THE LONG RUN (large scale)

S&P index ex dividends. Large scale. It seems an exp function. The problems that occur with
this graph:
- We are not able to see volatility due to the scale. The shaded bars represent
recessions. It could be argued that in the past the recessions were more frequent
(and the shaded bars ticker). >> problem linked to the scale. This is why at the
beginning of the sample you are not really able to evaluate volatility.
- We are not considering inflation, in fact, cost of life increases overtime and
purchasing power grows
- the index is considered ex dividend. this means that it represents just capital gain.

- The last two graphs. They are in a log scale. In this way, we could fix the scale
problem.
- After economy contracts, then it increases rapidly. There seem to be a relationship
between the gray bars and equity markets.
- The graph doesn’t solve the other problems (inflation and the exclusion of the
dividends)
 -
-

- The last graph is normalized to zero in 71 and considers inflation.

- Prociclity: economy goes up> sp500 up. So, covariance is a source of risk.

STOCK MARKET PARTICIPATION

These data represent the participation as fraction of population. For most countries it is low.
Indeed, stock market is risky and most investors are very risk averse.
Financial education, trust in financial markets (households particularly averse to risk).

HISTORICAL PERSPECTIVE: THE VERY LONG RUN

Main features of these returns:
- they are cum dividends
- they are real returns: it is considered inflation.
Real return risk free is the return of government bonds, usually Treasury bills (maturity one
year and issued by U.S. government).
the risk premium is the difference between the real return market index mean and the real
risk-free return.
The power of compounding is demonstrated by the fact that even small differences on
returns will create huge differences in the final values. Over a long period of time this will
produce huge differences.
Short run bonds are less risky than longtime bonds because of the low variation of return
rates overtime.
REAL ANNUALIZED RETURNS ON EQUITIES BERTHS BONDS AND BILLS INTERNATIONALLY

CAVEAT 1
 Most of the evidence comes from the US or, at most, from advanced and developed
economies. We are not considering countries with less attractive performances.
 The risk is that conclusions suffer from the, so called, survivorship bias: i.e., the risk
of excluding from the sample just the countries with less attractive performances,
there by biasing upward the estimates.
 Jorion and Goetzmann 1999 look at 39 countries (developed and emerging), for the
period 1921-1996, and observe that the risk-premium (real and net of dividends) is
the largest for the US (4.3% per year), while the median is only 0.8% per year.
Returns of built in Italy are negative because inflation consumed their returns.
Survivorship Bias during WWII

Survivorship bias or survival bias is the logical error of concentrating on the people or

things that made it past some selection process and overlooking those that did not, typically
because of their lack of visibility. This can lead to some false conclusions in several different
ways. It is a form of selection bias.
Survivorship bias can lead to overly optimistic beliefs because failures are ignored.

CAVEAT 2
 In the long-run, even small differences in average returns have large consequences.
Therefore, it is very important to take into account confidence intervals.
 For example1, consider a market with E(R) = 6% and σ = 20%. How many years (N) do
we need to safely conclude that the average return is ≥ 0 according to standard
significance levels?
 We can use the standard formula for a t-test with confidence level set, for example,
to 5%.
 Therefore, we need to find the value of N such that the statistic is ≥ 2 (i.e., the latter
is approximately the corresponding critical value for the t-distribution):

  If the average return was 3%, we would need N = 178 years of data!

HISTORICAL PERSPECTIVE
This is a shorter time series column from 1988 to 2020. The government is less likely to
default compared to corporations This is why, the bonds issued by the former are less risky
then the ones issued by the latter. When economy contracts call mom the black line
increases while the redline decreases. The black line goes up because of a sort of
substitution: investors flight to quality, meaning they fly to safe investment (so, their
demand goes up as well as their price).
The black line represents an asset which is a very good diversification because it behaves
like an insurance. Instead, the red line increases the volatility of the portfolio. We are able to
do this statements because we evaluate the covariance of the returns (assets and overall
market).

RISK VS. RETURN

There are several measures of risk. One of them is volatility. volatility and mean Returns are
positively correlated. supposed that financial returns are normally distributed, they can be
described just by mean and standard deviation. Actually, there are a lot of evidence that
suggest this doesn't represent reality properly.

RANGE OF RETURNS
In the worst-case scenario:
- T bill will have zero percent return
 - equity bonds we lose 25%

STOCK RETURNS HISTOGRAM

the returns are not perfectly symmetric like a normal distribution. In fact, they have fat tails.

TIME-VARYING MEANS
The means move over time. Means are not constant.
TIME-VARYING RISK Also volatility moves, risk changes.

RETURNS CORRELATION MATRIX

This table reports the correlation coefficients (measures of movement of two random
variables) between the return on government bonds and the return on other stocks.
Ro=1 ro=-1 ro=1
Diversification is important and it is linked to the reduction of portfolio's volatility.
The numbers are the averages of a sample between 1988 and 2019. The diagonal has one as
values.
Equity and government bonds move in opposite directions while corporate bonds and
government bonds move in the same direction.
The same table estimated during the Great Recession between 2007 and the 2009. Equity
bonds are represented by S&P 500 and small capitalization stocks become even more
correlated.
S&P 500 and corporate bonds are still negatively correlated and the same happens with
Treasury bills.
All the risky securities become even more similar and all the safe securities become even
more similar.

Correlation heatmap: normal times

On the axis there are assets. In normal times:
- the red squares correspond to the different asset classes
- The rest of the space is pretty much of light color. This means low correlation. So, if I
build a portfolio with different assets, I will diversify My Portfolio.

Correlation heatmap: crisis times

When crisis comes there are two asset classes
- risky securities
 - safe securities

RISK-ON – RISK-OFF (RORO)

Motivated by the evidence that the correlations across many assets increase in bad times,
HSBC developed the idea of risk-on – risk-off and a related index (i.e., the RORO index).
The intuition is that in bad times some assets become risk-on (i.e., risky assets) and others
risk-off (i.e., safe havens) regardless of idiosyncratic factors.
The risk-on – risk-off world is bi-polar, with two subsets of assets (risky & safe) and a high
correlation across assets within each group, and a negative or small correlation between
assets across the two groups.
Risk on means positive correlations while risk-off means negative correlations. This is
motivated by the fact that correlation between assets increases during crisis.
Roro index
SUMMARY STYLISED FACTS

Conclusion
 Performance of some ”asset class” very interesting.
 Why aren’t we all buying stocks (in the US, in 2007, 50% of households, or 23% of
high-income households, held no stock)?
 Risk increases with returns.
 Inflation eats up much of the returns from risk-less instruments.

Money market
 also cash,
 subset of the fixed-income market,
  short-term debt instruments (≤ 1 year) generally very liquid,
large denominations ⇒ but there exist money market mutual funds (MMFs) that are
easily accessible (i.e., it is possible to purchase shares also in small denominations
and with little to no transaction fees) and that contribute to the provision of liquidity to
the market.
 Government short-term bonds: in the US T-bills (e.g., Treasury Bills), in Italy
BOT, etc.,
 Certificates of Deposit,
 Commercial paper (and more recently ABS commercial paper),
 Repos and reverses,
 Key money market rates: Federal Funds Target rate and ECB main rates
(marginal rate, refinancing rate, overnight, etc.),
 Key interbank rates: LIBOR, EURIBOR, EONIA, etc.

Repos
Supposed that bank A has a government bond worth $100 and needs cash.
Bank A can sell the bond to bank B today and at the same time commit to
repurchase the same bond tomorrow at a slightly higher price.
This kind of transaction is called Repo, or repurchase agreement.
The standard maturity of a Repo contract is overnight (even though it is possible to
find Repos with alternative longer maturities).
A Repo contract is like a collateralized loan: if tomorrow Bank A does not have the
money to repurchase the bond, Bank B can keep the bond that covers the value of
the defaulted loan.
Note that typically Repos are over-collateralized, i.e., the value of the bond is larger
that the cash lent.
The percentage difference between the value of the collateral posted and the cash
lent is called ”margin”, or ”haircut”.
in the price, the rollover of repos would allow the financial institution
to raise only 85.5 in the wholesale market, which is insufficient to repay
Repos, haircuts, and fire sales
its liability of 90.
Brunnermeier and Pedersen (2009) consider these two effects in a dynamic model,
where some agents (customers) are faced with a liquidity shock in their endowment
and have to sell their assets. These customer trading needs are accommodated by
investment banks that
Figure: Main components of the money market

Commercial paper
 Short-term unsecured debt notes issued by large and well-known companies
 Often backed by bank lines of credit
 Maturity up to 270 days
 Issued typically in multiples of $100K
 Small investors invest in commercial paper through MMFs
 Low risk
 Recent trend: asset-backed commercial paper (ABS) issued by financial
institutions (the funds are used to purchase assets that then serve as
collateral).

LIBOR: details
LIBOR stands for London Interbank Offered Rate.
It’s the rate at which large banks are willing to lend to each others in London.
Typically, this rate is quoted in USD and it’s the main reference rate in the European
money market.
You can also find other LIBOR rates that are quoted in different currencies (Euro,
Yen, CHF, etc.).
The EURIBOR (European Interbank Offered Rate) is the rate at which banks in the
Eurozone are willing to lend funds one another.
LIBOR is designed as a survey among the biggest banks to estimate what are the
rates at which they believe they can borrow from other banks.
Over time, many contracts have been based on the LIBOR rate (for example, $350
trillion of derivatives have payoffs linked to this rate!).
However, bear in mind that LIBOR is not an effective rate: it’s only an estimate that
comes from a survey. This feature increases the risk of manipulation.

