Capital One Fraud Researchers May Also Have Done Some Fraud
bloombergview.com
bloombergview.com
[0] http://www.wsj.com/articles/SB100014240527023042799045795161...
There are fundamental things (like limited number of sufficiently high-level insiders) restricting competition in conventional insider trading to minimize the rent charged from insider "research".
Economic theory does lend some support for the idea that markets will perform better when the public is confident enough to participate in them broadly, and broad participation should drive valuations higher. (The latter is a dubious benefit.) But there is likewise plenty of theoretical justification for the harm done to markets by insider-trading laws. Markets are basically mechanisms for aggregating and analyzing information, and insider trading laws restrict the supply of basic facts. Mispricing ensues.
There are other justifications for banning insider trading, albeit all rationalizations post facto, as far as the legal history goes. Allowing a manager to trade her company's shares (particularly to short them) can put her incentives out of step with those of the company's other shareholders. But the managerial misbehavior that could provoke is already illegal under various other regimes. And the principal-agent problem is no argument at all for setting up a system like we have, one which bans trading by knowledgeable people who are in no position to influence corporate decisions.
But the economics are almost beside the point. The promise made by insider trading laws is that any Joe Schmoe can trade on the same footing as sophisticated professional investors, and that promise is a farce. It's a farce not only because the incentives will always drive some people to break insider trading laws, but because ordinary people do not have ready access to all of the legally "public" facts anyway; many of them are locked up in obscure or expensive databases. And if Joe Schmoe did have all the public facts, he would still lack the other big piece of the informational puzzle, which is the ability to interpret the facts properly. That more than anything requires real expertise.
In my opinion, the public-confidence angle is the strongest component of the case for outlawing insider trading, but I think it's a rather weak case in the end. The perverse truth is that if insider trading laws inspire the public to think they can play the markets on equal footing with sophisticated investors, then insider trading laws have succeeded in installing a false confidence; they have hoodwinked the public.
----
0. This has remained a main justification (at least rhetorically) for maintaining and strengthening insider trader laws over the past eight decades. For example, an act amending the rules in 1984 has this preamble: "Insider trading threatens ... markets by undermining the public’s expectations of honest and fair securities markets where all participants play by the same rules."
Unlike the case where a very small group of investors privy to truly secret information can collude to keep the secret information out of the stock price, to their eventual gain.
The promise made by insider trading laws is that any Joe Schmoe can trade on the same footing as sophisticated professional investors, and that promise is a farce.
If you want to see how good a trader someone is, look at how much they lose in a bad market rather than how much they make in a good market.
As they are not part of the portfolio they can avoid reporting on the performance, so you don't necessarily get to see the full picture of how their decisions affect their investors.
Random selection tends to bias towards smallcap/microcap, as the sheer quantity of those companies is higher.
You can't have it both ways, you either allow people to trade on private information (thus indirectly making the information public by moving the stock price) or you don't.
[0] http://www.nytimes.com/2013/07/21/business/a-shuffle-of-alum... [1] http://money.cnn.com/2014/11/20/investing/goldman-commodity-... etc...
Seriously, why not?
It's possible that releasing hourly negates the effect of the information withholding; but I don't think it's necessarily a foregone conclusion. Instead of trading on longer term signals you're trading on "did company X sell as much this morning as expected" - which I'll grant you is a ludicrous thing to base that companies worth on. Perhaps it means people who have less effect on long-term signals get to take part in insider trading though - if you know there's a new advert or offer coming out you can predict a revenue shift.
If there weren't laws against people taking their employer's or contractee's stuff, employers would be hire hit squads to prevent it from happening. Despite vision of libertarian utopias, that wouldn't be an improvement.
Also, it makes no sense to base judgement of the law on the impossibly high standard of 100% compliance. Indeed, any law that does achieve a compliance level that approaches 100% will almost certainly be accompanied by an enforcement regime so terrifyingly intrusive that it's hard to imagine a threat to civil society that could be greater.
