The precision is the tell. Regulation Z rarely asks for a time zone. Requiring the lock's expiration to the hour, and naming the zone, is an acknowledgment that this is a market price with a shelf life measured in days, and that a dispute about whether a lock had expired would otherwise turn on whose clock was being read. The same paragraph requires a separate expiration date for the estimated closing costs, so two clocks can be running at once on the same form and they need not end together.
What the lock fixes, first: the figures the lender may still revise. A creditor may use a revised estimate in place of the original for certain enumerated reasons, and one of them is exactly this situation. 12 CFR 1026.19(e)(3)(iv)(D) lists as a changed circumstance that "the points or lender credits change because the interest rate was not locked when the [Loan Estimate] … were provided", and then imposes a deadline: "no later than three business days after the date the interest rate is locked, the creditor shall provide a revised version of the [Loan Estimate] … with the revised interest rate, the points disclosed pursuant to § 1026.37(f)(1), lender credits, and any other interest rate dependent charges and terms." So an unlocked quote is explicitly provisional as to price, and locking converts it into a document the lender owes the borrower a corrected version of within three business days. A borrower who locks and receives nothing in writing within that window is missing a disclosure the regulation requires.
What the lock fixes, second: the date the loan is judged against the market. Several consequential tests in Regulation Z are measured not at closing but at the moment the rate is set. The Official Interpretations put it directly: "a transaction's annual percentage rate is compared to the average prime offer rate as of the date the transaction's interest rate is set (or 'locked') before consummation", and where the rate is set more than once, "the creditor should use the last date the interest rate is set before consummation" (comment 35(a)(1)-2). That comparison is what decides whether a first-lien loan on a principal dwelling is a higher-priced mortgage loan, and higher-priced status carries two concrete consequences: an escrow account must be established before consummation (12 CFR 1026.35(b)(1)), and a prepayment penalty is not permitted at all (12 CFR 1026.43(g)(1)(ii)(C)). The same date also fixes which conforming loan limit applies, since 12 CFR 1026.35(a)(1) measures the principal obligation against "the limit in effect as of the date the transaction's interest rate is set". Locking in late December rather than early January can therefore change which threshold the loan is tested against.
What the lock fixes, third: an expiration, with what happens after it left to the contract. If the loan does not reach consummation before the lock ends, the rate is no longer committed. Whether an extension is available, what it costs, and whether the borrower may take a lower rate if the market has fallen, a feature usually sold as a float-down, are all matters of the lock agreement. Regulation Z fixes none of them, and published figures for extension pricing come from lender marketing rather than from any source that can be checked, so the only reliable answer is the one in the agreement in front of you.
The practical asymmetry worth naming. A lock protects the borrower against a rising market and removes the benefit of a falling one, unless the agreement says otherwise. That is not a criticism of locks; it is what buying certainty means, and it is the same trade a fixed rate makes over thirty years rather than over thirty days. The decision is about how much delay risk sits between the quote and the closing table, and delay risk is largely a function of the things the underwriting, the appraisal and the title work can turn up.