The LIBOR scandal

LIBOR scandal: recently it was uncovered that banks participating to the survey were
using collusive strategies to influence the LIBOR rate to their advantage (according
to some testimonies, this behavior was going on since 1991!).
Several banks received large fines: Barclays $450m, UBS $1.5b, RBS $612m, etc.
Alternatives
Alternative risk-free rates are being set up in many countries:
 For the US dollar Secured Overnight Financing Rate (SOFR)
 For the Euro, the Euro Short-Term Rate (e STR)

TED and OIS spread:

The TED spread is the difference between the interbank rate (3M LIBOR) and the
short-term government rate (3M T-bill) and is considered a measure of default and
counterparty risk.
The LIBOR-OIS spread is the difference between the interbank rate (3M LIBOR) and
the overnight index swap rate (or OIS) on the interbank rate. The LIBOR-OIS spread
measures liquidity risk and default risk on the money market.
The OIS is a rate on a derivative contract on the overnight rate (on the interbank
market): to be more precise, it’s an interest rate swap where the variable part is
equal to the geometric mean of an overnight rate for each day of the reference
period.
MMFs and the break the buck risk (for example, look for the story of the crisis of the
Reserve Primary Fund, which held a lot of Lehman debt at the time of its default).
TED spread

LIBOR-OIS spread
Is the money market risk-free?
 The risk of a run on MMFs during the financial crisis convinced the US
government to extend a public guarantee to all the MMFs.
 New regulation after the financial crisis that took effect in 2016 (Dodd-Frank)
says that MMFs must hold at least 99.5% of their portfolio in cash or
government securities and repos collateralized by these instruments to be
exempt from fees and gates in times of stress (to avoid fire sales and
contagion).

The fixed income market

Debt securities with a maturity longer than money market instruments:
 Government bonds (T-bonds, BTP, etc.),
 Inflation-linked government bonds,
 Emerging market bonds,
 Corporate bonds,
 Municipal bonds,
 Mortgage-backed securities example MBS .
MBS

ABS
MBS Consider a bank that makes a $100K loan to a subprime borrower that wants to
buy a house1.
Assume that the borrower will default with a probability p = 10%.
In case of default, the bank can recover 50% of the face value of the loan (e.g.,
$50K) by foreclosing the house.
This is a risky loan!

By tranching the loan the bank can offload some of the risk.
The bank creates two tranches of equal face value (e.g., $50K):
1. senior: paid first in case of default,
2. junior: paid only after senior creditors are paid in full.
Clearly, the senior tranche has zero risk: even in the event of default, creditors are
paid in full.
The junior tranche is risky: with 10% probability creditors lose everything.
Since the senior tranche has zero risk, it will typically receive a very good credit
rating (e.g. AAA) and can be sold to investors looking for safe investments (cf. Ben
Bernanke and the global savings glut).
back

Can the bank do better?

Yes, by tranching + pooling.
Suppose the bank pools together two subprime loans with same default probability
and face value $100K.
The bank can create two tranches:
1. senior: $100K.
2. junior: $100K.
Note that the senior tranche is risk-free: even if both loans are in default, senior
creditors are paid in full.
What about the junior tranche? If both loans default at the same time, we go back to
the example without pooling.
But what if defaults are uncorrelated ...
Suppose the loans’ probabilities of default are independent of each other.
Recall Bayes rule in the case of two independent events: P (A ∩ B ) = P (A)P (B ).
In our example there is 1% chance (i.e., 10% × 10%) that both loans are in default
and 18% chance that only 1 loan is in default (e.g. 2 × 90% × 10%).
The bank can further tranche the junior tranche into:
1. senior mezzanine tranche,
2. subordinate equity tranche.
Now the mezzanine tranche loses money only 1% of the time, when both loans
default.
As you probably have guessed, this process goes on if the bank can do more
pooling and tranching.
In the limit, the bank could claim that 90% (i.e., 1 minus probability of default on
individual loan) of the MBS are super-safe (no default risk) and receive a AAA credit
rating.
Typically, the super-safe tranche is sold to investors looking for safe investments
(e.g. MMFs).
The bank then needs only 10% of the loan value to stay in business.
Note that if defaults are perfectly correlated, there could never be more than
50% of the mortgage pool that is safe.

Equity market
Stock and bond market indices
DJIA (Dow Jones Industrial Average):
 30 large stocks (blue chips),
 simple average (∑30 P /30). j=1 j
Standard & Poor’s Composite 500:
 500 stocks,
 market capitalization-weighted average. 􏰀 Additional indices:
Nikkei (Japan),
FTSE (UK),
FTSE MIB (Italy)
DAX (Germany),
Hang Seng (Hong Kong).
MSCI (several countries).

DJIA
Example: Data
Example: DJIA
⇒ equivalent to a portfolio containing 1 share for each component of the index. 􏰀 ⇒
the simple average weights more stocks with a higher price (price-weighted
average).
⇒ Index and portfolio have the same %∆ every day.
Let’s consider a hypothetical version of the DJIA with two stocks:
Portfolio
Index
Initial value = $25 + $100 = $125
Final value = $30 + $90 = $120
%∆ of the portfolio value = -5/125 = -4% Initial value = (25+100)/2 = 62.5
Final value = (30+90)/2 = 60
%∆ of the portfolio value = -2.5/62.5 = -4%

Stocks with the higher price dominate a price-weighted index like the DJIA.
Of the two stocks in the portfolio, the first had a price increase of 20%, the
second a price drop of 10%.
However, the final effect on the index is dominated by the drop in price of the most
expensive stock.

Example: S&P 500

Consider now a hypothetical version of the S&P 500 with two stocks.
The initial capital is allocated to the two stocks according to the relative market
capitalization.
With our data, the index assigns a weight to ABC 5 times as large as that on XYZ
The return on the index is:
500(30 −1)+ 100( 90 −1) = 0.15 = 15%.
Note that we get the same result if we compute the percentage variation of the
value of the index:
690 − 1 = 0.15 = 15%. 600
600 25 600 100

Market-value weighted indices

With market-weighted indices, the final effect is dominated by the stocks with the
largest market capitalization.
Market-weighted indices are the most common, since they avoid the intuitive
limitations of price-weighted indices.

Tracking error volatility

A standard measure used to estimate how far the returns of a fund are with respect
to a given benchmark is the tracking error volatility:
TEV = 1/(T-1)∑(rt −rB,t)2, t=1
where rt is the realized return in period t, and rB,t is the return on the benchmark in
the same period.
TEV is commonly used by index funds and exchange traded funds (ETFs) to
implement an automatic portfolio rebalancing.
At the same time, TEV is a measure commonly provided to investors in most mutual
funds.

How firms issue securities

Firms sell or float shares when they need to raise capital.
These new instruments are marketed to the primary market.
There are two types of primary market issues of common stock:
1. Initial public offering (IPOs).
2. Seasoned equity offerings (in Italian, ”aumento di capitale”).
There are two types of primary market issues of bonds:
1. public offering.
2. private placement.
 Primary market ⇒ offers in this market determine the number of financial
instruments that are traded
 Secondary market ⇒ trading in this market produces a transfer of ownership
of a financial instrument, without affecting the number of financial instruments
that are traded in the market.

If you had invested just $1,000 in Wal-Mart’s shares at the time of its IPO in October
1970 you would be rich: at the end of 2006 you would have accumulated
$1,370,000!
You would have had similar great performances if you had invested in Home Depot,
Intel, Cisco, etc. (over time, we will be able to tell if it was a good idea investing in
recent IPOs like Facebook, Zynga, Twitter, etc.).
Unfortunately, these are only lucky examples.
Not all companies that do ipo are successful stories. The issues:
If you are investing in a company that is doing an ipo, you have some risks.

When the company goes for the ipo: this is changed over time. when Microsoft went
public>> start up would always as soon as possible do an ipo to do big investment.
Why did these companies an ipo very soon? in that time there was not a large
venture capital industry. The founders listed their companies very early. On one
hand you receive more risk at the beginning, on the other hand there is more (?).
If a company list itself later, investor miss the opportunity. Regulation prohibits to
small investors to invest in non-listed companies.
Mutual funds specialized in start-up investors>> they pool their savings.
Facebook class A shares>> what does it mean? There are different types of stocks.
These different shares have different rights. When these companies went public, the
founders wanted to maintain control (you need a lot of shares or special shares).
Sometimes investors are happy that the founders have these special shares or a
large number of basic shares.

LONG RUN RETURNS OF IPOs

The apr from IPOs is relatively small. On average, it is not a success story (take a
look at the first three years). After 4-5 years, the companies become more similar.

Types of markets
They are ordered in increasing volume of trading.
1. Direct search market (e.g. Craigslist, Ebay). Not so quick to buy, it takes some
time. the market works when the volume of traded is relatively small. Ipos is a direct
search market, also a private equity could be it. You have to search for the
opportunities. Non organized market.
2. Brokered market (e.g. real-estate, primary). the broker doesn’t face the inventory
costs (risk of inventory). The primary market is a brokered market: there are
companies that help other firms to sell their shares.
3. Dealer markets (e.g. bond OTC). You have also this intermediated, however the
intermediate has inventory, they own some shares.
4. Auction market (e.g. NYSE). In which you have a computer algorithm that
matches the order in order to give to market orders to buy or market orders to sell
always the best prices.

