Economics of Market Making by Robert A. Schwartz and Bruce W. Weber Zicklin School of Business Baruch College, CUNY Università degli Studi di Bergamo Corso di Laurea Specialistica in Ingegneria Gestionale Intervento di Enrico Pellizzoni (R&D - Borsa Italiana SpA) per il corso di Analisi dei Sistemi Finanziari I Prof. S. Paleari, Ing. S. Vismara Presentazione tratta dal sito: www.nasdaqtrader.com/headtrader Slide 1 Session 1 Overview of Market Making And Introduction to the Head Trader Simulation Slide 2 1
Purpose 1) Gain experience with market maker operations 2) Understand how prices are set in financial markets 3) Have fun Slide 3 Simulation is an Important Learning Tool Simulations can provide a real-world context for analyzing trading Slide 4 2
An Important Distinction Simulations can be based on either 1. "Canned" data from past events 2. Computer-generated data Head Trader is based on computer-generated orders and price changes not canned repeats of past events Slide 5 A Trading Simulation Requires Prices can be affected by your actions You have some basis for anticipating future price movements Prices do not follow a random walk Simulation runs can be replayed, changing some parameters while holding all else constant Computer generated data are superior to canned data Slide 6 3
What Drives the Simulation? Informed P* Is p*>offer or p*<bid? 3 Sources of Orders Liquidity Order Flow Quotes, Prices, Volume Momentum Is there a trend/ pattern? Trading Mechanism Competing Market Makers Slide 7 Market Maker Costs What Were Your Costs in Head Trader? Cost of carrying an unbalanced inventory Acceptance of diversifiable risk "Realized spread" can provide a return Cost of ignorance Were you "bagged" by better informed traders? Profits from liquidity traders offset the cost of dealing with information traders Can a dealer market exist without liquidity traders? Slide 8 4
Market Maker Revenues As a market maker, how do you get your profits? The bid-ask spread Short-term trading profits Knowing your customers Knowledge of the order flow Trading the order flow Slide 9 Market Simulation - Step 1 Enter Slide 10 5
Step 2: Log-In Enter a market maker ID for yourself E.g., On actual Nasdaq display, GSCO=Goldman Sachs Slide 11 Step 3: Choose Level of Complexity Enter Slide 12 6
Step 4: Customize Enter Slide 13 Five (5) Cent Tick Size Enter Slide 14 7
Hide Dealer Positions Enter Slide 15 Terminology Your BID quote is the price you will buy at when an order is sent to you Slide 16 8
Terminology Your OFFER quote is the price you will sell at when an order is sent to you Slide 17 Your Market Maker Quote-Setting Recap: a quote consists of A bid to buy An offer to sell Quote size is in terms of round lots of 100 shares As a market maker, you set quotes and trade from your own inventory when customers want to buy or sell Remember, future price changes are uncertain Slide 18 9
More Terminology Your quantities are the minimum you will buy or sell when an order is sent to you; you may be able to buy or sell more Slide 19 Terminology Your bid and offer quantities can be set at either 10, 20, 30, 40, or 50 Slide 20 10
Your position and performance As you buy and sell, your trader summary will be updated here Slide 21 You Receive a Customer order... An order has been sent to you; You can sell either 10 or 16 round lots Slide 22 11
What should you do, buy 1,600 or 1,000? Check Trader Summary: You are long 1,000 shares Profit? Check Avg Paid for shares in user s current position Slide 23 How are you doing? Realized Profit = $0.10 profit on 1,000 shares = 100 Slide 24 12
You receive an order from a Dealer... An order has been sent to you; You can sell either 10 or 20 round lots Slide 25 Market makers use quote adjustments to control their inventory positions In Head Trader, after a market maker, who is short, receives a buy order (increasing his/her position) he/she may do any of the following: Match or better the existing best bid quote Move off the best ask quote Raise their bid and ask Buy from another market maker Alternatively, market maker who is long and receives a sell order may: Match or better the existing best offer quote Move off the best bid quote Lower their bid and ask Sell to another market maker Slide 26 13
How did TERI perform? Total Score is a function of six performance measures Slide 27 The Head Trader Simulation Are You Ready?... The Market is About to Open Slide 28 14
Session 2 Market Maker Operations: Inventory Control, Transparency, and Competition Slide 29 Market Maker Services Price discovery Liquidity Immediacy Slide 30 15
Inventory Control What made you a successful market maker? Inventory control Trading the order flow carefully Ability to hide/disguise large positions Knowledge of customers (source of the order flow) is also important in practice Slide 31 Where Does P* Come From? Ask Bid S(p(t)) P* Ask Bid D(p(t)) Buys Sells Buys Sells Slide 32 16
Typical Pattern of P* and Best Bid and Offer $28.00 $27.00 $26.00 $25.00 $24.00 $23.00 P* Ask $22.00 $21.00 0 1 2 Trading Days 3 4 5 Bid Slide 33 Price Discovery Definition: Finding P*, the value that best reflects the market s underlying demand to hold shares of an asset Locating P* is not a simple matter Slide 34 17
Price Discovery 1. A dynamic process 2. Accuracy is difficult to achieve 3. Accuracy depends on Market Structure: Definition: the rules, systems, and protocols that determine how orders are entered and translated into trades 4. Also depends on the trading behavior of individual participants Slide 35 The Other Two Market Maker Services Liquidity Immediacy Slide 36 18
Liquidity 1. Perfect liquidity: definition With perfect liquidity, a stock can be traded immediately at its equilibrium value, P* Slide 37 Liquidity 2. Attributes of a liquid asset Breadth: orders on the book exist at an array of prices in the close neighborhood above and below the price at which shares are currently trading Depth: orders are of large size Resiliency: price changes due to temporary order imbalances quickly attract new orders to the market, thereby restoring reasonable share values Frequent trading Slide 38 19
Liquidity 3. Attributes of an illiquid asset Little breadth, depth, or resiliency Accentuated short-period price volatility Trading cost, the shark, can wipe out all of your investment return Slide 39 Immediacy Market maker practices are designed to facilitate the rapid execution of customer orders However, large orders are commonly traded patiently (i.e., without immediacy) Upstairs negotiation of large block trades Breaking up large orders for submission over time Slide 40 20
Inventory Control If P* jumps above your offer, your customers will, on net, be buyers and your inventory will fall As your inventory falls, you raise your bid and offer The higher bid attracts sellers and the higher offer discourages buyers What happens to your inventory if your bid is raised above P*? Your inventory is controlled by adjusting your bid and offer relative to the unobserved P* Slide 41 Quote Revision & Inventory Change $26.50 $26.00 $25.50 $25.00 $24.50 $24.00 Inventory Ask 600 500 400 300 200 100 0 $23.50 $23.00 Bid -100 Do you know where P* is? -200 Slide 42 21
Machine Dealers Inventory Control Rules Inventory control links P* and mechanical dealer quotes Market makers have no strict position limits - rather, responses are triggered when inventory exceeds a multiple of 2,500 shares Inventory > 1x Trigger: Match or improve best bid (when short) or best offer (when long) Inventory > 2x Trigger: Also, buy or sell with other market maker to reduce position Inventory > 3x Trigger: Move bid one price increment below ask, or ask one increment above bid Slide 43 Transparency Why do we care? * Better order handling for investors Trading requires making tactical decisions Observing market maker quotes and recent trade prices and sizes facilitates proper timing and sizing of orders Knowing the quotes and trades enables customers to monitor the quality of their order executions Slide 44 22
Transparency As a Market Maker, How Transparent Do You Want the Market To Be? After you acquire a large inventory in the process of servicing a customer, you must work off that position You are now in the position that the public customer had been in "Shares sold to a market maker are still for sale You do not want your inventory revealed by a trade publication Delayed trade reporting may be desirable in a dealer market Slide 45 Transparency Why do we care? 1. Better market quality Liquidity (?) Stability (?) Fairness Price discovery Slide 46 23
Optimal Transparency There is an inverse relation between transparency and market maker-provided immediacy Few markets are totally transparent Whether or not to reveal your orders is often a choice given to public traders. Institutional investors in particular need this choice Slide 47 Competition In the simulation, did you compete via: The location of your quotes? The width of your bid-ask spread? What effect does preferencing have on price competition? Slide 48 24
View From a Market Maker s Desk Assume SHORT has a short position and wants to buy Dealer Bid Dealer Ask SHORT 26.00 GREAT 26.20 SWIFT 26.00 SHORT 26.30 TERI 25.90 SWIFT 26.30 GREAT 24.80 TERI 26.30 Should SHORT raise its bid to $26.10? Slide 49 Would SHORT Benefit From Bidding $26.10? The higher bid attracts a public seller for whom the value of the stock is between $26 and $26.10 (low probability) The tighter spread increases public interest in the stock, and hence order flow to the dealers (possible) but the next order does not go to SHORT (with preferencing, this is quite likely) Conclusion: Little Benefit: SHORT s incentive to raise the bid to $26.10 is weak Slide 50 25
So, How Do Market Makers Compete? Knowing their customers Offering an array of services Developing customer relationships; this results in Preferencing Quote matching A market spread that is greater than it would be in an order-driven environment Slide 51 Market Maker Preferencing In the simulation, what effect did preferencing have on The volume of orders you received? Your inventory control? Your profitability? Under which regime would you prefer to operate: Preferencing, or Strict price and time priorities? Slide 52 26