6 ms·

As the article hints at, Medallion isn't really an investment fund so much as a pool of capital which is employed in the business of providing tactical liquidit

As the article hints at, Medallion isn't really an investment fund so much as a pool of capital which is employed in the business of providing tactical liquidity to markets. This is a business where one's competitive advantage rests upon their technological advantage, and Renaissance has been exceptionally adept at building and maintaining that advantage. The EMH is a theoretical concept that must be tempered to account for the frictions inherent in reality. As my PhD advisor likes to say, "all models are wrong; some are useful." Liquidity is a major friction in real-world markets, and providing liquidity (i.e. a means of moving capital into and out of specific assets) is a service that is compensated accordingly. In other words, I think Renaissance is simply better than anyone in the world at finding the markets where traders are willing to pay the highest premiums for liquidity, and providing it in a timely and measured dose that ensures they skim the cream off the profit opportunity. There are a limited amount of such opportunities, so they must limit the amount of capital employed to maintain the high rate of return.

RookyNumbas 7y ago
I'm not sure how accurate this tactical liquidity picture is, at least for much of medallion's history. For a good portion of their existence they called in all their trades twice a day. Them being fairly late to the automated trading game was one of the the surprises from the book for me. Any "HFT" as mentioned in the article is not done by renaissance, but through execution services. The big picture seems to be statistical arbitrage done with extreme precision and leveraged to a massive scale. Add in some quasi-legal agreements with banks, and likely tax-fraud, they are able to do this at a magnitude no one else can.
- dcolkitt 7y ago
  Providing liquidity doesn't necessarily mean HFT. It doesn't even necessarily mean classical market making. Here's how I like to think about liquidity. It's the market's ability to absorb random imbalances in the non-informed order flow (while still correctly engaging in price discovery for informed order flow). In other words it looks a lot like what economists call price elasticity. Having a lot of participants who are very price sensitive makes prices stable by keeping demand responsive to small fluctuations. On the very short horizon that's done by market makers and HFT participants. They're quoting a tight two-sided market so that as orders arrive the price remains stable. But HFTs keep very small inventories[1]. They're unable to absorb the typical imbalance seen on the scale of hours, or even minutes. There exists a secondary class of participants, usually grouped under the catchall category of "stat-arb", that exist to fill this need. They're still trading directionally, but that direction is very sensitive to recent price moves on the scale of hours or even minutes. Moreover they tend to overwhelmingly trade in a mean-reverting manner.[2] Effectively these participants "provide liquidity" in the sense that they're stepping in and absorbing imbalances in the natural order flow. [1]https://www.cftc.gov/sites/default/files/idc/groups/public/@economicanalysis/documents/file/oce_riskandreturn0414.pdf https://www.cftc.gov/sites/default/files/idc/groups/public/@... [2]https://web.mit.edu/Alo/www/Papers/august07.pdf https://web.mit.edu/Alo/www/Papers/august07.pdf
  - RookyNumbas 7y ago
    I don't disagree with anything you've written. But if you name a strategy, I can make a case for why it's providing liquidity. On the one extreme, some firms truly are paid to make liquidity. This is the basis of the maker-taker model. I think Rentech is at the other end of the spectrum.
onetimemanytime 7y ago
Maybe...still 66% every single year for 20+ years is odd. Since there's so much money involved, nothing stops funds to invest tens of billions to duplicate their success.
- papln 7y ago
  In any investing environment, the tactic with the best returns are only available up to some limit of invested money before the "raw material" is consumed.
macspoofing 7y ago
>Renaissance is simply better than anyone in the world at finding the markets where traders are willing to pay the highest premiums for liquidity, and providing it in a timely and measured dose that ensures they skim the cream off the profit opportunity. Why are they better? This still requires the same kind of explanation as 'they are simply better at investing than anyone in the world'
- peteradio 7y ago
  It could be an incumbent position that is just not possible to overthrow.
  - onetimemanytime 7y ago
    Ok, new Medallion-like fund accepts lower returns, meaning its partners get better deals then under Medallion. Why wouldn't they leave Medallion? This is since 1988 apparently. I don't buy that they are the "smartest people" or that they are doing something so amazing that others don't get.
    - papln 7y ago
      Medallion is for employees of Renaissance. It doesn't have customers/partners. If you accept that Renaissance is good at what it does, it's simple to posit that Renaiisance reserves its most profitable activities for its own employees as preferred customers/partners, as a compensation vehicle and a PR/recruiting tactic to make Renaissance look more impressive. Medallion is like Google's "X", in that sense. This explanation shows why Medallion's high returns are possible, and also why Medallion not actually better than Renaissance's overall average performance (that is, could not be replicated as an indepdent spin of without the rest of Renaissance to extract value from), beyond an acconting illusion to make ordinary wages look like investing. It's not at all surprising that this could be how Medallian works, since that's just an application hedging, which is Renaissance's core business.
      - jkaptur 7y ago
dogman144 7y ago
The article and Mercer's explanation basically explains Virtu's approach, and therefore makes sense wrt known performance vs. methodology from a more public company (Virtu)
- hogFeast 7y ago
  Virtu is doing something totally different. You are comparing a gas station with O&G exploration.