Types of orders
Market orders (auction market). They are almost immediately executed. They are
executed at the current market price. The big advantage>> speed of execution.
 executed at the current market price (usually contingent to a maximum
number of shares),
 bid-ask spread. Supposed you look at the Bloomberg terminal, you have the
price if you want to buy and the one if you want to sell.
 speed of execution,
 best price quote.
Price-contingent orders. They are not immediately executed. If the price becomes
the same, I put in my order, it will be executed. if this doesn’t happen then my order
won’t be executed.
 buy order at or below a stipulated price,
 sell order at or above a stipulated price.
 stop-loss and stop-buy (common types of orders for short sellers) . stop buy>
order executed when the price passes some value. It is used by short seller.

Market orders:
- executed at the current price (usually contingent to a max numbers of shares)
- bid-ask spread
- speed of execution
- best price quote
price contingent order:
- buy order at or below a stipulated price
- sell order at or above a stipulated price
- stop loss (sell securities when they go below a specific price) and stop-buy
(short sells)
they provide liquidity, their fees are higher. While market orders absorb liquidity.

Limit order book: this one is for FedEx stock on NYSE market.
If you want to buy (using a market order), your order will go on the right side (ask
side). If you want to sell> bid side.
If you enter a price contingent order to sell your order will end up in the bid side. The
order is not executed immediately.
Market order> your order will be immediately executed absorbing liquidity.
Supposed you send a big market order to buy>> you’ll accept from the ask side
more than one selling offer.
Market orders will absorb this liquidity. Limit order book changes all the time
continuously: currently 9 out of 10 contingent orders are canceled.

This image shows how the market is created.

Order book> contingent order. Provide liquidity to the market. In normal times they
would take advantage between the bid and the ask. They buy and after sell at a
higher price, providing liquidity.
Market order> absorbs liquidity
You can cancel your price contingent order anytime you want. In fact, most of them
are actually canceled. This is not strange. Market conditions change all the time, so
investors change their orders. Nowadays, 60-70% orders are placed by algorithms.
They continuously change the orders and replace them.
There are different views related to pros and cons:
Pro algorithms can provide a lot of liquidity to the market. Difference between ask
and bid. More liquidity, less risk. It is a measure of cost.
Con: The risk: some investors can use the protrunning (?). They can see your order
faster than you. Indeed, they could delete the first order, before your order is
executed, so you will pay them more money.

The deepness of the orderbook> advantage: the liquidity is improved. Overtime the
measure of liquidity is shown by this graph.
The number of shares offered bid ask has increased since 2003.
This figure tells that smaller stocks are less liquid.
Price contingent orders
Column> conditions
Righe>> actions
You buy when the price is below the limit>limit buy order
Stop-buy: hedging. You buy when the price is above the limit. Short sell.
Stop-loss: you sell when the price goes low.

Electronic trading. The shift happens for very specific reasons.

 During the October 19, 1987 ”Black Monday”, when the NYSE lost
approximately 22 percent of its value in one day, allegedly several brokers did
not take calls from investors trying to sell their stocks.
 Authorities responded by forcing computers to start doing the job of brokers ...
 Regulation in response to the Nasdaq scandal of 1994: Nasdaq accepted to
insert in its
displays the bid/ask prices from other ECNs.
 Regulation NMS (2005): the objective is to digitally connect the different
exchanges. The idea was: let’s put in competitions the exchanges.
 In the US, the fraction of shares that are exchanged electronically went from
16% in year 2000 to 80% in the most recent years.

Black Monday

ECNs. Electronic communication networks. Now it is an exchange.

 Fully automated ECNs have significantly increased their market shares (e.g.,
Direct Edge, BATS, NYSE Arca, etc.).
 matching (or crossing) of the orders from the different ECNs.
 The rise of ECNs is linked to the growing importance of high-frequency
trading: a trading strategy that tries to exploit arbitrage opportunities across
the different quotes of the same stock in different ECNs.
 A word that became very popular is latency: the time to accept, process and
submit an order (buy/sell). For example, BATS advertises an average latency
time of 0.000082 seconds (82 microseconds) with a 10 Gig connection!
 For those of you who are interested in algorithmic trading, a must read book is
Michael Lewis, Flash Boys.
Exchanges compete in terms of prices and time of execution.

This figure shows how the regulation changes the market structure. Light blue
line>>NYSE was the main market. In the recent years others markets started to
grow.
Figure: Market shares of trading in NYSE-listed shares.

New trading strategies

 Algorithmic trading (high-frequency trading is a type of algorithmic trading).
 High-frequency trading: liquidity vs. risk (flash-crash May 2010, cross-market
arbitrage, front-running, level-paying field, etc.).
 Co-location.
 Dark pools.
 It is estimated that half of the daily volume of stock exchanges is initiated by
an electronic algorithm!

Flash Boys: details

Figure: Stock markets in the world by market cap (Note: Borsa italiana market cap is
$694 billions).

Buying on margin
Buying a stock on margin means that a fraction of the investment is financed by a
broker.
The margin is the fraction of the purchase price that is contributed directly by the
investor (the remaining fraction is borrowed from the broker).
In the US, the minimum initial margin is fixed by the SEC to 50% (Note: SEC stands
for Security and Exchange Commission).
Brokers in turn borrow from banks at the call money rate to finance investors.
The shares bought on margin must be maintained with the brokerage firm and are
used as collateral for the loan.
Since I am borrowing my investment is risky: supposed I buy stocks and the price
drops. I lose money and have to pay the broker. No the shares you bought act as
collaterals.

Example: Buying on margin

Initial value of investment: $10,000 (100 shares x $100).
Loan from broker: $4,000.
Equity: $6,000. Your own money.
Initial margin % = = 60%. Equity/investment
The initial margin is also simply defined as equity.
The account balance of the investor is:

. . . If the value of the stock drops to $70, the investor’s account balance is:

The margin % goes to: $3,000/7000 = 43%. You lose 3000 equity.
Suppose the price continues to drop:
⇒ If the value of the stock drops below $4,000, the equity of the investor becomes
negative.
To avoid this possibility, the broker sets a maintenance margin:
if the margin is below the maintenance margin, then the broker issues a margin call,
in this case, the investor is required to put new cash (or shares) in his account,
if the investor does not follow suit, the broker can sell securities (assets) from the
investor’s account in order to pay enough of the loan to restore the percentage
margin. Two possibility: the broker can either ask for additional collaterals or sell the
investor’s securities.

Suppose the maintenance margin is 30%. How far could the stock price fall before
the investor gets a margin call?
 P is the price of a stock,
 the value of an investment is 100P,
 the equity is 100P - $4,000,
  the margin % is (100P − $4, 000)/100P .
(100P −$4,000) /100P = .3 ⇒ Pmin = $57.14.
Equity/ value of assets=0.3

Investors buy on margin when they want to invest an amount of resources greater
than their own money allows (leverage).
The investor can achieve a greater upside.
At the same time, the investor is exposed to greater downside risk.

Example 1: Leverage. Excel file week 3

 Suppose an investor expects the price of the IBM share to increase in
following year with respect to the current market price of $100.
 For example, assume the investor expects a 30% increase (ignoring
dividends) in the price and has $10,000 to invest.
 The expected return of an investment in 100 share of IBM (no leverage) is
then simply 30%.
 What is the expected return if the investor borrows $10,000 from a broker and
buys 200 shares in IBM?
 Let’s assume the interest rate on the margin loan is 9% (call money rate +
spread).
 The expected value of the 200 shares in IMB is $26,000.
 After repaying the principal ($10,000) and the interest ($900), the investor is
left with $15,100.
 The expected return of the margin investment is then:
($15, 100 − $10, 000) /10000 = 51%.

What happens if, after a year, the price of IMB shares is below investor’s
expectations?
 For example, suppose the price of the shares drops by 30% to $70 per share .
 In this unfortunate (for the investor!) event, the return from the margin
investment is:
($3, 100 − $10, 000)10000 = −69%.

Example 2: Leverage
From 2006 to 2009 house prices in the US fell 30%. As Mian and Sufi (2014)
explain:
 the collapse in house prices hit low net-worth households the hardest
because their wealth was tied exclusively to home equity.
 the fact that low net-worth households had very high debt burdens amplified
the destruction of their net worth.
The amplification is the leverage multiplier.

Consider a home owner with 20% equity in a home worth $100K.

In this case, the loan-to-value ratio is 80% (e.g., the mortgage has a face value of
$80K).
 If house prices fall 20%, what the % decline in home equity?
 When prices drop, the house is worth only $80K.
 The home equity has been wiped out, for a −100% decline!
 The leverage multiplier is 5: a 20% decline in house prices lead to a 100%
decline in home equity (i.e., leverage is assets over equity!).
Synthetic leverage: options don’t cost a lot. These options give me the option to buy
in the future.
Short sales
 With a short sale, an investor first sell shares (that he doesn’t own) and then
buy the shares. If the price go up it will be very expensive.
 Why should an investor short sell some shares? Because he wants to profit
from a possible decline in the price of the shares.
 How can an investor sell something he does not have? Simple: by borrowing
the shares.
 Eventually, the investor must buy the shares and cover the short position (and
give back any dividend that the stock might have paid).
 Nowadays, most short sales are settled with a cash exchange.
 Revenues from short sales must be kept in a special account with a broker
and cannot be invested till the position is not covered.
 Short-sellers are also required to deposit collateral or cash in a deposit with
the broker to cover potential losses in the event of an increase in the price of
the shares.

