I feel sad that so much brain power goes into market arbitrage. They produce nothing usable by anyone else. Imagine if these people built companies that made new advances and new products.
I completely agree with you, but think your comment should be moderated down from 'They produce nothing usable', to 'They are paid exorbitant amounts compared to the value the produce for society as a whole'.
"but they manage your retirement fund", really? do they? last I checked there are algorithms running the entire show and making sure I and every other little investor get screwed for every 10th of a cent so that a few investors at the top who already have loads of money can have more. They've made a game out of moving money and skimming a bit off the top all the time. Let's not forget that they are actually taking money from your 401k every time they make these trades. Let's not pretend the financiers care about you, the little guy.
Another comment mentioned that YC is in finance, but let's be clear, YC is financing start-up businesses and helping them to grow. They're small finance. When I buy a share in Facebook, I'm not helping it grow, I'm betting on it's growth and the big finance industry is just the casino.
Spreads on liquid stocks are a penny now. Most limit orders are price improved by a few tenths of a cent -- a better price than you asked for. These are both courtesy of algorithmic trading. If a HFT trader does arb between two diff exchanges he's not hurting you. You're conflating it with front running your orders. That is not really a problem for a retail trader who isn't moving large block trades.
some market players are allowed to trade at sub-pennies, others are not. when one party can trade inside while leaning on the guy taking the risk to make the market, something is very very wrong with the "system". over time, the guys taking the risk are gone, and the whole market is weakened.
getting a price better than your limit isn't better than your asking price. you state i will accept nothing less, which is typically priced lower than or close to the market price. so your asking price is really the market price but with a safety net. it's not really a courtesy at all that limit sales go for higher than the limit.
and oh goodie, i made an extra $50 on my trade because of some algorithm. thank god for that.
Actually, the liquidity providers who are filling orders passed from your broker will give you a better price than the market NBBO (or your limit) by a couple tenths of a cent. You can google it to read more. It's generally called "price improvement". The main reason they do this is to capture order flow so they can use that knowledge in their own trading.
What would a FB IPO have looked like without the expectation of a deep and liquid secondary market for the lifetime of the company?
If the secondary market wasn't so big and efficient, it would be more expensive for everyone to raise capital, from large corporations to one man seed rounds.
I don't know about these guys, but according to Cathy O'Neil, D. E. Shaw profited by anticipating (c.f. index front-running) pension funds in a way that sickened her to the point of quitting.
Think about happens if indexes don't pre-announce their changes. Is their market impact larger or smaller than if people know about it and "front-run"?
Suddenly someone wants to trade 10x daily turnover in a small-cap stock right on the closing bell. It could be a dumb whale index fund, or it could be an insider trader who knows about a merger, or a sharp hedge fund who thinks the stock is undervalued.
If you're a liquidity provider or speculator, do you want to stand in front of this freight train at all? If you offset the huge liquidity imbalance, how much will you charge to cover your losses if it was an informed trader? Is it more or less than if you know it's an index fund?
Remember, even if there is some edge in this type of anticipatory trading, competition will reduce its margins significantly. Imagine I announce tomorrow in the Wall Street Journal that I'll buy 100000 off-lease Honda Accords next week from whomever will sell them to me cheapest. Everyone rushes out to buy them on Craigslist or eBay, but only those who make the least on the deal end up selling to me.
If you invest in ETFs, you are certainly being helped by automated arbitrage traders. These products rarely trade more than a cent from their true value because computers can calculate the hedging basket almost instantly. Yes, they have a profit motive, but competition keeps their profits small: https://meanderful.blogspot.com/2013/01/hfts-dirty-little-se...
It always costs money to transact. The alternative to trading against an ETF arbitrageur's quote isn't a costless transaction at the fair price, it's manually paying the spread and transaction costs on hundreds of stocks and risking slippage.
FWIW: YC probably earns more than even the biggest automated trading businesses.
> making sure I and every other little investor get screwed for every 10th of a cent so that a few investors at the top who already have loads of money can have more.
Do you have a reference for this? It runs counter to my intuition.
This is actually how algorithmic trading works. Computers are committing multiple trades in a fraction of a second in order to skim the tiniest of profits thousands of times.
The fact that computers are committing multiple trades in a fraction of second to make tiny profits does not mean that they are screwing you out of every 10th of a cent. In fact, my intuition is that you would be worse off without them because whoever you're trading with would be making big profits rather than tiny profits. It's the competition between computers that makes that profit so small.
> It is a statistical certainty that a percentage of trades will be losers. You are establishing a position with an unknown outcome. Sometimes they go your way, other times they go against you.
> How is it possible that one of the largest high-frequency trading firms executes millions and millions of orders for four years without ever having a down day? The short answer is what they do is not trading -- it is skimming. I call it legalized theft. High-frequency trading is a tax on investors, encouraged by the exchanges, allowed by the SEC. It is prima facie proof that something is amiss.
It seems to me that this author has difficulty with taking the limit of 1/sqrt(n) as n grows without bound.
I think we're discussing trading here. I'm not prepared to give anyone financial advice, but I'm probably not personally a huge fan of retail bankers, investment bankers, wealth management, expensive mutual funds, and so on.
"I feel sad that so much brain power goes into market arbitrage. They produce nothing usable by anyone else."
The simplest description of the financial sector is that they ensure that money is invested and reinvested. In general, this is good for the economy, and helps everyone. This pays for peoples wages etc.
Let's imagine a situation without financial services. All money just sits on a bank account. That's it. No return of investment. Zippo. Nada. Now, imagine that money is reinvested - now the money flows into a company, that can use the money to hire someone... etc.
The world with finance is a much more rich and varied world than a one without finance. The fact that it plays such an important role, and that it's not immune to human fallacy, leads to the pathologies we've sadly all become familiar with.
