A bipartisan group of senators has introduced a new immigration bill (“Border Act”) backed by the White House. The bill is a mixed bag. In general, the main purpose of the bill is to spend tens of billions of taxpayer dollars on detaining and deporting more immigrants from the border rather than finding ways to let them come legally.
These provisions will lead to more chaos and even more demands for more money. But some elements of the deal would improve the system, making legal immigration slightly easier for some groups.
Reckless government spending
The overall legislation—which includes foreign aid to Israel, Ukraine, and Taiwan—will cost $118 billion. This money is designated as emergency supplemental funding, meaning that Congress does not need to find a way to pay for it. It also occurs outside of the normal budgetary process, which means that Congress will completely disregard all tradeoffs between this spending and other alternative uses for the money. The immigration‐related portion is $20.1 billion—$6.5 billion more than the White House requested.
This money will mainly go toward the most well‐financed law enforcement agencies in the United States. The federal government already spends more on immigration and trade enforcement than all other federal law enforcement combined. For context, $20.1 billion is roughly double the FBI’s annual budget. Regardless, as my colleagues Romina Boccia and Dominik Lett write, “rushing funding out the door is more likely to lead to wasteful spending.”
The increases proposed by this bill and the timeframes in place to spend them (usually September 2026) effectively guarantee gross financial mismanagement:
- Customs and Border Protection (CBP) is adding nearly $7 billion to a $18 billion budget, and most of the money is going to the Border Patrol, which has a budget of about $5 billion.
- Immigration and Customs Enforcement (ICE) is receiving $7.6 billion—nearly double its annual budget.
- US Citizenship and Immigration Services will add $4 billion to a $313 million budget for asylum processing.
The bill also plans to fund permanent infrastructure and permanent positions with supplemental appropriations, making it likely that this one‐time supplemental becomes a regular occurrence. Aside from the new expulsion authority, the centerpiece of the bill’s policy changes is to make USCIS handle all asylum processing from the border, costing nearly $4 billion upfront and involving the hiring of 4,338 Asylum Officers.
But USCIS is usually fee‐funded, so this supplemental is effectively like creating a new appropriated subagency. If Congress thinks these appropriations are necessary, Congress should find a way to pay for it with offsets from elsewhere in the budget.
Expulsion authority trigger
The key provision of the bill would allow the president to expel immigrants regardless of the viability of their legal rights to remain in the United States if average daily encounters at the southwest border hit 4,000 in a week (about 120,000 in a month) or 8,500 in any single day. This includes people who apply to enter legally at ports of entry.
Since March 2021, the daily average for a month has always exceeded 4,000. Expulsions would be mandatory if the daily average for a week exceeded 5,000, or 8,500 in a single day, and expulsions would have to continue until the rate declined for two weeks to a daily average of less than 3,000 (a pace of 90,000 for a month) if the 4,000 threshold was used or 3,750 (112,500 for a month) if the 5,000 threshold was used. Under this provision, expulsions would have been mandatory almost continuously since March 2021.
Expulsion authority will not significantly deter crossings. This provision purports to resurrect the Title 42 expulsion authority in use from March 20, 2020 to May 10, 2023, albeit with some important differences as described below. During that time, Title 42 was used primarily against single adults from northern Central America and Mexico. From April 2020 to April 2023, they accounted for 83 percent of all expulsions and were expelled 94 percent of the time with the remaining 6 percent likely criminally prosecuted. As Figure 2 shows, total arrests of this demographic rapidly grew during the Title 42 era before falling again. It would be strange to resurrect such an ineffective policy.
Ending expulsions reduced known “gotaways” (successful evasions of Border Patrol). During the Title 42 expulsion era, the number of detected successful evasions of Border Patrol known as “gotaways” exploded to the highest levels in well over a decade. Gotaways rose from 13,696 in February 2020 to a peak of 73,463 in April 2023, just before Title 42 ended. Since Title 42 was terminated, successful evasions of Border Patrol have declined 79 percent to a daily average of about 500, or 15,500 per month, in January 2024. Evasions increased because Title 42 quickly returned people to Mexico where they could try to cross illegally again and because, with no right to request asylum, they had no reason to turn themselves in. Reinstating expulsion authority would harm border security.
The expulsion authority in this bill is significantly more complicated than Title 42 expulsion authority because the bill provides for the screening by asylum officers of immigrants who express a fear of torture or persecution in the country to which they’re being expelled. Involving asylum officers will make this provision significantly more difficult to enforce. The bill also places a time limit on its use in any given year (270 days in year 1, 225 days in year 2, and 180 days in year 3). This will undoubtedly cause more confusion and surges of individuals attempting to enter the United States. The expulsion authority would thankfully expire after three years.
New asylum screening
Even when expulsions aren’t in effect, the bill finds other ways to try to accelerate removals. Current law says asylum officers should screen out border crossers without a “credible fear” defined as having a “significant possibility” that they could establish eligibility for asylum. Officers also must screen for eligibility for the Convention Against Torture deportation relief, which uses a higher “reasonable fear” standard, which is somewhat higher than the “credible fear” standard. This is the standard that will now be applied for everyone.
This higher screening threshold for release after an asylum interview has effectively already been in place since May 2023 when the Biden administration implemented its so‐called asylum ban rule that creates a presumption of ineligibility for asylum for people crossing illegally. It has led to a modest reduction in the overall “fear found” rate, but still about half of all applicants are approved and are released to pursue their applications.