Timing

Example: Short sales

 The price of Dot Bomb is $100 per share.
 Suppose you ask your broker to short sell 1,000 shares of Dot Bomb.
 The proceeds from the short sale are $100K.
 Assume the broker asks a margin equal to 50% (i.e., initial equity) and that
you decide to use $50K in T-bills to cover it.
Therefore, you must have at least $50,000 in cash or securities in your account.
  Suppose the price of Dot Bomb falls by $70 per share.
 You buy 1,000 shares and you hand them to the broker.
 Your profit is $30,000 ⇒ ∆P× (number of shares).
You receive a margin call if the price of Dot Bomb goes up above a given threshold.

 Suppose the margin is 30%.

 How much can the price rise before you get the margin call?
 Equity = Assets - Short sale position.
 Equity/value of stocks = 150,000−1,000P/1000P = .3
Pmax = $115.38.

Short sales and buying on margin

 With buying on margin, the investor borrows some funds and his/her debt is
independent from the price of the purchased stock ⇒ when the stock price
changes, the value of the debt does not change.
 With a short sale, the investor first borrows some shares, and then will have to
give them back ⇒ when the stock price changes the value of the debt
changes as well.
 Use of stop-loss and stop-buy orders.

Short sales
 Securities lending is a huge business: the global volume of loaned securities
in 2008 was $2.3 trilioni.
 Who are the main lenders in this market? Mutual funds, ETF, pension funds,
etc.
 Who are the main borrowers? Hedge funds.
 What are the profits in this business? Consider as a reference that a popular
ETF like iShares Russell 2000 Index Fund in the fiscal year 2008 made
almost $15 milioni from security lending, about 10% of the total profits.

What’s the cost of borrowing securities?

There exist indices that track these costs. For example, the S&P Securities
Lending Index Series.
What’s the information that we get from this cost (demand/supply)?
What’s the risk? Typically, the risk is considered very small as borrowers must offer
collateral in the form of cash or securities for a value at least equal to the value of the
borrowed funds.
However, since collateral is generally made of securities, changes in their prices
expose the securities lenders to a default risk.

During the subprime crisis, the SEC issued a ban on short selling (September 19
2008) on 799 financial companies ”to protect investors and markets.”
During the Eurozone debt crisis, in the EU was issued a ban on short selling
sovereign bonds and sovereign CDS.
Short sales
Can short selling have real effects on asset prices?
 Information and equilibrium prices.
 Market efficiency.
 For an interesting discussion on the effects of short selling you can look at the
blog
of Kenneth French e Eugene Fama and search for a post titled Securities Lending.
Short sales and real effects
The evidence that short sellers know something more than the market is weak at
best.
And even if they had more information, the action of short sellers would push prices
faster towards the new equilibrium.

Short sales and real effects

 Consider a situation in which 50% of the times short-sellers know nothing, and
50% know something about the price of a stock.
  If all the other investors don’t know for sure if the short sellers are informed,
or not, then stock prices do not go down enough when short sellers are
informed, and go down to much when they are not.
 Over time, stock prices converge to their equilibrium values, but in the short-
run there could be real effects.
 These real effects could be catastrophic for some firms: for example, think at
financial firms that have a maturity mismatch between assets and liabilities.
Note how, in the example of the previous slide, we have real effects when short
sellers do not have any special information.
Costs and benefits of short-selling.
What about over-optimistic investors?
Appendix
Flash boys
 Micheal Lewis, in the second chapter of Flash Boys, describes brilliantly how
Brad Katsuyama, then working for RBC, first realized what HFT traders were
doing.
 Think about the figure describing the limit order book for FedEX we have seen
before. go to figure
 In the first row, we read that 400 shares are offered at 90.05: This means that
it is possible to buy those 400 shares at 90.05.
 Suddenly, Brad realizes that this was not possible anymore: once he had
pushed the ”buy” button, immediately the offered shares would disappear
from the limit order book so that he might be forced to buy at the next best
offer (i.e., 104 shares at 90.06 and 296 at 90.07).
 What was going on? back
First thing that Brad does is to call tech support.
Their answer is: ”it was because I was in New York and the markets were in New
Jersey and my market data was slow” and ”you aren’t the only one trying to do what
you are trying to do”.
 But Brad is not convinced: he did not see the shares on offer change unless
he hit the buy button.
 And in fact he tells tech support ” You see, I’m the event. I am the news.” back

 Over the years there had been several changes on stock markets.
 In 2002, 85% of all stock market trading happened on the NYSE and a human
being was behind every order.
 Stocks basically traded either on the NYSE, or the NASDAQ.
 In 2005, exchanges became for-profit public corporations competing with
each other.
 In 2008, there were already 13 exchanges and basically every stock traded on
all of these exchanges.
 The exchanges compete, for example, in speed of execution and fees. back

In 2002, the exchanges simply charged every Wall Street broker who submitted a
stock market order the same simple fixed commission per share traded.
Then things got a little trickier with the market-taker model:
 Say you want to buy a share in Apple that is quoted 400-400.05.
 If you enter the market and buy directly the share for 400.05 you cross the
spread.
 The trader that crosses the spread, is a taker (of liquidity).
 If, instead, you rested your order to buy Apple at 400 and at some point
someone
decides to buy your shares at that price, you are a maker (of liquidity).
 In other words, taker is another word for demand and maker for supply.
 Exchanges charged takers a small fee, paid makers a small fees, and
pocketed the difference.

The fee-structure was not the same across exchanges: for example, BATS paid
takers and charged makers.
back

Brad realizes then what was happening:

1. When the market showed, say, 10,000 shares of Apple offered at 400 it was
bundling the orders of several investors;
2. Some had the ability to jump out of the queue, by cancelling their orders. 3. The
orders where spread across different exchanges.
”I think it’s because we’re not arriving at the same time”.
Creation of ”Thor”: a code to insert delays into orders so that they arrive at different
exchanges at the same time. Interesting question: when you realize you have
figured out what goes on, what do you do with this information?

 According to Lewis, the origin of the HFT problem lies in Regulation NMS that
was passed in 2005 and implemented in 2007.
 The aim of Reg NMS is simple: introduce competition between exchanges
and guarantee to that investors receive the best possible price.
 For this reason, with Reg NMS regulation moves from ”best execution” to
”best price”.
 ”Best price” is defined as National Best Bid and Offer, or NBBO.

NBBO implies that:

suppose an investor wishes to buy 10,000 shares of Apple,
suppose 100 shares are offered at 400 on BATS, while the full 10,000 are offered on
the other exchanges at 400.01,
the broker is required first to buy the 100 shares on BATS, and then move to the
other exchanges.
How is the NBBO computed? By an algorithm running in a computer called the
Securities Information Processor, or SIP.
SIP is what investors see on their Bloomberg terminals. back
 Suppose HFT are able to compute the NBBO before SIP ...
 In this case, they see the market before everyone else.
 The bigger the time gap between public and private SIP, the greater the
opportunity for HFT.
 The bigger the volatility of a given stock, the greater the likelihood that its
price will change and that NBBO must be re-computed giving an advantage to
HFT with private SIP.
 Ding et al. (2014) estimate that on an average day, the public SIP and that
computed by private traders, for Apple stock, differ 55,000 times!

GAME STOP AND PREDATORY TRADING

It is going to disappear because nowadays it is very easy to download videogames
and digital contents.
It was believed that it was the new blockbuster.
The stock was under-valued. So, investors started buying the stocks.
After, in a very short time the price of game stop increased a lot.
If you believe a stock is overvalued than you start a short sell.
For a lot of these small investors, it is hard to do leverage (because they don’t have
enough equity), so they use options. The small investors use the simplest option: a
call option which allows you to buy something in the future at a given price.
This story has been crusted a bit as David and Goliath.
The video is a seminar run by Markus Brunnermeier. While the speaker is one of the
founders of HQR.
You try to trade to manipulate prices. Short squeeze is a kind of manipulatory
trading. It involves short trading.
Short squeeze: the person who do the short sale, cannot find the stock to sell.
Social media platform’s consequences: meme; echo chambers phenomenon (allows
to learn from other retail investors); coordination and transparency.
Many small investors (David) vs Goliath
The transparency of short positions makes it easy the short squeeze to happen.
Game stop is an example, what is new is that social media allows to coordinate.
- Meme
- Echo chambers phenomenon. Information just travels back to you.
Transparency makes coordination easier.
There are two views about this investing.
1. Many davids small investors and one goliah. We have to fix the market
microstructure in order to eliminate the manipulations and the
inefficiency. First best. Very efficient market.
2. The second view: you just want to create a fair level playing field. The
argument is that the wall street insiders have an advantage in terms of
order execution. So, to have efficient markets you should allow retail
investors to collude. It is the theory of second best. You fight an
inefficiency with another inefficiency.
The second view is similar to Robin Hood. You allow poor to compete against the
rich because they have easy market access. The philosophy of the econd principle.
Robinhood sees the retail orders and passes this information on to other players.
You might say that this is very unfair because retail investors will be front run by
professional investors who buy the information about the order flow. It reveals non
informed trading of the retail investors to market makers. So, there is a trade-off.
This platform is more about entertainment and games.
Short interest>100% There were more stocks shorted than the total amount of stocks
in circulations. More demand than supply.
Naked shorts: usually you need two business day to clear the transaction. I short a
stock, but I can do something even more risky. I don’t borrow the stock, but I just
sell. In two days, I have to borrow a share at a lower price. If it is not possible than, I
can cancel the sell within those two days.
Borrow the stock and sell it. Plus you have to add 10 dollars margin to the broker.
Risks:The price rises you will have the margin calls
Recall of shares by the lender. You have to send back the shares.
Naked shorts delay in settlement. If you have to deliver. Particular special position
you can undo the transaction.