Basically, if we look this from a historical perspective, a world without finance is a world without industrialization. That's not to say finance (or industry for that matter) is always fair.
Money sitting in a bank account is invested via loans and gets returns via interest. If the bank is savvy and makes good investments they keep the spread between the interest paid and the interest earned.
From the perspective of the business owner it matters little of the cash comes from a bank or as an equity investment. The cash spends the same and they will prefer whichever they can get that is most favorable to them.
Even though public equities get all the news they are a relative latecomer to the party and are frankly a pimple when compared to the bond and cash market. Even if the entire public market disappeared tomorrow the vast majority of companies would have no issue raising cash either through private equity, bonds or bank loans.
Almost all (99.999999+%) trades on the public market have nothing to do with raising corporate equity (i.e. newly issued stock) and instead are transmitting price information about the markets current beliefs about potential future earnings. That is valuable in aggregate because it allows good companies with strong prospects to raise funds at a lower cost (if needed).
But is transmitting pricing information around the globe via dedicated links to shave off tenths of seconds really necessary? Does any company need to raise funds at a micro-seconds notice?
After all the markets close at the end of the day but somehow the world continues to function for the next 16 hours without sub-second pricing information at everyone's fingertips.
You don't. You use the profits from the transaction as capital that then eventually funds the operations of some third party, who then uses that capital to hire employees and to purchase services and materials.
Now, does it improve liquidity? Investopedia claims so, thus, making capital cheaper and making it (in theory) easier e.g. to hire new people.
In itself it's not evil and actually beneficial. Lower transaction costs and more liquidity means there are more resources to go around. What people actually choose to do with the amassed capital is another thing then entirely.
Does it cure cancer? No. But blaming smart people from not trying to be a superman is stupid. Very, very few alone are that smart that they alone would make a huge difference.
It's hard to see why a career in finance would be less ethical all around than a career in some company that just harvests clicks.
Science currently, as a career choice, needs burning enthusiasm or it will burn people out - I see no reason to chide people for not choosing it.
As a fairly low-income engineer I find it hard to see any other reason than jealousy to bash people in finance in the general context.
From your source:
"The opponents of high-frequency trading feel that whatever liquidity HFT creates is superficial because the securities are held for a very brief period (seconds or fractions of a second) before being sold back again into the market. Most of the time, securities are bought and sold very frequently between high-frequency traders until they are bought by an investor. Opponents say there is thus no ultimate creation of liquidity but a mere facilitation for order execution.
HFT results in what is called hot potato volume. Positions are being ping-ponged between high-frequency traders and the other marketmakers. Thus there is the creation of great volume and no concurrent depth. For orders to be absorbed, buyers must hold their positions for a longer time than just a few seconds.”
And then merely counters that with a argumentum ad populum:
"With more than a decade in existence, high-frequency trading is now more or less an accepted part of the stock markets. There is a consensus that, on average, HFT has added liquidity to the markets and reduced trading costs."
The evidence appears rather weak for that position.
That "mere facilitation" is extremely valuable. Do you think stock exchanges themselves are valuable and worth paying for? They bring buyers and sellers together in a centralized place, merely facilitating a pre-existing desire to trade, taking a small cut in the process.
Imagine a world without them. When you want to buy a share of Microsoft, you could call all your friends seeing if they want to sell any or know someone who does. That could take forever or you might never trade. Search costs are real.
Much like the stock exchange helps buyers and sellers find one another in the same security, HFT helps buyers and sellers find one another across exchanges, securities or risk factors. They basically create a meta-market for you, at very little cost. Typical HFT profit margins are fractions of a cent per share traded, similar in magnitude to what the exchange earns.
To give a real example, imagine you want to sell Valeant Pharmaceuticals after their latest scandal. You only have access to the US markets. There's an abundance of sellers in New York pushing the price down, but over in Canada, the price is a little higher. Someone in Toronto just read the latest report, thinks the price will rise, and put a bid in, but he only has access to the Canadian markets. The two of you wish you could find one another, but you can't.
An HFT algo with real-time data from Canada and the foreign exchange rate places a bid on the NYSE, tightening the spread. You sell to them. They turn around and hedge their stock exposure by selling to the guy in Toronto, and trade an FX futures contract in Chicago. This is "mere facilitation", but it helped two traders transact at a better price than they would have otherwise, and forces convergence between two related markets to make them more efficient. Automation lets him run this same trade over thousands of securities for very low margins, making a few dollars in each every day. How is that not a great thing?
"The evidence appears rather weak for that position."
Sure, and it's exactly not a scientifically backed site.
"Not obviously pathological" was the intended message from my part. I have no strong opinion on the matter. My weak belief is that it's non-pathological, and thus not worth chiding.
Any real world system that has enough degrees of freedom to function and not be too brittle and cumbersome is bound to have loopholes and such. Which are ok, as long as they are not grossly parasitical.
I think it's hard to say what exactly is the specific sin of high-frequency trading, and that it appears as an indistinguishable part of the current market.
Thus critizing high frequency trading appears of critique in general of parties who merely leverage the existing information disparity in the market to the fullest.
The main benefit of the market is that it removes transaction costs. Money has lower transaction costs than barter, bank transfers have lower transaction costs than coffins of money, etc.
Inducing some mechanism that would guarantee full information parity between all agents in the market sound to me like it would eat up lot of the benefits of the lowered transaction costs. Although, I'm not an economist and can be persuaded with better information.
My original query was how do you use capitol that you have for less than a second to hire someone (or do anything for that matter). I'm not aware of a mechanism that would allow HFT to create a job. (Except at a HFT firm perhaps)
You could make a similar argument about a lot of the social media sites, there are engineers at Facebook making huge money, just to work out how to make you spend an extra 1-2 minute a day on their site.
Sure they were originally doing something of value with that, but now days they're just trying to make people more addicted.
