Sorry, you are mixing and matching concepts from reading too many articles on HFT.
> the dark pool operators make money from the HFT's not in terms of the commission, but in terms of collocating fees,
No. Dark pool don't really charge collocation fees. Barclays LX didn't. Most dark pools are in Weehawken and it's just a simple $500 cross-connect from your existing trading system.
> market data fees
It's a dark pool. There is no market data to charge a fee from.
> , and the right to be one of only a few HFTs allowed in the dark pool via collocating, ie exclusivity.
Not in the case of Barclays LX. They let every HFT in at very low rates which is the entire point of the article.
> Barclays was advertising LX to high-frequency traders by offering them more information, lower fees, and faster connections than it gave to institutional investors.
The "more information" being the key part - it sounds like the pools weren't as dark as their customers were led to believe. Your other points are spot on, though.
If Barclays was displaying resting orders to HFTs in LX, the AG wouldn't beat around the bush about it in the complaint. It would have been the headline issue. Doesn't seem likely that was the case.
I agree. It helped me when I was a bit rusty with data structures and was looking for a new job recently. And if you can try and re-implement the data structures in a different way. For example, the JDK HashMap uses chaining so try and build an open addressing one. Not only does it teach you about data structures themselves but it gets you practicing coding very quickly for these toy coding problems that get thrown at you during interviews.
How does Wall St always seem to end up with the blame on these threads? They priced the Facebook IPO too high and they get the blame, they priced the Twitter IPO too low and they get the blame. The company just went public and raised $2B+ at a higher valuation initial valuation then anyone expected. IPOs are tricky things to get right. Give it a few months for the hype to settle down before you start screaming about "joe public".
So because there are some corrupt people in Silicon Valley, one is simply not allowed to distrust bankers who have proven to not be trustworthy? Life must go well with logic like that.
Pricing an IPO is more or less an auction. When pricing an IPO they set a range and then meet with institutional investors to sell their book (in this case 70M shares. The original price was 17-20 per share, and there was excess demand at this price, so the bankers and company, pushed the price to $26 per share.
WS is the ticketmaster of the financial world: public companies, private corporations, startups, VCs, angel investors, investment firms, hedge funds, and HNWIs all blame wall street for some reason, yet they thrive on the services WS provide (and depend on the money made available by WS)
"Working for free" after 40 hours never resonated with me. It comes out of hard fought labor battles, but an assembly line is difficult to relate to when I'm eating free snacks in my comfy Aeron chair and working on things that generally interest me.
I think the biggest issue here is misaligned expectations between boss and worker. Personally, I've always gone into a job understanding what it was going to take to be successful. My comp expectations are adjusted accordingly. In trading bonds, which is nearly a 24 hour market, it does require >8 hours on the trading floor. At the end of the year my boss doesn't really care about my hours as long as our customers are happy and I've produced PnL.
I'd like to add that the statement "If you’re salaried, you’re working for free" is inaccurate in two very important dimensions:
1) Equity stake. The 40 hour workweek was invented for laborers, not owners. Would the author admonish a business owner to only work 8 hours a week? At what level of ownership stake does it become acceptable to work more than 8 hours?
2) Compensation increase through recognition and achievement. Working extra hours may not increase one's immediate paycheck but after many years in this business I feel comfortable saying that consistent hard work generally results in rather substantially increased compensation and opportunity in the long term.
I had the opposite experience. Working long hours never got me a raise or promotion. Successful projects got me raises and promotions, but I generally don't need to work even 40 hours to make that happen. Once I lifted my head up out of the code and started taking a more business approach to my work rather than engineering approach, that's when my value and income skyrocketed. The guys I know that rack up those 50+ hour weeks every single week get the same 3% raise they've always gotten and most of them are pretty bitter about it.
This is why hitching yourself to the wrong boss can be extremely detrimental to your career. A good boss would compensate and promote any 50+ hour rockstar performer for making him look good. As the organization gets bigger, the manager with the most senior people often wins, regardless of competency. If a boss is not giving out raises and fighting for promotions, he is going to have to play out more political capital in order to get anything done, or to cover his ass when shit doesn't get done.