PREDATORY TRADING
The premise of pred trading is that demand does move prices and prices move
stocks’ prices, but this is not a source of profit. There is a misunderstanding.
Trading 20 at the beginning investors buy and push prices to 420.
The average price is 220 probably you are buying at something above 220. You will
think that if you buy at this price you are earning. However, if you sold it the very next
second you will push the price down. Let’s say below 220, and they will lose money.
The price collapses even before you start selling so you will end up selling at less
than 20. So, the fact that you can move the price is a trading cost. Investors try to
avoid moving the prices.
Predatory trading: you try to trade or exploit that somebody is forced in this case to
buy or tricked into buy (short squeeze or gamma squeeze). Price is 20 you push it up
like 200 and at that point you have forced the short sellers. They can’t substain the
risk so they will close their position. They repurchase the shares by buying them. So,
the idea is that it’s the short sellers who push the price up, and now you will exit the
trade you bought between 20 and 200. You sell at a price between 200 and 420 and
then the price collapses afterwards.
You can do that in two ways: forcing the short sellers into short squeeze or pump
and dump. You push the price up and then you trick other people that price is
continuing going up. In the end, you sell to the people you tricked.
David against Goliath kind of story some media call it like that. It describes a
situation in which small investors could all decide that something was worth a lot of
money and then push the price up and make a lot of money. Patterson here is
making simple point, but often misunderstood, that this is not the way you make
money because if by buying, you increase the price, it means that the price goes up
not because there is more value but because of liquidity. You have increasing
demand so sooner or later it will go down and you will lose money.
The short squeeze or the sort of pump and dump schemes
in the short squeeze the prices go up to 420 in this picture because the sellers are
desperately trying to close their positions and they don't find this the stocks anymore.
it will be the price after that, and the investors, in this case, are able to make a lot of
money at the expense of the short sellers.
Pump and dump scheme is a little bit similar. It is very common for in crypto currency
markets right now. Small group of investors are basically are able to use social
media to trick other investors in buying a stock, so the prices go up and then all of a
sudden, the price collapses and again the people that bought the stocks are going to
lose a lot of money. now these schemes are illegal.

SPILLOVER EFFECTS
Spillover effect: you have six short-sellers. The red line tells if the short sellers are
going to survive. If one of them runs out of capital the line moves up and the three
below the line are also forced to liquidate. When the price is pushed up then one guy
is out of capital. As a consequence, the fact that he liquidates push up the price even
more and the other three guys will run out of business and so on. There is kind of a
chain effect, systemic risk. Multiple equilibria: fragility.
Retail buyers pushed up the price that led to a short squeeze. Short-sellers were
forced out of their positions, and so in terms of this picture. The retail buyers talked
on reddit and social media. They buyed game stocks in the stock market, and they
call option (an other way to bet that the price will go up). It is a way that has
embedded leverage: if you have 100 dollars you can buy shares or 30 40 times more
exposure to that stock by buying an option. Gamma squeeze: the market maker and
person who sell this implicit exposure to the stock he will hedge his exposure in the
stock market so he will be buying the share and he is pre-programmed to buy more
share as the price goes up.
And then there are hedge funds who short game style and the largest and most
discussed is called melin.

Open high low close.

January 8th the wildest day.
The low is a 112the high is 483 and it opens here and close here. In one day, the
price moves from 112 to 483. A typical volatility is 2-3% a day.
Now price is about 40.
Volatility: in a log scale you can see how large was the volatility (January).
Above 300% percent if annualized.
It was one of the most traded stock in the world. At the peak we have around more
than 200 percent turnover. In a single day the stocks were traded more than twice.
All the shares traded twice in a single day.
Some of the brokers, like robin hood, imposed trading restriction the day before the
peak. That is why it goes down.
Why people board and sold? Retail investors discussing on reddit. Retail sentiment.
Gamification of trading. These gamify traders wanted to buy game stop stock (a firm
that sells itself games).
Game sellers where not going well. This was a dead industry. They wanted to fight
against people who went short to save game stop.

How much were retail investors buying on these key days?

Net: buys minus the sells. Matt levine a journalist.
Key Monday buying. Price pushed up.
Selling: Tuesday wendsday and Thursday.
Why? Potential stories:
1. Maybe these investors were buying call options: so the market makers
has to have that by going out and buying the stock. They realized very
large embedded leverage. The more the price go up the more the delta
on the option. So, the more market makers need you buy them.
2. Short sellers were eventually forced to close their positions. They were
pushing the price themselves.
Delta: how much the option price change as a function of the stock price. How much
you need to hedge.
Gamma: how much that is changing over time or as a function of the stock price.
3. Institutional investors could be buying. They know that market
manipulation is illegal if they call each other or email each other. The
sec will come after these institutional investors. They don’t dare to this.
Retail investors trying to do that. Coordination devices. It could be that
retail investors were a coordination devise but then a bunch of
institutional investors started buying into this.
Why robin hood restricted trading.
They told market manipulation is illegal. The ceo testimonied.
Robin hoods itself had margin requirement that they had difficulty imposed . the
clearing house itself has significant risk.
Imagine some investors use options to bet on the rising price of a stock.
a. buy (long), leverage
b. buy call option: right to purchase a sock in the future at a pre-agreed
price. With a little equity, potentially, you can take a long position. This
is attractive to risk addicted investors (aggressive): because you can
have a little equity. The worst thing that can happen is that is one
month the price is less than 20 (the one the investor negotiated), the
options are useless. The guys that sell call options have to hedge to
buy and sell the stocks.

Pedersen is explaining the business model of robin hood:

Margin requirement of robin hood. 27>> 100 million dollars. 28>> 1,4 billion dollars.
The reason: when robin hood buys games stock than it takes two days to actually
pay the money and get the stock. A lot can happen to the price. The clearing house
has to control they make good on that transaction.
The set of the level of the margins. It is based on the value at risk.
There was also a capital premium charge of 2.2 billion. They did have 1.4 billion so
then they imposed this restriction that people couldn’t buy. They didn’t have enough
money to pay the clearing house.

The risk the clearing house is taking is proportional to the settlement period.

PAYMENT for ORDER FLOW

Robin hood pays money to market makers to have executed the orders. Citadel
securities is a broker that help other brokers to execute the trade. (Citadel runs a
hedge fund that is traded in the stock market. Citadel made an investment in Melvin
hedge fund.)
Payment for order flow: market makers pay robin hood to execute the orders.
Market makers want to earn the bid ask spread.
Bid at three and ask at 3.01. a market maker wants to buy at three and sell at 3.01.
earn a small spread for providing liquidity. They could lose money if someone is
selling to citadel at 3.01. and people keep selling, eventually citadel will pass the
orders to the market. The orders will hit the market and move it up. Citadel long
position. Market moving up. >> they are losing money.
Informed counterparties: the market makers will lose money.
Market makers don’t like having great counterparties.
They love retail investors hey are not informed and not large. It allows the market
makers to earn the bid ask spread. So, they would pay for it, if robin hood sent the
orders to the market. Any market makers could provide liquidity.

The retail investors are paying zero simply because the transfers of resource is
between the market maker and robin hood. Beneficial for small investors

BACK TO GAME STOP CASE

Short squeeze can happen for technical reason. The short sellers are forced to close
their position.
In the game stock case, there is the belief that this is not what happened.
The securities lending market was affected by a large turnover. Short sellers they
could have continued to short. But they simply could no substain the losses they
were facing. If you buy a stock and the price goes up. The size of your position goes
down and tipically the volatility will go down. Much bigger price moves>> the size of
your position in dollars and the volatility goes up. Plus, you are losing money at the
same time. big losses and big increases in the size of their position and volatility. So
they started reducing their positions.
Estimating short positions.
Dark line: estimating shorting level which is estimated with error. It was 100 and
goes down just over 20 over a week.
Price increased when short sellers were trying to close their positions.
Synthetic shorting: with derivatives
Real shorting: this showed.
Short sellers can have two effects:
reduce the prices when they think the stock is overvalued. They will short it and this
push down the price. They keep the prices in check. Short sellers can reduce prices
and they do that when it is the efficient price. But they can also push up prices. They
make market more efficient. In principle they are good. The macroeconomic effects
could in fact be increase in prices.

When there is less shorting demand the price of shorting goes down.
They had this large position when the price went up and then, when the price
collapses they made money on a much smaller base. So, they lost 14.
In January they lost 14 billion. Then, in February they make 6 billion dollars.

These buyers were just buying and selling.

The short sellers were not pushing down the prices. Some of them closed their
position but they were replaced.
Old owners who sold at this astronomical prices, were the winners.
Prices tend to equilibrium it is a matter of time it was not the short sellers.

IMPROVED INFORMATION TECHNOLOGY

Comments:
As for game stop, the price all of a sudden started to increase. When people talk
about short sellers and their actions’ effects, there is always the question: are they
real effects? In the long term nothing happens to companies or investors. The
evidence is that AMC was actually able to profit due to the increasing prices. They
were able to sell at a higher price, to issue more equity.>> transfer of resource to the
one who buys the stock and shareholders before the issue.>> old owners

1926: market manipulation is illegal. However, the law doesn’t say what
manipulations are.

