Asylum Was Built for Persecution, So Why Does Waiting Count as Winning?

Let me describe a deli. It is an ordinary deli with a number machine by the door, and the rule posted on the wall says that once you take a number, you may stay in the shop until your number is called. The shop has a clever policy: the line moves slowly, because the counter is understaffed, and nobody is ever asked to leave while they wait. Pretty soon you notice something odd. Some people come in with a number and stand there for hours because they want a pastrami sandwich. Others come in, take a number, and sit at the table by the window all afternoon because the shop is warm, the chairs are comfortable and nobody is ever going to ask what they ordered. Now put the people who are really hungry in the same line, behind a crowd of the second kind. The hungry are hurt worst by the slow counter. Everyone else is just enjoying the free heat.

That, in a sentence, is the case against a system in which waiting counts as winning. Asylum was designed for people fleeing persecution. When a claim takes years to hear and the person is released into the country while the claim is pending, the delay itself becomes a reason to file, and the people who suffer most are the genuine refugees stuck in the same line. I want to make that argument carefully, because it is a serious one, and because it can be done without turning our backs on anyone fleeing for their life.

The strongest case on the other side

Start where an honest argument should start, with the best version of the opposing view. It goes like this.

One. The 1951 Convention Relating to the Status of Refugees, and its 1967 Protocol, which the United States joined in 1968, bind us to a basic principle. Article 33 of the Convention says that no state shall expel or return a refugee to a place where his life or freedom would be threatened on account of race, religion, nationality, membership in a particular social group or political opinion. That is called non-refoulement, a French word for a plain idea: you cannot send a person back to the people who want to kill him.

Two. The Convention also says, in Article 31, that states should not penalize refugees for illegal entry or presence if they come directly from a place of danger and present themselves to the authorities without delay. The reasoning is that a person running from a regime does not stop at the consulate to collect a visa.

Three. Congress wrote the Refugee Act of 1980 to bring American law into line with these standards. The Supreme Court, in INS v. Cardoza-Fonseca (1987), read the "well-founded fear" standard generously: a person can have a well-founded fear of persecution even when persecution is less than likely to occur. The point was that you do not need to prove the worst will happen. You only need a fear that is reasonable.

Four. The delay is not the claimant's fault. If the immigration courts are backlogged, that is a government resource problem. It is unjust to treat a person worse because the state is slow.

Five. Detaining families and asylum seekers is expensive and harsh, and release while waiting is both cheaper and more humane.

Those five points are not a straw man. Points one through three are correct statements of law. Point four is correct as to fault. Point five has real weight on cost. A conservative who cannot state them fairly has not earned the right to disagree.

What the argument gets wrong

Here is the answer, in numbered premises.

One. Non-refoulement is a rule against sending a person to persecution. It is not a right to enter without permission and wait where you choose. The Convention is satisfied if a person is not returned to danger. It does not say which country must host him, and it does not stop a state from removing him to another safe country. The United States has an agreement with Canada, in force since 2004, under which each country generally sends back to the other the asylum seekers who arrive at their shared land border, on the principle that a person should ask for protection in the first safe country he reaches. The statute itself, in section 208(a)(2)(A) of the Immigration and Nationality Act, allows removal to a safe third country under an agreement where the person would have access to a full and fair procedure.

Two. Article 31 speaks of refugees who come directly and present themselves without delay. It is a principle with conditions attached. A system can honor it while still asking whether a person passed through other safe countries and whether he filed promptly.

Three. The incentive is real, and you do not have to impugn anyone to see it. Take a hypothetical, labelled as one. Suppose a claim takes four years to reach a final hearing. Congress provided in section 208(d)(2) that a work permit cannot be issued until 180 days after the application is filed. So the applicant may lawfully work for the time between 180 days and the decision. Four years is 4 × 12 = 48 months. The 180 days is about 6 months. That leaves 48 − 6 = 42 months of work authorization, three and a half years, before a judge ever rules. Whatever the merits of any claim, a system that produces that result has created a reason to file that has nothing to do with persecution. Some claimants have real fears, and some have both a real fear and an economic motive. The system should be built so that the second motive does not do the work of the first.

Four. The people hurt most by the delay are the true refugees. A woman who has been tortured and whose case sits for years is not helped by humane release. She is helped by a decision. Her family may remain separated. She cannot plan. Meanwhile the line behind her fills with people who have weak claims and strong incentives to wait.

Five. Cost cuts both ways. Detention is expensive. So is a system that adds years of benefits and eligibility to every case that takes years. Faster decisions save more money than any release policy.

The point to take from this is simple. Everyone in the debate says they want protection for the truly persecuted. Only one policy design delivers it: decide quickly.

What the law actually says about speed

This is the most underrated fact in the whole argument. The statute already sets a standard. Section 208(d)(5)(A)(iii) of the INA says that, absent exceptional circumstances, final administrative adjudication of an asylum application, including any appeal, shall be completed within 180 days after the application is filed. The people who wrote the law in 1996 expected claims to be heard in about six months. The gap between that sentence and today's reality is the story. Check the Executive Office for Immigration Review's own published statistics on pending cases and processing times, which the Justice Department posts, before you quote any number to a neighbor.

