, but our choices are high deductible, or very high deductible.  You likely have the same (unless you’re a public sector employee).  When my first son was born, we paid a $10 copay.  That was it.  Oh, and a $300 upcharge for a private room.  Later, when he needed speech therapy, we paid copays, then were issued a refund check, because, it turned out, there was a no copay, it was first-dollar coverage.

But let me backtrack:  the original HMO concept was staff-model.  Its name, Health Maintenance Organization, was adopted because of the focus on preventive care, in a manner that wasn’t the norm in traditional insurance, which, at the time, did not necessarily cover ordinary annual check-ups and the like.  HMOs came about in the 1970s as doctor practices which were affiliated with particular hospitals; they were also called “prepaid healthcare” and the idea was that they were not an “insurance” product but you simply paid in advance for all your healthcare from that particular healthcare system.

What happened to them?  Some time ago, I tried to figure out the story and there is no good book on the matter. Perhaps that’s now changed.

In the 70s and 80s, they became popular — not mainstream, necessarily, but popular.  In some cases, they became too popular — doctors filed lawsuits in areas (e.g., small towns) where an HMO dominated medical practice, and pushed for “any willing provider” laws as their medical practice was limited to the portion of the population not a part of that HMO.  I vaguely recall that that it wasn’t just about losing clients to the competition but that they ended up with the less-desirable customer base.

At the same time, major insurance companies established their own version of “HMOs” which promised customers (and employers) that they could have their cake and eat it too — medical care with the low cost-share requirements of a staff-model HMO, but with thick booklets of participating providers rather than specific medical clinics to visit.  We were for a number of years in the late 90s and early 2000s enrolled (via our employer) with HMO Illinois, a Blue Cross Blue Shield of Illinois “product.”  We had to choose a primary care physician and women choose an OB/GYN, and an “Independent Practice Association,” a collection of doctors and one or more affiliated hospitals.  (In-between my first and second child, the doctor’s practice I was at, switched from an IPA associated with the hospital down the street to one a half-hour away, which was a nuisance; later, they left the HMO entirely as the networks shrank and, in our last year in the HMO, I had an annual exam with a doctor whom I had picked somewhat randomly from the provider listings.)  This worked on the basis of “capitation” — the IPA was paid a fixed fee per patient, but rather than resulting in a focus on preventive care and health maintenance, each visit consisted largely of handing out referrals to specialists to churn patients out.  This wasn’t sustainable.  (Why didn’t it work?  These weren’t groups of doctors who had come together to provide managed care, but were purely financial arrangements — and specialists and hospitals were not a part of this system in any case so shunting a patient to a specialist was a financial gain, not a loss.)

The end point of this pathway was the movement from HMO to what was called HMO-POS, where the POS was “point of service” and it referred to the creation of an out-of-network reimbursement level, and to the PPO, what we’re now generally used to today, with networks but with the requirement for referrals having been abandoned.  Was it planned, or foreseen, when BCBS and other providers set up their HMO competitors, that this would be the outcome?  Surely not.

And traditional HMOs have not entirely disappeared — Kaiser still remains, having built itself up during the 70s and 80s to such a point that laments about “the provider list is too narrow” are not relevant.  I, again, tried to dig into their story more as well at some point, to understand why there are not dozens of other competitors with their business model, but concluded that it’s just not possible for a plan to become a truly-integrated staff-model HMO in this environment.

But –

in the meantime, we are witnessing the ever-increasing consolidation of hospitals and doctors’ practices.  Locally, my nearby hospital has a growing list of urgent care centers, sites for lab work, and affiliated doctors’ practices.  They had been a wholly-independent hospital but are now themselves merging with a large hospital chain in the area.  There are many similar networks, and growing numbers of them.  As I had watched this trend, I had thought that this would make it possible for a new type of staff-model HMO, one in which alongside their usual roles in the community as medical care providers to anyone who showed up, with any sort of insurance or none at all, the entire network could offer a prepaid/self-insured “medical care product” in which care within that system would be coordinated, with doctors and hospitals alike sharing the objective of providing the best and most cost-effective care — with care while travelling or for rare circumstances requiring even greater levels of specialization being managed through a re-insurance product.  Over time, if coordinated care produced the best outcomes for patients, more patients would switch.

But there was a missing piece.  Employers want to offer their employees medical care that’s reasonably one-size-fits-all.  If you have employees scattered across the country, or even across a wide metropolitan area, it adds one more layer of complexity to your process of providing employee benefits.  In order for the way health insurance works to change, the relationship between employers and health insurance has to change.

And yes, finally, I get to why I think that the ICHRA has the power to reinvigorate health insurance — if increasing numbers of workers are “shopping” themselves, and without the constraints of Obamacare plans which are obliged to use a very small number of tools in their toolbox (high deductibles, narrow networks based on doctors willing to accept low reimbursements), then we might eventually get to the point where a hospital network might find it financially feasible to offer a coordinated care product.

Yes, that’s a big if.  I’m not an expert, and I suspect that, even if somewhere, someone is looking at taking that step, there are likely too many regulatory hurdles in the way.  But it’s a start.

 

Image: http://www.dodlive.mil/2017/10/03/usns-comfort-how-the-hospital-ship-helps-during-disasters/(U.S. Air Force photo by Staff Sgt. Courtney Richardson)

Jane the Actuary Retirement

Forbes post, “Do’s And Don’t’s For Social Security Reform – Does A New Proposal Have The Answers?”

Originally published at Forbes.com on November 24, 2020.  Is it outdated four years later?  If we had reformed the system in those years, it would been, but as it is, it’s just as relevant today!

 

Just in time for Joe Biden to carefully scrutinize — or toss into the circular file — comes a new report from The Heritage Foundation, authored by Rachel Greszler, with recommendations for modernizing the Social Security old age program in order to improve its benefit structure and resolve its solvency problems. Are they on the right track? Yes — and no.

This brief report frames its recommendations in the form of Do’s and Don’t’s.

Don’t enact the Social Security 2100 Act. This House legislation matches up with many of Biden’s campaign promises, most notably that it increases Social Security’s minimum benefit to 125% of the single-person poverty level, per person, and applies FICA taxes to income above $400,000, unindexed. But Biden’s plan failed to fully-fund Social Security because benefit increases ate up most of the added revenue; this bill would have also increased the payroll tax by 2.4 percentage points, from 12.4% to 14.8% of pay. This would boost total taxes to 68.9% for the top bracket in the highest-tax state. The report notes that recipients of all income levels would see benefit boosts, even millionaires, and concludes, “Middle-income and upper-income workers do not need higher Social Security benefits to keep them out of poverty in old age, and workers of all income levels would fare better by keeping the money that the Social Security 2100 Act would take from them.”

Don’t “expand Social Security’s purpose.” Greszler criticizes proposals to use Social Security as a “piggy bank” to fund student loan debt or paid parental leave. These proposals, such as Marco Rubio’s 2018 proposal, would have, more or less, enabled new parents or young adults “borrow” against their future Social Security benefit and repay the funds by deferring the start date.

Do shift Social Security to a flat benefit. Regular readers will know that I’ve touted this reform from the start. Rather than having Social Security try to serve multiple purposes — an anti-poverty benefit as well as a pay-replacement benefit for the middle class — we should recognize that it can accomplish the former purpose much more effectively than the latter. How large this benefit should be, the report doesn’t specify.

Don’t raise or eliminate the tax cap. Greszler cites data on the impact of such a tax hike, and writes, “Even workers not directly affected by the higher taxes could experience reduced incomes as a result of lower capital that makes workers of all income levels less productive.” In fact, this argument is relatively weaker when advocating for a flat benefit; once you’ve removed the connection between the benefit formula and pay, the justification for limiting taxes to a given pay level becomes much weaker.

Do reduce the payroll tax rate. Gradually shifting to a flat benefit would enable a drop in FICA Social Security taxes from 12.4% to 10.1% while also becoming solvent. Of course, ending the cap would enable an even greater reduction.

Do reduce costs by increasing the eligibility age and indexing it to life expectancy, adopting the chained-CPI, and modernizing spousal benefits. This call for a change in the CPI used for Social Security benefits, which would tend to reduce benefits over time, relative to the current CPI-U measure, is quite the opposite of Biden’s call for adopting the CPI-E, an experimental measure which would boost benefits. With respect to spousal benefits, Greszler does not spell out a specific provision but in a separate article notes that the current survivor’s benefit disproportionately benefits wealthier women, and suggests that shared benefits or childcare earnings credits would improve the system. (Yes, there is a point of agreement with Biden’s plan here!)

Do let workers opt out of the Social Security earnings test. This is a proposal I have made in the past as well; Greszler correctly observes that the earnings test is perceived of as a tax but it isn’t, really, and suggests that workers have the option of keeping it, or keeping their full benefit instead and forgo higher benefits if they would be eligible due to recalculations from higher wages.

Do let workers opt out of a portion of their taxes and future benefits. This is in many respects the same proposal we’ve heard before: divert some of one’s payroll tax to an investment account instead, namely, in a system managed through the federal government, similar to the Thrift Savings Program for federal government workers. Greszler doesn’t specify what proportion that might be or what the corresponding reduction in ultimate benefits might look like, but touts the benefits of an “ownership option.” She also acknowledges that the math cannot be a simple matter of proportions because of the need to fund current retirees, though she notes that “That portion—similar to a legacy tax—would decline over time if policymakers enact reforms to put Social Security on a path to long-term solvency.”

Readers, this is where the math becomes challenging, and, to be honest, raises questions of fairness. Even in the existing system, higher earners subsidize lower earners because of the “bendpoint” nature of the benefit formulas. In a flat benefit system, that’s all the more clear. Is it reasonable for a higher worker to be able to use some of those funds meant to subsidize others, to use for him/herself instead? Would it not be more sensible to simply say that the lowered tax rate is what enables individuals to save more in individual accounts?

But in any case, if conservatives who would otherwise object to a nationwide autoenrollment retirement savings program find it attractive if it’s part and parcel of a broader reform package, does that point to a way forward?

Or is all of this simply several years too late, as the Democrats’ new objective to expand Social Security’s benefits doesn’t square with this at all? After all, this proposal imagines a trade-off of increasing benefits for the poor while reducing them for the middle class (gradually over time), but Biden promises we can have our cake and eat it, too, with benefit hikes for everyone.

 

 

December 2024 Author’s note: the terms of my affiliation with Forbes enable me to republish materials on other sites, so I am updating my personal website by duplicating a selected portion of my Forbes writing here.

Jane the Actuary Uncategorized

Forbes post, “Yes, Poor Retirees Pay More Effective Marginal Taxes Than The Rich. What Can We Do About It?”

Originally published at Forbes.com on November 16, 2020.

 

Reformers have long bemoaned the manner in which poor workers pay too high a “marginal tax rate” when both “regular” taxes and government benefit phase-outs are combined together. A recent article at Accounting Today summarized the situation:

“About a quarter of lower-income workers effectively face marginal tax rates of more than 70 percent when adjusted for the loss of government benefits, a study led by Atlanta Fed Research Director David Altig found. That means for every $1,000 gained in income, $700 goes to the government in taxes or reduced spending. In some cases, there are no gains at all.”

That article summarizes the conclusions of a May 2020 paper, “Marginal Net Taxation of American’s Labor Supply,” which took data from the American Community Survey to look at the situation across all surveyed households, to calculate the Marginal Tax Rates when all kinds of taxes, including federal/state/local income taxes, FICA taxes, and so on; and all sorts of transfer programs, including SSI benefits, Food Stamps/SNAP, Medicaid or ACA/Obamacare subsidies, subsidized housing, childcare benefits, and so on. Adding all of these up, the marginal tax rate for the lowest fifth of earners, as a whole, was 37.8%, a rate actually slightly higher than all but the highest 20% of earners, whose marginal tax rate was 41.3%.