March 2010: flash crush. For a very short amount of time the market got crazy,
stocks were traded for 1 cent and others 100 dollars. The prices were crazy. After
few days there was an agreement between brokers that decided the transaction of
that hour were canceled. Algorithms created this dislocation of markets. One London
trader was arrested for market manipulation.
Two issues:
3. the flash crush started in the future market. The problem was giving a
detailed definition in both equity and future markets.
4. The definition is so precise, that there is not a single time that more
that one investor was accused of manipulation. It is not so easy to
accuse someone, because you need to prove several things.
SUMMARY from Pedersen’s seminar
Predatory trading is trading that induces and/or exploits the need of other investors to
reduce their positions. This leads to price overshooting and a crisis that spills over across
traders and across markets. Demand moves prices of stocks but is not in of itself how
investors typically make money — market impact is typically a cost. Rather, predatory
trading is when predators force buying or trick buying and make money on the price
increase caused by their orders. Spillover effects occur when the price keeps getting pushed
up and more people have to liquidate.
During Gamestop, retail buyers pushed up the price which led to a short squeeze by hedge
funds. On one day, the price opened at 112 and closed at 483, which is an astronomical
increase in a day. Volatility by the end of January was above 300% annualized and the stock
had over 200% turnover at the peak.
People who bought were not just retail investors discussing the stock on Reddit but also
others. Retail investors were possibly driven by retail sentiment, including gamification of
trading, GameStop belief and nostalgia, and the sense that shorting is “wrong”. Other
buyers could be option hedgers, short sellers closing their positions, institutional investors,
and other retail investors.
Robinhood restricted trading due to difficulty meeting their margin requirements.
Robinhood has to post margin capital to clearing houses which grew from $125 million to
$1.4 billion on January 28th due to the high volatility and an excess capital premium charge.
They had to raise $3.4 billion from existing investors to cover this increased requirement.
Payment for order flow is the idea that Robinhood is paid money by market makers, such as
Citadel, to execute their trades. Market makers want to earn the bid-ask spread and could
lose money if there are informed investors about the fundamentals of the stock or there are
large counterparties. Retail investors are therefore an attractive counterparty for market
makers. This allows Robinhood to give retail investors the lowest trading cost although is
bad for institutional traders.
Shortsellers had to liquidate their position because they could not sustain the losses as the
price continued up. Volatility and size of their position increased as the price went up and
therefore they started to reduce their position. Shorting costs for hedge funds were
probably about 30% annualized in January. When the price collapsed, they made money on
a much smaller base. The price fall could have been due to recent buyers and previous
owners, but the price drop was only a matter of time.
Learned that demand moves prices, demand can be irrational, there are shorting
complications and predatory trading, and the power of social media and IT.

Holding-Period Return (HPR)

Rate of return over given investment period

Rates of Return: Measuring over Multiple Periods

• Arithmetic average
• Sum of returns in each period divided by number of periods
• Geometric average.
 • Single per-period return
• Gives same cumulative performance as sequence of actual
returns
• Compound period-by-period returns
• Dollar-weighted average return
• Internal rate of return on investment

Arithmetic average of HPR is 8,75%, but this is not capturing the timing of the
returns.
Geometric average does take into account the timing. It is below the arithmetic
average. So, in the geometric average the order of the factors counts. The
permutation is not possible. The geometric average doesn’t take account of that.
Dollar weighted takes into account the dollar you invest. 1 billion is the initial
investment. A negative return is not balanced by an identical positive return.
RATES OF RETURN
Annualizing Rates of Return
• APR = Annual Percentage Rate. It is the more common.
• Per-period rate × Periods per year
• Ignores Compounding
• EAR = Effective Annual Rate. ( (1+i(1/m))^m)-1
• Actual rate an investment grows
• Does not ignore compounding
Inflation and The Real Rates of Interest
Nominal Interest and Real Interest
1 R
1 r 
1 i
where
r  Real Interest Rate
R  Nominal Interest Rate
i  Inflation Rate
Example : What is the real return on an investment that earns a nominal 10%
return during a period of 5% inflation?
1  .10
1 r   1.048
1  .05
r  .048 or Nominal
Equilibrium 4.8% Rate of Interest
Fisher Equation
R = r + E(i)
E(i): Current expected inflation
R: Nominal interest rate
r: Real interest rate
t-bill has been mostly above inflation but not so much. A lot of returns from t bills are
erosed by inflation.
U.S. History of Interest Rates, Inflation, and Real Interest Rates
Since the 1950s, nominal rates have increased roughly in tandem with inflation
1930s/1940s: Volatile inflation affects real rates of return

Risk and Risk Premiums

Scenario Analysis and Probability Distributions
• Scenario analysis: Possible economic scenarios; specify likelihood and
HPR
• Probability distribution: Possible outcomes with probabilities
• Expected return: Mean value
• Variance: Expected value of squared deviation from mean
• Standard deviation: Square root of variance

Expected returns are the weighted averages of hpr, where the weights are the
probabilities.

The Normal Distribution

Transform normally distributed return into standard deviation score:
𝑠𝑟_𝑖= (𝑟_𝑖−𝐸(𝑟_𝑖))/𝜎_𝑖
Original return, given standard normal return:
𝑟𝑖=𝐸(𝑟𝑖 )+𝑠𝑟_𝑖×𝜎_𝑖
Normality over Time
• When returns over very short time periods are normally distributed,
HPRs up to 1 month can be treated as normal
• Use continuously compounded rates where normality plays crucial role

Deviation from Normality and Value at Risk

• Kurtosis: Measure of fatness of tails of probability distribution; indicates
likelihood of extreme outcomes
• Skew: Measure of asymmetry of probability distribution
Using Time Series of Return
• Scenario analysis derived from sample history of returns
• Variance and standard deviation estimates from time series of returns:

Value at risk (VaR):

• Measure of downside risk
• Worst loss with given probability, usually 5%

Risk Premiums and Risk Aversion

• Risk-free rate: Rate of return that can be earned with certainty
• Risk premium: Expected return in excess of that on risk-free securities
• Excess return: Rate of return in excess of risk-free rate
• Risk aversion: Reluctance to accept risk
• Price of risk: Ratio of risk premium to variance. Risk premium/
variance

The Sharpe (Reward-to-Volatility) Ratio

• Ratio of portfolio risk premium to standard deviation
Mean-Variance Analysis
 • Ranking portfolios by Sharpe ratios

The Historical Record: World Portfolios

• World Large stocks: 24 developed countries, ~6000 stocks
• U.S. large stocks: Standard & Poor's 500 largest cap
• U.S. small stocks: Smallest 20% on NYSE, NASDAQ, and Amex
• World bonds: Same countries as World Large stocks
• U.S. Treasury bonds: Barclay's Long-Term Treasury Bond Index

The risk premium: is it enough to say that stocks are better than bonds? No, you
should see the standard deviation and then the sharpe ratio.
This is an empirical distrubution.
Some values of the sample are far from the mean (different from t-bill). Normality is
not a perfect assumption.

ASSET ALLOCATION ACROSS PORTFOLIOS

Asset Allocation: Portfolio choice among broad investment classes
Complete Portfolio: Entire portfolio, including risky and risk-free assets
Capital Allocation: Choice between risky and risk-free assets

The Risk-Free Asset

• Treasury bonds (still affected by inflation)
• Price-indexed government bonds. The return is adjusted by inflation.
• Money market instruments effectively risk-free like repos.
• Risk of CDs and commercial paper is miniscule compared to most
assets
Capital Allocation Line (CAL)
• Plot of risk-return combinations available by varying allocation between
risky and risk-free
Risk Aversion and Capital Allocation
• y: Preferred capital allocation

5.6 Passive Strategies and the Capital Market Line

Passive Strategy
• Investment policy that avoids security analysis
Capital Market Line (CML)
• Capital allocation line using market-index portfolio as risky asset
Cost and Benefits of Passive Investing
• Passive investing is inexpensive and simple
• Expense ratio of active mutual fund averages 1%
• Expense ratio of hedge fund averages 1%-2%, plus 10% of returns
above risk-free rate
• Active management offers potential for higher returns
Vedi il video third week su learn
File excel trova appunti

EFFICIENT DIVERSIFICATION
Slides che non ha caricato
Senti audio tommaso

Dalla slide 29:

two source of portfolio:
- Risky securities
- Risk free securities
It is easy to compute the return.
y>> shares invested in the risky portfolio
expected return=y*return of risky sec+(1-y)* risk free rate
So, the volatility is sigma=ysigma(rp). Every portfolio with sigma equal to zero must
have the risk-free rate as rate of return.

E(r) is a function of volatility. We need to plot this.

We will put this

Into:
the one highlighted is the sharpe ratio. So:

INVESTMENT OPPORTUNITY SET

In P, y=1. In f y=0. In a portfolio you cannot leverage, you will be in between F and P.
Volatility you obtain by combining risk free securities with a risky portfolio.

ASSET ALLOCATION ACROSS PORTFOLIOS

Capital allocation line (CAL): plot of risk return combinations available by varying
allocation between risky and risk-free.
Risk aversion and capital allocation:
Y: preferred capital allocation.
Y= available risk premium to variance ratio/ required risk premium to variance ratio
Merton’s formula:

A: denotes the aversion to risk. It is high when the investor is very averse to risk.

PASSIVE STRATEGY AND THE CAPITAL MARKET LINE

Strategy that replicates some (equity) index for example. I don’t do individual
analysis. I passively replicate something.
When it replicates the overall equity market. The line takes the name of capital
market line.