What does make sense is widespread use of the kind of statistical analysis that makes patterns of insider trading very hard to conceal, along with enforcement that includes far-reaching chains of disgorgement in cases where convictions do stick.
Basically, if you're you're foolish enough to do business with a fund that is producing returns that are inexplicable absent reliance on insider trading, you could find yourself loosing large sums to the government if and when the fund's operators are taken down.
In big business, quarterly is often as close to "frequent" as one can get. Contracts take weeks if not months to negotiate, net 60 or 90 payment terms, hold backs, terms on goods in transit, etc. For some complex businesses, teams work on nothing but figuring out quarterly filings.
And then: if I am a shareholder, and I own part of the company, why can't I have access to the internal not-yet-audited general ledger status? It's my company, right? It should be feasible to give me access to a secure site where I can get the information...
I believe the real reasons, rather than technical infeasibility, are mainly two:
- Extreme "corporate governance": If shareholders can react on every minute transaction that you do, and they effectively micromanage the company, there is still more incentive for short-term thinking (a problem that a lot of listed companies are already very familiar with).
- Strategy: If competitors can buy one of your shares, they would get access to all your information. You can still do the same, though: on public companies, information is either public, or private. But who is going to start? Who is going to rush to be open to the competitor?
Check out: http://www.itg.com/intelligence/itg-investment-research/cove...
Also check out this article from the WSJ from June 2010 on Majestic Research: http://blogs.wsj.com/economics/2010/06/30/using-starbucks-du...
Using Starbucks, Dunkin’ Donuts to Track Economy As economists look for clues on the direction of consumer spending, they may want to look into how much Americans are willing to spend on their coffee.
Consumers have been more willing to spend since the lows of the recession, but recent declines in retail sales and confidence have sparked worries over whether spending can continue to grow in the second half of the year.
Enter the coffee indicator. A “tell-tale sign of how consumers feel about employment, income and the future is where they buy their coffee and whether they step up for the more expensive concoction,” wrote Majestic Research economist Steve Blitz in a recent research note.
Majestic Research tracks anonymous credit-card data, and can see how much consumers spend by category and store. Blitz broke out the average dollar transactions at Starbucks and Dunkin’ Donuts. The data show that during the worst of the recession consumers spent less at the two coffee outlets, but as the employment picture started to improve people were willing to spend more per transaction.
AFP/Getty Images Starbucks may hold a key to consumer spending. The trend reversed at the beginning of April when transaction size turned down. To be sure, much of that change is likely seasonal. Transaction size at Starbucks, especially, takes a big spike around the holidays as shoppers buy coffee baskets and mugs for those caffeine addicts on their lists. In the last two years, it has bounced back a bit through the late winter, turning down in April and then moving back up in the late summer/early fall.
So far, this year’s transactions at Starbucks and Dunkin’ Donuts is following the pattern. If that bounce back materializes in the late summer, it could indicate that consumers are still willing to open their wallets. But if the average transaction size levels off or continues to decline, it could indicate a more thrifty consumer will dominate the second half of this year.
They probably have thousands of employees who can access this information, I think it would be foolish to think that absolutely none of them does it for personal purposes.
My interactions with the criminal legal system as an expert witness since then have only reinforced my belief in Harry Silverglate's axiom that pretty much every US adult is merely an un-indicted Federal felon living their lives at the mercy of the thousands of AUSAs who could plunge them into a living hell.
If searches for X brand is way up one month then you might be able to predict that x band will have a good quarter.
> Surely lots of Wall Street firms -- Chipotle is followed by 31 analysts -- and asset managers are doing tons of research to try to estimate Chipotle's sales. They're visiting branches and calling investor relations and talking to pork suppliers and surveying consumers and generally getting paid a lot of money to build a robust estimate of how many burritos Chipotle is selling. One more piece of data -- one credit card company's charges at Chipotle -- would be helpful, but come on, not that helpful.
Could we get this data any other way? How about putting cameras in front of a few flagship Chipotle stores and using CV to track the number of people going in/out? That's legal, AFAIK.
Then everyone would have the maximum possible amount of information, which would lead to maximally efficient pricing.
