Streetwise Professor

Bloomberg published a long article by an Icelandic hydrologist on AI trading bots. In some respects it is a contribution to the AI doom literature. It worries that many developers using the same AI platform (Claude Code being explicitly mentioned) and the same training data will cause their bots to enter into highly correlated positions that are subject to destabilizing unwinds in the event of a market shock. (Hedge fund basis trade unwinds are an example of such an unwind.)

At a superficial level this makes sense, but the analysis fails to think through the equilibrium effect of AI trading bots. Indeed, the low entry cost that Darri Eythorsson correctly identifies as the distinguishing feature of AI trading applications mitigates the effects he frets about.

In essence, AI bots can be viewed as privately informed traders who extract predictive signals from masses of highly dispersed data. Moreover, the low entry cost means that there can be–and will be, if there are not already–large numbers of such informed traders. These will be competing intensely.

Market microstructure theory predicts the effect of intense competition between many informed traders. Specifically, the model of Holden and Subrahmanyam (J. of Fin. 1992) predicts that these informed traders “compete aggressively and cause most of their common private information to be revealed very rapidly.” I italicize “common private information” because that is exactly what Eythorsson posits that AI trading bots will produce (because they are developed using the same tool–Claude Code–and the same data). Thus, the trading bot scenario fits the HS model almost precisely. Moreover, HS state that in a continuous trading environment (which characterizes virtually all equity and derivative exchanges) “market depth becomes infinite and all private information is revealed immediately” with a large number of entities trading on the same information.

This is in contrast to the situation (as in the canonical Kyle model) where there is one (or a small number) of informed traders, in which case information is incorporated only gradually in prices and market depth is finite.

The increase in depth is arguably a very salutary effect, which should be considered when weighing the benefits and costs of a trading bot army–including the costs of position correlation.

There are other implications. Prices that rapidly and fully reflect private information limits the incentive of the bots to accumulate large positions. Moreover, the competition between a large number of bots means that although their positions may be correlated, they will not be concentrated. A small number of concentrated positions (“whales”) pose, in my opinion, a greater systemic risk than a large number of correlated small ones.

The intense competition also drives the economic profit of bots towards zero. This is a self-limiting feature that constrains the aggregate scale of bot operator positions.

It should also be noted that truly smart bots will make trades based on information about other bots. As an example of this “reflexivity,” bots will attempt to predict the positions held by other bots, and will have considerable data to allow them to do so. This information is relevant to assessing the risk of cascading liquidations of correlated positions, which would reduce the amount of piling on by bots that (as they should) take both risk and expected return into account.

Of course, big hedge funds and quant firms hate this development. It essentially commodifies their expertise and human capital. This should make you treat any scare stories about trading bots with a healthy dose of skepticism, and indeed cynicism.

The main legitimate concern about AI trading bots is one that is common to all privately informed trading. Specifically, what social value does it produce? Is investment in private information for trading purposes primarily rent seeking that generates private profit (at someone else’s expense) but not commensurate welfare gains?

Privately informed trading extracts wealth from the uninformed, who trade for portfolio adjustment/risk management reasons (or for speculative thrills). This is purely a transfer, and one that requires investment in real resources to acquire and utilize the private information. In the case of AI bots, these real resources are the cost of developing and operating the bots. This is costly rent seeking, pure and simple.

Whence does the social benefit come? It is an unintended consequence of the rent seeking: more informative price signals that lead to better allocations of resources (e.g., better capital allocation or production decisions).

Financial economists have tended to trumpet the beneficial effects of more informative prices and “price discovery” without quantifying them, and discussed the rent seeking aspects of informed trading sotto voce, if they discuss it at all. Thus, a true cost-benefit analysis has not been undertaken.

In some respects this is understandable, because calculating the benefit of more informative prices is effectively impossible. (Maybe AI can someday make this calculation!) As is the benefit of potential speed-ups of the revelation of private information by a few minutes or seconds. But color me skeptical that it does.

So what it comes down to is a comparison of the real resource costs of privately informed trading by AI bots vs. the current set of privately informed traders, hedge funds, quant traders, investment banks, and the like. Here, the very cheapness (and hence ease of entry) of AI-based trading tools strongly suggests that rent seeking costs will decline relative to those in the current system.