I feel sad that you think this about finance. I believe I've done a lot of good for the world and have gotten the opportunity to work with some very smart people along the way.
I first started working in finance because it was one of the few industries that paid engineers a fair salary without trying to trick us with monopoly money.
Of course, the world has changed and now tech companies are doing the same.
I genuinely want to know what you think the good you did for the world is in investment/trading. I really don't believe that the amount of money the financial industry is making is commensurate with its value provided to society.
> I genuinely want to know what you think the good you did for the world is in investment/trading.
My job has literally been to ensure that, out of every invested dollar, more money goes to the investment and less money goes to traders and other intermediaries.
It may not sound like a big deal, but this is literally billions of dollars a year that are invested in productive industries instead of going to someone's bonus or, worse yet, just vanishing due to lack of care.
> I really don't believe that the amount of money the financial industry is making is commensurate with its value provided to society.
I don't know if it's because I'm uninformed, but I would agree with you. It's not really clear to me why investment bankers earn such huge fees. I suspect that it's because they're basically able to pay their friends with shareholders' money. I don't remember where I first heard this idea, but I can't help but think that it rings true.
My hope is that some day we'll be able to do to banking what we've done for trading.
> It's not really clear to me why investment bankers earn such huge fees
You can think of investment bankers as salespeople. They sell very high value products in very competitive markets.
Whether you think that salespeople do an important job and/or a job that should be paid commensurate with the value of the product is a matter of opinion.Having spent a lot of time with salespeople when I was last in a startup, I have a lot of respect for folk who could do a job that I would most definitely, spectacularly fail at. YMMV.
> You can think of investment bankers as salespeople.
I agree. That's exactly how I think of bankers.
You actually have the same problem come up with other sales people too. If I'm buying a $100 million product from you, I'm probably buying it with other people's money.
I don't care that $5 million of that is going towards the sales person's bonus, taking me out to dinner, etc. I'd rather pay $100 million and have you buy me a $1000 steak dinner than pay $95 million and not even get a birthday card. Of course I want my HBS classmate to handle my account.
Do you not having savings you put into a retirement account? 401(k)? Who do you think the clients of the investment/trading firms are? Do you keep your money under the mattress?
> I really don't believe that the amount of money the financial industry is making is commensurate with its value provided to society.
I am not going to argue with your beliefs here, but just want to point out the amount of money someone gets paid or how much money an industry makes is not determined by "its value provided to society" as what would that even mean?
I live in LA, the entertainment industry here also seems to make a relatively large amount of money.
While the financial services industry is quite large and made up of many different types of companies, it is not a government public service or anything like that, where does "value to society" come from? Who gets to decide it?
Edit: I just want to add that we are using Hacker News which is part of YCombinator - YC is an investment company and part of the finance industry too.
You are satisfying a market demand. There are millions of companies who need to risk manage their interest rate, FX and commodity expodures. Banks offer a service to their customers and charge for that, just like any other. How is that not providing value?
I find it only natural. The free-ish market is one of the greatest things about humankind. I find the complexity, information movements etc. highly fascinating. Trying to beat the market essentially means that you're very interested in how mankind works at a core level. Academically that's very interesting and it's nice that the tokens you use to keep track of how successful you are happen to be money.
Additionally at least in theory they do help allocating capital to better uses which has giant leverage effects on the economy as a whole. I'd argue more efficient capital allocation and innovation (in the Schumpeterian sense) are both very valuable. I couldn't make a value judgment of what's more valuable tbh
I have been thinking about the "we provide liquidity and accurate pricing services and that's good" argument in the context of housing costs and gentrification.
The housing market is very illiquid—it's much harder to buy or sell a home than to rent a place to stay for the night. In the context of gentrification, though, this seems to benefit the preexisting landowners: the difficulty of moving provides downward pressure on supply, increasing prices, and also makes it less likely that longtime residents will sell before prices peak.
But, as a thought experiment, imagine if there was perfect liquidity: you could buy and sell a house at any time at the click of a button with ready financing and (somehow) free moving services, etc.
If that were true, many more residents would sell much earlier in the gentrification wave, to slightly richer people, who would sell to people slightly richer than that. This would result in the original landowners only getting a tiny fraction of the peak land value, and most of the returns going to the richest people.
This makes me wonder if this isn't generally true: that high liquidity primarily benefits the capital class at the expense of small asset holders.
There's probably also a corollary about late stage VCs here somewhere.
They generate returns for their investors, which are generally institutionals such as pension funds and asset managers. How is that not providing value?
As far as I'm concerned, markets should be put on a stepped clock with randomized queue processing. Said queue blind until processing. Then put the resources spent chasing the clock to better use.
A one minute interval seems appropriate. Plenty real-time enough to deal with most real world events and needs, while eliminating arbitrage based upon momentary, naval-gazing analysis.
Markets already belie the trope of an (entirely) "free market." At least level the playing field between moneyed quants and the rest of the world.
I for one don't feel sad, I feel good about that. I loved the movie "Good Will Hunting" and that's the message I got from it: the pursuit of the personal happiness trumps petting your mind to work for the greater good. Let's imagine, just as you said, that these brilliant mind did work towards the new advances and new products. We would get the new generation of smartphones one year sooner, and we'd be so moch richer for that, I guess. But if that comes at the expense of these people being one epsilon less happy than they currently are, that's not a trade off we as a society should be willing to make.
technically they are providing value. they have better knowledge to more accurately price assets, therefore through buying and selling they are correcting the pricing of real companies.
Liquidity that brings with it random, often unexplained volatility... But yes, I would probably buy the brick, and can understand why people lucky enough to have colo servers do this.
I was going to make the same point. At lower timescales, things look very different. Also the number of orders, cancels and corrects has exploded in the last couple of decades, purely because of algorithmic speculation. This makes the market significantly less transparent, and the weird and wonderful matching algorithms employed by exchanges often get gamed by colo'd traders with the relevant knowledge.
