HSAs and the privatization of pay for health care

At one point I was brooding over the idea that the privatization of pay for health care could work for Canadians (see comments section).

Well, I'm decisive now: such a scheme is not in the best interest of Canadians. Since my last post on this subject I've dropped some hints on how I feel about the application of a free market system on health care services in general.

My research since then is consistently showing me the same result: Canada's labour force and demography is not well-suited for the kind of privatized health care schemes that have been most widely proposed. I've barely scratched the surface of this subject, but perhaps I'll post some of my preliminary research here at some point. That said, nothing should be expected from me too soon -- I'm a nit-picker.

Today I just want to comment on Health Savings Accounts (HSAs). The GAO has recently released a paper which states that HSAs act as tax shelters for the rich (h/t Mark Thoma). I have two comments on this and on HSAs in general.

First, the GAO report isn't too much of a surprise to most people, but it highlights another piece of the wage/productivity puzzle posed by Prof Stephen Gordon (which he later re-visited, as noted on the linked post). He has said that wages are growing at the same rate as worker productivity in Canada, while U.S. wages are growing at a lesser rate than U.S. productivity. HSAs may offer yet another answer for this discrepency between the two countries. Since the U.S. introduced HSA's in 2004, this optional benefit has represented yet another reason for U.S. wages to face downward pressure as deductions on paycheques increase for high-income earners especially.

My second point is closely related and has to do with the distortion of data used to illustrate the rich-poor gap. Consider employer-provided benefit packages held by a range of income earners in the U.S.A. To the highest earners, health fees are an insignificant proportion of their income. They will spend their money on whichever services are best. Mid-to-high income earners are more likely than low income earners to opt for HSAs and accept the greatest possible coverage offered by their employer, most especially if they see this to be beneficial to their individual needs. Meanwhile, studies have shown that low-income earners are less likely to accept the same wide coverage from their employer (although if they work for the same firm as upper-income individuals, they may not have a choice under employer benefit nondiscrimination rules), or they may decline altogether. And, of course, they are relatively less likely to opt for HSA's.

Given this, what do we see? If we forget about the highest income-earners, we should be seeing a decrease in wage differentials once benefits are deducted. Yet studies have shown that non-benefit wage differentials within this group of the population are not decreasing in the U.S., they are increasing. Imagine that employer provided benefits and HSAs were accounted for. What do we see now? I see a magnification of wage differentials. Have I got this wrong?

Economist Marc Lee recently stated on the REP blog:

It is fascinating to me that in the wake of the Chaoulli decision by the Supreme Court private options are becoming more commonplace in Canada, just as more and more sensible people in the US are calling for a Canadian-style universal public insurance model.

I imagine that I'll be nodding my head in agreement more vigorously as I delve deeper into this subject.