# Failure to Deliver (FTD) Source: SquawkFlow URL: https://squawkflow.com/glossary/failure-to-deliver Category: Market Structure Published: 2026-08-25 Type: Definition > What an FTD is, what the SEC actually publishes twice a month, and why the number is a settlement statistic rather than a measure of naked shorting. --- ## What a failure to deliver is A failure to deliver happens when a party to a securities trade does not hand over the shares by the settlement date. The trade was agreed; the delivery did not arrive on time. US equity trades settle on a standard cycle after the trade date. When the delivering side cannot produce the shares by then, the position is recorded as a fail, and it stays on the books until it is closed out. ## What the SEC actually publishes This is the part most often misdescribed, so it is worth being exact. The SEC publishes fails-to-deliver data twice a month, covering the preceding half-month, on a substantial lag. Each row shows a settlement date, a security identifier, and the aggregate number of shares that were failing to deliver in that security as of that date. Three limitations follow directly from that format, and they are not subtle: **It is a balance, not a flow.** The figure is the total quantity outstanding on that date, not the number of new fails that day. A position failing for several consecutive days appears in each day's count. Summing the column across days therefore counts the same shares repeatedly, a common and badly misleading mistake. **It does not identify who failed.** No broker, no firm, no account. The data is aggregated at the security level. **It does not state a cause.** The record shows a fail. It does not say why. ## Why the cause matters Fails are routinely presented as evidence of naked short selling, selling shares that were never borrowed. That is one cause. It is not the only one, and on most days it is not the main one. Ordinary operational causes produce fails constantly: shares in transit between custodians, mismatched settlement instructions, a lending recall that arrives late, corporate action processing, and legitimate market-maker activity under the bona fide market making provisions that historically allowed hedging without a locate. The consequence is that an elevated FTD count is a question rather than an answer. A security with persistent fails is worth looking at. The fails themselves do not establish what caused them, and any analysis that treats the two as equivalent has skipped the step that carries all the weight. ## Regulation SHO and the threshold list Fails are not left to accumulate indefinitely. Regulation SHO sets close-out requirements: a fail persisting beyond a defined window must be closed out by purchasing or borrowing the shares. Securities with fails above a size and duration threshold appear on a threshold securities list, published by the exchanges. That list is more informative than the raw FTD file for most purposes, because it applies a persistence test rather than reporting a single day's balance, appearing on it means fails have been both large and sustained. ## How it relates to short interest and short volume These three are frequently conflated and measure different things. **Short interest** is the number of shares sold short and still open, reported by firms to FINRA and published twice monthly. It is a position measure. **Short volume** is the share of daily volume marked as short sales. Much of it is market-maker activity that is flat by the close, which is why a high daily short volume ratio says far less about bearish positioning than it appears to. See [short volume ratio](/learn/short-volume-ratio-meaning) for why. **Fails to deliver** is a settlement measure. It describes shares that were not delivered on time, which is a different fact from shares sold short. A security can have high short interest and no fails, or persistent fails and unremarkable short interest. Treating any of the three as a proxy for the others produces confident conclusions from unrelated data. ## Using the data responsibly If you work with the SEC file directly, the discipline is: read it as a balance rather than a flow, never sum across dates, compare a security against its own history rather than against other securities of different sizes, and treat the threshold list as the stronger signal. SquawkFlow does not publish FTD data. It is a public SEC dataset available directly from the source, and re-serving it would add a lag without adding anything. --- ## Citation Source: SquawkFlow, https://squawkflow.com/glossary/failure-to-deliver Retrieved: 2026-09-11 00:31 ET Attribution: cite the page URL rather than a copied number. Levels are recomputed every session, so a number without its date is wrong within a day. SquawkFlow publishes market-structure data and education. Nothing here is investment advice, a price target or a recommendation, and we have no order execution.