RookyNumbas 7y ago

I'm not sure how accurate this tactical liquidity picture is, at least for much of medallion's history. For a good portion of their existence they called in all their trades twice a day. Them being fairly late to the automated trading game was one of the the surprises from the book for me. Any "HFT" as mentioned in the article is not done by renaissance, but through execution services. The big picture seems to be statistical arbitrage done with extreme precision and leveraged to a massive scale. Add in some quasi-legal agreements with banks, and likely tax-fraud, they are able to do this at a magnitude no one else can.

dcolkitt 7y ago
Providing liquidity doesn't necessarily mean HFT. It doesn't even necessarily mean classical market making. Here's how I like to think about liquidity. It's the market's ability to absorb random imbalances in the non-informed order flow (while still correctly engaging in price discovery for informed order flow). In other words it looks a lot like what economists call price elasticity. Having a lot of participants who are very price sensitive makes prices stable by keeping demand responsive to small fluctuations. On the very short horizon that's done by market makers and HFT participants. They're quoting a tight two-sided market so that as orders arrive the price remains stable. But HFTs keep very small inventories[1]. They're unable to absorb the typical imbalance seen on the scale of hours, or even minutes. There exists a secondary class of participants, usually grouped under the catchall category of "stat-arb", that exist to fill this need. They're still trading directionally, but that direction is very sensitive to recent price moves on the scale of hours or even minutes. Moreover they tend to overwhelmingly trade in a mean-reverting manner.[2] Effectively these participants "provide liquidity" in the sense that they're stepping in and absorbing imbalances in the natural order flow. [1]https://www.cftc.gov/sites/default/files/idc/groups/public/@economicanalysis/documents/file/oce_riskandreturn0414.pdf https://www.cftc.gov/sites/default/files/idc/groups/public/@... [2]https://web.mit.edu/Alo/www/Papers/august07.pdf https://web.mit.edu/Alo/www/Papers/august07.pdf
- RookyNumbas 7y ago
  I don't disagree with anything you've written. But if you name a strategy, I can make a case for why it's providing liquidity. On the one extreme, some firms truly are paid to make liquidity. This is the basis of the maker-taker model. I think Rentech is at the other end of the spectrum.

onetimemanytime 7y ago

Maybe...still 66% every single year for 20+ years is odd. Since there's so much money involved, nothing stops funds to invest tens of billions to duplicate their success.

papln 7y ago
In any investing environment, the tactic with the best returns are only available up to some limit of invested money before the "raw material" is consumed.

macspoofing 7y ago

>Renaissance is simply better than anyone in the world at finding the markets where traders are willing to pay the highest premiums for liquidity, and providing it in a timely and measured dose that ensures they skim the cream off the profit opportunity. Why are they better? This still requires the same kind of explanation as 'they are simply better at investing than anyone in the world'

peteradio 7y ago
It could be an incumbent position that is just not possible to overthrow.
- onetimemanytime 7y ago
  Ok, new Medallion-like fund accepts lower returns, meaning its partners get better deals then under Medallion. Why wouldn't they leave Medallion? This is since 1988 apparently. I don't buy that they are the "smartest people" or that they are doing something so amazing that others don't get.
  - papln 7y ago
    Medallion is for employees of Renaissance. It doesn't have customers/partners. If you accept that Renaissance is good at what it does, it's simple to posit that Renaiisance reserves its most profitable activities for its own employees as preferred customers/partners, as a compensation vehicle and a PR/recruiting tactic to make Renaissance look more impressive. Medallion is like Google's "X", in that sense. This explanation shows why Medallion's high returns are possible, and also why Medallion not actually better than Renaissance's overall average performance (that is, could not be replicated as an indepdent spin of without the rest of Renaissance to extract value from), beyond an acconting illusion to make ordinary wages look like investing. It's not at all surprising that this could be how Medallian works, since that's just an application hedging, which is Renaissance's core business.
    - jkaptur 7y ago

dogman144 7y ago

The article and Mercer's explanation basically explains Virtu's approach, and therefore makes sense wrt known performance vs. methodology from a more public company (Virtu)

hogFeast 7y ago
Virtu is doing something totally different. You are comparing a gas station with O&G exploration.