David E. Shaw and the Ultimate College Hedge
nymag.com
nymag.com
I grew up in Israel near the Technion, our MIT equivalent which produced a few Nobel Prize winners. It's not a big deal to get in, if you have poor high school scores you take a year long prep course, classes are hard but most people push through. Having a Technion degree might give you a slight edge getting your first job but that's that - you'll be judged by your performance moving on, not on your alma mater.
When I lived in the US for a few years, I couldn't understand how people allocate such importance to a person's high school performance (which is basically what determines their alma mater in the best case scenario, before family donations etc). To me it signals a risk-averse mentality that fears taking a chance on a potential hire.
One day he pitches to the gnomes at JP Morgan, and they tell him they aren't interested. But... since you are here anyway, do you have any ideas about solving these kinds of problems with computer science? And they shared some of their issues in the early 80s. Shaw went back to his office and a week later, they called and offered to add a zero to his salary if he would come work for them. Well, yes!
Shaw described the basic concept of statistical arbitrage, where each effect on a predicted price added some variance. If someone knew enough to more accurately reduce the variance on the models, they could essentially do arbitrage while other market participants could not be sure the transaction was riskless. He described the early experiments at JP Morgan, where they needed to calibrate a parameter in their model, so one day JP Morgan did 2500 trades one way, and 2500 trades another way, and by measuring the market response, Shaw had his parameter. He estimated JP Morgan spent about $5 million or so to measure that parameter, and then used that number for years for a many, many fold return on investment. He also said that the next market participant trying to measure that parameter had to spend $100 million to get the same parameter.
He made it clear he was not looking for investors, that they push the effects as much as they can without distorting the market and ruining their advantage. Over time, the ideas leak out of the firm and stop working, but his team is always coming up with new ideas. He claimed a riskless return year after year of about 15% at the time. He was there at Cornell trying to recruit the top CS students. He said that those quants could sit in a room alone, working on technical problems, never have to manage anyone or do anything business related, and share in the profits of their ideas, with returns for some topping $1 million/year.
almost exactly the story but with a few additional twists, specifically having access to detailed information on client trading flow. Many of the kings of statarb had a past with a direct link to the electronic order desks of major brokerages, PDT and Shaw with MS and the list goes on worldwide. This gives them the insight needed to build models that can detect statistically the difference between trending informational flow and random fleeting uninformed noisy flow which reverts.
As of today these statarb incumbents will be fairly difficult to challenge as they have invested in having superior expensive raw informational data sets and additional the size of their trading gives them private information. However their costs are likely too high so the new game will be to find 10x cheaper approaches in costs as the techniques and math are not as advanced as they claim. The ugly truth is most of their PhDs are twiddling their thumbs and they are effectively buying up anyone interested in the field in hopes to deprive smaller upstarts from talent at an affordable price.
Founder of mega titan firms like DE Shaw will certainly be exceptional individuals and will obviously cary with them huge egos which will blind them to the degree which luck and being in the right place at the right time played.
It's a big puzzle to me why this doesn't happen more often. In the past it did. Fighting inequality or discrimination, especially at the professional level, started when someone said, "Okay, I'm just going to make a school for my own then." My favorite example of this is the dismantling of Jewish exclusion at New York medical schools, which happened by creating new medical schools that did not discriminate. [1]
The idea that prestige is entrenched, basically, is wrong. Maybe that idea is the main obstacle to education reform.
Although maybe this is already happening for rich people. Alt Schools come to mind, although its story has come to an end.
I don't think it's possible for the superrich, as a class, to guarantee admissions for their kids at elite universities, because there's only so much space. So why haven't they built their own school?
[1] http://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC18080...
It is about signaling, not education.
And for that, you need time for the college "image" to become elite.
I've taken community college CS, and I've taken CS50 (online). I did not find this to be the case.
Leland Stanford's reason for starting Stanford in his son's name was to honor his son's memory (who died when he was nine). Perhaps if his son was alive and wanted to attend college, he may have donated to Harvard or Wharton instead.