For some reason I can't quite put my finger on, I don't expect this to happen any time soon.
Which is one reason I remain skeptical that 'efficient pricing' has ever been a genuine goal.
It would do all sorts of weird things to the theory of the firm forcing internal transactions to maintain information asymmetry.
https://gigaom.com/2010/08/18/parking-lots-help-predict-earn...
Analysts already do this with satellite data for parking lot occupancy rates for businesses.
And then the employees wouldn't be guilty of insider trading. They would only be in violation of some agreement between them and their employer. Right?
I am not quite sure how one would go about an alternate method for assessing the actual value of user data, but it has always seemed rather obvious to me that it is hugely downplayed by all the self-interested parties and their defenders.
Would it, in this case maybe be the change in market cap pre and post sales figures? But even that, as pointed out, has some increases baked in based on alternative research to estimate performance. I guess you would need to find a very specific company that really does not lend itself to outside, tangential research and that keeps its performance measures and metrics under wraps really well between announcements. Anyone have any idea of an industry or company that fits such a profile?
Does anyone else realize we are really in and moving deeply into an era that is going to add significant opportunities for corruption and economic capture.
You can back this argument out to less extreme versions and see that it's very difficult to economically value a fact. E.g. what if the traders named in this article had 100X the capital on the line for the same trade. Is the data now 100X more valuable because of the added economic value they derived from it? What if their trade netted them more than the change in market cap?
A credit card processor that charges 0% fee on transactions would have a huge advantage when pitching retailers.
However, that information would still be non-public and likely run afoul of insider trading. That said nothing prohibits companies from making this information public in real-time and maybe it makes sense for the SEC to require publicly traded companies to make this information public in real-time.
Whether a card processor's handling of financial transaction flow would be viewed as non-insider is something that I don't know. I can make reasonable arguments either way.
Exploiting a pricing failure eventually corrects it: buying up undervalued options eventually raises the price to where they're no longer undervalued. An important question is how much you can buy up before they're priced correctly and how much you can make off doing so.
It's possible that these two gentlemen were already exploiting the price difference to with an order of magnitude of its potential. Being able to reliably make ten million or so a year is pretty cool, but not enough to run a sizable business off of -- it might not even be worth it for the 'big boys' to exploit. If you could pull in hundreds of millions or billions of dollars a year, then you're talking -- but you'll be splitting it N ways, with everyone else who decides to exploit the same trick.
So why, you may ask, do we have such high fees for credit card transactions? And believe me: they ARE high. No more than a hiccup when debit card fees were restricted to roughly $0.15/transaction but CREDIT card fees (which use the same systems) continue to be a substantial fraction (a few percentage points) of every purchase.
It's because no one is pitching retailers. The cards are sold to the consumer (who pays nothing). The retailer pays, but the contracts they are offered do not give them space to do much other than refuse to take a type of credit card (and give up a bug chunk of their business) or to accept them and pay the fees.
If you make the product really attractive to merchants, then you will have a problem getting customers. You will have to compete with the existing credit card products. Banks make a lot of money on those existing products and are highly motivated to ensure that the product you propose never, ever gets off the ground. History shows that they are both willing and able to (a) use this money to "buy" customers over with bonuses that you won't be able to match because you aren't charging the merchants so much; and (b) aggressively try to keep you out of all the infrastructure - there are many 'moats' that they control, including the large settlement systems, technical infrastructure installed at merchants (forget about using the same POS terminals to read your cards and Visa/MC, even if it's technically easy), etc.
There have been and are many attempts to make new alternative card products, but they aren't realistic to succeed. EU considers a wish for such a card system every couple years, but it turns out to be unfeasible even given the combined financial resources of the interested governments. Right now Russia has a strong motivation to support and subsidize such a system, but again, they don't want it that much to warrant the huge expenses for the relatively low expectations of success.
It probably saves us thousands to tens of thousands of dollars each year and we don't even have a huge amount of revenue.