I don't understand what you mean about "lucky". Anybody can. I have had a rack of colo servers which I use for personal stuff for the last 15 years. In an AWS world it's in some ways absurd, but some things are easier this way.
Precisely, and money isn't the only factor here. Your contacts, your firm's contacts and your trade volumes (amongst other things) can get result in you getting preferential treatment.
As someone who works in the industry I've found that analogies are distinctly unhelpful. That's not really unique to finance [0]. Can you explain what you mean by "literally wiggles them out"? I cannot say that I have observed a phenomenon that I would describe with those words, but I'm always interested in learning more.
That increase in volatity pushes prices up and down without real logic behind, in the short term, causing the small retail pundits to hit their conservative stop losses and take them out of the market. That's what I meant by "wiggle them out". Does that make sense?
Thank you for clarifying. You're talking about retail speculators getting "stopped out".
Do you have a reference showing that is happening with increasing frequency due to HFTs? It runs counter to my intuition because HFTs lead to less "gappy" markets and markets that respond to news events more quickly. For the purposes of this comment I will assume that this is true.
Is the amount that retail speculators lose to increased frequency of getting stopped out greater than the amount they gain from tighter markets? I don't have a clear opinion on this.
Is the net loss to retail speculators greater than the net gain to all other market participants? I suspect not because retail speculators are such a tiny portion of the market.
If this kind of volatility has increased with HFTs, it's caused by an overly small tick size. Markets will go back to how they were if you increase the tick size back to 1/16 of a dollar. This wouldn't help price takers other than retail speculators, but that's how you fix it if it really is a problem.
I'm working in the industry, and came from astrophysics, and intend to return to it one day. I would have stayed, but when the salary is 10x or more and you have a family, it's an easy decision. A part of me always feels dirty doing what I'm doing when I see news about Hubble, New Horizons, etc.
And as for 'super smart people', I'd just say quants in general are 'super smart' at the very narrowest of mathematical problem sets, but shouldn't be let within a mile of a compiler. And they shouldn't be invited to parties either.
> Eventually the scientists went so far as to develop an in-house programming language for their models rather than settle for a numbercentric option such as ASCII, which was popular at the time.
That was a nice eye roll on that line. Sadly the whole article was pretty pointless. It was one of those wowsa wow wow wow woop woop article, by the end of it you know nothing more than they take speculative risks on short/medium term using futures. Which any CFD trader does.
I'm curious to know what the original statement was that got mangled into this. Maybe the author heard ASCII instead of C? Not sure where numbercentric would come in, though.
In another article [1] it says "Currently Renaissance has three funds: Medallion, Renaissance Institutional Equities Fund and Renaissance Institutional Futures Fund. The latter two have been open to investors, but losses and withdrawls have left the funds wilted ($30 billion to $6 billion in three years)."
> Initially successful with annualized return of over 21% (after fees) in its first year, 43% in the second year and 41% in the third year, in 1998 it lost $4.6 billion in less than four months following the 1997 Asian financial crisis and 1998 Russian financial crisis requiring financial intervention by the Federal Reserve, with the fund liquidating and dissolving in early 2000.
If there's a language that you know best and highly prefer, then use that; but if you have no preference then Python is very suitable at the moment, as it's both easy to get started in it and also easily extending to powerful numeric computation and machine learning frameworks.
If you explicitly want to move into deep learning, I'd start with Python, since it has excellent math libraries (pandas/scipy) and neural libraries (keras).
For those questioning the returns, it's important to remember this fund's strategies clearly have capacity constraints. The article mentions a cap of $10B and a decade ago it was half that amount.
Why cap the size? What this means is usually whatever strategies RenTec employees are using in their Medallion fund likely do not work at larger dollar amounts.
Even worse, attempting to run a quant strategy above its size constraint can break it (this includes exposing too much about it to others). This would lead to it not ever working again even if they went back to only using small amounts.
Once while applying for jobs I had an interview for a dev job with a firm like this. They refused to tell me exactly what they did, or even the company's name. I had to sign an extensive NDA before I could even interview (it was weird having to sign an NDA when it wasn't clear who I was signing it for)!
I was fresh out of school and flubbed the interview, but will never forget the experience. All they told me was, I would work around the clock, but if I came and did my job I would likely make a mint.
I've been dreaming about building stuff like this (on a much simpler scale) for a long time.
Perhaps a model that takes in past quarterly reports, commodity prices, executive performances and news regarding a company and calculates 'tones' surrounding a company.
(for example appearance of 'good performance', 'meeting targets', 'vast improvements' would possibly yield a positive tone where as 'disappointing returns', 'failed endeavours' etc would yield negative). Then gauge past price sensitivities to various tones and see if there is any way to train the machine to speculate merely on these 'tones'.
RT started off as a CTA. If you look at CTAs a good Sharpe is something like 1-1.25. You can buy a book that will get you most of the way there, and it's really simple (Author Andreas Clenow, I forget the title). I have actually verified this myself, coding it up in Python.
If you simulate a random walk that has a sharpe of 1, you'll find there's plenty of periods where it's down or sideways. So you need investors. They have to pay you management fees so that you can run the thing while it doesn't make money.
Simons then changed tack. In the early 1990s the fund industry was very different. Most funds were discretionary funds where the boss decided what to punt on, and investors were used to trusting them to do so. So changing tack to another systematic strategy, in equities, could be sold to the investors.
Now this is significant, because there's a lot more stuff you can do with equities. There's industry structure, there's individual company information, there's analyst predictions for each firm, each firm has its own price series, there's corporate actions, and so on for thousands of stocks. Contrast that with CTAs where you have a couple dozen liquid things, maybe 40, in a few categories (bonds, equities, commodities, FX).
If you find a strategy that works for a commodity and it gives you some Sharpe, maybe 0.1, and we assume for simplicity they are uncorrelated (which they aren't!), you can roughly sum them up to sqrt(n). So there's a big difference between 40 and say 2000. This is just a vignette, don't take it literally. But the point is you can find strategies that in aggregate are much better in the equities space. Purely my opinion, I don't often talk to people about this, so no feedback other that my collaborators.
So back to the story, Simons builds this thing and it starts printing money. Now he's insanely rich, and the thing never loses money. So what does he need investors for? They are getting a free ride until the fund hits capacity.
Now if he had just gone straight to the Medallion fund, the thinking is the same. Why on earth would he need investors? If he didn't need investors, why would he have his name out there? He wouldn't, and neither would anyone who'd discovered anything similar. There's probably a few groups like that out there. They're unlikely to have built out the capacity, because they never became hedge funds. But I'm thinking they're out there.
he's stated in the past they generally fade direction. the strategy isn't scalable, so it's pretty obvious most of what they do is liquidity provision, market-making, and short term arbitrage.
Eventually the scientists went so far as to develop an in-house programming language for their models rather than settle for a numbercentric option such as ASCII
Wow.
It's amusing to imagine what the people at the fund told the journalist which ended up being turned into that.
A perfect example of the "noisy channel" idea in information theory.
I was kind of imagining they told it exactly like that, as a joke, to see if it ended up in print without getting picked up by someone who knows what a computer is
Are you sure? Okay say you found out about Renaissance at 14--were you really going to spend the next decade and a half of your life working towards getting into and then completing a Phd in physics or math from a prestigious program just for the chance to interview at Renaissance?
I upvoted you. I think your account is genuine because it was created on the date of your username, and the other date username accounts are a joke as they were created today.
> Eventually the scientists went so far as to develop an in-house programming language for their models rather than settle for a numbercentric option such as ASCII, which was popular at the time.
I might be missing something obvious, but this makes no sense to me. ASCII is not a programming language. And although UTF8 has taken over, I think ASCII is still very common.
> (To this day the company’s website, rentec.com, looks like it dates from the Netscape era.)
That's not very fair. I think the design is quite modern and minimalistic, and the website has definitely had some recent updates. It uses jQuery 1.11.3, which was released last year.
> When the bill came, they would pull out a special calculator that could generate random numbers. Whoever produced the higher number picked up the tab.
I can't figure out why the author decided to include this anecdote. I also can't figure out why the scientists didn't just flip a coin.
Another random thought: It's interesting to think that whenever I've bought or sold some stock in the past, that little piece of data has made it's way through all of their millions of lines of code, and all of their neural networks. They might have even made a few fractions of a cent based on my behavior.
> By studying cloud cover data, they found a correlation between sunny days and rising markets from New York to Tokyo.
Wow. That's incredibly interesting, and that's just one example. Alright, their code has probably read all of my tweets, too. And everything else I post about online. Especially on all the subreddits for investments and stocks.
> I don’t know, like 90 Ph.D.s in math and physics, who just sit there looking for these signals all day long.
Oh geez. Yeah, they're looking at everything. Traffic, Yelp reviews, Pokemon Go, Netflix ratings, flight delays. I don't know, maybe even newspaper horoscopes. Maybe the newspaper prints something that says Virgos are going to get lucky with an investment. And then they watch all of the online activity of all the virgo investors who still read newspapers.
Oh. And all of those companies selling your data? They're buying that, too.
This sounds like a fascinating place to work. I want to know some of their secrets.
One more random thought: They knew Donald Trump was going to win. I guarantee it. And they probably made a ton of money on all those stocks that went up.
>Renaissance currently caps Medallion’s assets between $9 billion and $10 billion, about twice what it was a decade ago. Profits get distributed every six months.
If it was $5b in 2006 and $10b in 2016 then you can speculate how often they raise it. Your prediction that $1 million would turn into $4.5b, and the fact that the fund sits at $10b clearly indicates they distribute profits more than they raise the cap.
No. So let's say it was $5b in 2006 and they were letting it compound. By 2016, at 35% interest compounding it would be worth over $100b. So by the simple fact that they raised the cap in 2016 to $10b (and the fund was not worth $100b) tells you that they are not letting it compound and as the article states, dividing the insane profits.
That we have an article that discusses people providing no value to society yet working the system to extract billions of dollars (and the political power that comes with it) as if it were fascinating and cool rather than obscene tells us how fucked up our society is right now.
Imagine reading a newspaper in 1840 reporting on a group of scientists who are making a killing in cotton because they figured out, using statistics Moneyball style, how to maximize profits by buying slaves that nobody else wants. Imagine the article tells it as a story of smart guys who amazingly did what no one else could do, with nary a comment or even awareness about the evil of slavery.
Don't worry, they provide liquidity. Surely you remember how dysfunctional and the financial markets were before we had sufficient liquidity? I mean heck, it could take minutes, MINUTES, to execute a trade.
I'm not gonna argue that financiers are curing cancer, but it sure does bother me that so many intelligent people who understand nothing about the industry have just grown to assume that it provides no value to society.
I don't think they don't provide value - just that the value provided isn't commensurate with their payment. It's as if they extract every bit of surplus value they create, which means they produce no net value to society. It's as if they are an "employee" of Society Inc., but is paid exactly what they produced.
I actually think it's a net negative, in the current US.
Now there are good things that capitalism and markets do, but the way it's currently functioning no longer provides those benefits, because algorithmic trading has driven the signal to noise ratio in to bedrock.
Wall St at this point is basically just a social noise amplifier stuck in an ugly feedback dynamic with the rest of the country.
We long ago stopped hearing music from the underlying economy, we're on to mic static screaming at full volume.