COST AND BENEFITS OF PASSIVE INVESTING

This strategy became so popular over time because it is easy and cheap. The main
reason for this is that the fees that are asked by passive assets manager are a
fraction of the ones asked by mutual funds (and hedge funds, which are even more
expensive).
Active management: potentially it could offer higher returns. In practice, the evidence
shows that the average mutual fund before fees it makes exactly like the passive. Of
course, there is a distribution, some funds do it better others worse. Potentially they
could do better.

Diversification and Portfolio Risk

Some risks affect all assets>systematic. Others not.
- Market/Systematic/Nondiversifiable Risk.Risk factors common to whole
economy. Level of interest rates, for example. It is possible to reduce it but not
eliminated.
- Unique/Firm- Specific/Nonsystematic/Idiosyncratic Diversifiable Risk. Risk
that can be eliminated by diversification.

Figure 6.1 Risk as Function of Number of Stocks in Portfolio

By increasing the number of stocks, you can reduce risk but at some point, you can’t
anymore because there is systematic risk.
If you have one stock, you will have sigma as risk. The return doesn’t reward the
investors because it is very easy to diversify that risk.
Figure 6.2 Risk versus Diversification

Asset Allocation with Two Risky Assets

Covariance and Correlation
Portfolio risk depends on covariance between
returns of assets

Covariance and correlation coefficent Calculations

A measure of co-movement.
Spreadsheet 6.1 Capital Market Expectations

Two securities: a stock fund and a bond. 4 scenarios. Comparing the results in the
boom and recession, the two stocks are complementary and one the opposite of the
other in terms of risk and returns.

Spreadsheet 6.2 Variance of Returns

Spreadsheet 6.3 Portfolio Performance

Suppose you build a portfolio 60% bond fund and 40% in stock fund.
The volatility obtained 6,65% is smaller than both the volatilities of the stocks that are
in my portfolios.

Spreadsheet 6.4 Return Covariance of portfolio

Most of the time to estimate these objects:

Using Historical Data.
- Variability/covariability changes slowly over time. the estimate is a random
variable.
- Use realized returns to estimate
- You cannot estimate averages precisely
- Focus for risk on deviations of returns from average value

Rate of return (RoR): Weighted average of returns on components, with investment

proportions as weights

Expected rate of return (ERR): Weighted average of expected returns on

components, with portfolio proportions as weights

Variance of RoR:

The lower is the correlation coefficient the lower is the variance. If it is negative, it will
reduce the variance. The larger possible value is 1.
If ro is 1.

In order to reduce volatility, ro has to be smaller than 1. You can do better than the
simple weighted average.

Risk-Return Trade-Off
• Investment opportunity set
• Available portfolio risk-return combinations
Mean-Variance Criterion

Portfolio A dominates portfolio B

Not always the solution is so immediate and simple.

Spreadsheet 6.5 Investment Opportunity Set

s>stock b>bond. S has larger return but also larger volatility. We can compare them
with the sharpe ratio. The only requirement is that the sum of the weights must be 1.
We allow the weights to be negative (short sale) and larger than one (levered
position). Short sale if w<0. Leverage if w>1.

Figure 6.3 Investment Opportunity Set> the set of standard deviation and
expected return you can obtain by combining the two risky securities.
Opportunity Sets: Various Correlation Coefficients

The Optimal Risky Portfolio with a Risk-Free Asset

• Slope of CAL is Sharpe Ratio of Risky Portfolio

• we use it to find the Optimal Risky Portfolio: Best combination of risky and safe
assets to form portfolio

SLIDE 19

Two Capital Allocation Lines

You want to allocate your capital between a risky portfolio. And a free-risk security.
CAL: capital allocation line. CAL MIN: it goues through the minimum variance
portfolio. The slope is:

Should you be satisfied by this CAL or not? Supposed we choose A. A will bring you
to a better CAL: on the CAL that goes through A you have larger returns for the
same volatility. CAL(A) dominates CAL (min). you are going to stop when you reach
the CAL that is tangent to the opportunity set. The slop of this CAL is the highest
possible Sharpe ratio that you can obtain given the combination you can achieve
with these two securities with this risk-free rate.

Figure 6.7 The Complete Portfolio

The optimal portfolio is independent to your aversion to risk. It will be the same for
everyone. Cl will be the same (optimal portfolio), while C will be different (complete
portfolio).

Figure 6.8 Portfolio Composition: Asset Allocation Solution

6.4 Efficient Diversification with Many Risky Assets

Efficient Frontier of Risky Assets
Graph representing set of portfolios that maximizes expected return at each level of
portfolio risk. Markowitz model.
Three methods
• Maximize risk premium for any level standard deviation
• Minimize standard deviation for any level risk premium
• Maximize Sharpe ratio for any standard deviation or risk premium

Figure 6.9 Portfolios Constructed with Three Stocks

Slide 25
Figure 6.10 Efficient Frontier: Risky and Individual Assets

6.4 Efficient Diversification with Many Risky Assets

Choosing Optimal Risky Portfolio>> Optimal portfolio CAL tangent to efficient frontier
Separation Property implies portfolio choice, separated into two tasks
1. Determination of optimal risky portfolio
2. Personal choice of best mix of risky portfolio and risk- free asset

6.4 Efficient Diversification with Many Risky Assets

• Optimal Risky Portfolio: Illustration
• Efficiently diversified global portfolio using stock
market indices of six countries
• Standard deviation and correlation estimated from historical data
• Risk premium forecast generated from fundamental analysis

Figure 6.11 Efficient Frontiers/CAL: Table 6.1

The black line is the efficient frontier when short sales are not allowed. When shorts
sales are allowed, we can do better. The blue line dominates the black one (larger
return at the same volatility). This is not surprising.
6.5 A Single-Index Stock Market
Index model: Relates stock returns to returns on broad market index & firm-specific
factors
Excess return: RoR in excess of risk-free rate
Beta: Sensitivity of security’s returns to market factor
Firm-specific or residual risk: Component of return variance independent of market
factor
Alpha: Stock’s expected return beyond that induced by market index

Excess Return
Statistical and Graphical Representation of Single-Index Model>> Security
Characteristic Line (SCL: Plot of security’s predicted excess return from excess
return of market
Algebraic representation of regression line

Correlation index: Ratio of systematic variance to total variance

Figure 6.12 Scatter Diagram for Ford

Figure 6.13 Various Scatter Diagrams

6.5 A Single-Index Stock Market

Using Security Analysis with Index Model

• Information ratio: Ratio of alpha to standard deviation of residual
• Active portfolio: Portfolio formed by optimally combining analyzed stocks

6.6 Risk of Long-Term Investments

Why the Unending Confusion? Vast majority of financial advisers believe stocks
are less risky if held for long run
• Risk premium grows at rate of horizon, T
• Standard deviation grows at √T
• Sharpe ratio, 𝑆1√𝑇, grows with investment horizon

Table 6.4 Investment Risk for Different Horizons

CAPM AND APT

7.1 The Capital Asset Pricing Model
Optimal portfolio: where CAL and efficient frontier meet. The portfolio is made up by
a risk-free security and a risky portfolio.
Capital Asset Pricing Model (CAPM). It is an equilibrium model and tries to make a
forecast of the expected return of an asset.
Security’s required rate of return relates to systematic risk measured by beta and the
risk of the market portfolio. Each security’s return is related to beta and to the risk of
the market portfolio. An asset will have higher or lower rate of return depending on
the co-movement between the asset itself and the portfolio.

this is the risk premium

A: it is a measure of aversion to risk. The o^2 is the variance (sigma).
Market Portfolio (M): it is a portfolio that contains each security that exist in the
market weighted by the corresponding market capitalization.
Each security is held in proportion to market value. Each security will contribute to
the variance of the portfolio by its beta, its sensitivity to the overall market. Beta: the
contribution of the security to the overall variance of the market.

7.1 The Capital Asset Pricing Model: Assumptions

Price takers: the individual decisions will not affect the equilibrium price. This doesn’t
happen in reality. In fact, there are large investments that affect prices. This will have
a market impact. Suppose a mutual funds decides to sell its holding, so this may
have an impact on price. So, this assumption requires that investors are small.
3. strong assumption. We know that many securities are not traded. There are firms
which are not publicly owned and listed.
Transaction cost: bid-ask spread for example.
The last assumption is linked to the CAL. Tipically borrowing is more expensive.
Single-period horizon
Homogeneous expectations
Rational and mean-variance optimizer (averse to risk).

Hypothetical Equilibrium
• All investors choose to hold market portfolio because they are identical and
rational. Model with a representative investor. Clearly, they will have the same
portfolio which must be the market portfolio.
• Market portfolio is on efficient frontier, optimal risky portfolio. It contains every kind
of capital (this is why we have the assumption “every security is traded”). If everyone
owns this market portfolio then it has to be the optimal risky portfolio.
• Risk premium on market portfolio is proportional to variance of market portfolio and
investor’s risk aversion
Risk premium on individual assets. The model will be able to tell the risk premium
on every single asset in the market.
• Proportional to risk premium on market portfolio
• Proportional to beta coefficient of security on market portfolio
Figure 7.1 Efficient Frontier and Capital Market Line

IMPLICATION
Passive Strategy is Efficient. A passive strategy is an investment strategy in which
we track (replicate performances) of the market portfolio. Buying all securities
weighted by their market cap. In practice, transaction costs exist so you will want to
have a minimal number of stocks, enough to be closed to the market portfolio.
Two techniques: Physical replications and synthetic ITS that buys stocks.
Mutual fund theorem: All investors desire same portfolio of risky assets, can be
satisfied by single mutual fund composed of that portfolio.
• If passive strategy is costless and efficient, why follow active strategy? Why anyone
should do active strategy if passive strategy is costless and efficient? But how can
we get information about the stocks if people just follow passive strategies. Most
investors follow passive strategies. As a result, there is less incentive to gather
information on individual stocks. Then if there is not enough information, then one
assumption falls. There is enough reward for investors to do the security analysis. In
this way you can think of hedge fund. They study single securities and markets in
order to spot inefficiency. Their action should bring efficiency because they change
prices. Pedersen: short sellers bring information to the market.
• If no one does security analysis, what brings about efficiency of market portfolio?

