Via Mark Thoma, a good counter-intuitive point by Paul Krugman about the American welfare state:

We like to think of ourselves as rugged individualists, not like those coddled Europeans with their oversized welfare states. But as Jacob Hacker of Yale points out in his book "The Divided Welfare State," if you add in corporate spending on health care and pensions - spending that is both regulated by the government and subsidized by tax breaks - we actually have a welfare state that's about as large relative to our economy as those of other advanced countries.

The resulting system is imperfect: those who don't work for companies with good benefits are, in effect, second-class citizens.

Since I have Hacker's book on my desk, here are the numbers: In 1995 Germany and Sweden spent about 27 percent of GDP on after-tax social welfare expenditures—and two percent of that was "private" spending (i.e., by employers). The United States, meanwhile, spent 25 percent of GDP—and 8 percent of that was private.

Granted, the American "divided welfare state" is comparable to or better than Swedish-style socialism for those workers who happen to have stable jobs with good pensions, 401(k)'s, corporate day care, and gold-plated health care benefits. But for everyone else, it's inequitable, regressive, and a source of uncertainty for those increasingly at risk of losing their jobs. (Indeed, it's becoming a worse deal even for many workers with stable jobs, as they face greater cost-sharing for health care or as companies default on their pension funds, or what not.) In the old days, though, businesses loved it, and some of them even backed the creation of entitlements like Medicaid and Medicare as ways of reinforcing the status quo.

But now many companies look ready to shift more of the welfare burden onto the government, and move towards a European-style welfare state, although Daniel Gross has noted that there are two types of companies here—those that, like GM, want the government to pick up its health care and pension costs, and those, especially newer, high-tech companies, who still want tiny government. Meanwhile, 60 percent of workers still get their health benefits through their company (although that number's declining), so it's not clear how many voters actually want to shift to a European-style welfare state, at least right now. Change won't be easy, although the opportunity is certainly there. Also, those companies that no longer want to be on the hook for, say, health care costs aren't necessarily going to push for single-payer, or France-style health care. They could just as easily be convinced to agree to the Bush administration's horrible HSA proposal, which would shove people onto the open insurance market. That, I think, is going to be a major fight.

One other miscellaneous point: It's worth noting that most of the legislation that expanded the employer-centered "private" welfare state was passed with very little public debate. The Revenue Act of 1978, for instance, slipped in an obscure provision to create 401(k) plans, which was estimated to have a "negligible effect on budget receipts" by the CBO—today it costs the government over $100 billion a year, for a program that mostly benefits the well-off. It would be naïve to pretend that public opinion has an all-powerful effect on public legislation, but even by those standards there's been a breathtakingly small amount of oversight on the expansion of the private welfare state.