DHS Proposes a $103,265 Fee on Every H-1B Cap Petition

Published August 24, 2026 | By Parikh & Prasad, PC

DHS wants to charge $103,265 for every H-1B cap-subject petition — due at filing, on top of every fee you already pay. The proposed rule went on public inspection August 24, 2026 and publishes in the Federal Register August 25, 2026. Nothing changes today; this is a proposal. But DHS puts the price tag on employers at roughly $8.8 billion a year, and you get 30 days to say something about it. If your company files H-1B cap cases, read this one.

What DHS is proposing

The rule is called Fee for Certain H-1B Petitions (RIN 1615-AD20, CIS No. 2861-26, DHS Docket No. USCIS-2026-0298). It adds one paragraph to the USCIS fee schedule, at 8 CFR 106.2(a)(3)(xii): an additional fee of $103,265 for all H-1B cap-subject petitions, including those eligible for the advanced degree exemption.

Two words in there do most of the work.

The first is additional. This does not fold into the existing H-1B petition fee at 8 CFR 106.2(a)(3)(i). DHS kept it as its own line item so the money can be tracked, allocated, and reported separately. You would pay it on top of the Form I-129 fee, the Asylum Program Fee, the fraud prevention and detection fee, and everything else on the USCIS Form G-1055 fee schedule.

The second is filing. The fee is due when the petition goes in, not when it gets approved. Denied, rejected, withdrawn — you paid it either way.

Which petitions would be covered, and which would not

All H-1B cap-subject petitions. That means the regular 65,000 numerical limitation and the advanced degree exemption at INA section 214(g)(5)(C), 8 U.S.C. 1184(g)(5)(C) — the allocation for beneficiaries holding a qualifying U.S. master's degree or higher.

Employers often assume the master's cap gets gentler treatment. Not here.

Cap-exempt petitions under INA sections 214(g)(5) and (7), 8 U.S.C. 1184(g)(5) and (7), fall outside the proposal entirely. DHS explains the reasoning: a lot of cap-exempt H-1B workers sit at institutions of higher education and at nonprofit and governmental research organizations, and exempting those employers lines up with how DHS already treats them under the Asylum Program Fee at 8 CFR 106.1(f)(2) and 106.2(c)(13)(i).

Where the $103,265 figure comes from

It is not what an H-1B petition costs USCIS to adjudicate. It is long division.

DHS added up roughly $8.8 billion in costs it wants recovered across six federal agencies, divided that by the number of cap petitions it expects to receive each year, and set the fee at the result. The figure measures nothing about what your petition costs the government to process. It is a revenue target with a petition attached.

To its credit, DHS does not pretend this is cost recovery in the ordinary sense. The preamble concedes that "H-1B cap-subject petitioners may object to paying a fee that recovers costs far beyond the direct adjudication cost of their individual petition." Its answer is that the fee schedule has always absorbed the cost of services provided free or below cost, and that ability to pay has long been a factor in setting immigration fees. DHS looked at spreading these costs across all benefit requests and passed, concluding that H-1B cap-subject petitioners are "most willing and able to pay an additional fee".

What the money would pay for

Here is what separates this from a routine fee adjustment: the revenue does not stay at USCIS.

DHS proposes splitting it across six agencies. USCIS would keep roughly a third. Another third would go to the immigration courts at the Justice Department. The remainder is divided among the Department of Labor, ICE, the State Department, and Customs and Border Protection.

Read that again. About two-thirds of the money raised from H-1B employers would fund agencies that do not adjudicate H-1B petitions — immigration courts, consular operations, labor enforcement, and immigration enforcement.

DHS and USCIS would set all that up through reimbursable agreements under INA section 286(n), 8 U.S.C. 1356(n). How reimbursement gets requested, on what timeline, with what documentation, in what amounts, subject to what limits — all of it would be worked out before a final rule takes effect. None of it is in the proposal.

DHS is upfront that this is new ground. DOS and DOL have never independently used the section 286(m) authority to set fees covering their own adjudication costs. And before the $600 Asylum Program Fee DHS created in 2024, the agency "had never directly transferred the costs of one program to another."

The legal authority DHS relies on

Mainly INA sections 286(j), (m), and (n), 8 U.S.C. 1356, which let the Secretary set fees for adjudication and naturalization services "at a level that will ensure recovery of the full costs of providing all such services." DHS also cites INA section 103(a), 8 U.S.C. 1103(a); INA sections 214(a)(1) and 214(c)(1), 8 U.S.C. 1184(a)(1) and (c)(1); and sections 102, 402, and 428 of the Homeland Security Act, 6 U.S.C. 112, 202, and 236.

Whether that authority reaches far enough to fund immigration court proceedings, consular processing, labor standards enforcement, and port-of-entry activity out of a single nonimmigrant petition fee — that is the question the comment period exists to test.

How this interacts with the $100,000 proclamation payment

It does not replace the $100,000 payment required by Presidential Proclamation 10973, 90 FR 46027 (September 19, 2025). DHS says the proposed fee is separate, rests on different legal authority, and that a petitioner caught by both "would be required to pay both amounts."

The real picture is messier. The proclamation payment applied to certain H-1B petitions filed on or after September 21, 2025 and before September 21, 2026. On June 8, 2026, the U.S. District Court for the District of Massachusetts vacated the agency guidance implementing it in California et al. v. Mullin, No. 25-13829 (D. Mass.). The government appealed to the First Circuit on June 11, 2026, and DHS says that appeal is still pending. If the order comes off, DHS says it would collect under the proclamation's terms — and it notes in a footnote that the proclamation, unless extended, expires before the fee proposed here could ever take effect.

This is not the only fee change aimed at H-1B employers

The proposal catalogs the rest, and the stack matters more than any single piece.

The CBP 9-11 Response and Biometric Entry-Exit Fee Final Rule, 91 FR 51360 (August 10, 2026), is already final. It requires covered employers to pay the statutory 9-11 Biometric Fee on all H-1B and L-1 extension-of-stay petitions — no matter whether a fraud fee applies, no matter whether there is a change of employer. "Covered employers" means more than 50 employees in the United States with more than 50 percent of them in H-1B or L-1 status. The statutory fee sunsets September 30, 2027 unless Congress extends it.

The EOIR Fees interim final rule, 91 FR 35369 (June 11, 2026), brings new and increased filing fees before the immigration courts, and DOL's Improving Wage Protections proposed rule, 91 FR 15454 (March 27, 2026), is still pending. DHS says it found no conflict or duplication across these rules, and invites comment on that point.

What DHS expects this to do to filing volumes

Fewer filings, by its own math. DHS assumes annual cap-subject volume will settle at 85,000, below recent years, and admits USCIS "may see a reduction in the number of H-1B cap registrations and some employers, including small entities, may file fewer petitions as a result of this proposed rule."

It still expects demand to hold, largely because the program is heavily oversubscribed today and because DHS reads employer demand for H-1B workers as relatively insensitive to filing costs. The rule also softens the number by stretching it across time, measuring the one-time fee against everything an H-1B worker earns over a full six-year run rather than against a single year's salary. Same fee, very different impression, and DHS chose the longer frame deliberately.

The 85,000 assumption is the weakest part of the rule, and it is worth understanding why, because everything else rests on it.

85,000 is a ceiling, not a floor. The statutory cap limits how many workers may be granted cap-subject H-1B status in a fiscal year. It has never guaranteed that 85,000 employers will file at any given price. DHS used the statutory maximum as the denominator in its fee calculation, which quietly converts a legal limit on approvals into a forecast of demand.

DHS's own attrition math points the other way. The rule states that USCIS "typically selects more than 85,000 beneficiaries/registrations in the lottery, and receives more than 85,000 H-1B cap-subject petitions, to account for denied, rejected or withdrawn petitions." Historically, in other words, some selected employers never file. Yet DHS projects filings under the fee at exactly 85,000. That requires attrition to fall to roughly zero at the precise moment a $103,265 charge makes walking away most attractive.

The timing makes it worse. The fee is due at filing, and filing happens after selection — a petitioner gets a window of at least 90 days to file once a registration is selected, under 8 CFR 214.2(h)(8)(iii)(D)(3). So the real decision point is not the cheap registration in March. It is the moment a selected employer has to move six figures out the door, with three months to reconsider.

And the denominator lands suspiciously well. DHS's projected volume comes to rest on exactly the statutory cap — which is exactly the figure that makes the agency's cost total divide out to the fee DHS proposed. A genuine forecast produces a range, not a point estimate that happens to equal the number the arithmetic requires.