If they were a little savvier, they could have used their data from CapitalOne to decide which businesses to research and then create shill research, much like the FBI's "parallel construction" of evidence. I wonder whether that would have put them in the clear; when the SEC came knocking, they could have pulled out their research, saying "Look, we figured this out fair and square."
I'm thinking whether a firm running city centre CCTV could do this with some computer vision systems. Mobile phone operators and ISPs that serve businesses must have similarly closely correlated signals to hand too.
>"Wonder how closely footfall at high-street traders equates to their revenue - pretty close I'd imagine?"
To put it another way: I was thinking that the number of people who go in to shops in the high-street might correlate well with the revenue for those same shops.
Huang and Huang had access to a db which is not open to everyone, granted, but they had to extrapolate the stock direction based on data from a subset (customers who buy Chipotle with a CC), of a subset (with a Capital One card)... and then compare that to analyst expectations, etc. but then, would it be insider trading if I stood outside a Chipotle polling customers who exited on the dollar amount spent? That is also proprietary information, and one I can use to trade stocks on. I'd like to know if they did any trades where the return went south. I know of sector investment funds which pretty much do this all day long, forecasting all sorts of industries, and it's not ilegal.
I'd like to know what others think.
Nevertheless, one may still be committing insider trading by trading based on non-public information even if the bet was wrong. Though is such a circumstance they may not have come up on the radar.
The point of Levine's post is that for the markets to work, there have to be traders trading at an advantage. The point of insider trading laws isn't to level the playing field --- that's exactly what you don't want. It's to eliminate a class of agency problems.
We can (probably should) want to deter credit card companies from trading on data mining payment data without believing that insider trading laws are the right vehicle to do that with. Privacy regulations and mandatory confidentiality agreements could set up an effective civil deterrent, rather than sporadic and incoherent SEC and criminal investigations.
I can see why professional traders and speculators might not want to level the playing field, but I don't see why long-term investors would feel that way. When I invest in stocks, I intend to hold for a long time. I'm betting on the business success of that company or group of companies, not on inequalities in the availability of information about them.
Isn't this what insider trading laws try to do? The law obviously can't keep insiders from learning information first, so the law prohibits insiders from acting as investors until the info is public. At worst this creates a lag between information and price, but that wouldn't matter much to long-term investors.
Not that insider trading laws can create total information awareness for every investor. But at least it gives every investor a more equal opportunity to find and use information.
minor edit for clarity
Anyone with the time and means to do so, anyway. Anyone could do these things; but most don't. For the people who do, what is protected is their investment in gathering and analyzing the data. The law does not grant Person A access to the data set of Person B--true--but the question is whether Person B could do their own work to gather the same data set.
Whereas information that is only available, in any form, to Person A would be considered non-public, and not legal to trade on.
If all traders were equally informed (perhaps by a law that any information you wanted to trade on had to be public), there would be no way for people to profit by digging up new information. We (society) like it that analysts have a financial incentive to figure out the truth about how well a company is doing - whether that be by coming up with a better model of how one industry affects another (which will ultimately lead to better allocation of resources), or doing the legwork to realise that a particular company is a massive fraud (which leads in a more direct/obvious way to better allocation of resources).
But we don't want them to just bribe insiders - that causes agent-principal problems, gives certain market participants unfair advantages, and all the rest of it. Hence the law, where figuring out these things through research is encouraged, but getting them from insiders is illegal.
However there are other ways that requirement can be met for instance SEC v. Dorozhko found that a hacker using stolen information to trade was guilty of insider trading. Obviously he had no fiduciary duty however he was found to be deceptive.
If the SEC prosecute this case I imagine they'd try to extend the argument presented in Dorozhko.
No. US v O'Hagan did not establish misappropriation as "insider trading". It established it as fraud, and as an SEC §10(b), which covers a broad range of securities fraud that is not describable as insider trading.
The syllabus for the case itself seems to be at pains to distinguish misappropriation from insider trading.
The importance of the distinction is that the Capital One case is something the spirit and letter of the law wants to deter, while satellite imagery of commercial activity isn't, despite it too being an example of market actors leveraging their own access and assets to gain an advantage in the market.
