Market Making Interview Question

J

JamesTrader

Dear readers,

Soon I will apply to various prop traders. I have actually prepared quite a bit already. However, despite extensive research online, I am yet to be convinced about an optimal way to solve the ''make me a market'' questions.

What I know is:
- They ask to make a bid and ask
- They may or may not define the spread you are allowed to use
- They give or ask you the confidence level of your answer
- Remain calm, confident but not arrogant, decisive and ask smart questions

Theoretically I know the approach, but I have not witnessed it in practice yet.

I therefore would really like to ask two scenarios with the in between steps of finding the answer:
- Make me a market on the number of people working in this office and give me your confidence level and why?

- Make me a market on the people living in India, the spread if maximum 100 million. What is your confidence level and why?



A friend told me that knowing the context is also important, because games like these are not how the real markets work. For example: what would be my incentive to make a market and trade, I do not know the other trader and he may have more information than I do. If this latter is the case he will never trade if its unprofitable for him to do so, in that sense me making a market would be pointless as there is no gain for me.

I immensely appreciate it, as this is the only thing I am currently stuck at.

I hope you all have a nice day.
 
Last edited by a moderator:
JamesTrader said:
what would be my incentive to make a market and trade, I do not know the other trader and he may have more information than I do. If this latter is the case he will never trade if its unprofitable for him to do so, in that sense me making a market would be pointless as there is no gain for me. I immensely appreciate it, as this is the only thing I am currently stuck at.
More...

I have worked in an inst environment where people are making markets near me, although personally I wasn't responsible for making any markets.

You ask what would be your incentive to make a market? Stating the obvious your M.O. is to make a net profit long term. It depends on the markets you are going to trade as each market has it's own participants who each have their own M.O. You state above that a trader with more information will never take a trade if it's unprofitable, i take this to mean negative expectancy. This is where you need to shift your understanding. Market Makers can make money by taking the orders (providing liquidity) to non price sensitive traders. These non price sensitive traders may have more information than you however they are in the market for another reason. For example a producer who needs to hedge, a fund that needs to roll etc. A lot of these participants just need to get business done and dont care about the price so much.

MM's to some extent tend to have a financial incentive to trade in the form of exchange rebates. This can be directly volume linked or in some cases a stipend for hitting volume targets which is popular with new challenger/emerging products/exhanges.

MM's also look to profit by trading using stat arb strategies. Volatility bands can be used. To get into these markets you need to work for an inst or a prop firm that specialises. GL.
 
Propwarrior,

Thank you very much for this insighful and enlightning reply. It has provided me with a more down to earth understanding what is actually going on.

Would this be a valueable thing to point out during the interview process? For example in the financial world no one is ever making a market on the population of India, the information is available for everyone.

In a sense it would be wise to ask the interviewer wether or not he or she is someone who is not price senstive?
 
Questions like these are pretty retarded as you cannot apply the principles of market making in rates to the population of a country. I would therefore use 2rosy's suggestion for the interview Q but understand that making a market on a desk is different. Your decision there will be based around current vol, sentiment and whether it will reduce or extend your current exposure.
 
Thank you for the suggestions guys, really appriciate it.

So if I have it right by using 2rosy links I must start with a sample based on probability, earning the premium and then fluctuate my market based on what the interview does (buying and selling) and try to balance my books in the end?
 
"- Make me a market on the number of people working in this office and give me your confidence level and why?"

Market 2 / 100

Why: I see two interviewers. So by bid is solid. I'm not gonna get run over because -I can see u-. Offered at 100 because you're just on 1 floor, and not that awesome.

"- Make me a market on the people living in India, the spread if maximum 100 million. What is your confidence level and why?"

No.

Why: I dont know (or care) about the population of india, but I think 100M is too narrow to quote it anyway. Reckon it's somewhere between 200M and 500M. So. My information is to crap to quote that tight. No market. No cares. Call someone else first and i'll dime him. Maybe.

Edit: If india is mentioned in an interview, RUN. Your tech support will be shithouse.
 
You obviously have to make a market. The point of the question is to understand if you understand what market making is as well as to see if you are capable of committing to an answer where there is either bad information or information asymmetry.

My answers would be:
for the office - a wide market because there is obviously massive information asymmetry and I will only get picked off. If I have to make a market I would make one based on some assumptions and guesses but I would skew to the upside because I could understate the number - especially when you consider that MMers tend to have a lot of ops people. I would make the market very small.

For india - I would make a market of 1Bn-1.1Bn. The reason is that I think India has 1Bn people in it but if I am wrong the number is more likely to be slightly higher. I would make it a normal size.

In practice the MM business is very competitive - spreads are almost intolerably low. Most of the money is made in the risk management, not in the bid/offer spread. It takes huge infrastructure to monetize the small spreads in liquid markets. They want to see if you are okay with this. It's not natural.
 
1. Make me a market on the number of people working in this office and give me your confidence level and why? First estimate the number of people in the office. Let's say your number is 60. Then your bid can be 30, and ask is 120. Confidence should be high in this case, meaning the actual number is somewhere between your bid and ask. But it's a safe bid that will probably not get transacted because the spread is too high but will avoid loss if transacted.

2. Make me a market on the people living in India, the spread if maximum 100 million. What is your confidence level and why? First guess the population of India. India was chosen because you probably don't have that number memorized. I think (without checking) it is about 800 million. You then bid for 700M and ask for 900M. Since you are not sure of the actual population in India, your confidence will be low. But you can add that population data is readily available in census data, is easy to forecast, and won't fluctuate much month to month. So your actual confidence level can be high.
 
You must log in or register to reply here.
Back
Top