As an extreme case, if creation and use of AI bots is free (which it isn’t, but consideration of this case is informative), we would get more informative prices at zero cost. Indeed, there would be significant cost savings due to the elimination of costly hedge funds, quant firms, etc. Indeed, this would represent a real world refutation of the Grossman-Stiglitz proposition that fully informative prices (“strong form market efficiency”) is impossible. Yes, there are costs, but if costs are low we can approach fully informative prices.

The main lesson is that if AI is a cheaper way of acquiring and trading on private information than the incumbent system, their displacement of incumbents would represent a welfare gain (though not a Pareto improvement!). This gain would come in two forms: lower rent seeking costs and more informative prices.

But that’s where the debate about the benefits and costs of AI trading should focus. The welfare effects of privately informed trading are ambiguous, at best, and such trading has a strong rent seeking component. If AI trading bots reduce rent seeking costs, their development is salutary. If not, it isn’t.

The very feature that Eythorsson highlights–the ease with which anybody can create a trading bot–suggests that AI will reduce rent seeking costs, and microstructure theory suggests it will improve price efficiency. Lower costs and higher benefits than the incumbent system represent a decided improvement.

Combine that with the likelihood that, as argued earlier, the equilibrium effects of competition among AI bots is likely to have little effect on systemic risk (and potentially may reduce systemic risk by reducing position concentration), well, don’t worry: be happy. Unless you are an incumbent privately informed trader that is. In which case, the trading bot reaper is coming for you.

The BBC recently ran a story hyperventilating over “the insider trading suspicions looming over Trump’s presidency.” This is just one piece of a genre: there have been numerous other stories during the period of the war pointing out that there have been oil trading volume spikes shortly before Trump Truth Social posts regarding war-related news.

The BBC has examined trade volume data on several financial markets and matched them to some of the president’s most significant market-moving statements.

It found a consistent pattern of spikes just hours, or sometimes minutes, before a social media post or media interview was made public.

The BBC then stated: “Some analysts say it bears the hallmarks of illegal insider trading, whereby bets are made by people based on information that is not available to the general public.”

Either the BBC is mischaracterizing what “some analysts” told it, or these analysts are full of it.

Even in securities markets in the United States, under Rule 10b-5, trading on “information not available to the public” is not illegal. Instead, the information must have been obtained illegally, or trading on it represents a breach of a duty.

With respect to commodity futures markets, “insider trading” was not even a thing prior to the passage of Frankendodd in 2010. That piece of legislation incorporated a new anti-manipulation rule that essentially copied 10b-5. Since insider trading is illegal under 10b-5, the insider trading is plausibly illegal in commodity futures markets as well.

The CFTC implemented the authority established by Dodd-Frank in its Rule 180.1. The rule elicited considerable commentary, on the issue of insider trading generally. This was framed as a discussion of “material non-public information” (“MNPI”):

The Commission recognizes that unlike securities markets, derivatives markets have long operated in a way that allows for market participants to trade on the basis of lawfully obtained material nonpublic information. This final Rule does not prohibit trading on the basis of material nonpublic information except as provided in the following paragraph or otherwise prohibited by law.

And the following paragraph?

Depending on the facts and circumstances, a person who engages in deceptive or manipulative conduct in connection with any swap, or contract of sale of any commodity in interstate commerce, or contract for future delivery on or subject to the rules of any registered entity, for example by trading on the basis of material nonpublic information in breach of a pre-existing duty (established by another law or rule, or agreement, understanding, or some other source), or by trading on the basis of material nonpublic information that was obtained through fraud or deception, may be in violation of final Rule 180.1. [Emphasis added.]

That is, trading on MNPI is a necessary, but not sufficient condition to violate Rule 180.1. The information must have been obtained by fraud, or theft, in order to rise to the level of a violation. And the norm in these markets is that trading on MNPI is NOT insider trading.

What is the likelihood that these conditions are met in any of the events set out in the BBC article? Small. Very small. For Trump or anyone connected to him? Even smaller.

Consider the March 9 event, where Trump was interviewed by phone by CBS, and the network released the content of that at 1829. Trading volume spiked about 47 minutes earlier. We don’t know the time of the interview, but presumably there it was considerably before the story ran. Meaning that there were people in CBS News who knew about it. Presumably several. Meaning that it could have escaped into the wild and sparked the trading volume noted by the BBC.

Meaning that if anyone violated 180.1, it was likely someone tipped of by CBS. Was this obtained by fraud? Was there a duty by anyone at CBS not to disclose this information?

The BBC also calls out a blip in trading activity (smaller than on March 9) on March 23 about an hour prior to Trump backing off his threat to attack Iranian power plants and bridges. This spike was smaller than the one on March 9.

But the decision that Trump announced was surely not made up at the instant he started tapping his post into his phone. And many people would have known about that. And DC leaks like a sieve.

You don’t leak information you are going to trade on. And the very volume of the trading that attracted the attention of the BBC and other news outlets makes it highly unlikely that any of those in the administration who were privy to this development were the traders responsible for the volume spike.

Eyeballing the BBC graph there were around 5000 lots traded during the spike. In round numbers, about $570 million worth of oil. The March 9 spike was even bigger, around 8000 contracts (~$800 million). So either a lot of people/firms had the news and were trading, or a couple of big institutions were doing so, or both. Nobody in the government would have the ability to trade in that size, and if they did, they would be total idiots to do so.

That is, the facts are most consistent with leaks leading to trades by someone outside the administration. And the leaks are not necessarily from within the administration: in the March 9 case, the information plausibly came from CBS.

Yes, it was non-public information (in the sense it had not been publicly disclosed), but as noted above that is not sufficient to establish an insider trading violation. And it certainly does not establish that Trump, or anyone in the administration, profited from this information.

Yes, there could be a violation here. One could argue that a leak from within the administration was a breach of a duty. But what if the information had not been classified? Would its revelation be a breach? Would anyone who acted on it have obtained it by fraud, or in violation of a duty?

It has long been known that securities prices often move well prior to public announcements, such as takeover announcements. Information is hard to keep contained, especially when related to things that involve many people. Such information leakage results in insider trading prosecutions only in extraordinary circumstances.

In other words, what has the BBC and others reporting breathlessly in the walls-are-closing-in genre is pretty much the norm in financial markets. Information tends to flow prior to public disclosures, and seldom constitutes a crime (or civil violation) when it does. If anything, the story just illustrates how pervasive leak culture is in DC. But as beneficiaries of that culture, don’t expect the BBC or any other media organization to call it out.

As the Iran War rolls into its sixth week, the media (especially the virulently anti-American UK rags FT and The Economist), the Democrats, and the Europeans have declared that the US is losing. Which means things must be going pretty well!

The US and Israel are allegedly losing because the regime has not toppled, and the Straits of Hormuz are largely (though not completely) closed. The first principle of war is “the objective,” and the losing narrative implicitly assumes that regime change is the objective.

But as the US made at the outset, and as Trump reiterated in his speech earlier this week, although that regime collapse would be a great outcome, that is not the objective. Instead, the objective is to destroy the ability of whatever regime controls Iran to use ballistic missiles and especially nuclear weapons to attack neighboring and non-so-neighboring states–including those in Europe, and perhaps eventually, the US proper.

It is difficult to get a definitive assessment on the campaign’s progress in achieving this objective. The administration claims it has made substantial progress, and a few more weeks of bombing will suffice. Given the intensity of the campaign, with now approximately 20,000 precision weapon strikes on strategic targets, that is very plausible. But bomb damage assessment is an uncertain exercise, so definitive judgment is difficult, especially for those (which consists of just about everyone outside select precincts of the Pentagon and the CiA) who don’t have access to all the intelligence in the first place.

A “war of choice” narrative is wrapped up with the “losing” narrative. Many on the left and the fringe right claim that the threat from Iran was not “imminent,” so the attack is premature and reckless.

The idea that a threat must be imminent to justify preemptive action is idiotic. Hell, what does “imminent” even mean? Those blathering about imminent never say.A day before Iran decides to launch a nuke? The day before it develops the capacity to launch a nuke?

By then it would be too late. And even if Iran’s nuclear development were to behind its ballistic missile development, one of the factors that features in the losing narrative–Iran’s continued ability to launch ballistic missiles and the imperfect ability and high cost to intercept them–makes it plain that if anything the US and Israel waited too long. Every month that would have passed would have seen a large increase in Iran’s missile stockpile, and an increase in their range and payload. In not too many months this would have made any attempt to attack Iran vastly more difficult and costly, and likely impossible. And the end result would be that Iran had North Korea-like immunity from attack, and could therefore act with impunity.

Imminent, in other words, was months ago. At best, February 28 was just in time.

Insofar as Hormuz is concerned, for the US it is an economic inconvenience, with mainly distributive effects. Yes, Americans pay higher prices at the pump (but not for natural gas and hence power, because the maxxing out of US LNG export capacity means that the demand shock from the loss of Qatari supply can’t hit the US), but American producers profit accordingly. And ~$100 oil is not an economic catastrophe–been there, done that (e.g., after the Russian invasion of Ukraine, and other times before, especially if you look in real terms).

The often expressed idea (especially in the media) that the US did not anticipate Iran’s action, or that it had no plan to deal with it, is absurd. I think that the plan was exactly what the US is doing. That is, Trump and the Pentagon realized that a closing of the Straits would be a cost of taking action, but the cost of opening it by force is greater, as is the cost of not taking on Iran at all. If anything, the US approach to the Straits demonstrates what the US perceived to be the cost of not acting.

There are no free lunches, and no cost free wars. The enemy always gets a vote.

And with respect to Trump personally, he had to be cognizant of the potentially steep political costs he might pay–including another impeachment ordeal. You might disagree with Trump’s decision, but that is different from saying that he made his decision without due consideration of the costs (military, economic, political, personal). Revealed preference logic implies that he perceived these costs to be less than the cost of inaction. Saying “he didn’t consider the costs” really means you disagree about what the costs are.

Hormuz is also a military challenge, no doubt. The threat can be (and has been) reduced by attacks on Iranian anti-ship missile batteries, attack speedboats, minelayers, and conventional navy, but even residual capability would pose a substantial threat.

Europe and Asia also feel the impact of the blocking of the Straits, and more so than the US given that they are net energy importers. Europe huffily refuses to do anything about the closure (well worse than that actually–see below), saying it is not their war.

Well, to use a phrase attributed to Trotsky, “You may not be interested in war [Europe], but war is interested in you.” And the refusals betray sunk cost fallacy thinking. The war happened and you have to optimize subject to that fact: saying that you didn’t choose to be there is irrelevant. I know where you want to go, but you can’t get there from here.

It is likely that Europe’s indignant refusal to do anything reflects not so much moral superiority, as utter incapacity. Just what could they do, given their shriveled military–and especially naval–capabilities? Moral indignation is therefore most likely a way to avoid confessing powerlessness.

Furthermore, Europe is doing things that are distinctly contrary to its interests. Most notably, several European nations (Spain, France, and perhaps Italy) are restricting US use of bases in their countries, and limiting overflights. Which will not stop the US and Israel from continuing the campaign, but which might prolong it, thereby prolonging Europe’s economic misery. This is masochistic posing.

Europe actually could do something that would not require it to employ (non-existent) military force. It could contribute the the US marine insurance reinsurance backstop. That would facilitate the reopening of the Straits. I have not heard this idea mooted, weirdly not even by the US, but definitely not by Europe.

As for the nations that at least have pretensions of global military power, the UK and France: in perhaps a reversal of traditional stereotypes, Albion is feckless, and Gaul is perfidious. British fecklessness was demonstrated by its tragi-comic inability to deploy even a single warship to defend its base in Cyprus, and its passive-aggressive denial of bases in the UK and Diego Garcia, at least for a while.

France–and in particular Macron–is objectively an American enemy right now. As popular as tertiary syphilis at home, where he can accomplish nothing, the little popinjay (whose wife apparently periodically pops upside the head) struts around the world attacking the US. But worse than his words is his active opposition–in conjunction with Russia and China–to a UN resolution advanced by the Gulf Arab states to permit the use of force to open the Straits.

Macron says that this would be impossible. Well, nobody is asking you to do it boy. Among the options lead/follow/get the hell out of the way, all you need to do is choose #3. But no. You actively decide to get in the way.

This is incomprehensible for many reasons. First, it represents a complete rejection of Arabist French foreign policy extending back decades. The reversal of French support for Israel in 1967 is a good estimate of the starting point for this policy, but through the years it has seen, for example, strong French support for Saddam, e.g., building the Osirik reactor that Israel destroyed, and including being a major arms supplier to Iraq during its gruesome war with Iran.