If I could wave my magic wand, Id switch to a single exchange which resolves all outstanding bids/orders once every 10 seconds in a single transaction. (You could create a single virtual exchange out of many actual ones using smart contracts.)
I just don't believe Wall St would stop generating any real value by being forced to slow down and take a breath (and allow humans to act on relevant time scales). Instead, we have a noise amplification arms race.
"As long as people aren't burdening society" doesn't apply to an organization that has extracted $55 billion of 'rent' from the society - if there's nothing comparable that the rest of society is getting for that fee, then it is a burden that they've managed to put on everyone else.
In the grill bar example, the rest of society has received a lot of tasty burgers and pleasant dinner experiences. Where is the $55bn-sized beneficial service that the rest of stock market has received from Medallion?
Let's first tackle something a bit easier than absolute value, which is relative value. If we can't agree on the answers to these questions it's unlikely we can agree on anything:
1. Does it make sense that four hedge fund guys got more money than all the kindergarten teachers in America combined? [1] (I used the word "got" instead of earned because "earned" presumes an answer.)
2. Does it make sense that Zuckerberg is worth tens of billions and Albert Einstein maybe had a net worth of one million at death? Or that Einstein earned a modest professor's salary for his enormous contributions to science, but his estate earns $10-12 million on royalties for his name and likeness (photo, Baby Einstein license, etc)?
You can't answer that it makes sense because that is what the market determined, because that would be begging the question (whether the market reasonably and fairly compensates people).
Don't first tackle something a bit easier on my behalf, but if you need to then we can.
Like the definition of value, which I asked you about above, whether something "makes sense" to me does not mean it will makes sense to you, and vice versa. I enjoy the chance to discuss and learn more about this stuff, but there is a ton of subjectivity here.
You seem to be saying value is intrinsically tied to money and how much money someone makes or how much money their estate is worth...?
That is not a priori true. And many examples where it appears value and money don't interact at all. One example sort of connected to what you asked: I am not a full-time kindergarten teacher like you brought up, but I do volunteer with an after-school math tutoring program. Now this is nothing to brag about or anything, just one person for a few hours a month playing a very small role helping students as they learn math. But this is volunteer work and I don't get any money for doing it - I don't even get a cookie or Capri sun like the kids sometimes get.
Does it make sense I am paid all of $0.00 for this time I spend volunteering helping students learning math?
So now let's just get to the real question you ask: does the market "reasonably and fairly compensate people" for work they are currently doing as well as for any future impact their previous work may have led to?
My opinion and my gut feeling is that in general no it doesn't.
But why would it?
So I'm not an expert with in-depth knowledge about every occupation and employee differences between them, and nobody knows the future or what future positive impacts someone's work may lead to. I also don't claim to have some specific set of target amounts in my head for what would be fair compensation for each respective type of work in the market...
The unfairness here is something I feel, but I know everyone's feelings and views on this topic will be unique. I have not come across a system that is objectively more fair. Have you?
I am not trying to play games, I really don't know.
So what system do you suggest would be more "fair" to everyone? and again how do you define providing value to society?
Another way of thinking about this is marginal value. Imagine a world where person X has a heart attack and the next best person qualified to do their job steps in and paid their wage demands. The fall in social benefits less change in wages is their marginal contribution.
For Einstein, this is presumably astronomical.
For the best primary school teachers this will also be high (it might be low/negative for some of the worst).
These are not directly relevant to the question of whether the quant traders have a positive marginal contribution to society.
It gets even worse when you see how those billions of dollars are then used. One of the founders of Renaissance is Robert Mercer [1]. He invested millions in the Heritage Foundation, Breitbart and Trump's campaign.
Whether it is an appropriate analogy depends on whether you think it obscene that they make so much, especially for what they do. See my other comment: https://news.ycombinator.com/item?id=13034873
Makes me wonder why they don't just invest all the money under management in that fund.
Edit: They are capped and take out profits...so I wonder if the cap is the magic or if they use better algorithms for their private fund (only natural since it's their own money). They would have the same inputs (their data sets, infrastructure, programming language etc.) so I wonder if they have two very different models (<=10 billion and >10 billion) or how those are linked.
Because liquidity is finite. Profitable investment opportunities don't scale indefinitely. When you discover that an asset is being offered for cheaper than what you think it's worth, pretty soon either you or others buy up all the cheap supply, and the price catches up with your valuation.
This sounds like the fictitious company in the Series "Billions". How did these guys make 98.5% in 2000 and 33% in 2001 in the depths of the .com collapse of Wall Street?
They use big data techniques to find weak correlation in data and invest moderate amounts of money in them. They never make a large return on any given investment, but they make very persistent returns; over the long run this adds up.
The limiting factor behind what they do is that it doesn't scale.
> How did these guys make 98.5% in 2000 and 33% in 2001 in the depths of the .com collapse of Wall Street?
The name "hedge fund" comes from the concept of a hedge; a strategy that will underperform in good times, but over perform in bad ones. 2001 is the exact time you'd expect a hedge fund to do well, especially one focused on dispassionate data analysis.
It's also how profitable sports betting syndicates work. Lots of small bets with relatively low expected value. 1% expected return per week is 70% return in a year. You can't scale the syndicate above certain limit.
There are many ways of achieving extremely high Sharpe ratios (e.g. exploiting market microstructure inefficiencies), with following caveats:
1) They are not scalable (for each such strategy, market turnover capacity is very limited, and increasing it requires coordinated efforts of many smart people). This is the reason they are not taking additional investments and strictly invite-only.
2) Operations efficiency becomes the key. Execution capability has to be consistently at top of the market, and margin for error becomes nearly nonexistent. It is very easy to screw things up. Hence the need for top technical and engineering talent.
Actually that is the least suspicious. They knew it was a bubble and bet against it. Like the guys in The Big Short. You make money when the market is far from reality.