And, yes, the same is not only true when looking specifically at retirees, and splitting them out by wealth levels, but even more extreme. (Thanks to Mr. Altig for sending me the data here.) The average marginal net tax rate for lowest-wealth-quintile retirees, on a lifetime basis, works out to 55%. For the middle-wealth folks, it’s 44%. And for the wealthiest, it’s in the middle, at 48%.

What does it mean to say “on a lifetime basis”? That’s a calculation that takes into account the double-taxation we pay when investing our savings and receiving interest income or capital gains. And, yes, someone with little income, saves little of it, but still saves some, and someone without sophisticated investment strategies gets lower investment returns, but still gets some.

Here are two graphs, first the effective marginal tax rates by wealth quintile, considering only the current year’s income, and, second, considering the effect over the individual’s lifetime due to double-taxation on savings, for all people over 65 in the survey:

And here, in table form, is a breakdown of the calculations of extra tax paid for the lowest fifth of the country, in terms of wealth, based on a baseline and the recalculated numbers for earning an extra $1,000 — for both “normal” taxes as well as the loss of government benefits, again, on a current-year and lifetime basis. To explain the abbreviations (especially for non-Americans):

  • TANF is traditional welfare for the unemployed poor,
  • SNAP is Food Stamps, that is, food vouchers,
  • Section 8 is subsidized housing,
  • CCDF is Child Care Development Fund, the name for government childcare subsidies for the low income (yes, over-65s can be eligible if they are taking care of children, such as grandchildren, while still working),
  • Medicaid is medical care for the very low income;
  • Medicare is medical care for everyone over 65;
  • ACA means subsidies to purchase private-sector health insurance;
  • SSDI means Social Security Disability Insurance (not based on income); and
  • SSI means Supplemental Security Income, or benefits for low-income people over age 65 or who are disabled, for whom Social Security Old Age or Disability benefits are insufficient to keep them out of poverty.

Remember, too, that these are averages. In the same way as, for all individuals, Altig’s research found that some workers were far more impacted than others, the same is likely true here as well. The 75th percentile person in the age group 60 – 69, in the lowest-wealth group, had a 74% marginal lifetime tax rate; the 25th percentile person, only 33%. For those age 70 – 79, the 75th percentile tax rate was 74%, and the 25th percentile, 34%.

So what’s to be done with this information?

When it comes to younger folk, the call to remedy these high marginal “tax” rates, taking into effect loss of benefits, tends to produce two reactions: some people shrug these calculations off with the response that there is not really any alternative way to design benefit programs, and others insist that these impacts are not relevant because the poor so sincerely want to move away from government dependency that potential benefit losses don’t factor into their choices in any case. Whether this is true or not, the picture gets even more muddled when it comes to older Americans.

On the one hand, however much we think of the over-65s as retired, so that “income” doesn’t matter as such, many older Americans continue to remain in the workforce, even if only part-time. But there is no promise that, even if their paychecks are mostly wiped out by taxes and benefit losses, in the future, with pay raises to come, it’ll be worth it. And the “cost” of working, for someone over 65, is, often enough, greater than for their younger co-workers, even if just from the physical impact of time spent on one’s feet at a cash register.

And at the same time, much of the “income” of the over-65 set is in the form of pensions and the spend-down of tax-deferred retirement savings. And here there is a question that only a few experts are talking about: to what extent is it worthwhile to prod the lowest-income workers to save more for retirement, if they do so at the cost of accruing more debt, in the here-and-how, and if their future retirement income is less than all the retirement calculators predict it will be, due to the loss in benefits?

And, on the third hand, there’s what James Meigs, writing in a recent article at City Journal, called the “Chump Effect.” The article is particularly timely with discussion of a potential student loan elimination program being floated as achievable as an Executive Order, producing anger from people being made to feel like chumps for having saved for their children’s college tuition, or recent graduates, for having worked while watching peers partying, or choosing less-expensive schools than those now complaining about student loans. It is valuable, for the well-being of society, for those who worked and saved, not to feel they are being made into chumps if they think their efforts were not worthwhile.