DHS does not really dispute the fragility. It concedes the rule generates the necessary revenue only "provided filing volumes do not fall short of those projected herein," and says plainly that "the success of the USCIS fee model and this rulemaking in generating the necessary revenue depends on the filing volumes of cap-subject H-1B petitions not falling short of those projected herein." The entire rule is conditioned on an assumption DHS has not tested.

Which creates a structural risk in the design itself. The fee is cost divided by volume. If volume comes in below 85,000, the fee under-recovers, and DHS's own methodology would point toward raising it — which would suppress volume further.

Will employers actually pay it?

Set the modeling aside and ask the practical question: when a selected employer has ninety days to decide whether to write this check, what actually happens?

For a small employer, in most cases nothing happens. The petition does not get filed. We covered why below, but the short version is that a fee due at filing, non-refundable, and owed whether the petition is approved or denied is not something a company of that size can plan around. It is not a cost that discourages participation at the margin. It ends it.

The more interesting question is the one DHS seems not to have asked: whether large employers will pay it either.

DHS's reasoning treats big filers as the group that can obviously absorb the fee. That confuses having the money with being willing to spend it, and it overlooks several things that make large employers more likely to balk, not less.

Volume multiplies it. The fee is charged per petition, every year. What looks survivable on a single hire becomes a permanent, recurring, seven-figure line item for any company that staffs through the cap in real numbers. Nobody approves that quietly. It goes to finance, and then it goes to the executive team, and the question stops being "can we afford this candidate" and becomes "should we be running a cap program at all."

It buys a chance, not a hire. The money is due at filing. The petition can still be denied. The beneficiary can withdraw, take another offer, or leave within the year. Large employers are not more tolerant of paying six figures for an outcome they do not control — they are usually less tolerant, because their finance functions are more disciplined about exactly that kind of spend.

The alternatives get easier as companies get larger, not harder. This is the part DHS's logic misses completely. A large company is precisely the one with foreign subsidiaries, existing offshore engineering or research operations, and the legal budget to use categories this fee does not touch. When the cost of a cap petition crosses six figures, the multinational is the employer best positioned to put the role in another country instead — and to move the person later through a route the rule never reaches. Ability to pay and ability to avoid paying run together.

And the fee arrives on top of everything else. It is not being proposed into a quiet environment. It stacks on the existing filing fees, the fraud fee, the Asylum Program Fee, the expanded biometric fee on H-1B and L-1 extensions, and whatever the proclamation payment litigation ultimately produces. Employers evaluate the total, not the increment.

DHS's model assumes this fee filters out the hires an employer does not genuinely need. That is not what a large, non-refundable, upfront charge filters on. It filters on whether a company is willing to commit cash to an uncertain outcome when a cheaper option exists — and for the largest employers, a cheaper option usually does. The result is not that the program keeps its 85,000 filings and sheds only the marginal ones. It is that filings fall at both ends: small employers because they cannot pay, and large employers because they have somewhere better to put the work.

What DHS says the benefit is — and the question it never asks

Every significant rule is supposed to weigh costs against benefits. DHS quantified the cost side to the dollar. The benefit side is thinner than you might expect.

The Regulatory Impact Analysis section titled "Benefits of Proposed Rule" runs about six lines and quantifies nothing. It says the rule "would better align fees with costs of administering the lawful immigration system across multiple agencies" — which restates the rule's purpose rather than identifying a benefit to the public.

The substantive claim sits in the preamble instead, and it is a single sentence:

DHS believes that U.S. employers, if required to pay an additional $103,265 fee when filing an H-1B cap-subject petition, would be less likely to hire an H-1B worker over a qualified and highly-skilled American worker unless the need is legitimate and they have no alternative for obtaining the specialized skills of the employee.

DHS adds that because demand exceeds the cap, the fee "could also have the indirect benefit of better protecting the wages and job opportunities of U.S. workers, as the H-1B program is intended to do."

That assumption — deterred H-1B hire becomes American hire — is the whole public-interest case for the rule. And it leaves out an option employers actually have.

An employer priced out of an H-1B petition has a third choice beyond hiring domestically or going without: place the role outside the United States. The team gets built abroad, and the work goes with it. Nothing in this rule reaches an engineer employed in another country.

We reviewed the full 67-page proposal. It does not address that possibility anywhere — no discussion of relocation, offshoring, or work moving abroad in response to the fee. For a rule whose sole claimed benefit is protecting U.S. workers, the alternative that would defeat that benefit goes unexamined.

That gap has legal significance beyond the policy argument. An agency must consider the important aspects of the problem in front of it; a rule that does not can be challenged as arbitrary and capricious under the standard set in Motor Vehicle Manufacturers Association v. State Farm Mutual Automobile Insurance Co., 463 U.S. 29 (1983). DHS priced the burden precisely and left the benefit as an untested assumption. Employers who believe the realistic response to this fee is relocation rather than domestic hiring should say exactly that, with specifics, in the comment docket — that is how the point gets preserved.

The categories this fee does not touch

The proposal taxes one door. It is worth seeing which doors stay open, because the contrast shapes how employers will respond.

The fee reaches cap-subject H-1B petitions and nothing else. Cap-exempt H-1B filings under INA sections 214(g)(5) and (7) are excluded outright. L-1 intracompany transfers, O-1 classification for individuals of extraordinary ability, and the other employment-based nonimmigrant categories are not covered by this proposal. Neither, obviously, is employing someone abroad.

The L-1 interaction deserves particular attention. L-1 generally requires a qualifying corporate relationship and a period of qualifying employment with the foreign entity before transfer. So an employer that responds to this fee by standing up an engineering or research operation overseas is not only avoiding the fee today — it is building the corporate structure and the employment history that make later transfers possible through a category this rule does not reach. A fee designed to keep jobs in the United States creates an incentive gradient pointing the other way.

None of that makes an alternative category the right answer for any particular employer. Eligibility for L-1, O-1, or cap-exempt status turns on specific facts about the company, the role, and the individual, and getting that analysis wrong is expensive in its own way. But employers who have never seriously examined those options are likely to start.

What it would mean for smaller employers

There is no small-employer exemption in this proposal, and no discount either. The fee applies, in DHS's words, "irrespective of their size and nonprofit status."

The Regulatory Flexibility Act made DHS run the numbers on small entities, and the numbers are not subtle.

About half the employers who filed cap-subject H-1B petitions in FY 2025 were small businesses. DHS estimates the fee would exceed 1 percent of annual revenue for roughly three-quarters of them — more than 11,000 companies. It formally concluded the rule would have a significant economic impact on a substantial number of small entities, and proposed it anyway.

Set that against the whole filing population and the scale becomes clearer: by DHS's own count, the group it flagged as significantly harmed is close to two of every five employers who use the program.

Why this would push small employers out of the program entirely

A fee of this size does not trim small-employer participation at the margin. In our view it forecloses it, and the reasons are structural rather than sentimental.

The money leaves before anything is decided. The fee is due at filing, not on approval. An employer whose petition is denied, rejected, or withdrawn has still paid $103,265. A large filer can absorb that across a portfolio of petitions and treat it as a cost of doing business. An employer filing one or two cannot diversify that risk at all — every filing is a concentrated bet with no refund attached.

Above 1 percent of revenue, this stops being a line item. For the employers DHS identified, the fee is not a hiring expense in any ordinary sense. It is a capital decision, sitting alongside a facility lease or a senior hire, and it has to clear against every other use of that money. Companies at that scale do not have a budget line this fits into.

DHS's own reframing works far less well here. The rule softens the number by measuring it against everything the worker will earn over a full six-year H-1B run. That framing assumes the employer keeps the worker six years and can carry the cost against future payroll. A smaller company has neither the retention certainty nor the balance sheet to smooth a six-figure outlay across six years. It pays in full on day one, from cash it has now, for a worker it may or may not still employ six years from now.

And the alternatives are not evenly available. Cap-exempt status turns on institutional affiliation, not size. L-1 requires a qualifying foreign entity and prior employment abroad. Both routes favor organizations that already have scale and structure. A ten-person company has the fewest ways around this fee and the least capacity to absorb it.

