*In the space of a few minutes, roughly a trillion dollars of value evaporated and then largely returned. Some shares traded at a penny; others at a hundred thousand dollars.*
On the afternoon of the sixth of May, 2010, the American market experienced something that had no real precedent. Over the course of a few minutes, the Dow Jones Industrial Average fell by roughly a thousand points, a decline of some nine percent, and then recovered the majority of that fall within approximately twenty minutes. Around a trillion dollars of value disappeared and substantially returned in less time than it takes to eat lunch.
The details of individual securities are stranger than the index figure suggests. Shares in some well-established companies traded at a single penny; others, briefly, at prices near a hundred thousand dollars. These were not reporting errors. They were actual transactions, in the shares of substantial enterprises whose businesses had not changed in any respect during the minutes in question.
The causes remain partially disputed, but the broad mechanism is understood. A large automated selling order entered a market already under stress. Automated participants, which normally stand ready to buy and sell and thereby provide the continuous availability of a counterparty that investors take for granted, responded to the resulting conditions by withdrawing. Some stopped trading entirely; others continued but at prices so far from the prevailing market that they were effectively absent.
What happened next is the essential fact of the episode. With the usual counterparties withdrawn, orders arriving in the market found almost nothing on the other side. An order to sell at whatever price is available will be executed at whatever price is available, and if the only bid remaining in the system is a penny, then the transaction occurs at a penny. The prices printed during those minutes were not judgements about value. They were simply the residue of what happened to remain in the order book when everything else had gone.
This reveals something about prices that most investors never have occasion to confront. A price is not an inherent property of a security, like its mass. A price is the outcome of a transaction, and a transaction requires that someone be willing to take the other side. In ordinary conditions there are always many participants willing to do so, and this abundance is so reliable that investors treat the availability of a price as a permanent feature of the world. The Flash Crash demonstrated that it is not a feature of the world but a service, provided by participants who may withdraw it.
The withdrawal is not irrational on the part of those who withdraw, and this is worth understanding rather than resenting. A participant whose business is standing ready to transact does so on the assumption that they can manage the risk of doing so. When conditions become sufficiently disorderly that they cannot determine what anything is worth or what is happening, the rational response is to stop, and every such participant reaches this conclusion at approximately the same moment for the same reasons. Liquidity therefore tends to disappear collectively rather than gradually.
The episode carries a direct practical implication concerning the mechanics of how an investor places an order, and it is one of the few genuinely actionable pieces of market structure knowledge an ordinary participant needs. An instruction to transact at whatever price is available will be honoured, and during those minutes it was honoured at prices no rational person would have accepted. An instruction that specifies a price beyond which one will not transact simply goes unfilled in such conditions, which is an entirely different and considerably better outcome.
Many of the most extreme transactions from that afternoon were subsequently cancelled by the exchanges, on the grounds that they were clearly erroneous. This provides less comfort than it appears. The cancellation was a discretionary decision made after the fact, according to thresholds determined at the time, and an investor whose transaction fell just outside those thresholds received no relief. Relying on the possibility that a bad outcome will be reversed by someone else's discretion is not a form of risk management.
The broader lesson concerns the relationship between price and value, which the episode separates with unusual clarity. During those minutes, the businesses whose shares traded at a penny were exactly as valuable as they had been an hour earlier. Nothing about them had changed. The price had changed, dramatically, and the price and the value had simply parted company for a period. This is ordinarily invisible because the gap between them is small; the Flash Crash made it enormous and therefore visible.
An investor who has internalised this will treat a falling price with more care and less reverence. The price is telling them what someone was willing to pay at a particular instant, under whatever conditions prevailed at that instant, and those conditions may have had nothing to do with the enterprise in question. It is information, but it is not necessarily information about the business.
A further point concerns how quickly this understanding fades. The episode is now sixteen years old, and the mechanisms introduced in its wake have functioned well enough that most investors have never experienced anything resembling it. This is the ordinary pattern: a rare event occurs, prompts reflection and reform, and then recedes from memory precisely because the reforms appear to have worked. The danger is that the absence of a recurrence is taken as evidence that the underlying vulnerability has been removed, when it may only mean that the conditions producing it have not been assembled again. Liquidity is still a service provided voluntarily by participants who may withdraw it, and no rule compels anyone to stand ready to buy what an investor wishes to sell. The mechanism that produced that afternoon has not been repealed.
At VESTFY™ the Flash Crash is presented as the clearest available demonstration that a market is a mechanism rather than an oracle. It functions well nearly all of the time, and its reliability is easy to mistake for a law of nature. It is not one, and the investor who understands this will be considerably less inclined to accept whatever the mechanism prints as a verdict on what they own.