I eagerly await correction. :)
http://www.sec.gov/answers/insider.htm (search for "lawyers")
http://en.wikipedia.org/wiki/Insider_trading#Misappropriatio...
http://meyersandheim.com/how-to-win-an-insider-trading-case/ (misappropriation section)
http://www.sec.gov/news/speech/speecharchive/1998/spch221.ht...
also clarify the relationship, which according the SEC revolve around the obligations of "trust and confidence".
I got the alternate impression from reading the O'Hagan syllabus and 10b-5 itself. Oh well!.
I've pushed this point far past a point at which I'm comfortable defending it. :)
[1] http://www.sec.gov/answers/insider.htm
[2] http://www.justice.gov/usao/nys/pressreleases/December13/Wei...
The difference in this case seems to be they broke customer confidentiality rules to obtain the data, so use of that data for profit is now illegal.
According to the article, the big problem here was that the data being used here belonged to Capital One, and these researchers were running their own thing on the side using that data.
Even if we decide it isn't insider trading, Capital One is on the hook for what it's employees do. Chipotle or whoever could sue CapOne for breach of contract, and CapOne could sue these guys for that. So even if we decide as a society it's not criminal, another group of guys doing the same thing somewhere else might not get to keep all that sweet dough they made purely on civil grounds. I don't know. It's interesting.
[1] https://supreme.justia.com/cases/federal/us/521/642/case.htm...
I'm not sure that is the exact same thing. These guys had to do some research.
Consider, for example, a large pork supplier. If they see orders from Chipotle trend up by (say) 30% over a quarter, are they free to buy call options on CMG? What if you're a feed supplier to this pork supplier, and you know that Chipotle is their biggest customer, and see that this pork supplier's demand for feed has gone up by 30%; can you then buy call options?
This is a gray area, and I'm not sure the line is really that clear cut.
Your examples are not novel issues, certainly there would be case law on point. I can not cite the case law, but I would tend to believe your examples fall within the classification of insider trading (i.e. trading based on a companies non-public information). I know for example as an attorney if I am working on a merger and I buy/sell stock or inform someone else who buys/sells the stock that would be insider trading.
Generally, think of the Martha Stewart allegations, she was informed by a friend/corporate officer that the FDA was rejecting the companies cancer drug. Based on that non-public information her brother sold the stock and then the FDA decision became public and the stock dropped. It is a stretch from your hypothetical, but the non-public information paradigm would seemingly still apply.
Apparently, you can. Read the following: http://sloanreview.mit.edu/article/when-is-it-legal-to-trade...
Under the legal theory of Respondeat Superior, Capital One would generally be liable for acts of its employees. However, liability will generally not extend to the employer when employees are acting outside the scope of their employment. Certainly if the employees were breaking the law it can be presumed they were acting outside the scope of employment, but even if it is found the employees were not breaking the law it would appear they were acting outside the scope of their employment. In short Capital One would have affirmative legal defenses.
That being said, I highly doubt Capital One is "on the hook" because they were breaking company rules and would have been terminated. Further, I am pretty sure the employees were breaking several laws as well as Capital One's policy (at least according to the contracts I signed and seminars I had to sit through), Capital One would not be directly responsible if they attempted to prevent such behavior (which they do).
In this case, Capital One could only be on the hook if they benefited or distributed individuals data, they actually did neither. They had only general sales data and only the two employees illegally made money.
I heard of people at a renowned analyst company who would trade on their findings before release. Supposedly that was all clear as they only used public data.
Even at that, you have, what is essentially insider information on the raw sales data of a company that way since you are essentially seeing the same information that the company itself sees, but just a rather linear proportion of it.If you have the Capital One portion of Chipotle sales, and you know the electronic transaction market share of Capital One cards, and you know the proportion of cash to electronic transactions; I would not be surprised one bit if you could essentially predict the sales figures down to within 1% margin.
I had tons of non public info as an auditor for Arthur Andersen. No way in hell I would have traded on any of that information. Job > Jail.
Additionally, I have worked for companies which prohibited transactions in any derivatives or options and barred shorting the stock.