Risk Premium of Market Portfolio

• Demand drives prices, lowers expected rate of
return/risk premiums. This is a model of supplies and demand. Supply is fixed,
demand will move prices. Changes I demand will change the price if the investors
expect a lower return they will demand less, and price will fall and turn to risky
securities. All investors are happy in equilibrium. All stocks have a specific rate of
return.
• When premiums fall, investors move funds into risk-free asset
Equilibrium risk premium of market portfolio proportional to
• Risk of market
• Risk aversion of average investor

Expected Returns on Individual Securities

Expected return-beta relationship
• Implication of CAPM that security risk premiums (expected excess returns) will be
proportional to beta

Security with beta=0. It doesn’t move when market moves. This security will have an
expected return equal to the risk-free rate.
Beta=1 it moves exactly like the market. It will have the same risk premium of the
market. Expected excess returns will be equal.
Beta>1 it amplifies market’s movements. This is a risky security and will have a
larger expected return than the market.
Beta<0 it will have a lower risk premium than the market. It moves against the
market, so it reduces the variance of portfolio.
The equation has a very strong implication: the only difference between securities is
the beta.

The Security Market Line (SML)

Represents expected return-beta relationship of CAPM
Graphs individual asset risk premiums as function of asset risk
Alpha
Abnormal rate of return on security in excess of that predicted by equilibrium model
(CAPM). It is a measure of mispricing between the price implied by the capm and the
expected return you estimate. Alpha is random, this is why it is not a pure arbitrage.

Figure 7.2 The SML and a Positive-Alpha Stock

All stocks must be of the blue line.
You are an active investor: you can estimate the beta for a stock (you can use as a
benchmark) then you look in the recent past the expected return of that stock, that
difference (between the expected return of the model and the historical effective
expected return) is called alpha. It is a pre money that you can make. There is only
one kind of risk in the model. Security that pays more: alpha>0. If you are a hedge
fund you will compare the exp return of your calculation (based on for example past
data) with the expected return implied by the model. Supposed you buy the stock.
Overvalued securities: below the blue line. You will short this security.
Undervalued securities: above the line. You will go long this security. Alpha is
positive in this case. The additional return (not implied by the model) is a kind of free
return because of some mispricing on the market (like money left on the table you
can grab). However, this is not pure arbitrage because that would be something like:
there is one stock, and I can immediately sell and buy at different prices. Here
instead is a risky arbitrage because it is an estimate and because you are buying an
undervalued security. There is a trade-off. You want to buy more the stock (risky
money eft on the table) but by doing so you reduce the diversification of your
portfolio. Trade off: alpha- idiosyncratic risk. Information ratio alpha/idiosyncratic
volatility.

Applications of CAPM
• Use SML as benchmark for fair return on risky
asset
• SML provides “hurdle rate” for internal projects

7.2 CAPM and Index Models

Alpha is the intercept. This is additional return that I get wjhen the contribution of the
market is zero. Risky arbitrage. e(it) is random noise. The larger the variance of e the
less you want to buy the stock. The random noise is the idiosyncratic risk of the
stock.

7.2 CAPM and Index Models: SCL

• Security Characteristic Line (SCL)
• Plot of security’s expected excess return over risk-free rate as function of excess
return on market
• Required rate = Risk-free rate + β x Expected excess return of index
We are talking about return of index because the market portfolio is very big. It is
much easier to replace the market with an index (S&P500 e.g.).
7.2 CAPM and Index Models
• Predicting Betas
Mean reversion: over time the more it is the time the stock is around the more the
beta of the stocks is closed to the beta of the market 1. It is possible to use models
able to adjust beta.
• Betas move towards mean over time
• To predict future betas, adjust estimates from historical data to account for
regression towards 1.0
Adjusted beta: you estimate the beta with a regression (row beta), then you add to it
an adjustment linked to the specific security.

7.3 CAPM and the Real World

Strategy to test capm:
We could see if it is true if all investors would built the same risky portfolio. Extimate
the expected return. Finally, I would make a plot to check whether is true the
combination of E(r) and beta are on the sml. If there are too many alphas, then capm
doesn’t work very well and doesn’t predict very well expected return.
CAPM is false based on validity of its assumptions
• Useful predictor of expected returns as many oter models. It is not a perfect
description of reality. It is untestable because the market portfolio is not observable.
• Untestable as a theory
Principles still valid. From capm we can acquire that:
• Investors should diversify. Only systematic risk is priced. If you don’t diversify you
will not be paid for sustaining idiosyncratic risk. It is not rewarded.
• Systematic risk is the risk that matters. It is rewarded by financial markets. It is the
variance of the market portfolio.
• Well-diversified risky portfolio can be suitable for wide range of investors. It means
that each investor holds the same risky portfolio. Young and old, very risky averse
and not so much risk averse, rich and poor, will be holding the same portfolio.
For most individual the fraction of wealth invested is very small. Most of the wealth is
outside the equity market. If you are using this model you are forcing systematic risk
to be related to stock market. Stock market is correlated to additional sources of
systematic risk. Earlier studies on capm noticed that usually small size stocks
(stocks with a small maret cap) tend to be above the sml. While big size stocks tend
to be above the line. The size could be a proxy of additional risk.
Capm concentrates on one source of systematic risk where there are additional risk
factors.
The principle that well diversify portfolios is the same for all the investors is still valid
in other models.
Untestable theory (Rolls) he arguments that it requires the market return of market
portfolios. But we can’t have this return, we have to estimate an index’ return. We
can’t observe the market portfolio, in practice. Market portfolio contains any assets
existing in the world. It is a claim on all the wealth existing in the world. It doesn’t
exist a portfolio like this.

7.4 Multifactor Models and CAPM

They are extensions of capm. They consider
• Multifactor models
Two factor model. Factor related to size. 2 betas: Beta that capture the exposure to
excess return on the market; the second beta is related to size. You will plot the
graph in three direction.

Fama-French Three-Factor Model

It is widely used. It is an extension of capm.

Estimation results
Three aspects of successful specification
• Higher adjusted R-square
• Lower residual SD
• Smaller value of alpha
We have to additional factors:
- Smb: small minus big. This is the size factor. It captures the differential return
of small stocks versus big stocks. Small> higher return for beta.
- Hml: high minus low. High> high ratio between book value (value of a
company from balancesheet) and market value (share pricexnumber of
shares). High> a company with market value<book value. Low> market
value>book value. They are referred to value stocks>> low price compared to
book value. Grow stocks>> high stock price compared to book value. On
average value stocks tend to have larger returns for a given beta. A good
candidate for growth stock: tesla. Value stock> more traditional sectors, coal
producers, iron producers.

Table 7.2 Multifactor Models and CAPM

The value of r square is higher in three factor model.
Intercept is smaller in three factor and it is less significant (number in brackets)
Excess return both significant and similar.
Smb e hml not very significant.
Single index model
Alpha negative: an investor should go short. Supposed that the market return goes
up by 1%, 1,3 beta. Actually, you will bring home 0,97 but you will lose 1,3. You don’t
want only to go short. You are still exposed to the systematic factor. You can
neutralized the effect of the market. Go short and go long for the market index. You
can use leverage to completely neutralize the market movement. Long short strategy
to hedge systematic risk and bring home alpha.
In the three factors model you would have to go long additional factors and go short
for the stocks of ford.

7.5 Arbitrage Pricing Theory

It established that return of well diversified portfolio are exposed to a number of
limited factors.
Arbitrage
• Relative mispricing creates riskless profit
Arbitrage Pricing Theory (APT)
• Risk-return relationships from no-arbitrage
considerations in large capital markets
Well-diversified portfolio
• Nonsystematic risk is negligible
• Arbitrage portfolio
• Positive return, zero-net-investment, risk-free portfolio
Apt with one factor

e>> goes to zero. Non-diversified risk is negligible.

Table 7.5 Portfolio Conversion

Steps to convert a well-diversified portfolio into an arbitrage portfolio:
By arbitrage I want to hedge the effect of the market. I can go short with a size betap
the market. My contribution would be -betapRm. Then I invest in the risky asset 1-
betap-(-betap). I go long an asset, and with an other I go long (the same quantity).
Well functioning market> no arbitrage opportunity. Alpha should be zero. If they are
not working properly then you can do an arbitrage.
Figure 7.5 Security Characteristic Lines

Rm > source of risk.

Single stock not diversified.

7.5 Arbitrage Pricing Theory

Multifactor Generalization of APT and CAPM
• Factor portfolio
• Well-diversified portfolio constructed to have beta of 1.0 on one factor and beta of
zero on any other factor
Two-Factor Model for APT

Table 7.9 Constructing an Arbitrage Portfolio

Constructing an arbitrage portfolio with two systemic factors
Risk of an arbitrage portfolio
According to apt theory there is no risk. E(p) is equal to zero> well-diversified. You
can bring home alpha for sure. When market is efficient this won’t work.