So what can be done? I’ll remind readers, again, that a “basic retirement income” like that of the UK, the Netherlands or Australia, would solve at least some of these problems, though I admit that I seem to be a voice crying in the wilderness on this point. So how ‘bout it, America?

 

December 2024 Author’s note: the terms of my affiliation with Forbes enable me to republish materials on other sites, so I am updating my personal website by duplicating a selected portion of my Forbes writing here.

Jane the Actuary Retirement

Forbes post, “Social Security Benefit Accruals Stop After 35 Years Of Work History. Is That Fair?”

Originally published at Forbes.com on November 11, 2020.

 

Imagine that you’re a blue-collar worker who starts out with an apprenticeship directly out of high school and works until your full retirement age of 67 — for a working lifetime of 49 years.

If you had a pension in the private sector, say, 1.5% of pay per year, you’d get a pension benefit of 73.5% of your average pay.

On the other hand, imagine you’re a white collar worker who attends college for 4 years, then grad school for another three, with a gap year travelling Europe somewhere along the way. That’s a working lifetime of only 41 years. For many college graduates, add another year, because college often takes longer or an unpaid internship gets tacked on somewhere along the way. With the same pension plan, your benefit would only be 60% of average pay. If you decide to take early retirement at age 62, that would put you at a 35 year working lifetime, or a hypothetical private-sector pension of 52.5%.

But in any of these three cases — the person who works 49 years, or 40 years or 35 years, if their highest-35-year average salary is the same, indexed for wage increases, then their Social Security benefit — or, strictly speaking, their Primary Insurance Amount before any early retirement reduction, is the same.

Is that fair?

Is it fair that someone who continues to work after reaching that 35 year marker, even if they take a pay cut (say, in a semi-retirement part-time job) so that the new work history won’t actually boost their Social Security benefits, must continue to pay FICA taxes?

Yes, framed this way, it sounds pretty outrageous. It suggests that blue-collar workers are being cheated out of money they should be getting, or that white-collar workers are getting money that they don’t deserve.

Of course, that’s not actually the way the 35 year averaging works. (I admit: I initially wrote “the purpose of the 35 year averaging” but realized I can’t make a claim as to the intentions of the formula’s designers.) Effectively, the system presumes that workers who are not in the workforce are missing for valid and appropriate reasons, whether it’s schooling, or periods of unemployment or caregiving for children or parents. In some Social Security systems, say, France, for instance, the system requires a full “working lifetime” and grants “credits” for justified reasons such as these, but no such “credit” for periods spent on the beach in Fiji.

Or, the system’s design has the effect of saying, “anyone who worked ‘enough’ years over their lifetime deserves the same benefit-relative-to-income as anyone else who worked at least that much.” Social Security, in this point of view, is not about hard work, but just about being “a worker.” There’s nothing particularly worthy about working more years beyond this minimum, no reason to get rewarded or for someone who didn’t do so to lose out, because, after all, Social Security is not truly “earned” in the same way as a private-sector pension but is social insurance, which operates entirely differently. This is similar to Biden’s promise that any individual who works at least 30 years (and, remember, to earn sufficient credits requires earning $5,640 per year, as of 2020, or 15 hours per week at minimum wage) would be promised a benefit of 125% of the single-person poverty line, not because they had “earned” it by their contributions but as a matter of social insurance.

But this is, again, the problem with our ideology about Social Security. If we wished to reward people who worked more years, without increasing costs for the system, it stands to reason that we’d need to somehow reduce benefits for people who had worked fewer years. And we’ve talked ourselves into the premise that the system is fundamentally unalterable, rather than designing a system which is fundamentally flexible enough to respond to changing conditions.

 

December 2024 Author’s note: the terms of my affiliation with Forbes enable me to republish materials on other sites, so I am updating my personal website by duplicating a selected portion of my Forbes writing here.