I'm sympathetic to your signaling point of view, but this statement is definitely wrong. CS curriculum varies a lot between the top schools and a typical public university.
For example I just now looked at the labs for Operating Systems at University of North Carolina and compared that to the work I did at CMU. UNC has students implement malloc and free. At CMU I implemented a process scheduler and virtual memory system.
Each program has its own strengths, and simply comparing between similarly titled classes is an unfortunately poor metric. (Eg. At UNC, OS is really an introduction to C and OS concepts class. The Operating System Implementations class sounds more similar to CMU's OS Class.) Academic programs need to be compared holistically, and only if one is consistently worse than the other do you have an answer. (Otherwise I could go and claim that UNC is better than CMU because our real-time group is much better.)
Of course, that's too much analysis for most people (including me), so I would argue that looking at where graduates work is typically an okay proxy for undergraduate educational quality.
That being said I've spent a lot of time in my career recruiting students out of undergrad. And I've noticed significantly more rigor and in depth work from students from some of the top schools compared to most everywhere else.
My undergrad OS course didn't even have us implement malloc or free... CMU's sophomore systems course is a lot more rigorus, let alone their OS course.
And that's in systems courses where the variance is actually a lot smaller than in other subfields. There was simply no course in my undergraduate institutions curriculum where you could learn what a Turing machine or even DFA was! We never saw real asymptotic complexity analysis (our algorithms course spent most of the semester reviewing basic data structures as interview prep). Obviously no AI or ML course. Even where there was coverage the material covered was so different. For example, the PL course was all about using a functional PL and didn't include any of the type theory, logic programming, language implementation work, etc. It was a watered down version of SICP, which is an intro course.
If you think the difference in quality between UNC and CMU is large, consider the massive difference between branch campuses and top-ranked public research universities. For all practical purposes, "CS" means something very different at regional branch campuses than at top tier research universities like UNC and CMU.
No. Not even close. Unless you're taking about flagships like uiuc and Berkeley (which are not local for 99%).
This may be true in the humanities but is nowhere near true in the mathematical sciences.
My final year analysis course at college only covered a fraction of the content in the FIRST year advanced math sequence at Harvard.
As someone who went to one of those truly local state schools then taught at a #1 CS school: the quality difference in CS educations is absolutely insane. See also the "Java Schools" rant.
I went to school just as Java hit .. saw the old C days and then Java (in my OS course .. the horror!). I feel Java for OS stunted me to this day. Curious what your thoughts are on this.
(Also, don't read anything into my user name. I just happened to make my first comment about something related to java and then never logged out.)
He, in my opinion, wanted to guarantee for his kids - the networks / access / prestige that come from attending elite schools Yale, Stanford, Harvard, Horace Mann, etc.
The motivation for David Shaw isn't fighting inequality or discrimination but ensuring the best possible outcome for his kids.
Another alternative is to fund a lab or think tank with no students.
I'd argue most newish schools are set up because their founders wanted to promote a new kind of education (Olin College's hands on engineering program, Liberty University's Christian environment, etc.) not just start Yet Another Prestigious University
> I hate articles like this. I used to work on wall street, and these sort of articles were rampant up until the burst: they glamorize financial "wizards", hedge funds, trading life etc. They're financial porn. They glamorize what, in my opinion, is a dubious field. All the while attracting all sorts of people into a field that has already proven it's "value" or lack there of to society.
https://news.ycombinator.com/item?id=2813902
Edit: It's also funny how the top reply by btcoal describes this current article perfectly:
> These articles typically have a set format: 1. Intro into the mysterious culture of the hedge fund 2. Background on the precocious child prodigy turned market whiz 3. Description of firm's hiring practices of PhD's etc 4. Questioning value of Hedge Fund's to society 5. In conclusion, this hedge fund makes a lot of money...
