Not at all. In fact this is where my answer will differ from the bulk of people who entered this industry purely for the money. First, I speak for HFT:
1. It is not cheap to be in this field, and it's extremely competitive - I say this despite having a very talented team who have come from very prestigious positions and still have the credentials to go where they choose. For every 1 buy-side trader or developer who is well-paid, I know 3 kids who are making 1.3~1.5x as much in the tech industry - I have employment statistics from my college/grad school's alumni associations. If our goal was to make money by doing something brainless, we would have gone to Silicon Valley, which to be honest, is full of less sophisticated VC investors (e.g. Thiel) who have yet to be stung by a Madoff-class fraud, and even worse failure rates than hedge funds.
2. If what we're doing is speculative, brainless, and unintelligent BS, then we're buying into a bubble and we've been trained to get out of it quick.
Then I speak for a typical investment bank: Being able to move capital between those who need it and those who have it is a service that the general public undervalues. Just as it's inconvenient to buy a new house with a briefcase full of bank notes; it's the reason why we enjoy very low interest rates for loans. (In Genoa, they used to charge 100% interest rates...)
It's best to put this in the perspective of someone who has to structure a product. Suppose you are structuring a reverse mortgage product for Goldman Sachs/Morgan Stanley/JPM. What you're doing is a very genuine service to humanity. Let's talk about this point of view of a typical middle class American. I know I'll die one day, so why should I have assets locked up in my principal residence? Sans inheritance split to my relatives, I would want to take as little money as possible into my grave. The only problem is that I need somewhere to live up till the day I die, and there are two main solutions: I could sell away my home right away, then lease it from the buyer. Or I could enter in a reverse mortgage contract. The reverse mortgage pays you the future valuation of your home, and you get to live in it. Come a day you pass away, the bank goes to your estate executor and takes ownership of your home.
Right, sounds easy enough. Why does an investment bank deserve to be paid much for this? Well, think about it carefully, chances are, the persons whom you have to take title of the home from are probably the person's sons/daughters. There are three scenarios where this can go terribly wrong:
1. The price of the home plummets by the time you take ownership.
2. The price of the home skyrockets by the time you take ownership.
3. The person's sons/daughters are middle-age to senior citizens by the time you take ownership.
You realize there's an asymmetric risk between (1) and (2). If the price of home plummets, there's high probability that the estate will just hand the home over to you because they got a good deal out of the reverse mortgage. If the price of the home skyrockets (you paid $200,000 for a house that is now worth $2,000,000), do you think the estate is going to hand it over? No. There's a much higher probability that they will head to a lawyer and sue you for scamming the poor granny or grandpa.
1. And I want to tell you, it's not very nice to be representing Goldman Sachs in front of a jury, because juries like to judge everything ex-post. The house is now worth $2,000,000 when it was worth $200,000, so you must have scammed them! (How would we have known it would become $2,000,000?)
2. Even if the jury does award the decision in your favor, it's also not very nice to be representing JP Morgan and being interviewed on CNN for "scamming" senior citizens and kicking them out of their suburban Pennsylvanian homes.
It's a lose-lose situation. I don't think it's an easy job at all. Of course, my job is to model the probability that they will head to a lawyer a random number of years from now - so there's both default risk and longevity risk. And I am certain few traders even know how to price these.
I've mentioned this on another thread that this is a matter of definition of scope - what I'd call "HFT" is different from what another person would. However, a reasonable estimation is 55% of the average daily volume on the ES.
I want to divide my answer into two parts: its effects on society vs effects on retail traders. It definitely provides a net benefit to society, as it is responsible for lowering the risk premia for producers, pensions, mutual funds and sovereign wealth funds. Say in the futures market, you can find significant correlation between local risk constraints of HFTs and risk premia for producers - if hedging becomes cheaper, then inventory gets held up, and spot prices go up. The majority opinion is that HFT lowers trading costs for long-term institutional investors. I was at a conference where the CEO of a 3-digit-billion mutual fund portfolio said openly that their research found that HFT lowered trading costs more than it imposed damage. I believe a recent survey confirms this opinion across the industry, but I need to dig out the article again.
Is it a beneficial to retail traders? Yes.
