Finance-AT2 | Price Impact © By Kevin Mak and Tom McCurdy

| January 30, 2017

Question
AT2 | Price Impact

© By Kevin Mak and Tom McCurdy
In addition to the trades of TNX Corp that your sales team has generated, you note that the
conglomerate MegaCorp (ticker MC) has just released earnings. The release was very positive and
caught the street off guard since industrials haven’t been faring well lately. However, the firm also
announced that a restatement of previous quarter’s financial results will be necessary due to some
‘accounting oversights’.
MC is widely held by nearly every pension fund and the opinions as to whether this is good or bad
news varies depending on who you talk to. Ultimately, nobody knows, but everyone seems to be
taking a view, by either adding to their positions or selling some of their holdings. Your sales team
has been on the phone all morning and will let you know, when the market opens, the exact number
of shares that you will need to buy or sell.
The average daily volume of 1 million shares / day for MC is very low and institutional trading
today will dominate the market volume, and consequently, the market price. That means that if
there are more sellers than buyers in the market, you can expect the stock to fall throughout the
day, and vice versa if there are more buyers than sellers.
When you receive your instructions to buy or sell shares, complete the order throughout the day by
purchasing or selling shares of MC using market or limit orders. Your clients will measure your
execution skill based on the fill that you assign them versus the daily VWAP.
One of the junior traders is asking your advice on using “Marketable Limit Orders”. You explain to
them that a marketable limit order is a limit order to buy (or sell) a specific quantity of shares at
specific price that is higher (or lower) than the current ask price (bid price). The purpose of using a
marketable limit order is to effectively execute a market order and get filled at market prices, but
set a limit to how far through the limit-order book the order will possibly fill through. This will
prevent “fat finger” trades and bad fills from occurring due to a lack of liquidity. You’ve given them
the following example:
Initial Order Book:
Bid Volume
1000
4000
4500
10000
15000

Bid Price
9.98
9.95
9.90
9.88
9.84

Ask Price
10.00
10.03
10.07
10.42
10.45

Ask Volume
5000
3000
2000
18000
25000

If a student were to put a market order to purchase 13,000 shares, they will get fills from the first 4
orders on the ask-side, and be getting some shares at 10.42 (an awfully undesirable price). Instead,
if they put a limit order to purchase 13,000 shares at a price of 10.10, it will fill 10,000 shares at the
first three prices, and then insert a limit order to buy the remainder (3000 shares) at a price of
$10.10.
Page 1

Agency Trading Simulation #2 – AT2
When the AT2 simulation begins, you will receive a message instructing you to either purchase or
sell 100,000 shares of MegaCorp. Your goal, over the 10 minute period, will be to build a position
where the cost closely matches the VWAP for the entire period. The trading simulation is designed
such that an entire day of trading will occur over a 10 minute (real-time) trading period.
MC shares will begin trading at a price of $47.00. During the simulation, programmed liquidity
traders (ANON) will buy and sell shares on the open market. These orders will be small, relative to
the size of those from human traders. That is, the latter are participating in an illiquid market so
that, in this case, their trades will have a large effect on determining market clearing prices.
Discussion Questions and Follow Up:
(1) How does your average cost compare to the market average cost? What is the optimal
strategy to employ so that your average cost is close to the average cost of the market?
(2) How does the lack of liquidity affect your execution strategy? Does this affect your use of
limit orders and market orders?
(3) What information may be gleaned from the limit order book when executing your strategy?
What types of strategies can be employed to counteract or exploit these strategies?

Page 2

Teaching Notes:
This case is designed as a follow up to the AT1 trading case. The specific differences are as follows:
1. Instead of every student purchasing 100k shares, some students will be purchasing 100k and
some will be selling 100k. At the beginning of each iteration, they will be randomly assigned a
‘purchase’ or ‘sale’ instruction, each with a 50% probability. Due to variance of the
assignment, there will likely be an imbalance of buy and sell assignments across the class.
2. Traders have less information regarding the volume that will trade throughout the day.
Instead, they will need to observe the trading volumes that occur on the market and forecast
expected trading volumes based on known information.
3. Traders can forecast potential price changes if they can determine relative supply/demand
imbalances in the market (potentially by observing the order book).
4. Traders are not being instructed to specifically use limit or market orders, they should use
both depending on the situation.
Students should begin by estimating the volume of trade that will occur throughout the day. As stated
in the case, on average, there are 1 million shares traded by the ANON trader with itself. In addition,
they can expect that each trader will purchase or sell 100,000 shares. When one student purchases
100k shares from another who sells 100k shares, the volume traded on the market (reported as
trading volume) is actually 100k and not 200k (which some students may initially suspect). The
correct estimation of # of shares to be traded throughout the day is
( )
With this estimate, they can observe the volume of shares being traded on the market and roughly
estimate the % of daily volume that has occurred (similar to AT1). The main difference is that the
shape of the volume curve will be entirely dependent on what the students do. If very few of them
trade at the beginning, the volumes will be low, and other students will infer that more trading will
happen later in the session, and therefore wait to trade until later.
While trading is occurring, students should try to observe the limit order book and try to observe
whether there is an imbalance between buyers and sellers. If there fewer students that need to
purchase 100k than students who need to sell 100k, there will be a natural supply for the stock and
that will cause the market clearing price to be lower by the end of the simulation. This strategy is
much harder to execute than it is to explain, because while trying to determine whether an overall
supply/demand imbalance exists, there can be temporary imbalances simply reflecting the current
strategy of buyers and sellers1 and not necessarily reflecting the market’s overall share requirement.
Students should also decide whether it’s advantageous to use limit orders versus market orders. The
advantage of using market orders is that it disguises their trade intentions since their order does not
1

Consider a situation where there are 15 buyers and 5 sellers. By the end of the case, the price of the stock will
certainly be higher. But, during the beginning of the case, perhaps all 5 sellers are aggressively selling and only one
of the 15 buyers are buying. The observation is that there are ‘more sellers than buyers’, even though it’s simply a
temporary dislocation.

Page 3

appear in the limit order book (students can still observe the information in the time and sales
window). The information leakage has a cost to it that was not a problem in AT1 since all of the
students know that everyone’s order is to purchase shares.
Students are not being encouraged to trade ‘both sides of the market’ but some may do so anyways. As
discussion point 3 discusses information leakage and strategies to exploit or counteract strategies, one
common method will be for students to be regularly purchasing and selling shares on the open market.
If a student wants to accumulate 100k shares, they could do so by purchasing 200k shares and selling
100k shares throughout the day- the net position being 100k overall. This makes it less obvious to
other market participants that the student is purchasing, since they’re regularly seen on both the bid
and ask side of the market.
As a follow-up and more technical discussion that is not simulated by RIT, in real markets, traders
have various other options such as:
1. Hiding their trading ID from a trade (either executing through an anonymous ECN, or simply
not disclosing their trader ID when submitting the trade).
2. Using “dark pools” that allow anonymity of fills, and do not display bid or ask interest in the
market.
3. Internalizing trades from other clients and matching them internally.
The following is an evaluation guideline:
Within 15 cents of VWAP
Very Good
Within 20 cents of VWAP
Good
Within 25 cents of VWAP
Average
Within 30 cents of VWAP
Needs improvement
More than 30 cents from VWAP
Improper execution
* Note this guideline is much looser than AT#1 due to significant price movements caused by the
human traders. If the human traders are experienced, results should be significantly closer to VWAP.

Page 4

Get a 30 % discount on an order above $ 50
Use the following coupon code:
COCONUT
Order your essay today and save 30% with the discount code: COCONUTOrder Now
Positive SSL