Working any number unpaid hours is working for free, no matter how you spin it.*
1) Business owners aren't laborers. They work, and they may draw a salary for tax purposes, but they cannot be placed in the same bin as "employee," which is what "laborer" means in this context. As a matter of opinion on your aside, I'd argue that the "level of ownership" required would be one sufficient to imbue the equity holder with the significant responsibility, authority, share of the profits that other "major" equity holders have.
2) That compensation generally rises with recognition and achievement is: a) not in dispute; b) not relevant to the point, because it doesn't address unpaid hours worked.
* There's a certain argument that exempt salaried employees are paid for their expertise, not their time. Therefore, the hours they work merely divide the salary they earn, so that their effective hourly rate decreases as hours worked rises, but is never zero. Of course this isn't applied symmetrically--a salaried worker who completes his tasks "early" in the week is unlikely to be allowed to take the rest of it off without a reduction in pay (or, more likely, suspicion and being marked for termination).
In regards to #1 I understand your point. I am suggesting that a large portion of software developers in startups do have significant responsibility, authority and share of profits. Even at a large company: Consider a scenario of a Principal/Architect earning $150k salary and $150k in equity grants and working on a flagship product. Even at an extremely large company it appears to me, based on my own experience, that a meaningful connection exists between effort and reward.
In regards to #2, I don't think you can wave away this point. A raise is, in fact, compensation.
If the "Principal/Architect's" equity grant affords him a definitive say in the direction of the company, then he's an owner, not a worker. If it doesn't, then the grant is indistinguishable from an ESPP with a larger discount, in terms of raw compensation, and any hours he works over 40 is unpaid (i.e. 'free' work).
With respect to #2: A raise is not, in fact, compensation. It's a promise of future compensation based on future work. Calling it anything else is, frankly, less than honest.
Any equity stake is improved by work in the company. Your argument seems to be that the work isn't sufficient, but you're not being clear about where you draw the line.
WRT #2, the impact on income when viewing a career over a long period of time is clear. It appears you're being dishonest by refusing to acknowledge this rather obvious fact. You can say "it's not compensation" until you're blue in the face but at the end of the day people who do more will on average earn more.
I think the line is somewhat blurred, but I object to the notion that I'm not being clear: if the equity stake doesn't give a person the same, or comparable, authority and take of the profit as "major" shareholders, the person is an employee, not an owner. To take one specific case as an example, if a person can be fired by the "real" owner/equity holders, he's not an owner. So the "Product Manager" in your example, if he doesn't get paid for every hour he works, is working for free for some hours, regardless of the particulars of the ESPP program he's participating in.
You're conflating "compensation in year N+1" with "being paid for hours worked in year N." The compensation you're talking about is dependent on work done in year N+1, and is not some sort of "back pay" or "built in, amortized pay" for the work done previously.
A simple example may help illuminate what "working for free" means in the context of this discussion.
1) You earn $1000 per week in salary compensation.
2) You work 45 hours during a given week.
3) You are paid $1000 for work done that week.
It doesn't matter whether you get a raise the next week, or what the amount of the raise is. For the week you already worked you contributed 5 hours of unpaid, free labor to your employer.
Surely what matters is the overall, effective, relationship between work done and money received.
Consider the following "toy model", which is of course a deliberate oversimplification. Everyone starts at a salary of $10000/year, in exchange for which they are required to work 40 hours per week. If you choose, you can work 50 hours per week instead (all year), in which case your salary the next year will be 10% higher.
In this model, all the compensation-or-not-whatever-you-want-to-call-it you get for working extra hours is in the future, and does indeed depend on your working in the future. Just as you say.
On the other hand, someone who works 40 hours a week every week for 10 years gets $100k and someone who works 50 hours a week every week for 10 years gets almost $160k.
I don't think the fact that the compensation is in the future and subject to uncertainty makes it any less real. Imagine that it works this way instead: When you work extra hours one week, you get "paid" a token. If you collect enough tokens, you can exchange them for a salary increase. It seems clear to me that these tokens have value (would you rather have 100 of them than 50? Yes? Then they have value), and therefore being given them in exchange for doing extra work is a form of compensation.
Of course in the real world you don't know that your extra hard work will lead to a salary increase, and you don't know you'll still be working at that employer to take advantage of it. So let's say that those tokens are exchangeable, once you have enough, for stock options instead of salary increases. Again, clearly compensation (though you might choose to be skeptical about how valuable it is) and uncertain in very much the same ways.