Personally I agree with Matt that this is a not the usual kind of thing. Clearly Chipotle could anonymize their sales with dummy purchases but the actual numbers would still be there. Like a search engine, for the data stream (credit card charges) to work you are forced to put enough information in the transaction to identify it.
This is also something I see happening with IoT type technologies, when it becomes possible for someone to collect data on their own at thousands of locations for relatively small numbers of $, like the helicopters trying to estimate oil availability, you'll have data streams that can inform economic activity. Imagine something like a cellphone sized thing with a camera that just counts customers using OCV to note blobs at the counter, in a restaurant and texts a tally once an hour. Seems ridiculous but its quite possible to do, and much more cheaply than just repurposing disposed cell phones (although that works too).
Maybe anonymized CC trend data should be made public?
If I pay people to go count the number of customers in line at a representative sample of Chipotle restaurants during lunch every day, compare the results to the previous quarter where I was doing the same thing, and trade on that data, is that illegal? It's nonpublic information. Would my employees be prosecuted for this if they traded on it?
Prosecutors and the SEC attempt to stretch our laws every day. It doesn't mean that they are going to win these cases.
Anyone CANNOT query Capital One's transaction databases, which is the entire problem here. Make the transaction databases public (Bitcoin does) and open the casino to everyone else.
What if Capital One signed an agreement with Visa saying that in exchange for issuing cards, they wouldn't provide the data to stock traders, and would keep it safe?
What if they didn't trade on it, just sold the info to Taco Bell?
Seems like a situations I want to avoid, if I'm Chipotle.
If they did, then obviously these two would be in violation. But I have seen no mention of such an agreement.
Think about the success of the Oatmeal's Exploding Kittens card game on Kickstarter. Aside from the fact that KS's new payment processor Stripe is leeching an astounding amount of money from this campaign (where most pledges have probably been made with cash-like debit cards being charged at the rate of credit cards), there are also implications of aggregation inside that cartel.
How many playing card printing companies out there are eager to handle that order? What if you knew ahead of time who they were going to use, based on that "bank account" info you have to give out when you sign up for a campaign on KS? Private companies leverage this info all the time, and it's much easier for them to get away with it.
This stuff is already being done legally, with certain industries and within certain financial institutions. Just seems like these guys were just caught in an unfortunate legal snag - otherwise they just seem like good traders to me.
Now, using helicopters to track oil tankers - as described in the article - seems to be a blurry line, but in theory anybody could do that and then use that information, it is available to everybody... In principle at least.
This data was entirely private, so I would agree with being not fair.
That said, I hope these guys do not have to go to jail, but are simply forced to pay their gains (or a portion of those) back, maybe as a fine.
Who releases it? Every credit card company? At what frequency?
What if one of the nightly bulk loads into the transaction data warehouse fails with an integrity error, alerts the DWH team, is picked up by a junior person because of a staffing anomaly, who forces it through, and then the next day's data release is off by a couple of million, but the automated trading bots and manual trading desks don't know this, so they trade on this new signal...
....I'm sure you see where I'm going with that.
> You sometimes see insider-trading cases where someone makes like a thousand-percent return...
Am I the only one bothered by the author's writing? Am I getting old? Conversational is ok, I guess (though, Bloomberg?). But these "like"s? They're not 'like customer-support guys', they are customer-support guys (or else they are analysts).
Bah-humbug.
How did they get caught? Did the SEC see this account has 1,800% return on investment over 3 years and investigate who owned it? Or did Capital One discover some odd queries on their production database and report them to the SEC?
What would the downside be for releasing earnings data daily, or even weekly or monthly?
http://www.businessinsider.com/credit-cards-sell-purchase-da...
Brokerages are obligated to flag and report suspicious orders or trades to the [equivalent of] SEC.
If necessary, the SEC then asks the brokerage for more information and investigates it.
The definitions of "suspicious" can be somewhat arbitrary, but those >1000% returns would definitely be considered suspicious. Two people at the same workplace getting abnormally high returns would also be considered a red flag.