This is part of the reason I come to HN. A detailed post on a problem (and sometime solution) I am unaware of.
A newbie question. I understand how a pair tickets to say the superbowl would be more valuable than a single ticket, people want to go with their friend. Is there a practical example for equities? I will buy 100 shares $FB at $250 only if I can also get 50 shares of $SNAP at $35 at the sametime? If I can't get that combo, I will only pay $240 for $FB?
Is this aimed at equities that have less liquidity?
Great question! Complements in capital markets typically aren't as strong as other markets like event tickets. On the complements side, hedges are a good example. Market makers like banks are willing to quote much larger sizes for hedged transactions, e.g., an institution that wants to buy a large block of equity in one company while selling others with similar qualities as "factor hedges." The net notional changing hands is roughly the sum of the parts. Still, there's a big price difference between doing this trade atomically and as a series of transactions where the market maker has to "wear" the risk for some period.
Substitutes in capital markets are ubiquitous. There might be hundreds of candidate hedges in the example above, but given how trading workflows are, there's no way to communicate that amongst market participants. A market maker has no way of knowing if someone wants to buy SNAP (and thus potentially has market-moving information about it) or if they're using it as a hedge for a short position and would substitute something that the market maker wants to gross down on (and offer a more aggressive price because of that). As such, market making is a game of pricing under risk and uncertainty. Combinatorial auctions eliminate much of the uncertainty.
Do you envision scenarios where this new expressiveness is used in strategic but not market efficient ways? For instance, in the SNAP example, presumably the correlate to price improvement when purchasing as a substitutable hedge is a price premium when purchasing a specific equity, as market participants can deduce -- in a way that they previously could not -- that there's something inherent to SNAP that one (or the market) values. I don't know if it's possible under the mechanics of your ATS, but this seems to produce an incentive to obfuscate such a purchase of SNAP specifically, potentially in ways that detract from market efficiency. Am I off base? To put it more generally I wonder whether this higher dimensionality might not lead to more sophisticated game-theoretic posturing, rather than less.
This is a great question, and the full answer involves lots of mechanism design nuance. The short answer is that we have uniform clears per trading instrument, and bids are sealed, so there's no direct signaling game that would allow someone to try to pass off an alpha trade as a hedge. We plan to introduce a mechanism that will enable signaling through tokens (opaque identifiers). Bidders can attach whatever token they want, and other bidders can price discriminate against tokens (change their prices for, refuse to trade with, trade exclusively with) based on historical post-trade outcomes that we make known via an immutable audit trail. Participants can create and use tokens freely, so it's not segmentation. Instead, it's a means of inducing a repeated play game and a market for reputation.
The founders can answer better, but a lot of traders are not just looking at individual stocks but rather packages of stocks together.
Say you want to invest in the health + tech space, but there's some risk that covid ends and gyms come roaring back. So then you want to minimize the risk by putting some money in the gym industry as well -- getting an entire package of peloton, apple, and 24hourfitness stock is actually worth more to you than the individual stocks on their own.
Comments
This is part of the reason I come to HN. A detailed post on a problem (and sometime solution) I am unaware of.
A newbie question. I understand how a pair tickets to say the superbowl would be more valuable than a single ticket, people want to go with their friend. Is there a practical example for equities? I will buy 100 shares $FB at $250 only if I can also get 50 shares of $SNAP at $35 at the sametime? If I can't get that combo, I will only pay $240 for $FB?
Is this aimed at equities that have less liquidity?
Great question! Complements in capital markets typically aren't as strong as other markets like event tickets. On the complements side, hedges are a good example. Market makers like banks are willing to quote much larger sizes for hedged transactions, e.g., an institution that wants to buy a large block of equity in one company while selling others with similar qualities as "factor hedges." The net notional changing hands is roughly the sum of the parts. Still, there's a big price difference between doing this trade atomically and as a series of transactions where the market maker has to "wear" the risk for some period.
Substitutes in capital markets are ubiquitous. There might be hundreds of candidate hedges in the example above, but given how trading workflows are, there's no way to communicate that amongst market participants. A market maker has no way of knowing if someone wants to buy SNAP (and thus potentially has market-moving information about it) or if they're using it as a hedge for a short position and would substitute something that the market maker wants to gross down on (and offer a more aggressive price because of that). As such, market making is a game of pricing under risk and uncertainty. Combinatorial auctions eliminate much of the uncertainty.
Do you envision scenarios where this new expressiveness is used in strategic but not market efficient ways? For instance, in the SNAP example, presumably the correlate to price improvement when purchasing as a substitutable hedge is a price premium when purchasing a specific equity, as market participants can deduce -- in a way that they previously could not -- that there's something inherent to SNAP that one (or the market) values. I don't know if it's possible under the mechanics of your ATS, but this seems to produce an incentive to obfuscate such a purchase of SNAP specifically, potentially in ways that detract from market efficiency. Am I off base? To put it more generally I wonder whether this higher dimensionality might not lead to more sophisticated game-theoretic posturing, rather than less.
This is a great question, and the full answer involves lots of mechanism design nuance. The short answer is that we have uniform clears per trading instrument, and bids are sealed, so there's no direct signaling game that would allow someone to try to pass off an alpha trade as a hedge. We plan to introduce a mechanism that will enable signaling through tokens (opaque identifiers). Bidders can attach whatever token they want, and other bidders can price discriminate against tokens (change their prices for, refuse to trade with, trade exclusively with) based on historical post-trade outcomes that we make known via an immutable audit trail. Participants can create and use tokens freely, so it's not segmentation. Instead, it's a means of inducing a repeated play game and a market for reputation.
The founders can answer better, but a lot of traders are not just looking at individual stocks but rather packages of stocks together.
Say you want to invest in the health + tech space, but there's some risk that covid ends and gyms come roaring back. So then you want to minimize the risk by putting some money in the gym industry as well -- getting an entire package of peloton, apple, and 24hourfitness stock is actually worth more to you than the individual stocks on their own.