1. Companies need to raise capital.
2. Those companies want to raise capital at a rate closest to the true value of shares in their enterprise.
3. In order to do that, the people who buy those shares need to know that they can resell them later, at a similarly efficient price. This is called a "liquidity premium". If your shares cannot be quickly resold at an efficient price, you are incurring a negative liquidity premium.
4. Because of the foregoing, it is extremely valuable to society for markets to price things efficiently. This is a service that needs to be performed by someone. That someone is, for the most part hedge funds.
The ideal scenario (which we are tending towards) is that there are a very small number of extremely sophisticated players that keep markets efficient for everyone else. The more money David Shaw makes, the less money there is for other people to make. Which means we need to employ fewer and fewer people in the business of asset price discovery.
Price discovery is an important and necessary function, and the fact that there are these few people making tons of money from it is actually kind of a great thing for the world. All those billions David Shaw made represent many millions not going to other people in finance. Which means those minds are freed up to work on other productive things.
Yes, but let's say you are a venture capitalist. Ultimately, you want the company you allocate capital to to IPO, because the IPO is what allows you to recoup your investment later on. The more efficient the market is that your company IPOs into, the more directly your incentives are aligned with investing in good companies. That is why market efficiency, and its counterpart, liquidity, are important.
> If exchanges operated on auctions at every say 15 minutes instead of streaming prices would that matter in the capital allocation process?
No, that's a perfectly reasonable and good way to operate a market. It's called a "call market", and for most purposes it is in fact strictly better than a continuously trading market. Its primary downside is the lack of immediacy - you have to wait 15 minutes. Of course, you can make that window arbitrarily small, say, every minute.
I'm mostly arguing that things like statistical arbitrage, which is what DE Shaw does, are necessary and important. And DE Shaw would continue to operate just fine in a call market. AFAIK they are not primarily HFTs.
Consider high frequency trading. Does it really matter for capital allocation that you can sell your Google stock for a fair price in 10ms instead of 20ms? There are hundreds of PhDs working on algorithms and billions invested into infrastructure to make those 20ms into 10ms, which I'd argue is a byproduct of how exchanges work and doesn't serve humanity in any way.
That's not entirely true. Liquidity is often used to mean 'ability to exit your position at the most recently traded price', or something similar. However, i'm using a slightly expanded definition, which means: exit your position at its intrinsic value. The ability to sell your stock for what its actually worth is extremely valuable, and statistical arbitrageurs help you do this.
> Consider high frequency trading. Does it really matter for capital allocation that you can sell your Google stock for a fair price in 10ms instead of 20ms? There are hundreds of PhDs working on algorithms and billions invested into infrastructure to make those 20ms into 10ms, which I'd argue is a byproduct of how exchanges work and doesn't serve humanity in any way.
Here I agree with you, although I think it's important to understand that a lot of high frequency trading is actually making markets efficient. That is to say, the service being provided by HFTs is truly valuable. What is not super valuable is squeezing out that last marginal millisecond. The energy poured into the last millisecond is indeed deadweight loss, but I think the amount of energy being spent there, while very large in absolute terms, is not so large relevant to the financial industry writ large.
So yes, I agree that the competition at the margin in HFT is probably not productive, it is a byproduct of the necessary service that HFTs provide and the dynamics of the environment in which they provide it. I also think that any 'solutions' to the problem of that energy expenditure are likely to make everyone worse off over all, not better. You really want the HFT game to be a winner take all market, in the way that it currently is, precisely because competition is deadweight loss.
If you do things like introduce purposeful stochasticity into order submission (for instance), you're going to make the competitive equilibrium more multipolar, so that instead of one dominant firm taking it all, you have a bunch. And that's actually worse for everyone. What we want is basically 1 HFT firm that is the best and does a good enough job that nobody else tries to compete, and earns a reasonable profit for providing the efficiency and liquidity that they do. I think the competitive environment is actually converging on that, even though it may not look that way from the outside.
