The New Republic’s Jonathan Chait has penned an ode to the excise tax on health benefits. He says his "faith in the Cadillac tax remains ardent," even after being led astray by a faulty graph.  We hate to dim his ardor, but it’s a love gone wrong.  You should know in advance that it’s not a typical love story, though: there will be some math.

Chait sought to rebut criticisms of the tax from Allan Sloan and Bob Herbert.  He’s especially angry at Herbert for saying that "the dirty little secret behind this onerous tax is that no one expects many people to pay it."  He’s so angry, in fact, that he uses an entire paragraph – 98 words –  just to object to the phrase "dirty little secret."  (That’s a word ratio of more than 36:1; we told you there would be math.)

Then he turns his ire on Sloan for doubting that employers would return the money they save (by cutting benefits) to their employees.  He takes particular exception to this statement from Sloan:  "I … find it hard to believe that employers can work any harder than they already do to hold down health care costs."  Chait responds as follows:

First, Sloan doesn’t believe that employers could do any more to hold down health care costs than they already are? What exactly are they doing right now? And even if you think they are doing a lot already, surely scaling back a tax break that encourages lavish spending would encourage them to step up their efforts, right?

Let’s start with that "lavish spending" piece.  A definitive study published in Health Affairs analyzed actual employer health plans and discovered that benefits design (the allegedly "lavish" part) accounted for less than 4% of the cost difference.  Plans won’t be taxed because employers are spending lavishly.  They’ll be taxed because employees are older, sicker, work in the wrong industry, or live in the wrong part of the country.  The typical benefits given to people in these categories accounts for 96.3% of what makes a plan costly (or, to use  Chait’s word, "lavish.")

Singling these people out for demographic reasons isn’t a matter of identifying health system abusers:  it’s discriminatory.

Employers have experienced annual cost increases of 131% over the last ten years (Kaiser Family Foundation), and have increased employee premium contributions by 128% – and that’s not counting all the benefit cuts that have increased out of pocket costs for employees.  That’s how they react to cost increases. Since they can’t reduce them, they pass more of them on to their workers.

That’s not cost-reduction.  It’s cost-shifting.  That’s not "bending the curve."  It’s passing a bigger portion of the curve on to the middle class. Yes, they’ll "step up their efforts" – but their efforts only hit the family pocketbook.

Chait goes on to argue that, despite objections from many analysts and economists, the money saved will be passed on to the affected employees in the form of increased wages.  What he doesn’t tell you is that in two recent surveys, employers – real employers, not extrapolations from statistics – indicated overwhelmingly that they would not do that.  No way, no how. (Mercer, Towers-Perrin)

He then cites "a shelf of studies" on the wage/benefit connection, and quotes at length from Austin Frakt’s review of the literature .  (Frakt actually only lists eight, which is a pretty short shelf, and two are from one source, Jonathan Gruber.)  Frakt quotes two researchers (Baicker and Chandra) as saying that a 10% increase in health premiums reduces the likelihood of being employed by 1.2%, reduces hours worked by 2.4%, and has other effects.

Frakt goes on to say that "we don’t have to take just (their) word for it," naming seven more studies on wages and price (plus an aggregation of all of them).  But the other studies don’t confirm Baicker and Chandra! They just indicate that there is a relationship between wages and health care costs – a fact which few deny.

Furthermore, I did what Chait presumably did not do, and actually read their paper.  (Frakt may not have read it either; he only quotes the publicly-available abstract. I, sparing no expense for the public good, sprang five bucks for the whole thing.)  The authors used an interesting but unorthodox methodology involving medical malpractice rates  – but their malpractice database excluded dentists, and dental care would be disproportionately affected by the excise tax.

Assuming their methodology is essentially sound, however, it’s worth noting that the authors found their correlation to be especially strong in manufacturing, a shrinking industry.  And they also found that much of wage/benefit shifting occurred with married workers who already had health coverage.  We could debate the merits of this intriguing study endlessly, except for the fact that …

… it’s all moot where the excise tax is concerned.  The excise tax is not designed to find wasteful or excessive spending.  It’s just designed to reduce what health insurance spends.  Heck, you could reduce that to zero tomorrow.  Just outlaw health insurance!  The remaining costs would be covered by people out of their own pockets – to the extent they could afford care.

That’s what the excise tax does:  it shifts cost back to individuals.  So if Baicker and Chandra are right, some employees will work 2.4% more hours.  But that income will be taxed.  Instead of receiving a dollar’s worth of coverage, they’ll work to earn that $1 and only have seventy cents or so left after taxes.  That’s a net loss of roughly 30%, and they’ll have to work more hours to get it.

Such a deal.

And the excise tax will target only a tiny part of the overall cost curve.  We need reform that goes where the big dollars are – areas like overtreatment, chronic disease, and hospital care – not more marginal tinkering at the expense of the American middle class.

I’m glad somebody’s feeling "ardent’ about this tax, since I noted yesterday that most of its other supporters aren’t feeling the earth move.  But with the impact this tax is going to have, it’s worth asking:  What’s love got to do with it?