This is also where the small-employer question stops being a fairness argument and becomes an arithmetic one. DHS needs 85,000 petitions filed to raise the revenue the fee is calculated on. It simultaneously found that close to two in five filers face a significant economic impact. Those two findings sit in the same document and pull against each other: the population DHS identified as most likely to stop filing is a large enough share of the filing base to put the 85,000 assumption out of reach. DHS never reconciles them.

The alternatives DHS did not consider

DHS looked at exempting or discounting small entities and said no. Its reasoning is that doing so "creates a sizable perverse incentive for employers to avoid the fee" and would cause a revenue shortfall.

What is striking is how little analysis sits behind that. The Regulatory Flexibility Act requires an agency to describe the significant alternatives that would accomplish its objectives while minimizing the impact on small entities, and it specifically contemplates options such as differing compliance timetables that account for the resources of small entities, along with outright exemptions. See 5 U.S.C. 603(c). For a rule DHS concluded would significantly affect more than 11,000 small entities, that discussion runs about three sentences and addresses exactly two possibilities: exempt them, or discount the fee.

Several obvious middle options go unexamined — deferring payment until approval, phasing the fee in over several cap seasons, or tying relief to filing volume rather than to entity size. Note that DHS's stated objection does not reach them. The perverse-incentive concern is that employers would restructure to look small. An alternative that never asks how large the employer is — a deferral, or a phase-in — creates no such incentive at all. DHS rejected size-based relief for a reason that says nothing about size-neutral relief, and then stopped.

If you are a smaller employer, that is the most useful thing to put in a comment. Not that the fee is unaffordable in the abstract, but that a specific, workable alternative exists which DHS never evaluated, and what your company would do under the rule as written versus under that alternative.

Worth noting what the one real exemption actually turns on. Cap-exempt status is about mission and affiliation, not size. A large university pays nothing. A ten-person startup pays the full $103,265.

What happens next, and how to comment

Nothing takes effect now. This is a notice of proposed rulemaking; the fee applies only if and when DHS issues a final rule adopting it. The proposal states no effective date and does not say which cap season would be the first one hit.

Comments are due 30 days after publication in the Federal Register — September 24, 2026, working from the scheduled August 25 publication date. They go through the federal eRulemaking portal at regulations.gov, referencing DHS Docket No. USCIS-2026-0298, in English or with an English translation attached. Emails and letters to DHS or USCIS officials do not count as comments. Mailed and hand-delivered submissions are not being accepted.

One caution: every submission is posted publicly, without change, including any personal information it contains. DHS expressly advises limiting what you include.

DHS says the most useful comments point to a specific portion of the rule, explain why a change is needed, and bring supporting data or authority. A comment showing what a $103,265 per-petition charge does to a specific hiring plan lands harder than a general objection. So does one documenting what the realistic alternative to that hire actually is.

What to do now

Do not change your current filings. Petitions on file and petitions being prepared today are unaffected. No fee is owed under this proposal unless and until a final rule takes effect.

Model the cost into your cap planning. If you are making headcount and budget decisions now, run the $103,265 per-petition number through them — especially if you register a large number of beneficiaries.

Confirm whether you are actually cap-subject. Employers affiliated with institutions of higher education, or that qualify as nonprofit or governmental research organizations, may be cap-exempt under INA sections 214(g)(5) and (7). Cap-exempt petitions sit outside this proposal entirely. The analysis is fact-specific and worth doing properly rather than assuming.

Review the categories this rule does not reach. L-1, O-1, and cap-exempt filings are untouched by the proposal. Whether any of them fits a given role is a specific legal question, not a default answer — but it belongs in the planning conversation now rather than after a final rule.

Consider commenting. Thirty days is short, and employers — particularly the smaller ones inside that 76 percent — are exactly the parties whose data DHS says it wants.

Account for the fees already in force. The CBP biometric fee expansion is a final rule, not a proposal, and it applies right now to H-1B and L-1 extension filings by covered employers.

At Parikh & Prasad, PC, we work with employers across the country on H-1B cap strategy, cap-exempt analysis, and employment-based immigration planning. If you sponsor H-1B workers and want to know how this proposal — and the fee changes already in force — would hit your program, contact us to schedule a consultation.

This post is attorney advertising and is provided for general information only. It is not legal advice and does not create an attorney-client relationship. Immigration law changes rapidly; consult a qualified immigration attorney about your specific situation.

Related Practice Areas