One Arab state–the UAE–has already expressed its disgust with France’s action by withdrawing funding from a next generation Rafael fighter project. And France can be sure that the flow of Arab capital to the country generally will dry up. Which given France’s parlous fiscal and economic situation is quite harmful.

Second, anything that defers reopening of the Straits extends France’s economic pain.

Third, France, like the rest of Europe, has been in a panic about the US exiting Nato, or at least dramatically cutting its commitment to European defense. What better argument to make for the US to tell Europe, “lots of luck! You’re on your own! Practice your Russian!” than this active opposition to the US effort in Iran.

Macron’s remarks in South Korea–a country that exists by the grace of the US and 37,000 fatal American casualties–demonstrate his idiocy in this regard. He called for South Korea and other nations, including Japan which has also been the beneficiary of American military dominance, to form an alliance in opposition to the United States. Which would provide a perfect justification for the US to cut loose the members of this coalition of the weaklings. I’m sure that would work out great for all of them.

In World War I, Germany likened its alliance with Austria-Hungary as being “shackled to a corpse.” That’s what Nato has become to the US right now, and Macron is making the corpse’s stink all the worse. Unlike Wilhelmine Germany, the US can get out the bolt cutters.

Fourth, the Europeans–and France in particular–constantly bleat about international law. Well, Iranian interference with free navigation is a flagrant violation of one of the oldest and undisputed parts of international law. But rather than allow others to enforce this law, France actively opposes it.

This is pretty much spot on:

To spotlight the absurdity of Macron's position here, Elysee is voting alongside Moscow and Beijing to obstruct the GCC bid to establish a multinational coalition with a Ch VII mandate to enforce transit rights, enshrined under UNCLOS, and their collective rights to self defense,… https://t.co/YbVqKlXDVx

— dan linnaeus (@DanLinnaeus) April 4, 2026

The absurdity of Macron’s position is manifest. The explanation is not. I am at a loss to understand what he is doing, or what he thinks he is doing, or what he thinks he will accomplish. The best explanation I have is a cut-off-your-nose-to-spite-your-face attempt to thwart the US. But this is absurdly futile. France’s actions are an inconvenience, and will not dissuade Trump or Israel. The second best explanation is blackmail. Other than that, I got nothin’.

So wither the war? I think the most likely outcome is a 2-3 week continuation of the bombing campaign. After which, the US let the chips fall where they may. Will Iran continue to close the Straits when there is no pretext for doing so? Fairly likely, which would amp up the pressure on those more dependent on the Straits to do something.

The dispatch of 2 Marine Expeditionary Units and a brigade of the 82nd Airborne division to the region has led to all sorts of speculation about ground action. One surmise is seizure of Kharg Island, Iran’s oil export hub.

I consider this extremely unlikely. First, the USN is not going to sail a couple of amphib flotillas up a shooting gallery. Second, the force sent to the region (roughly 10,000 combat personnel) is too small. Third, even if the island is seized, it is an . . . island–making the logistical support of US troops on it difficult and vulnerable. Fourth, it is not necessary to seize Kharg to deny Iran the ability to export oil. Just mine the approaches. Turnabout is fair play.

It is more likely that Marines will be employed to seize the islands (held by Iran, claimed by the UAE) in the Straits themselves. This would likely be necessary, but not sufficient, to reopen the Straits in the absence of Iranian capitulation.

Insofar as the bombing campaign is concerned, the US does have escalatory options–and Trump has threatened and continues to threaten to use them. These include destroying crucial parts of Iran’s economic infrastructure, notably power plants and perhaps desalination facilities. The latter option has been widely decried as a potential war crime, but Iran has attacked such facilities in Gulf states, so that genie is out of the bottle.

The main reason for such escalation is to secure if not regime change, a deal with the regime that cements the US objectives and opens the Straits. And here we get into a game theory dilemma. If the threat does not achieve this, acting on it will be unlikely to do so. And Iran can credibly play crazy and refuse to cave to the threat. So in some respects this is crazy vs. crazy, or at least two parties who can credibly play crazy.

And ironically, a regime that is willing to accept such severe punishment is the best argument for the war in the first place.