> What is the one sentence describing who would use this and why though?
Today's market structure costs institutional investors (and by extension households) at least a trillion dollars annually (and Smart Markets hold the potential to eliminate that loss).
> Today's market structure costs institutional investors (and by extension households) at least a trillion dollars annually (and Smart Markets hold the potential to eliminate that loss).
How does one get to this estimate? That is ~5% of US GDP. Everything else was easy to follow - this seemed high, at least intuitively.
It's a huge number that only makes context when looking at the absurd scale of capital markets globally. BlackRock has written on the cost of liquidity [1]. Unfortunately, much of the institutional research on this topic is in a walled garden, so we plan on publishing on this when we have our own data. Treating it as a Fermi problem, the market cap of US equities is ~50T and 140T notional of US equities traded in 2021. The global market cap is 125T (I don't have trading volumes there). FICC is much larger than equities.
Portfolio returns compound exponentially, so even small inefficiencies matter big time.
You can get to big numbers on global capital markets, for sure. I was wondering whether you a consulting/VC-style estimate given how specific the statements was: "Smart Markets hold the potential to eliminate that loss" of "at least a trillion dollars annually".
How do you think about it? Let's say we expect half the benefit to come from equities.
>> 0.5T / 125 T = 0.004
>> Smart Markets would need to raise portfolio returns by an average of .4% (net of trading costs) annually.
Don't know if this is accurate but the entire online stock brokerage industry revenue appears to be about $14B.
That seems like a significantly lower upper bound to the market size here.
That said, what seems interesting here is to come up in advance with many potential arbitrages, and load them in advance for fulfillment if they occur. Risky but interesting than having to roll your own complex tool for this.
To clarify, 1T isn't what we're claiming as our revenue opportunity; it's what traders are missing out on annually in the form of portfolio returns due to market friction and missed Pareto outcomes.
(FWIW and not that it's the market that we're going after per se—our strategy is mostly blue ocean—the market for US equities electronic execution services across the whole stack of technology, market data, broker algos, etc., is $18B/yr.)
This isn't clear to me - are your buyer's institutional investors? Are they buying your technology to create trade options for their end users i.e. an individual investor? I don't know what an ATS is so I gather that I'm not a direct user of your technology - perhaps I would be an indirect user? Would E-Trade, for example, leverage your technology to provide me with a combinatorial buying option?
An ATS is like an exchange, so we match buyers and sellers. And you guessed correctly that the initial users are institutional investors - or more directly their brokers. So initially we'll have institutions creating and sending in "Expressive Bids" to improve their execution performance, and to express trades they currently can't via plain limit orders.
That said, we'd love to get to the point where E-Trade etc. are offering combinatorial bidding to retail traders, with us on the back end.
Thanks. Obviously I'm not a direct user of your technology and so maybe this is not intended for me but if you could translate your "A" and "B" into a hard, real-life example that I could understand I would be empowered to be an advocate for you. Best of luck to you.
No worries, happy to concretize this: the really easy example would be shoes. How much would you pay for just a right shoe or just a left shoe? A lot less than the pair, since you might not be able to find the other shoe in the right size, condition, etc. Same with the seller - they don't want to be stuck trying offload a single left shoe.
In stocks, A might be a company you invested in and B some ETF that you bought as a hedge for A. What if you sell out of A, and then the price of the ETF drops? There's value in being able to liquidate the full position - the single stock plus the hedge - at once.
There has to be a lot of additional data gathered at the time of 'intent to purchase or sell' - because otherwise your solution eats away a lot of powerful institutions alpha. And without 'novelly' expressive orders, there's no new place for them to go... that Trillion dollars doesn't just evaporate in today's world.
It's more like ensuring an opportunity to leverage the information that's already being gathered. As it stands, PMs have to construct concrete portfolios because they need to send the trading desk specific instructions on what to buy and sell. The portfolio they ship out for execution is effectively a low dimensional projection of a high dimensional decision process. That process has extensive substitutability (sizing and substitutability if something is going to be more or less expensive to execute than transaction cost models predicted), but there's no way to communicate that in today's trading workflows. That results in the market missing out on Pareto outcomes.
We've already seen this in sourcing markets [1]. Capturing more information at the time of bidding resulted in massive (40-60%) efficiency gains for both sides of the market.
Oh yes, I see the problem statement and agree from a PM perspective this is quite good.
That said, there are a lot of people who make good money making inferences from these current concrete dynamics - in some sense, you're just forcing the market to innovate (this is good).
I always like to know who I'm asking to change when building products -- and this one is a very interesting (read: fun and potentially lucrative) set of actors.
I am aware how big the markets are. I am also aware that transaction costs are miniscule.
By your own data above, if typical fees are $0.0009 per share traded, $1tr in costs implies notional value of instruments traded each year of approx $1x10^17, assuming average price of $100 / share.
Ah, agreed, but we're talking about two different things—direct transaction costs versus allocative inefficiency/missed Pareto outcomes. OneChronos is about unlocking Pareto efficiencies—situations in which two or more parties can trade to mutual benefit. An easy example is a (scaled down in price differences, scaled up in size) version of the complements example above, e.g., an ETF arb trading the basket against the underlying with a small tolerance for tracking error. An institution that can take the basket or the underlying as a hedge or as an investment position can interact with the arbitrager, creating economic gains for both parties in the process. At institutional scale, efficiency gains measured in bps and compounded exponentially add up.
> Today's market structure costs institutional investors (and by extension households) at least a trillion dollars annually (and Smart Markets hold the potential to eliminate that loss).
What is the subject and the verb for the problem you are solving and for whom? This is too vague.
How does this sit with other innovations such as all to all trading in OTC markets that are designed to match many buyer/sellers at, for example, a mutually beneficial mid price?
The products that still trade predominately {OTC, bilateral, non-electronic} do because non-price factors, trading conventions, and counterparty risk make it difficult or impossible to trade on a central exchange. Expressive Bidding and a mechanism that allows for matching market dynamics (OneChronos) will enable electronification and more active trading in these markets. Ten years ago, I would have said that there were commercial headwinds against this (dealers wanting to trade bilaterally). With banks increasingly focused on repeatable and less variable trading revs, such is no longer the case. The unsuitability of such products for double auctions is the limiting factor.
Thanks. It is worth noting a lot of traditional OTC products since 2008 crisis have moved to either electronic (exchange, ATS, MTF etc) with CLOB/RFQ/Auction style of execution and in some causes that's coupled with central clearing. A lot of this has come from regulation - DoddFrank, MiFID2 and its still on-going.
The most interesting aspect of this is that its enabled non-dealer <> non-dealer trading via certain venues.
This is a significant tailwind for the next wave of electronification that we're hoping to advance. US swap dealers, for example, now have a SEF reporting requirement, but most of the pre-trade is still in the screens. A Smart Market that allows dealers to control for non-price factors could change that.
That it is!
> What is the one sentence describing who would use this and why though?
Today's market structure costs institutional investors (and by extension households) at least a trillion dollars annually (and Smart Markets hold the potential to eliminate that loss).