You might want to take a look at a book entitled "The Quants" by Scott Patterson. He talks extensively about Medallion. Inception, speech recognition people, etc.
I can't say anything about the other book, but "The Quants" is a good read and dives into the history of these funds.
I also liked the style for what is not there. Usually, people addressing these questions try to make it look like magic and easy. I hate the 5 minute montage in movies with cool music that condenses 5 years of hardship. You can pick it up, read some pages and put it down if you don't like it.
The author seems to have had access to the people he talks about, and had help from the right people (Mandelbrot, Taleb, Ed Thorp, to cite a few). For me, this increases weight and it makes it cool.
An edX course about Discrete Time Signals? That's nice. Finding out that the instructor is Richard Baraniuk knowing that I used a paper of his (an adaptive optimal kernel) [to do time-frequency analysis for multiphase flow pattern recognition]? That's cool!
The only way they're making money like this is by cheating the system. I highly suspect they're using their public funds to feed data into their private one in order to siphon off profits. It's like a pyramid scheme except the mechanism of moving money is a form of market manipulation and this article is playing right into their hands by making them sound like geniuses ("it's all scientists!"). Clearly, if this business is run by "smart people" (scientists) then their other two funds--which are actually accessible--must be promising! Right!?
TL;DR: The investors in the public funds are likely being scammed.
No I think you got it wrong. They're using a lot of algorithms to filter out the noise in the market and make speculative moves (bets, gambles) with a certain amount of certitude (weight, size) with the goal of making money.
I believe they keep that fund private because the risk factors must be through the roof. And frankly, they don't need that external money to make a ton of more money.
Do they have low latency, very fast computers, yes. Is it what's making the difference, no, I don't think so. It's the models they use to assess and predict - current and future moves - that makes the real difference.
I would think they keep the fund private, and return profits annually, to keep the size of the fund limited. The strategies they employ probably cannot work as well at larger scale. Here's a rare interview with Jim Simons: http://ritholtz.com/2016/07/jim-simons-mathematician-cracked...
And yet, despite the risk factors for the private funds being through the roof, they never lose money, and as another commenter pointed out, the public funds are losing money. I'm sure it's all on the up and up. ;)
Once a fund grows beyond a certain size, its trades can have significant influence on the market. This, in turn, disrupts the models that led to its success.
On that scale, for any trades/substrategies you can easily reach a cap where extra capital is not useful.
If you can already run/extract all the deals that you want with your current capital, then taking in more money just distributes the same pie into smaller pieces.
> I believe they keep that fund private because the risk factors must be through the roof. And frankly, they don't need that external money to make a ton of more money.
I think the real reason is that the main fund can't manage more money.
The system itself is broken. Money is supposed to be a proxy for value added, yet these guys are adding nothing. Don't come at me with the "investing in winners" guff. They are taking real wealth.
Some companies are going to be overvalued, and some undervalued compared to how much value they will make in the future (e.g. how much profit they will generate over a ten year period).
If I manage to figure out that Company A is overvalued compared to Company B, and I short Company A in order to buy stock in Company B then by doing so, in a simple supply-and-demand model... Company A's stock will go down very slightly (because I shorted them), and Company B's stock will go up slightly (because I bought their stock).
A higher stock value helps a company - they can sell some of their own stock in order to raise more funds. Therefore, the value I've provided is I've made a calculated decision that the economy should be spending more on (for example) robotics and less on coal because robotics companies are going to more successful, and coal less successful.
Allocating funds to the more profitable companies and industries is one of the strengths of the market economy - it adds efficiency.
Price-finding in general adds value. Also market liquidity adds value. I wouldn't discount what these people do, even if they are narcissistic selfish assholes as a breed.
Billions? Pull the other one. Someone is doing real work, these guys extract it as fiat tokens without adding enough value by a long shot, then they use that to get real things produced with real skill and effort, goods such as meat, wine, cars.
The system imbues certain players with monetary tokens far exceeding value added.
Without being able to sell shares, a wine company that makes a profit can only grow by reinvesting their profit. E.g. it cost 100 tokens to set up the shop, it grows by 15 tokens a year, it has to wait 7 years before it can set up a different shop. With the ability to sell shares, once they prove a profitable business model, they can sell shares in order to grow the business into other areas.
Without arbitrage, a wine shop that returns 15 tokens for every 100 invested might be able to sell shares at the same value as a meat shop that only makes 8 tokens for every 100 invested. Even though the wine shop is more popular, more profitable and providing more value to the market, it is forced to grow at the same rate as the meat shop.
We need a system of allocating resources efficiently, and the economy will provide the most value possible. This is what arbitrage does. And you'd want pretty smart people doing it, as these decisions can cause the success or failure of companies.
To put it in other terms, the footsoldiers might do the actual fighting, but I prefer to be part of an army with generals that decide where to place troops rather than one composed entirely out of 'actual fighting people' (or as you call it 'real things produced with real skill and effort, goods such as meat, wine, cars').
There are ways of cheating the system, and I support strong financial regulations. But that doesn't mean these jobs shouldn't exist in the first place or they don't provide any value.
Banking floods fiat ahead of wealth creation to capture labour. These guys didn't create billions in value. They didn't enable it either. They are part of a system that forces us to go to them, pledge our labour for the right to exist. You have the cart before the horse. They also do not allocate resources efficiently. Most bank lending is into land. These guys ride that to steal labour.
They do not add this much value, they appropriate it.
Land value tax and state issuance of money via basic income funded by LVT, expansion of said money to be roughly based on projected wealth creation. No income tax. No sales tax or not much of one.
Or we could just stick with boom and bust where the banks clean up every time and the west is slowly sinking.
This has nothing to do with allocation of land. If you have a land value tax there'll still be property companies with shares and people who arbitrage them.
