Hi, I have only very limited knowledge (next to none) on trading, so maybe someone here can explain this to me: How large is the trading account that one trader at a BB is in charge of? ofc this depends on the asset class, the rank of the trader, his/her experience, etc. Also, please note I am talking about an agency/flow trader who makes markets for clients. I really have no idea how large such an account is. 50 million? 100 million? more, less? and how large is (on average) the transaction volume of a trader p.a.?
thanks for any explanations, as I really don't have a clue on this.
Do you want to define the "size" of a trader's account? Do you mean the stop, the risk budget (defined as a VAR tgt, most commonly), max drawdown or something else entirely? Regardless, no matter what measure you choose, it will vary wildly, and so will the transaction volume... Moreover, it's quite common that none of the metrics above are the limiting constraint and it's something else entirely that defines account "size". At any rate, it's simply not possible to generalise and you shouldn't even try.
Why do you want to know these things? What's the ultimate question you're trying to answer?
yes, you are probably right, I guess it is hard to quantify.
As for what I'm trying to answer: I'm trying to understand how many traders you need to execute a deal (yes, ofc this depends on how liquid the market is, which asset class, etc.). so for instance, if a large asset manager wants to buy very liquid bonds (e.g. US treasuries, to name the arguably most liquid bond around) and the transaction size is one hundred million (so thinking of a super large asset manager, e.g. blackrock or agi or pimco). how long would it take under normal conditions to execute that deal? how many bond traders would be involved?
im just trying to understand the dimensions of trading.
$100mm in a liquid bond (bear in mind not every US bond is equally liquid) is a very simple process, like Martinghoul said only one trader unless it's someone very junior. You might not even need the sales force, the phone or the chat. A window pops up in the screens of the traders in the desks with all the info needed (firm, trader, notional, price, number of dealers involved...) depending on whose bond it is (each trader covers different ones) the trader can accept it, reject it or quote another price (like a counter-offer). Takes under 20 secs.
The average BB flow trader has a stop loss of 5mm...aka...once they go neg 5mm on the year, most BB flow traders will get the tap on the shoulder that says "stop taking prop risk." This is the average. A new junior trader will have a stop-loss closer to 1-2mm. The average desk head will have a stop-loss of 15-30mm. These are "averages." For the few BSDs out there..the numbers are massive. I heard that Glen Hadden lost something like 300mm (i can't remember exactly..200mm??500mm?? something in that ballpark) at MS on one of his first big prop trades (a TIPS auction / curve trade back in early 2011 or late 2010...i can't remember). However, he made something like 1bln at GS before his move to MS...and MS let him take the risk and keep going. I know he recovered most of that P&L...but i don't know the rest of the details. This is an example of a flow BB trader who does NOT fit your "average" criteria.
ok thank you. that's quite interesting. especially the stop loss limits. yea heard of glen hadden too.
Interesting...how much revenue does a trader have to pull in to justify getting those risk limits you mentioned (1-2mm, 5mm, 15-30mm)?
Regarding your "large trade question"....if we just bump up the numbers to something meaningful...we can have a discussion. Lets say a customer (maybe the central bank of Russia) calls up the 5yr trader at MS and asks him to bid on 10bln 5yr notes. This is a sufficiently large trade (maybe 10% of avg daily trading volume in the 5yr) to warrant attention. The trader will bid on the securities...but his bid will be pretty far off market to account for the pretty massive risk. With the normal bid/offer spread of 1/4 tick...and avg liquidity of 100mm/1/4 tick...you might guess that the trader would bid 10-15 full ticks behind the current bid...but recall that the liquidity function gets thicker as price moves...so...100mm for the 1st 1/4 tick..200mm for the next 1/4 tick...300mm for the next and so on...(these are just estimates...but they are in the ballpark). Assuming a normal quiet mkt during NY hours...the right bid is probably 5 ticks back (unless the mkt is at the highs of the day..in which case you need to add a few more ticks for your risk). Trades of this size are not the norm...and are pretty rare. If a customer actually wanted to sell 10bln 5yr notes..they would normally split the trade into 10 lots of 1bln each. and they would probably get a much better avg price (unless they time the mkt perfectly in which case the dealer would be screwed...this is not the norm). In all these cases...the trades would still be handled by 1 trader, and take less than 20 seconds to execute. The hedge might take longer to get off...but that is a separate issue and gets more into the style of the market-maker.
its usually a sink or swim situation...the junior guy is expected to make 0-2mm (1st years usually lose 0-1mm)....the avg guy is expected to make 10mm...and the desk head is expected to make 20-30mm. Making slightly less than your expectation will only be allowed for so long before you get replaced....and losing will usually get you replaced pretty quickly. So with 8-10 guys on a desk...each desk has an expected revenue of approx 100mm. So when a desk only makes 50mm or less...that is a bad year..and more than 100mm is celebrated (as you'd expect). When a BB desk makes less than 30-40mm...you usually see guys getting replaced. Exceptions are for when you have a guy who made 300mm one year, and then loses 20mm the next...he will have a couple years of "bank" built up so that he can lose a little for 1-2 years before he needs to worry about losing his job. If this is your situation...you won't be getting relevant advice on a board like this.
The 1st year guy (his 1st time trading the banks money) gets his 1-2mm risk limit for free...he has accomplished nothing when he gets this...other than sitting on the desk and being a bitch while paying attention and trying to learn thru osmosis, for as long as it takes for a senior guy to say "i think he's ready." Sometimes this happens right away...sometimes it takes 1yr. Usually you will spend most of this time building spreadsheets and doing research projects for the more senior traders (this is how most traders learn the mechanics of their markets...by doing work / projects for the more senior guys)
So, for the junior guy (1st year)...just not losing money while "actively" trading is an accomplishment....and making 1-2mm is hitting your expectation. You are expected to increase revenue by 1-2mm / year...so by year 3-4 you are potentially a 10mm/year guy.
If you can beat these numbers...then you are a potential rockstar. If you come up shy on these numbers...then you are a potential layoff candidate.
Very informative, thanks for the info. Would give you a SB but I'm out. Hope your personal trading has been going well BTW (ran across one of your posts from earlier this year about that).
I find the annual stop limit thing is not true, at least it doesn't work like that in my experience (plus, that's not a stop limit), although I assume that it depends on the product you trade, but it's not that rare to lose 2 or 3 bucks IN A DAY, and 5 happens from time to time and I'm talking about someone who has been trading for 4 or 5 years, not a desk head. Although it's true that these people make money if you account for the whole year, otherwise you are in trouble. Having max drawdown limits on an annual perspective doesn't make that much sense, and as for the daily ones, they are probably around what you said, but again, I'm talking about daily.
I agree with the assessment you did for the juniors though, my MD told me when I started that they considered breaking even on your first year a good performance. That is first year trading, not first year on the job obviously.
i'm not talking about drawdowns...i'm talking about your annual P&L going -5mm (this number is specific to the firm / desk)
how a firm (and sometimes this is desk specific) handles annual stop-loss limits for the regular traders can obviously vary quite a bit. My last year trading for a bank...my stop-loss (on paper anyway) was 5mm...meaning if i ever went -5mm on the year...i should expect to get the tap on the shoulder. In a sense...this was the equity capital that i had to trade prop with (with unlimited leverage). My budget (or expected revenue) was 15mm..so in a sense the firm was risking 5 to make 15...(with the potential unlimited option in rare cases)....i didn't have any daily limits..other than to not breach my annual limits.
Sure, there were days when my P&L would swing around 1-2mm...but so long as i didn't go -5mm on the year based soley on my prop trading...nobody paid much attention. I would consider myself a middle of the road trader in that respect. Nobody was expecting me to make much more than 15mm. Another guy on my desk went -5mm and he got the tap..no more prop trading..just be a flow trader and handle customer flow.
As a flow trader (treasuries and MBS in particular...but i'm sure this holds for other desks as well)...customers can easily put you into positions that you would never take as a prop bet given your risk limits. If you clearly get run over by a customer and lose 5mm on a single trade in my example (i've seen this happen a bunch)...if you are close to zero before that trade takes place...its common to get a "mulligan" for that P&L hit and chalk it up to the cost of building a franchise...or put that trade into a desk book (sortof the same thing). But if you are just slinging around 100mm 10yr notes with no flow..and you go -5mm on the year..in my experience...you get tapped. Senior guys and desk heads obviously get much more leeway...but there is a point where mgmt will tap a desk head (these are a little more sensitive or course...but they happen..every firm has a max pain point for their traders)
I see. I was talking about total PnL, not purely prop. Just a misunderstanding, sorry man.
how much $$ BB gives you to trade? (Originally Posted: 01/02/2011)
I know trading really depends on individual (based on group, MD etc) but I am pretty sure that there is some consistent range of how much capital you are given (assuming BBs to make things easier) to trade according to your level (analyst, vp, director etc)
I worked at BB NY trading during the summer and one director told me he made 30 million dollar for the firm last year and got 3% as a bonus.. (The bonus % became lower after 2008-2009) but i didn't get to ask him how much capital he were given to make that return
Also, I was wondering how much capital analyst (assuming he got his book), associate, vp are given to trade on average.
It doesn't really work like that in a BB. Its more that you have certain risk limits as a trader but not necessarily a specific amount of capital. Ie as an analyst you are not told that you have 1 mil of capital, but rather that you have these risk limits x,y,z. As an analyst you will also probably be helping out with a book or just doing flow.
Also I know of one BB that charges traders for using capital.
I agree with derivstrading. No matter what asset class you trade, it all revolves around your risk, normally expressed in VaR (Value at Risk) and your vol (volatility).
For example on a Rates Desk, your desk will be given an overall VaR limit for your entire desk but there will also be specific risk limits for specific tenures/points on the curve. For instance, you will have specific limits on the amount of risk/exposure you have for each bucket (e.g. 2s, 5s, 10s, 20s and 30s). You normally want your risk to be flat (i.e. zero) for each bucket, unless your desk has a strong view on a point on the curve.
Many people tend to get caught up in "dollar notional" thinking (i.e. constructing a portfolio/book based on dollar value). The major flaw with this mindset can be expressed using a simple example with the "traditional" 60-40 stock-bond split for a portfolio. This allocation ignores the fact stocks tend to be much riskier than bonds. As a result of this misguided allocation, the majority of your risk lies in stocks (e.g. ~70% or more, depending on your test). Hence, the rationale for managing a trading book/portfolio based on risk as opposed to "dollar value".
Your duties as an analyst on a trading desk depends on your asset class. For example on an FX options desk, you'll help the senior traders price trades. Especially when you start, you'll mostly be monitoring the risk for the books (e.g. making sure delta, vega and gamma are within limits) and hedging when necessary.
Qui omnis optio qui itaque laudantium tenetur. Quam molestiae eaque sit incidunt repellendus distinctio. Velit aspernatur est et non dicta et repellat. Sed sed perferendis et harum. Corrupti inventore dolorem nihil.
Eaque quod possimus sapiente. Perferendis natus similique ut eos et accusamus. Veritatis qui fuga vitae quis aut repellat voluptas. Quaerat consequatur assumenda sint ut rerum autem maiores.
See All Comments - 100% Free
WSO depends on everyone being able to pitch in when they know something. Unlock with your email and get bonus: 6 financial modeling lessons free ($199 value)
or Unlock with your social account...