Skip to content

Bernies health care numbers DO add up

February 13, 2016

The latest attack on Bernie’s health care numbers by the  Committee for a Responsible Federal Budget (CRFB)gets the numbers wrong. This is not surprising if you consider that CRFB is a Pete Petersen project, devoted  to cutting Social Security, Medicare, and other social programs.  Normally I’d just point to a refutation rather than attacking the source, but this has widely spread throughout both old and new media, spun as from a non-partisan source.  And it is quite true that CRFB is staffed by both Democrats and Republicans who advocate cutting Social Security and Medicare.   That does not however make it a credible source; nor is it known for rigorous analysis.

Since it has been updated in the last few days to support instant hit pieces, well known economists such as Doug Henwood and Dean Baker have not yet had time to rebut it.  I no longer have the eyesight required for significant research, but a friend pointed me to a good some-guy-on-the-internet comment in a Reddit thread.  After printing it out in a large font, it seemed to me to be an excellent analysis.  So I’m quoting this rebuttal to be considered based on its own merit rather than on the reputation of analyst.  (Though, the author seems to be intelligent, thoughtful and informed, and I suspect that if I hung around Reddit, I would be very familiar with NonHomogenized. )

The first point NonHomogenized makes is that CFRB relies on estimates that revenue-maximizing top rates for capital gains taxes are about 30%. Wheres a Center onBudget and Policy Priorities (CBPP) study(pdf) “found capital gains rates that maximize capital gains revenues well above 30 percent. The CRS result implies that studies such as the new CBO-JCT paper may be picking up temporary spikes and declines in realizations around the time of a tax change — trends that cannot be sustained permanently.”

CBPP also quotes the CRS as saying:

in addition, JCT’s 28.5 percent estimate represented the rate at which capital gains revenues are maximized, not total revenues. If policymakers shrink the gap between the capital gains rate and the top marginal rate on ordinary income, ordinary income tax revenues are likely to rise, as taxpayers stop converting as much of their ordinary income into capital gains to take advantage of the differential. The capital gains tax rate that maximizes total tax revenues is thus likely to be higher than the rate that maximizes revenues just from the capital gains tax.

The capital gains rate that maximizes total revenues depends on both the capital gains rates and the difference between the tax rates on capital gains and the rates on ordinary income.

NonHomogenized makes additional point. Note that the blockquoate that follows is from NonHomogenized, not from CRS:

But only giving part of the picture is hardly the limit of the problems with their analysis:

The analysis being reported on also cited the CBO for estimates of how much revenue would be raised from an increase in income tax rates of 1%… but the CBO’s estimate wouldn’t include the payroll tax being paid on capital gains being treated as ordinary income, so the numbers shouldn’t be strictly comparable in the first place. Additionally, people would no longer be deducting their health insurance premiums and costs, and would instead be paying taxes on that additional income. The CBO scoring applies to a completely different proposal with different implications for revenue.

Similarly, there are issues with their analysis of the employer-paid payroll taxes. For one thing, the numbers on the CBO estimate for an increase in payroll taxes start in 2015 (at $45 billion), while the numbers for Friedman’s analysis start in 2017 (the CBO estimate for a 1% increase in payroll taxes, incidentally, is that it would raise $74 billion in 2017).

If you take the CBO average for 2015-2024, it would be $80 billion per year, but the CBO average for 2017-2024 is close to $86 billion per year. More importantly, however, that estimate assumes a “reduction in individual income tax revenues that would result from a shift of some labor compensation from a taxable to a nontaxable form”, but in this case, the payroll tax would in many cases be replacing the health insurance coverage employers already provide, which is generally already untaxed.

Later in their analysis, they talk about the Laffer Curve and Bernie Sanders’ proposed top tax rates, and repeat an obvious mistake I have seen many sources make: adding employer-paid payroll tax rates to the tax rate paid… but not including that same money in the compensation (which affects every tax rate, since that ‘compensation’ is untaxed). If your employer pays 6.2% of your wages as a payroll tax, and you make $100,000 per year, and someone wants to calculate the tax rate you’re paying, they can’t say “oh, you’re paying 6.2%”, because that $6200 needs to be included in your compensation, and it goes untaxed. That means you’re paying ~5.84%, and this adjustment needs to be made for every other tax you pay as well. It might seem like a small difference, but when you’re talking total employer-paid payroll taxes of 14% of income, it adds up: taking the number from their analysis (a top marginal tax rate of 77%), it would actually be nearly 10 points lower, or about 67.5%.

(My comment: Note that an effective top rate a bit under 70% that Sanders proposes (0nly for people with incomes in the millions) is generally considered to be in the range that maximizes revenue. Given the stimulative affect of overall spending increases in Bernies budget, we could expect the economy to grow faster that under other budges.)

The CFRB also cites Professor Kenneth Thorpe’s takedown(annoying Scribdb download) of cost estimates for Sander’s health care plan. In response Professor Gerald Friedman politely but vigorously rips Thorne a new one. I am tempted to quote Thorne almost in full as I did with NonHomgenized, but since Thorpe’s  rebuttal is far more easily accessed than a comment buried deep in a Reddit thread, I will content myself with a summary

  1. Thorpe does not adequately document all his assumptions.
  2.  Thorpe’s low estimate of administrative savings is less than the savings simply from replacing insurance company overhead by a Medicare like system. Additionally, to arrive at his number, Thorpe must be assuming savings of zero to providers in needing to deal with only a single insurance plan.
  3. Thorpe makes  a number of actuarial assumptions contrary to Sander’s actuarial assumption without providing justification.  Tharpe’s estimate derives from a study considering a not-single-payer plan in ONE state.
  4. Because Thorpe has underestimated administrative savings, he assumes cost savings must come at the expense of actual health care.
  5. Thorpe underestimates savings from negotiating large scale purchases of pharmaceuticals and from reduced medical inflation.
  6. Thorpe assumes an increase in utilization of 14% which would mean an increase in discretionary spending of 38% per year.
  7. These and many other justification by Thorpe are not only contrary to common sense, and in some cases contrary to basic arithmetic. They also go against real-life experience with single payer systems all over the world.


CFRB attack on Sander’s healthcare plan 

CRFB is a Pete Petersen shill organization, devoted to winning cuts in Social Security, Medicare and other social programs.

NonHomogenized rebuttal

Center on Budget and Policy Priorities Study

Thorpe attack on Sanders’ health care plan (annoying scribdb link)

Friedman takedown of Tharpe attack:







Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: