Elena Tenenbaum kisses her eight-week-old baby Zoe at their home in Providence, Rhode Island. Tenenbaum has been able to use Rhode Island’s paid family leave program, which started in 2014 and covers four weeks of partial pay.

It’s no secret that the United States is the only developed nation that does not have any legal right for employees to take paid leave. Given the urgency of the problem, it’s encouraging to see policymakers on both sides of the aisle on Capitol Hill put forth proposals for paid family leave. But a federal policy that truly helps families across the income spectrum requires careful deliberation and design. That’s why tacking on a paid family leave provision to the tax legislation now before Congress will do little to help the families that need it most.

The purpose of paid family and medical leave is to provide workers with at least some of their regular paycheck when they need to be at home to care for a new child, care for an ill family member, or recover from serious illness. These are times when we need to focus on families’ health and well-being, and when we shouldn’t have additional money worries. When an aging parent takes a fall or when a new child comes into the family, the last thing a family needs is to face the stark choice of caring or earning—the family needs both.

To address this need, a number of policy proposals are being considered on Capitol Hill. One of these proposals is based on the Strong Families Act of 2017, introduced by Sen. Deb Fischer (R-NE), which would provide a tax credit to employers who provide family and medical leave. A version of this proposal is included in the tax bill reported out of the Senate Finance Committee last week. The proposal would give employers a tax credit of up to 25 percent of the wages paid to employees while they are on family or medical leave.

While the intention is good, the evidence we have available shows that this will likely do little to solve the problem. The evidence instead shows that it’s unlikely to increase access to paid leave for workers in any meaningful or widespread way. Rather than incentivizing new firms to change their family and medical leave policies, the tax credit will most likely benefit the firms that already offer this paid benefit—firms that tend to be large, higher-paying, and would offer paid leave even without the tax subsidy.

This is especially troubling because only 6 percent of low-wage workers have access to paid family leave, compared to 13 percent of private-sector workers overall. This means that neither the gap in access to paid leave between high- and low-wage workers nor the gap across firms is likely to close as a result of this policy. Those workers who need it most aren’t likely to benefit.

Furthermore, a tax credit such as the one proposed in the legislation will be a large cost for employers compared to relying on a social insurance program modeled on state programs. In the three states that have implemented paid family leave, the program is paid for by a small tax only on employees. Under the proposed tax credit now before the full Senate, firms must pay 75 percent or more of their employees’ wages to qualify for the benefit. If a firm has many employees that need leave in a given year, for example, then it could be devastating to these businesses’ bottom lines. A social insurance program, however, requires a small cost every year, smoothing the expenses year to year, and unlike temporary disability, there isn’t an insurance market for paid leave to help employers smooth this expense should they want to go that route.

These viewpoints are shared among scholars across the political spectrum. Over the past year, I have participated in a working group led by the American Enterprise Institute and The Brookings Institution on how to develop a paid family leave program. Our conclusion, after looking at the available evidence, is that tax credits are not the way to go. As one conservative working group member put it, “there is no convincing evidence that they would effectively expand access to paid family leave for low-income workers.”

Instead, policymakers should look to a proven model in our own backyard: the states. California, New Jersey, and Rhode Island have successfully enacted their own paid leave programs, with New York state, Washington state, and the District of Columbia set to join them soon. A great deal of research speaks to the success of these state social insurance-based programs, proving that a well-designed national paid leave system can be beneficial to families, businesses, and the economy.

If we are going to enact paid family leave, let’s enact a program that we know can work. A new tax credit is not the right way to enact paid leave.