1. The majority of retail traders trade on time scales over which any particular HFT's trade cannot have an aggregate price impact on their trade. When a HFT goes down, it's because another has arbed away its opportunities. Vice versa, if a HFT flourishes, it has successfully competed away another HFT. Retail and long-term institutional investors should not worry.
2. The majority of retail traders value lower transaction costs over informational edge. (That's why they look for brokerages with low commissions and margins above anything else.)
3. Some argue that there is manipulative activity. From my experience, it's absolutely impossible to dislocate a market whose capitalization dwarves any attempt to corner it by virtue of your speed. There is actually a model-driven proof behind this. You could profit from structural loopholes, but this is no different from a greedy executive or congressman that exploits insider information - it does not mean that all executives or congressmen are inherently evil. Even if you could find those rare opportunities to walk the order book to your advantage, this is something that anyone with sufficient capitalization can do, and has very little to do with being fast.
There are a very few funds who claim this is what they do as a marketing punchline. From what I know, the space of strategies that genuinely trade on language processing is extremely small and practically non-existent. The only "evidence" people have of this are the job offers to NLP guys from some prop firms. These people are employed for their modeling skills, not actually to develop algorithms to trade on news.
Studies have shown that the markets are very efficient to news, even that which is not publicly available. For instance, when the Challenger disaster took place on Jan 28, of the three main contractors, only Thiokol's stock price took a significant hit and did not recover. This was immediately on the day of the incident, and nearly six months before the Rogers Commission Report on Jun 9. This was in 1986. So it's not immediately obvious how an algorithm that parses Twitter sentiment would be any different from other fat-fingered trades like Waddell & Reed's ES trade. It will be destabilizing no doubt, but the problem is not the algorithm or the speed of the trading, rather the ill-advised user.
Last edited by artemiso; April 30th, 2013 at 08:56 PM.
The following 8 users say Thank You to artemiso for this post:
In 2007, CDS contracts traded at premiums to low-delta equity puts, which yielded several opportunities in both the low-latency (e.g. for announced LBOs, equity puts could be traded against short-dated CDS and we could match contract expirations using equity leaps) or in the long-duration space (e.g. long FX and write CDS protection).
This is a measured, well thought and logical reply that argues many urban legends about HFT/algo stuff that ALWAYS brings out the pitchfork crowd screaming for blood. I doubt many on this board will take this and appreciate it for the depth that you provide. People prefer their information from ZH or some conspiracy website or mainstream media. After all, a lie told once is a lie, a lie told a hundred times is a truth, isn't it?
Still, very much appreciated.
The following 4 users say Thank You to Deucalion for this post:
In simpler words, I was describing a market relationship in very liquid markets that could be exploited both in a low-latency, event-driven strategy or a long-duration macro trade. I made it a point in my first post that I won't go into specifics of strategy implementation. I apologize if you've hoped for something with greater detail.
The following 2 users say Thank You to artemiso for this post:
Iím not seeing how the second scenario is a risk. The bank isnít entitled to the house they are only entitled to the amount borrowed and the interest accrued. The home only serves as collateral against the loan. Letís say after both principal and interest accrued the reverse mortgage has a 200k balance. The estate can either sell the house for 2 million to pay off the reverse mortgage netting them difference. If they want to keep the house they can easily get another mortgage to pay off the original reverse mortgage since the value of the house has greatly increase.
The following user says Thank You to quantarb for this post:
I should make it clearer that this is from the perspective of the person structuring a new product with these properties back when reverse mortgages didn't exist. The only feature of the contract that I've mentioned is the lump sum of the residual value of the house - meaning the bank keeps the call option on the house rather than the beneficiary, and therefore the litigation risk.
Indeed, you could deduct from the lump sum the value of a call option on the residual value with exercise price equal to the principal amount of the mortgage plus the accumulated interest at the time of the owner's passing, so that the beneficiary/estate receives a claim equal to that call option. It's not immediately obvious when you're structuring this product that there is litigation risk and that you could take away "for free" in the process of transferring that call option - although it's clear as daylight now that I've mentioned it.
Thanks for bringing this up, as it's a breath of fresh air that there's someone interested in the finance aspect.
