This is an options trading pit, but otherwise has nothing to do with anything.

Photographer: Scott Olson/Getty Images

Tricky Twins Spoofed Trading Computers

Matt Levine is a Bloomberg View columnist. He was an editor of Dealbreaker, an investment banker at Goldman Sachs, a mergers and acquisitions lawyer at Wachtell, Lipton, Rosen & Katz and a clerk for the U.S. Court of Appeals for the Third Circuit.
Read More.
a | A

I have in the past praised the Securities and Exchange Commission's enforcement actions for their fine sense of humor and dramatic tension, but occasionally one feels that they stretch the bounds of plausibility. Take today's enforcement action against Behruz Afshar, Shahryar Afshar and Richard Kenny. I can believe that a pair of lovable rogues teamed up with "their friend and former broker" to run some option-trading scams. But it goes a bit far that the lovable rogues are also twin brothers. And it is completely over the top that the centerpiece of their alleged scam consisted more or less of the twins pretending to be each other. Come on! That is the plot of a teen comedy, not a securities enforcement action.

But here we are. The twin-switcheroo scam, as set out in the SEC order, is delightfully simple and goofy. Options exchanges classify orders as either "professional" or "customer," and treat the "customer" orders better, giving them execution priority and charging them lower fees as part of an effort "to attract retail order flow and level the playing field for retail investors over market professionals." The Afshar brothers ran a professional options trading operation, both in the colloquial sense (they "employed at least four traders" who traded using their capital), and in the technical sense used by options exchanges, which involves, roughly speaking, submitting more than 390 options-trading orders per day.  (They averaged somewhere north of 3,000.)

But they found a clever, or stupid, or so-stupid-it's-clever way to avoid being treated as professionals. The test for professionals was done on a quarterly basis: If you exceeded 390 trades per day for a month in one quarter, you are a professional the next quarter. So the trick was, allegedly:

  • They set up two companies, one in the name of one twin, one in the name of the other twin.
  • They did essentially all of their trading in the first company, Fineline Trading, owned by Behruz Afshar, for a quarter.
  • The next quarter -- when Fineline would be a "professional" -- it took a time-out, and they did essentially all of their trading through the other company, Makino Capital, purportedly owned by Shahryar Afshar and "named after a sushi restaurant in Las Vegas, Nevada, that Shahryar frequented."
  • The next quarter, when Makino would be a "professional" trader, they switched back to doing their trading through Fineline, which was now back to being a "customer" by virtue of its three-month time-out.

So for instance, in the fourth quarter of 2011, it was Fineline's turn to be the "customer," and it placed 248,300 orders. Makino, the "professional," placed 80. The next quarter, they flipped: Fineline was now the "professional" and placed only 466 orders (for the quarter, not per day), while Makino was the "customer" and placed 214,541.

There is not much to say about this except that it is silly and obvious and you're not supposed to do it. I guess one other thing to say about it is that it was bizarrely lucrative: It allegedly brought them "over $2 million in transaction fees wrongly avoided and higher rebates wrongly received," in addition to the advantage of gaining priority in executing their trades.

But there's another alleged scam here, which the SEC calls "spoofing," and which maybe it is. It's weird spoofing though. Classic spoofing is like:

  1. Widgets are trading at $101.
  2. I place 100 orders to buy widgets for $100.
  3. I place an order to sell one widget for $102.
  4. Everyone thinks that there are a ton of buyers for widgets and race to buy ahead of them, pushing up the price to $102.
  5. I sell my widget for $102.
  6. I cancel my 100 buy orders, which "were not bona fide orders because they were not intended to be executed."
  7. I've made an extra $1 on my sale.

As we've discussed before, this seems dumb in theory, both because it is risky (if someone sees my 100 buy orders and sells me 100 widgets, I now own a lot of widgets I don't want!) and because it only works if other traders are idiots and interpret my fake orders as a sign of shifting demand that they should follow blindly. But being dumb in theory does not stop it from working in practice, at least some of the time, and it seems to happen a fair amount and be bad

The Afshars' and Kenny's alleged spoofing was different. It allegedly went something like this :

  1. Options were bid at $7 and offered at $9.
  2. Fineline, acting as the good twin, put in 18 hidden "all-or-nothing" orders to sell 10 option contracts each at $8 (i.e. a total of $1,440 worth). Under the rules of the options exchange (Nasdaq OMX PHLX, once the Philadelphia Stock Exchange), this type of order is not displayed to other traders, and can only execute in full. So if someone put in an order to buy one option for $8, the orders to sell 10 would silently wait to execute until more buy orders came in.
  3. Makino, acting as the evil twin, then put in one public order to buy one option for $8.
  4. That one-option order wouldn't execute against the hidden all-or-nothing orders, because it wasn't big enough.
  5. But lots of other people saw that the options were now bid at $8, and rushed to put in big buy orders at the same price.
  6. And those orders executed against Fineline's 180 options worth of sell orders, so Fineline ended up selling 180 contracts at $8 for a total of $1,440. (This is better than Fineline would have done without the alleged spoofing: It would presumably have had to sell at the bid, $7 per contract, or $1,260 total.)
  7. Makino's one buy order didn't get executed, and Makino cancelled it.

Unlike in classic spoofing, here the big orders are the ones that were intended to (and did) execute, and the tiny one-contract order is the one that was allegedly a spoof. So Fineline/Makino aren't accused of creating the illusion of massive demand. They are accused of creating the illusion of a wee bit of demand, the absolute minimum quantity of demand. 

Why would that trick anyone? Why would options traders, seeing an order to buy one contract, rush to put in orders to buy (at least 180 contracts? The answer comes in two parts, basically:

  • Those traders were algorithms, and
  • They wanted to get paid for providing liquidity.

The way PHLX works (and the way that many exchanges work) is that it charges traders a fee for "taking" liquidity (by, say, buying options for $9 when they are offered at $9), and pays them a rebate for "providing" liquidity (by, say, offering to sell options at $8, and then having someone come in and buy them for $8.) That is, it pays traders for posting bids and offers that then trade. So computer algorithms, seeing that the best bid for these options had changed from $7 to $8, rushed to bid $8 for them so that, if someone wanted to sell at $8, they would be first in line to trade and get paid the rebate. There was no analysis of changing demand here; there was just a desire to be first in line to be paid the rebate. 

But they were tricked, because actually there were already waiting sell orders that would provide liquidity at that price, and so their new bids executed immediately and were charged the fee. And Fineline, whose hidden orders "provided liquidity," got paid the rebate -- in this case $46.80, or $0.26 per contract.

A $46.80 rebate is not really much to write home about, compared to the $2 million that they allegedly made on the "customer"/ "professional" scam, but a buck is a buck, and I suppose they did this a lot. Also, Fineline's purpose in this alleged spoofing was presumably not just to earn the rebate, but also to actually sell all those options. Again, unlike in traditional spoofing, the big sell orders were the ones they wanted executed. And if they could, by bidding to buy a single contract, trick a bunch of algorithms into buying options from them at a higher price, they'd do it all day long.

Unfortunately, they'd also instant message each other about it all day long:

Behruz: i love getting ‘em with the hidden x-a anyhow . . . bring me such pleasure and joy . . . at times I roll over laughing
Sub-Account Trader: haha
Sub-Account Trader: yes
Sub-Account Trader: it’s a nice feeling
Sub-Account Trader: I love it when I use nasd [to place the small-lot order] to bring the bid/offer in and then get em

And:

Behruz: … I’d rather do the phlx aon on that and bring the offer in
Sub-Account Trader: ok . . .
Sub-Account Trader: 10 50 lots?
Behruz: yes
Sub-Account Trader: ok done
Behruz: when you’re done we’ll cancel the 1 lot
Behruz: that’s my offer
Sub-Account Trader: k
Behruz: come here kitty kitty
Behruz: they are afraid :)

And:

Sub-Account Trader: but i offered em on phlx aon and brought the bid in on phlc [sic]
Sub-Account Trader: i LOVE doin that lol …
Kenny: that is the finest... bringing in the bid or offer
Kenny: makes you feel proactive!
Sub-Account Trader: makes me feel like i was smarter than the computer haha
Kenny: true dat

He was definitely smarter than the computers! The computers saw a one-contract bid and joined it in huge size.  They thought they wanted to trade at $8, and then they did trade at $8, and they regretted it not because they were tricked about value or demand, but because they were tricked about whether or not they'd be getting a rebate. That is presumably why the SEC emphasizes that Fineline/Makino's alleged spoofing got them rebates (which it did), rather than that it moved the price (which it also did, and which was worth more money): It's hard to sympathize with the algorithmic trading firms for changing their price based on just one bid for one contract. Those firms can't really claim to have been deceived about the value of those options, or even about the amount of demand for them. (One contract worth of demand at $8 is not really much more than zero contracts worth of demand at $8.) All they can claim to be deceived about is the supply of rebates -- and that's more of a function of PHLX's (legal, disclosed) hidden all-or-nothing order type than it is a function of spoofing.

If you believe the allegations in the SEC's order, then Kenny and the Afshars clearly intended to manipulate the market, and succeeded in doing so. But despite those chats, they don't exactly come across as evil masterminds putting one over on a bunch of innocents. They come across as guys trying to outsmart computers in a race to earn trading rebates. The problems here, if they are problems, are about the structures and incentives involved in liquidity rebates and hidden order types. Kenny and the Afshars were just proactive about exploiting them.

  1. More specifically, from the SEC order:

    A non-broker-dealer person or entity that places more than 390 orders in listed options per day (on average) — whether executed or not — on any listed options exchange during any calendar month in a quarter will be designated as a “professional” for the next quarter. Conversely, a “customer” is a non-broker-dealer person or entity that does not exceed the 390-order threshold for each calendar month in a quarter.

