When is a policy not a policy? When it’s the policy you disagree with.

The AEI-Brookings Working Group on Paid Family Leave has published its report, “Paid Family and Medical Leave: An Issue Whose Time Has Come.” HT2 Timothy Taylor.

I’m disappointed. It’s not just that I disagree with its policy conclusion, especially in the context of large and growing budget deficits. More on that anon. It’s also the report’s misuse of language. At multiple times throughout the report, authored by Aparna Mathur and Isabel V. Sawhill, the authors state that the United States has no policy on paid parental leave. Since they note that some states do have such a policy, their point is that there is no federal policy. In their introduction, for example, Mathur and Sawhill write, “We all believe the United States needs a paid parental leave policy.” But the United States has a paid parental leave policy: it’s the absence of forced paid parental leave.

Imagine that you and I are discussing what to do today. You strongly want to go to the zoo. I strongly want not to. You say you have a policy of going to the zoo. That makes sense. But does that mean I don’t have a policy on going to the zoo? Not at all. My policy is not to go to the zoo.

Now to the merits of the issue. The authors of the study, based on inputs from the working group, which, by the way, includes Republican economist and former McCain advisor Doug Holtz-Eakin, argue that parents would gain from a policy in which the federal government pays for them to take time off with pay when a child is born. That’s true. Normally, though, when economists make that kind of point and argue for it, they do some kind of cost-benefit analysis. In this report, though, such analysis is absent.

Not that there’s no analysis. There is. To their credit, the authors and the other members of the group see some big problems with mandating that employers provide paid parental leave. They write:

We also discussed one other approach: an employer mandate. This approach is popular with the general public. However, we do not favor it for two reasons. First, it would be burdensome on employers, especially small businesses and those employing a disproportionately high share of likely parents. Second, it will likely lead to a reluctance to hire female workers of a certain age.

Later in the piece, they expand on this point slightly:

However, an employer mandate for paid family leave would invite damaging unintended consequences. A mandate is not free; it would directly hurt firms’ profits, especially small businesses less able to absorb the new cost. Firms would respond to the increase in expected labor costs with some combination of raising prices, cutting other compensation, and reducing employment. Furthermore, a mandate would create a strong incentive for hiring and pay discrimination against those most likely to need or take paid leave.

I’m not sure why they think small businesses are less able to absorb the new cost. But their point about other compensation is on target. I was disappointed that they didn’t cite one of Larry Summers’s best articles. Here’s what I wrote on the issue in “Paid Parental Leave Is Not a Free Lunch,” last summer:

Think of how employers would react to such a mandate. They would realize that the main people who would take advantage of paid parental leave would be women of child-bearing age. This makes those women less valuable to them as employees. So their demand–the amount they are willing to pay–for women in that category would fall. Women in that category, on the other hand, would be willing to work for less because the benefit is valuable. In economists’ jargon, in short, both the demand curve and the supply curve would fall. The wages of those women, therefore, would fall.

Benefits paid for by recipients

This analysis of mandated benefits is widely accepted among economists. One economist who laid it out clearly, in a 1989 article in the prestigious American Economic Review, was Lawrence Summers. Summers was more recently the Secretary of the Treasury under former president Bill Clinton and later director of President Obama’s National Economic Council.

So what policy do they want? A majority of the working group want a new payroll tax. They never say how high, but in the discussion of one such proposal, they seem to imply that the tax would have to be at least 0.4 percent of payroll. Some of the group want to fund the benefit by cutting other government spending or getting rid of some tax preferences.

Consider first the payroll tax. That’s profoundly unfair. Everyone would have to pay the payroll tax, including people with no children who will never get to use the benefit. I think I read the report pretty carefully, and I couldn’t find even a mention of this fact.

As for cutting other government spending, that’s preferable, but, given the tsunami of spending that will happen in the next few decades with Social Security, Medicare, and Medicaid (SSMM) is it really good policy to cut government spending in order to fund a new program? Wouldn’t it be better to hold the line on new programs so that government spending on some old programs can be cut when the tsunami hits?

The same goes for tax preferences. There are a lot of tax preferences in the tax code I would like to get rid of. But in the next few decades there will be pressure to get rid of them when the SSMM tsunami hits.

Indeed, along these lines, think back to the payroll tax increase that the majority proposes. When the SSMM tsunami hits, there will be strong pressures to increase the payroll tax. But that cupboard would be slightly more bare if the payroll tax is increased now.

When economists advocate new government programs, they typically at least try to point out some market failure. These economists don’t. To be sure, they argue that there are benefits. But there are benefits from a company policy of having a coffee machine. There are also costs. The employers and employees are the only players: they can jointly decide whether it’s worth having a coffee machine. If the gain to employees from such a machine outweighs the cost to employers, the employers will provide it and pay slightly less in other compensation, making both employer and employee better off. If the gains to employees from paid parental leave exceed the costs to employers, then employers will provide such leave, as some employers do. In pointing out that mandated benefits have the problems they do, the working group is actually making my implicit point in the above-referenced article: the benefits are less than the costs.

Postscript: Isabel V. Sawhill wrote an excellent item, “Poverty in America,” for my Concise Encyclopedia of Economics.