Internal relocation bar
The bill also requires asylum officers to screen out asylum seekers when they have “reasonable grounds for concluding” that they could have relocated to another part of their home country. This bar already exists, but it has been interpreted to require it to be reasonable in all circumstances, so it’s not clear exactly how this would be applied. However, the bar is not applied at the credible fear stage, but rather at the final merits decision stage, given the difficulties with assessing this ground. The Trump administration attempted to change this, but the Biden administration never implemented it.
Nonetheless, very few border crossers are even subject to fear screenings, so raising the standard and imposing the relocation bar will not matter without a seismic shift in border processing. Changing the screening standard will not lead to significantly more removals. The only way to increase removals is to increase screenings dramatically and then have the resources to expel people whose fear claims are rejected. The bill tries to rectify this with a massive infusion of money, but even these resources will be insufficient.
It is extraordinarily unlikely that USCIS, which already struggles to hire people, will find, hire, and train nearly 4,400 new asylum officers in less than three years. Even if it does, ICE will not have the detention and related resources to detain and remove them.
New asylum process
The bill creates a new process for asylum seekers that would completely remove their processing from the immigration courts and turn it over to USCIS. The government could choose to use this process whenever it wants (p. 116), but it would presumably be used during times when it cannot detain everyone who crosses or for groups such as families who are difficult and expensive to detain. Only people who express a fear of return are subject to this process. Anyone else would be subject to normal detained expedited removal proceedings if resources allow (if they don’t, they’re just released with an immigration court date).
This new non‐detained process allows people who express a fear of return to be released at the border under alternatives to detention electronic monitoring (p. 116) and receive a fear screening interview after 90 days (p. 121). This process will only be used as operational constraints allow (p. 189). The asylum officer may grant asylum in this initial interview if the applicant shows “clear and convincing evidence” of their eligibility. If they are found to have a fear of persecution, then they go onto a merits decision. If they get denied there, they can appeal to a newly created appeals board.
USCIS lacks the capacity to do any of this. Its asylum processing is quite possibly even more broken than the asylum processing in immigration courts. The bill appropriates $4 billion to change this, but it is impossible to imagine that USCIS would be able to do screenings for more than a fraction of arrivals for years. There are currently over one million pending asylum cases with USCIS, and it completes just about 50,000 per year. It has funding for 1,028 officers, and it has only been able to fill 760. The bill increases their wages and streamlines hiring, but it is highly unlikely it will ever reach a staff of 5,366 officers.
Legal immigration changes
- More green cards: The bill will increase legal immigration—green cards—for family‐ and employer‐sponsored immigrants by 50,000 per year for the next five years (250,000 green cards). This is the bill’s best provision. These caps have been barely touched since the Immigration Act of 1990. For context, this is about 14 percent of the annual cap, but it makes up barely half a percent of the 9 million applicant backlog. About 62 percent of the green cards would go to family‐sponsored immigrants. The caps were previously only increased based on unused green cards in earlier years. The figure shows how this increase compares with prior years’ caps.
- Afghan Adjustment Act: This provision allows Afghans who were evacuated from Afghanistan to adjust to legal permanent residence.
- Employment authorization for backlogged workers: H‑1B visa holders—and their spouses and children—who have a pending I‑140 employer‐sponsored green card petition filed on their behalf are authorized to work. This mandates the current spousal H‑4 employment authorization regulation and expands it to include the children of those workers. It is disappointing that this provision only includes H‑1B visa holders and their family members, while similarly situated workers with other visas are excluded.
- Documented dreamer protections: Children of H‑1B visa holders who had at least eight years in the United States in the status of an H‑4 dependent of the H‑1B worker will be deemed to qualify as a child of the worker, if the parent reaches the front of the green card line. They also would be eligible to keep their H‑4 status and receive work authorization. This is a great change but again, it’s incredibly disappointing that the senators decided to exclude children on other visas.
- Visitor visas for family of citizens and green card holders: Visitor visas are usually denied for family members of citizens and green card holders because they could be seeking to immigrate permanently. The bill would include Sen. Rand Paul’s bill that would create a process to issue them visitor visas.
- Mandates processing asylum seekers at legal crossing points: Whenever the expulsion authority is in effect—which can happen at 4,000 encounters per day—the government must process a minimum of at least 1,400 “inadmissible aliens” at ports of entry (p. 219). This is the current administrative limit for the CBP One phone app appointment scheduling process. The bill explicitly recognizes “a process approved by the Secretary to allow for safe and orderly entry into the United States” outside of the visa categories (p. 207). However, the bill does restrict the ability of DHS to grant parole at southwest ports of entry to quite limited categories (p. 188). This would prevent a CBP One entrant without fear from receiving work authorization by virtue of receiving parole. Asylum seekers would be unaffected because the bill requires them to see employment authorization.
- Asylum seeker work authorization: The bill requires that immigrants released from custody under the new asylum process are granted work authorization, which is valid for two years and renewable until the process concludes. They must be granted the work permit if they receive a positive fear determination (at the latest 90 days after entry) or if no determination could be made within 90 days (p. 321). This is a huge change from the current system that requires a six‐month waiting period, an application for asylum, and a separate application for work authorization, which has a filing fee.
It is worth noting that the deal doesn’t touch the highly successful parole sponsorship programs.
The Senate bill has some very positive immigration policy provisions, most notably greater legal immigration. The changes to border policy are either temporary or marginal. The biggest issue is the astounding sums that the bill appropriates outside of the appropriations process to fund fast‐tracked deportations and expulsions. These sums will have to become baked into the appropriations process going forward, meaning that this temporary supplemental will become a permanent, unprecedented increase in immigration‐related spending without offsets.