William Gale, co-director of the Tax Policy Center, has dusted off a five-year-old plan to convert tax deductions for retirement savings into flat-rate refundable credits. Gale’s new proposal, like the one he co-wrote in 2006 with Jonathan Gruber and Peter Orszag, would make 401(k) subsidies less skewed toward high-income households.

In the new paper, Gale proposes a version that would raise tax revenues by $450 billion over 10 years by reducing the proposed government match from a revenue-neutral 30 percent to a revenue-raising 18 percent. These savings, which would come primarily from cutting tax breaks for households in the top income quintile, could stave off deeper cuts to other government programs or fund the president’s proposed jobs bill.

This seems like a no-brainer, and from a pure policy perspective, it is. Current tax breaks are very poorly targeted. For the same dollar contribution to a 401(k), high-income taxpayers in the 35 percent tax bracket get a tax break that’s three-and-a-half times larger than the tax break received by moderate-income taxpayers in the 10 percent bracket. Combined with the fact that high-income households can afford to save more, the result is that an estimated two-thirds of these tax breaks go to households in the top 20 percent of the income distribution. And since most high-income households are already saving, they can simply steer funds into tax-favored accounts, which is why “tax break” rather than “tax incentive” is the more accurate term.

Gale touts his plan as a progressive one, and in a sense it is, since it would reduce the cost of a highly regressive tax break while potentially easing the pressure on essential government programs. But there’s a difference between “less regressive” and “progressive.” Most of the subsidies would still flow to high-income households who can afford to save more and who also tend to get more help from their employers.

An 18% government match will not change the fact that high-fee, high-risk, 401(k)s are not an affordable retirement vehicle for low-income or even middle-income families. And since the funds are still taxed when they’re tapped for retirement, the value of the tax break remains tied to investment returns. (It helps to think of the tax break as a no-interest loan from the government, the value of which depends on how much you make investing the funds). This also tends to favor higher-income households who can afford to take on more risk.

If Gale’s plan isn’t a solution to the retirement crisis facing our country, isn’t it at least an improvement over the current system? Certainly, as long as it’s not oversold. Tweaking a broken system can forestall more far-reaching reforms. (This might partly explain why Orszag, in his new role as a Citigroup vice chairman, is still free to tout the plan). A more imminent concern is the fact that Gale promoted his plan at a recent Senate hearing as a way to help soften the blow of what he characterized as unavoidable Social Security cuts.

It’s also too bad that Gale proposed a lower government match rather than reducing the contribution limit. The 2001 Bush tax cuts increased the contribution limit from $10,500 to $15,000 and introduced a higher “catch-up” contribution for older workers (these limits are now indexed for inflation). Few middle-class households can afford to contribute $10,500, let alone the current limit of $16,500 (or $22,000 for those 50 and older). Lowering these limits but keeping the 30 percent match would be more progressive and have a real incentive effect, as middle-income households might actually increase their retirement saving, as opposed to high-income households simply shifting money around.