The billionaire owner of Renaissance has done something similar. It's interesting. Rather than giving the money away to academia or other philanthropic causes à la Bill Gates, they choose instead to micro-manage it. Shaw and the Renaissance guy both formerly worked in academia, and I can see why this would make them skeptical about large donations to academic institutions (beyond the motives proposed in the article).
Think of how transformative that Yale slot could’ve been to someone not standing on a multi-billion dollar pile.
Statistically speaking, isn't it likely that someone that gets into Yale also gets into another elite school? Not defending it but most people "close" to getting into one (waitlisted, etc) end up getting into another. It's only "transformative" for someone that's not even close to getting in (like myself), but even that's hazy.
And yet you still have to work quite hard at those prep schools and elite colleges.
A friend of mine of similar status simply doesn't use the family name, went to an ok but not Uber-elite school and works a normal job.
> One of her areas of expertise is how to pay for college. In her writing, media interviews, and YouTube videos, she cautions parents not to “follow the herd with your donating dollars” or pin their hopes for their children on getting into brand-name colleges. “Don’t believe the hype,” she tells them. “You might find yourself obsessing over those annual college rankings. Don’t take them too seriously.” The sensible solution, she argues, is for families to “pick a few financial safety schools” — public universities close to home. A degree from an elite college, she reminds readers, may not translate into higher earnings in later life. “The Ivy League isn’t necessarily the gravy train.”
VS
>Shaw and his wife, financial journalist Beth Kobliner, have sent their three children to an elite prep school, supported them with hyperqualified nannies and tutors, and encouraged their extracurricular interests. But while the typical snowplow parent quietly eliminates potential obstacles by clearing the road ahead, Shaw and Kobliner have seemingly bulldozed an entire mountain.
>Starting in 2011, when the oldest of their three children was about two years away from applying to college, the Shaw Family Endowment Fund donated $1 million annually to Harvard, Yale, Princeton, and Stanford and at least $500,000 each to Columbia and Brown. The pattern persisted through 2017, the most recent year for which public filings are available, with a bump in giving to Columbia to $1 million a year in 2016 and 2017. The foundation, which lists Kobliner as president and Shaw as treasurer and secretary, has also contributed $200,000 annually to the Massachusetts Institute of Technology since 2013.
>The total donations for “general” purposes across seven years and seven elite schools are $37.3 million, which represents 62 percent of the foundation’s giving over that period.
It's possible that she's giving out advice that she would not apply to herself, perhaps to reduce competition for her own strategy.
I suppose the author is criticizing him, but I ended up liking Shaw from the article.
Can confirm there was a well known, published policy on going into David’s office and we got a half page memo from David detailing the importance and exactly how to white space D. E. Shaw & Co.
The company ran with excellence as the goal in everything. Even with that, there was a bit of riches to ashes to riches trajectory. I loved my time there, still work with several talented colleagues I met there, and think highly of David Shaw and what he’s created.
D.E. Shaw & Co.
D.E.Shaw&Co.
DE Shaw & Co
D. E. Shaw.
It's a branding consistency issue. The space and what it conveys (attention to detail) is incredibly important in retaining the trust of some clients who are likewise particular.
Some years ago I sent out a memo about the use of an apostrophe in a company name that was perhaps received similarly. Some think it doesn't matter. It matters to some. That's who it is done for.
Only hire the true hackers seemed to work :)
I would expect many schools would do a similar thing. Any school that would receive his kid's application would surely admit them under similar rules. He does not need to hedge his options.
It was theft because the money didn't go to the school but instead went to the facilitating lawyer and some corrupt employees.
In addition to risk aversion, the article also talked about his efficiency and value of his time.
it's like this thought i had about superman being buff - he's the last person that needs to lift weights
He is extremely risk-averse, even with all of his money.
Hence, if you judge his action trough the lanes of risk minimization, it will all make sense.
Perhaps he's extremely risk-averse _because_ he has all this money. Making more money would not have much impact on his life, while losing it would make a significant difference.