> state issuance of money via basic income
Again, nothing to do with arbitrage.
> No income tax. No sales tax or not much of one.
Not sure property tax will match income + sales tax, but again, nothing to do with arbitrage.
Look, I agree there are people with wealth who don't deserve it. But in my opinion, these are people with inherited wealth. Or the people who got wealth through ill-gotten means. Or the people who avoid taxes.
But the people who are in finance? By the time they die they're usually multi-millionaires, not billionaires. And the market as a mechanism is not fundamentally flawed - it just needs careful and judicious regulation.
Land value tax is aimed at suppressing land speculation. I'm not sure why you are contrasting my mentioning LVT to the original topic, as I put that in because the parent poster said, in a very supercilious way:
> Can you propose an alternative for capital allocation? Perhaps some magic central planning AI?
If as the GP suggests they're making moves based on certainty of the larger fund's move, they can often be making the larger fund's gains smaller, which is counter to the interest of their customers in the larger fund.
I believe you are conflating the concepts of "currency", "wealth", and "value."
Here is an example I used with my kids once. Let's say you went the Pittsburgh Pirates game in 1978[1] and you caught the first home run hit by then rookie Barry Bonds. If you sold it in 1978 you would have been lucky to get $50 for it, if you sold it in 2007 when he retired with 762 home runs you could have gotten significantly more for it, perhaps several tens of thousands of dollars, and if you sold it in 2011 after he was tainted by steroids you would get less.
But in all that time, the ball was just the ball. It's value went up, its value went down. As it went up and down the value of other home run baseballs also changed in value, but in ways completely unrelated to the value of Barry Bond's first home run ball. If you owned every single home run ball from the major league the value of any one ball was only related to that ball's history and not to the other balls.
Quantitative analysis of markets is the art of figuring out how to recognize the value of something is changing before others do, and capture that value by acting on that knowledge. Whether it is offering to buy Barry Bond's first home run ball because your model suggests he could become an all time home run hitter, or selling it in 2009 because your model suggests Barry is likely to be tainted by the steroid scandal in baseball.
You didn't steal any money, you didn't change the amount of "value" in the market, and the amount of "value" in the market it not controlled by a fixed number, its controlled by other events completely outside of your control. You bought something before others knew it would be valuable, you sold it before others knew it would be less valuable. Definitely not a zero sum game.
If you had a magic oracle and, say, bought a ball for the fair price of $1 the day before you knew it would spike to $100, you are denying the seller an honest opportunity to sell it for $100. It's the moral equivalent of keeping your mouth shut when a clerk rings up an item for less than it costs. It might not be illegal, but it's certainly not "creating value" - you're using an information disparity to create a transaction with someone who wouldn't agree to your terms if they knew what you knew. ("Fools and their money" or something along those lines.)
Of course, once you add in the realities of uncertainty and much longer time periods, you can make more subtle arguments about risk exposure, hedging, etc. But in the simplest case, trading on privileged information is absolutely zero-sum, at least in the sense that there's a clear winner and loser (and in the stock market, it's illegal!)
> you didn't change the amount of "value" in the market
> Definitely not a zero sum game.
Grandparent comment is right and you seem to misunderstand the idea of zero-sum. A transaction is zero-sum if it doesn't change the amount of value. The stock market is mostly zero-sum. The only non-zero-sum transactions are when firms sell their own stock to get spending money. The only added value of the stock market is choosing which of these transactions should happen. Since front-running doesn't affect these transactions, I think it's safe to say that front-running is zero-sum.
You are correct that I don't understand the definition of zero-sum that you propose. Can you break this down a bit?
Your definition is "A transaction is zero-sum if it doesn't change the amount of value." And I'm having trouble with that because I can't see the linkage between 'transaction' and 'value'. And then you state "The stock market
is mostly zero-sum." which I don't understand at all, how can a system be fractionally zero sum by your definition?
This is the way I see it, a zero-sum system (note that I think of it as a system and not a single transaction) is one where the sum across the system of all events is zero. So for example if you have have a series of financial transactions across an expense account there are a list of expenses journaled as negative amounts and a list of re-imbursements journaled as positive amounts and when you add them all together the answer is zero. (And when it isn't zero then you either have unpaid expenses (negative) or have been over reimbursed (positive)). Another example might be in chemistry when you are balancing a chemical equation and all of the electrons and molar weights of different elements have to be the same before and after the reaction.
And that is the reasoning I use for my definition of a zero sum system. A system is zero-sum when all inputs and outputs across all transactions are balanced, resulting in no net change.
The GP comment was short, it asserted "Yes it [stock trading] is [zero sum] when you don't create wealth" and that was a follow up to an assertion that these funds were "taking wealth". Both of the original and the follow up used the concept 'wealth' to express as a proxy for 'value' in a series of transactions which are measured in 'currency'.
One could argue that on the basis of number of shares in existence, trading on the stock market is "zero sum." I would agree with that. I don't find that a particularly useful abstraction but recognize it would be one way to look at it. If on the other hand you are talking about wealth creation, the stock market is very much not a zero sum system. That is because the at one level the stock market is a reflection of the economic GDP of the companies that compose the market, and as economic GDP grows, so does the value of those companies. It is by that basis that an individual investor can buy stocks (equity) in a company, hold it for a long time, and "gain wealth" simply by having partial ownership of an asset which is growing in value. Nobody was made "less wealthy" by that growth and that person holding on to their stock. It was not a zero sum system.
That value is created by companies whose stock is being traded, not by companies doing the trading at millisecond intervals (the necessity of which we're discussing here).
> and if you sold it in 2011 after he was tainted by steroids you would get less
To extend that analogy the trading company has a tap on the Reuters newsfeed and they sell the ball as soon as the steriod story reaches them, milliseconds before it reaches the newsrooms.
