The Senate immigration bill faces opposition from conservatives and liberals alike, but critics on both sides are missing a crucial problem with the legislation. The bill fails to address a fundamental flaw in our current immigration system: the arbitrary and unfair manner in which it restricts the number of green cards issued each year.

In addition to placing an annual limit on the number of green cards, the current system imposes a uniform per-country cap. Most countries never come close to reaching their limit, which is 25,620–or 7 percent of all family and employer-sponsored visas. But four countries persistently max out their caps: Mexico, India, the Philippines and China. For these countries the visa backlog for some family categories is twenty or more years–an impossible wait. For employer-sponsored visas the backlog for skilled workers is five or more years.

The cap on annual admissions from these countries is a major cause of illegal immigration and the deficit of skilled immigrant labor. The Senate proposal skirts the problem but cynically addresses its effects by increasing the number of green cards only for those with education and wealth while channeling lower-skilled immigrants into a guest-worker program.

Congress repealed the noxious national-origin quota system, in place since the 1920s, which blatantly discriminated against immigrants from Eastern and Southern Europe and Asia on grounds that they were racially undesirable. Individual attributes–family relations and job skills–were deemed more important in selecting immigrants than national origin or race. The per-country cap was established to keep the immigration stream diverse, to keep the system from being monopolized by a few countries. The logic appealed to Americans’ sense of fairness, that we should treat each country the same.

But this sense of formal equality obscures the fact that not all countries are the same: Some are larger and some smaller, some are more prosperous than others, some have special historical relationships with the United States that others do not. The one-size-fits-all approach doesn’t make sense in the real world. Why should Belgium and New Zealand, say, have the same number of visas as Mexico and India, if the former are underused and the latter are oversubscribed?

In the early 1960s Michigan’s Democratic Senator Philip Hart proposed an innovative bill that did not treat immigration as a zero-sum game but took into account both America’s needs and the needs of sending countries. Hart’s initial vision allocated 20 percent of all visas to refugees; 32 percent to countries in proportion to their size of population; and 48 percent to countries in proportion to their immigration to the United States during the previous fifteen years. It set a minimum and maximum for all countries and called for an update every five years. It was a complicated formula that eschewed a unilateralist approach in favor of one that balanced human rights, the emigration needs of sending countries, and Americans’ historical and familial ties abroad. The bill had bipartisan support but was bypassed by another bill, supported by the Kennedy Administration. That bill was ultimately signed into law as the Hart-Celler Act by President Johnson in 1965.

Hart’s ideas could be updated to solve the problems caused by the per-country limits in ways that don’t denigrate family ties or create an underclass of temporary guest workers. Since 1980 we’ve had a separate law dealing with refugee admissions. But we could allocate green cards to countries based on the relative size of their population and emigration demand; their ties to American citizens and institutions; and their supply of low- and high-skilled labor that we need. In other words, if we acknowledge that migration is driven by supply and demand and take into account the needs of the United States and other countries, we might have a system that is more realistic and fair.

We might also return to a regionalist approach. Few people know that before 1965 there were no numerical restrictions on immigration from countries of the Western Hemisphere, in keeping with the tradition of Pan-Americanism. When we imposed quotas on Mexico and the rest of the Americas after 1965, we got illegal immigration. Today the free-trade agreements in the Americas (NAFTA and CAFTA) ease the movement of goods and investments–especially those of multinational corporations–but not people. The European Union, by contrast, allows citizens of member states free movement to other member states. If we combined measures to strengthen the economies of our hemispheric neighbors with a more sensible immigration policy, we would go a long way toward solving the illegal immigration problem.

Finally, rather than rehash the agonizing debate over a mass-legalization program every twenty years, we might consider restoring statutes of limitations on prosecuting unauthorized presence. We used to have such a policy (one to five years) before Congress eliminated it in the 1920s. The underlying logic was much the same as the arguments for legalization today: Immigrants who work hard and sink roots in their communities effectively become part of our society and should not have to remain in the shadows forever.

We would be wise to rethink the unilateralist premises of US immigration policy. It makes much more sense for our policies to reflect how migration is driven by supply and demand in a world in which population and wealth are distributed unevenly. History shows that policies designed to categorically exclude–from the Chinese exclusion acts of the nineteenth century to the imposition of unrealistically low quotas on Mexico in the twentieth–are doomed to failure.