It is, for sure, true in a sense that for the extra-hard week you just worked you contributed some "unpaid, free labor to your employer". But you probably did it with the understanding that it would, on balance, lead to your employer paying you more in the future (either in recognition of your productivity and commitment to the company, or on account of being more successful through your efforts, or both). The difference between this and real unpaid work is much the same as the difference between lending someone some money (when they might just run away with it and never repay you) and giving it to them.
Now, if you weaken your claim a little and say that extra work is generally very badly paid, I'd be much more likely to agree.
> that consistent hard work generally results in rather substantially increased compensation and opportunity in the long term.
I agree. I don't think I'd be in the same place in my career today if I had a "working for free" attitude. Sometimes it was more hours and sometimes it was just a genuine curiosity about taking on challenges that are completely foreign to me.
Getting increases in comp is always a long-term game. There's no road map (i.e., work 2 more hours building this to get that experience which will increase my market value by X) and you'll go down a lot of wrong roads that will add zero to your market value. More importantly it requires delayed gratification. There can be a long delay between the work you put in and the payoff.
But as well as your comp expectations, your comp is also adjusted accordingly through a bonus program. I am not sure the advice provided applies to those with a bonus and equity incentive.
Bonus is part of an overall compensation package. Its not payment for 'overtime'. Its just a different risk/reward calculation then only salary. The key point in all of this is to have a mutual understanding of expectations with your manager.
> I have a finance background and previously worked for UBS and in retail and wholesale stockbroking. I fully understand the role of, say, market makers in the financial system (ie providing liquidity).
> I hear those same arguments being used for HFT but honestly they don't really stack up (IMHO). <snip> It must be a challenge and I'm sure you can make a lot of money, which is fine, but it doesn't really contribute anything to the financial system.
These two statements stand in direct contrast. If you understand the role of market makers then you should understand that the most popular HFT strategy is to be an electronic market maker. It's about competition for liquidity provision which tightens spreads which reduce the cost of trading for everyone.
> 2. Not allowing investment banks to fail ("too big to fail" like Merrill Lynch). Bailouts of investment banks could reasonably be described as welfare for investment bankers;
This would have caused the collapse of the global economy. Google would have suffered mightily from the following depression. Companies like ML might "deserve" to be punished and that process is now working it's way through the courts, but that bailout cost a lot less than 25% unemployment.
> 3. Allowing banks to take positions as well as operate as market makers. There really should be strict separation here;
By it's very definition, a market maker must take proprietary positions.
> 5. Strict regulation on commodities market to separate speculators from hedgers. Driving up food prices causing untold misery even death should result in criminal liability for the organization and key personnel.
Hyperbole. There's never been any evidence that speculation on commodities is "driving up food prices" let alone causing "misery event death".
> For any job that you're likely to get in your 20s, NYC finance jobs pay worse, have far longer hours, and are much more high-pressure.
If you're working in the front office then your pay is going to be a multiple of a startup job. And it's going to be significantly higher then any established company in the area (Amazon, Facebook, Google). A good rule of thumb is your base will be similar what established companies pay and your bonus is all gravy.
Pressure and hours vary by firm. It's hard to generalize.
I'm not sure how much bonuses software engineers get. I've heard it's great for bankers, traders etc. But do software engineers(who work on building infra/tools, not the kind who write trading algos) get paid good bonuses as well? By good I mean 30%+
The closer to the trade you are the higher bonus is as a percentage of your total comp. I did mention "front office" in my original post.
- Infrastructure is vital to most trades. Bonuses are going to more than 30% if you're on the cutting edge here - prop shops, some technology oriented hedge funds, and some banks.
- If by tools you mean GUIs then no.
- Engineers who write trading algos (or more likely algo components that are used by a trader at a higher level) will have a bonus that is more than 30%.
Common enough. Skills are so specialized thhough that it doesn't always make sense to switch though. Its a different career track, not a promotion. Top end trader will make more but top end developer will make more then the average trader. You should focus on what you are good at.
2) Shorting isn't free. It will costs money to borrow the stock. You have to borrow it for more than 6 months until the deal closes.