The State Department made its visa bond program permanent in a final rule published Monday, August 3, and effective the same day it published [1][2]. Applicants for business and tourist visas from covered countries can be required to post a bond of up to $20,000 as a condition of getting one [1].
The tiers come straight from the rule: the exact amount is "either $10,000, $15,000, or $20,000, based upon the applicant's circumstances as determined by the consular officer, unless the bond requirement is waived" [2]. The pilot's ceiling was $15,000, with steps at $5,000 and $10,000, and the Associated Press reports the final rule raises the maximum to $20,000 and removes the $5,000 low end [3]. The rule's own text confirms the removal and shows $15,000 surviving as the middle step [2]. Beginning October 1, 2027, and every seven years after, the $20,000 ceiling adjusts upward for inflation using CPI-U, rounded up to the nearest $1,000 [2].
The Department's account of its own pilot is the reason to read this rule closely. "The Department initially anticipated that approximately 2,000 B-1/B-2 visa applicants would be required to pay a bond to travel to the United States during the one-year Pilot Program. However, travelers from 50 countries were ultimately subject to the Pilot Program, and approximately 20,000 visa applications were determined to require a visa bond payment." [2] Ten times the projection.
What those 20,000 did with the requirement is the finding. "close to half of those applications have resulted in a bond payment, for a total temporary monetary cost to the public of about $115 million," the rule states, and elsewhere that "nearly half of the 20,000 applicants subject to the bond in the pilot ultimately chose not to pay the bond" [2]. The arithmetic is consistent: roughly 10,000 paid bonds against about $115 million works out to about $11,500 each, which sits inside the pilot's $5,000 to $15,000 range.
Issuance collapsed alongside it. "Compared to the same 10-month period the preceding year, visa issuance rates declined by 83 percent as of July 2026," the rule reports [2]. The Department does not present that as an unintended cost. Its words are that "some applicants appear to self-select by not paying a bond," and that it "expects that this final rule will contribute to the continued reduction of demand for B1/B2 visa applications from nationals of countries subject to the program" [2]. Reduced demand for lawful travel is stated as an expected result.
On overstays, the rule reports a change of a different magnitude. "In FY 2024, there were 45,488 overstays from these 50 countries. In the first 10 months of the pilot, the number of overstays was fewer than 50." [2] The two periods are not the same length. Prorated, 45,488 over twelve months is roughly 37,900 over ten, so fewer than 50 against roughly 37,900 is about one-tenth of one percent of the prior rate.
The strongest version of the Department's case is stronger than the issuance number alone would suggest, and it is worth working out. An 83 percent decline in issuance leaves about 17 percent of the prior traveler volume. If that remaining group had overstayed at the old rate, ten months would have produced something on the order of 6,400 overstays. The rule reports fewer than 50. Volume collapse does not account for the gap. Something else is in these numbers, whether the bond's own deterrent effect on the people who did travel, the rule's requirement that compliance run through commercial air arrival and departure at a U.S. port of entry or CBP Preclearance location, or how a bonded traveler gets counted in the first place. The Department's claim that bonded travelers complied at high rates is not an empty one.
The cost side is in the rule too, stated in the Department's own accounting. Roughly 10,000 people declined to post $10,000 or more and did not make the trip. A qualifying applicant now spends an estimated two hours arranging the bond, which the Department values at $49.02 using the 2025 Bureau of Labor Statistics median hourly wage of $24.51 [2]. That multiplication checks. The bond itself is refunded on departure, which the rule stresses, making the $115 million a float rather than a fee. A traveler who cannot raise $10,000 in the first place never reaches the refund argument.
How the program became permanent matters as much as what it does. The Department invoked the Administrative Procedure Act's exception for rules "involving a foreign affairs function of the United States," 5 U.S.C. 553(a)(1), which also lifts the statute's usual requirement that a rule wait 30 days before taking effect [2]. That is why publication day and effective date are the same day, and why there is no comment docket attached to a permanent program covering 50 countries. The Office of Personnel Management published four civil service rules the same morning; those ran a 30-day comment period and drew 1,252 comments. This one drew none, by design.
One date is worth pinning down, because the reporting and the rule use different formulations. The Federal Register issue carrying the rule is dated Monday, August 3, 2026, and the rule states plainly that it "is effective August 3, 2026" [2]. The Associated Press account says the notice "is to be formally released on Monday, the day the program becomes permanent," without naming a date [3]. August 3 is a Monday and August 4 is a Tuesday, so the two accounts agree. Anyone reading a secondary summary that pairs "Monday" with August 4 has an internal contradiction, and the Federal Register controls.
What the permanent rule buys, then: 50 countries that AP reports are mostly African [3], a ceiling that starts at $20,000 and indexes upward from October 2027, about $115 million held in bonds, and a record in which the clearest measured effect is on people who applied lawfully and either paid or gave up.