  2. The SEC says that "in fact Behruz had an ownership interest in both companies," though it doesn't allege that Shahryar owned any of Fineline. Behruz seems to be, in the SEC's eyes, the ringleader here.

  3. Specifically, as the SEC says:

    All orders for multiple accounts beneficially owned or controlled by the same person or entity, and all sub-accounts of a person or entity’s master account, must be aggregated when determining whether the 390-order threshold has been exceeded by that person or entity.

    If in fact Behruz Afshar controlled both Fineline and Makino, then he should have aggregated them, and by pretending that Shahryar was the sole owner of Makino the twins evaded that requirement. 

  4. Oh fine, there is one more thing to say about it, which is, look at this alleged instant message exchange between Behruz Afshar and one of their contracted traders:

    Behruz: opening bank accounts, trading accounts, etc ….
    Behruz: i think we have about 10 llc here all tied to sushi names
    Sub-Account Trader: LOL
    Sub-Account Trader: you and your fish man
    Sub-Account Trader: too funny
    Behruz: i’m not even the one that came up with these damn names
    Sub-Account Trader: haha who made em up
    Behruz: my bro and rich

    Really naming your trading firms after sushi restaurants is as reasonable as anything else.

  5. That's how the SEC describes the alleged spoof orders in this case. "Spoofing" is to some extent colloquially defined, though it is now sort of codified in the commodity (not stock options) markets as "bidding or offering with the intent to cancel the bid or offer before execution." That semi-definition is a bit troubling; as the law firm Milbank, Tweed, Hadley & McCloy has pointed out

    The CEA’s statutory definition (“bidding or offering with the intent to cancel the bid or offer before execution”) arguably leaves additional room for interpretation and argument in that it could encompass legitimate and common practices.

    i. For instance, many traders hedge with the use of stop loss and other types of orders that are put in place as a precaution but that the trader hopes and expects to unwind without execution.

    ii. Moreover, there is an argument that any order genuinely exposed to the market (and thus to a risk of execution) is inherently legitimate. In this way of thinking, spoof orders are unlike (for example) wash sales or other classic market manipulation techniques that create the illusion but not the reality of a change of ownership. That the spoofing trader intended to cancel his orders does not change the fact that his orders each represented actual – and potentially actionable – market activity.

  6. This is based on the SEC's "Specific Spoofing Example" on page 14 of the order, though I have omitted and rearranged some steps (in particular, they started offering only 120 contracts, and later raised the offer to 180). The options were $11 strike November 2012 Ford options. An options contract is for 100 shares, and prices are expressed in cents per share, so for my own convenience I have just multiplied the prices by 100 to get the price in dollars per contract instead of cents per share. If you are an options trader I am sure this drives you nuts.

  7. At this point Fineline was the "customer," giving it priority and saving on fees, and Makino was the "professional," used for the spoof-ish order "presumably to avoid raising any suspicions of a wash trade and to decrease the likelihood of an execution (due to the lower priority of 'professional' orders)." Both Makino's and Fineline's orders were allegedly being placed by Kenny at this point.

  8. I don't entirely understand why it didn't execute; perhaps enough "customer" orders got ahead of it to prevent execution? It doesn't matter much economically; Fineline is buying 180 contracts, so it's pretty irrelevant if it buys one of them from itself. I'm not sure it matters much legally, either. If Makino's one-lot trades never executed against Fineline's big trades, they look like spoofing, which is illegal. If they did execute, though, they look like wash trades, which are also illegal. Either way, the actually important question is whether they were illegally tricking other market participants, not whether they were intended to execute. That is, the fraud here is, or isn't, manipulation; the intent to execute is mostly beside the point.

  9. The SEC says "other market participants submitted buy orders at $0.08 in sufficient quantities to completely fill all eighteen AON orders," meaning at least 180 contracts but quite possibly more. The avalanche of orders occurred over about 40 seconds, "between 9:56:03 and 9:56:43."

  10. The SEC actually quotes the head trader at an algorithmic trading firm, who "emailed individuals at the PHLX about his concerns":

    [W]e have encountered some strange trading behavior recently on PHLX. It appears like we are trading against hidden AON orders, and we believe that someone might be manipulating the market. Here is one example from today that we found in GE, all timestamps are CST. Before the trades happened the PHLX BBO was .07 bid at .08. The volume on the .07 bid was 1 contract. We tried to join the .07 bid for a size of 130 contracts, and we immediately traded 13 times, each trade was for 10 contracts. We are particularly concerned that a market participant is entering an order to buy 1 contract at .07 (not AON), and then they are layering many orders to sell at .07 using an AON contingency.

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.

To contact the author of this story:
Matt Levine at mlevine51@bloomberg.net

To contact the editor responsible for this story:
Tobin Harshaw at tharshaw@bloomberg.net