The same 1996 law described in The 1996 Immigration Law Behind Most of Today's Border Fights also built the front door to the process. Here is how it works in order.

1. A person arrives or is found without valid documents and is placed in expedited removal. Under INA section 235(b)(1), an officer may order removal without a hearing.
2. If the person says he fears return, he gets a credible-fear interview. The standard in the statute is a "significant possibility," taking into account the credibility of his statements and other facts, that he could establish eligibility for asylum. It is a screening threshold, deliberately lower than the standard for winning.
3. If he passes, he moves on to full proceedings. Typically he is served a Notice to Appear under section 239, and the case goes before an immigration judge, where he can apply for asylum and related protection. The next steps are described in Inside a Removal Case From Notice to Appear to the Plane Ride.
4. While waiting, he is detained or released. The statute calls for detention of arriving people pending the credible-fear decision. But detention space is limited, and a settlement from 1997, the Flores agreement, as interpreted by a federal court, restricts how long children may be held, which is generally summarized as about twenty days. The practical result in a system short on beds is that many families are released into the country to await a hearing. Look up the current text of the settlement and the court's orders for the details.

None of this requires a villain. A system with a low screening threshold, limited detention capacity and a thin court docket will produce delay, and delay will produce incentives. That is arithmetic, not malice.

Reforms that speed decisions without gutting protection

Here are the proposals that fit the argument. They are mostly about capacity and procedure, not about shutting the door.

• Staff the front end. More asylum officers and immigration judges, with a set goal of decisions in months. The 180-day sentence is already in the law; fund it.
• Hear recent arrivals first. Put the newest cases at the front of the docket so a decision follows quickly, rather than letting the freshest claims join the back of a multi-year line.
• Enforce the one-year filing deadline. Congress wrote it in section 208(a)(2)(B). The exceptions for changed or extraordinary circumstances protect people who truly could not file in time.
• Use safe-third-country agreements where the partner country provides a full and fair procedure, and where it is willing, so that people who passed through a safe country can be sent there to ask for protection.
• Match detention to capacity, or build alternatives that work. If a person is going to be released, the system should have reliable ways to make sure he shows up. Missing hearings is where removal orders in absentia come from, and neither the claimant nor the public is served by that.
• Keep the standard for winning intact. The statute's definition of persecution on one of five grounds is worth defending. The goal is speed, not a narrower definition.

I would add a note on tone. Conservatives lose the argument when they sound as if every claimant is a fraud. The right line is: most claimants deserve a prompt answer, whichever way it goes, and a system that takes years is unfair to everyone, including the taxpayers and the neighbors.

The part that should make everyone uncomfortable

Here is a fact that the left and right both avoid. Under the law as written, the credible-fear screen is intentionally generous, and the full asylum standard is intentionally demanding. Many people pass the first and fail the second. Both parts were designed by Congress. The first part says: do not send away anyone who might have a real claim without a proper look. The second says: grant protection only to those who show persecution on a protected ground. When the distance between the two takes years to cross, a person who will eventually be turned down has already spent the intervening years here. That is not a reason to lower the first bar. It is a reason to shorten the distance.

Whoever tells you the system is "broken" only because of cruelty, or only because of loopholes, is selling half a story. The system is slow, and slowness is the problem. If you want a perspective on the border side of the equation, Towers, Sensors and Scanners, The Technology Half of Border Security covers the technology at the line, and Asylum, Refugee, Parole and TPS, Four Statuses People Mix Up explains how asylum differs from the other statuses people mix up.

The strongest objection, one more time

A critic will say: "Fine, but speeding up hearings sounds like a way of denying them faster. If you hurry, you will make mistakes, and the mistakes will send people back to danger."

Here is the honest reply. Speed and care are not opposites when the system is staffed to do both. A hearing that happens in six months with a proper record is better than one that happens in six years with a faded memory and a lost file. The risk of error is real, which is why the right design includes legal review, the right to an attorney at the claimant's own expense, appeals to the Board of Immigration Appeals and the federal courts of appeals, and a screening standard that errs on the side of a second look. A conservative who believes in due process should want all of that. What he should not accept is a system that treats delay as the safeguard.

The bottom line

• Asylum exists for people fleeing persecution on one of five grounds. The 1951 Convention and 1967 Protocol bind us not to return them to danger, and conservatives should say so plainly.
• Non-refoulement does not require years of release while a case waits. Section 208 already calls for a decision within 180 days and allows safe-third-country removal.
• The incentive is simple arithmetic: a four-year wait with a work permit after 180 days means about 42 months of authorized work before a decision. Fix the delay and the incentive disappears.
• Fund asylum officers and judges, hear new cases first, enforce the one-year deadline and keep the standard for winning where Congress put it.
• When you argue this, argue for the real refugee. He is the one waiting longest in a line that is too long for everyone.