Hey Artemiso, thank you for clearing up your post. I have questions regarding the operations of running a hedge fund. Iím currently running automated trading strategies with my own funds. Friends and family members are looking to invest with me given my success so far. My plan is to start with managed accounts from Interactive Brokers since theyíll be taking care of the back office and billing.
My goal is to build a track record and then eventually launch my own hedge fund. Recently, a founder of a startup hedge fund told me people with managed accounts never get anywhere. Is there any merit to his statement, and if so how come?
The following 2 users say Thank You to quantarb for this post:
However, we set up a master-feeder structure with international investors and I can give you the other side of the picture.
The problem is that you can't go very far with reasonable rates of return and even 2/20 fees. I've mentioned this another thread before that even $5m AUM/person is really pushing it for a CTA, much less a complete fund structure. You will draw less salary than you would in employment, and will not afford the talent, research and infrastructure that is necessary for you to run a sustainable business.
You have two alternatives:
-- Run strategies that have very high rate of risk-adjusted returns and allow you to bootstrap your employment/infrastructure overhead. But this leads to a catch-22 situation: in today's environment, you need sophisticated infrastructure and high turnover/low commissions to achieve these returns, but you need to achieve these returns to afford the infrastructure and trade large volume. It's not that it's impossible - I don't want to say so because I want you to dare to dream, but if you account for the barriers-to-entry (firms, talent that you're competing with in this area), it's extremely unlikely. More likely, you will have to compromise for strategies that can be run on commodity off-the-shelf hardware and 1 or 2 persons, and you have to keep surviving year-on-year hoping for that eventual combination of HNWs that get you to the $10m AUM/person mark, which is extremely unlikely. The strategies that you can trade via IB infrastructure and commissions are very limited, which make it even more unlikely for you to succeed with the list of constraints above.
-- Raise a large amount of seed capital. This is the likely route - that's why most funds are offshoots of buy-side traders previously employed at reputable firms. However, your best market will be institutional investors, who have extraordinarily high standards of due diligence. I said two(?) posts ago that this field is extremely competitive and costly and I meant it. As an example, just 2 months ago, a relatively wealthy VC investor misunderstood that we were running a technology startup (we're very ambiguous about this) and wanted to contribute $2m to us but we turned him down. Partly because this is a pittance by asset management standards, but also because we'd have to navigate red tape that we decided was more costly and time-consuming than an additional $2m AUM would earn us. It showed us that it's very easy to get funding as a startup in any other industry, because none of this red tape would have impeded us. But this leads to another catch-22 situation: if HNWs aren't cost-effective, then institutional investors are your only option, but they have much higher standards. But the latter is the lesser of two evils.
Which leads to a question, what are "high standards of due diligence"? I'm assuming you are the most stellar manager already, you still need:
-- At least two prime brokers, because they are functional regulators and you don't want to get stuck when one of them pulls your leverage.
-- A third party administrator with SAS 70 Type II. You need auditable returns.
-- Multiple custodians and the (non-IB) infrastructure to handle multiple brokers and managed accounts.
-- To facilitate tax reporting for your investors.
-- To be insured.
-- To iron out all the terms and conditions on paper: redemption fees, equalization, administration fees, custodial fees, secretarial fee, registered agent fee, annual incorporation fees, shareholder services, currency denominations, lock-up periods, hurdles, issuance of securities, cross-collateralization in share classes, so on and so forth.
Unfortunately, "high standards of due diligence" simply translate to "you need to have deep pockets to afford all the legal and service costs". (I think the standards are misguided, but it's how things work.)
The next best alternative you have is to find a prop trading firm (not a trading arcade) that may allow you to eventually become an equity partner, and/or give you flexible employment terms to run and develop your own strategies using the firm's infrastructure, and/or run your own personal account while not competing against the firm.
The following 7 users say Thank You to artemiso for this post:
1. There's a lot I'll still like to talk about, and I'll try to answer any remaining questions (1 more page) with my remaining time.
2. I think this is a good-natured audience to talk to - though you guys should feel free to be critical of my posts because I've gone back to read some of them and found a few responses to be lacking. Anyway, I really appreciate all the questions and hope you've all gotten something useful out of it.
The following 3 users say Thank You to artemiso for this post: