The initial corporate diversification decision: Empirical ...
Markus K. Brunnermeier LECTURE 1: INTRODUCTION EMPIRICAL ...€¦ · Lecture 01 Intro: Empirical...
Transcript of Markus K. Brunnermeier LECTURE 1: INTRODUCTION EMPIRICAL ...€¦ · Lecture 01 Intro: Empirical...
FIN501 Asset PricingLecture 01 Intro: Empirical Regularities (1)
LECTURE 1: INTRODUCTION EMPIRICAL REGULARITIES
Markus K. Brunnermeier
FIN501 Asset PricingLecture 01 Intro: Empirical Regularities (2)
Money, Bonds vs. Stocks
$0
$5
$10
$15
$20
$25
$30
$35
$40
$45
$50
19
72
19
75
19
78
19
81
19
84
19
87
19
90
19
93
19
96
19
99
20
02
20
05
20
08
20
11
SP500
US 1y Tsy
US 10Y Tsy
FIN501 Asset PricingLecture 01 Intro: Empirical Regularities (3)
Monthly Returns
-25.0%
-20.0%
-15.0%
-10.0%
-5.0%
0.0%
5.0%
10.0%
15.0%
20.0%
1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010
SP500
US 10Y Tsy
US 1y Tsy
But what is the right horizon?
FIN501 Asset PricingLecture 01 Intro: Empirical Regularities (4)
Compared to Recessions
10
100
1000
10000
4/1
/19
53
4/1
/19
55
4/1
/19
57
4/1
/19
59
4/1
/19
61
4/1
/19
63
4/1
/19
65
4/1
/19
67
4/1
/19
69
4/1
/19
71
4/1
/19
73
4/1
/19
75
4/1
/19
77
4/1
/19
79
4/1
/19
81
4/1
/19
83
4/1
/19
85
4/1
/19
87
4/1
/19
89
4/1
/19
91
4/1
/19
93
4/1
/19
95
4/1
/19
97
4/1
/19
99
4/1
/20
01
4/1
/20
03
4/1
/20
05
4/1
/20
07
4/1
/20
09
4/1
/20
11
Recession
SPX Index
FIN501 Asset PricingLecture 01 Intro: Empirical Regularities (5)
Stock Market Predictability
• Random walk hypothesis:
– Stock market prices evolve according to a random walk, and therefore cannot be predicted.
• Changes in stock prices can only be attributed to:
– News on future cash flows
– Change in “Risk premia” – the “dark matter” of finance
– Shifts in Behavioral Bias
FIN501 Asset PricingLecture 01 Intro: Empirical Regularities (6)
Testing Market Predictability• Cross Section vs. Time-Series
– Cross-Sectional studies refer to data collected at the same point in time, or regardless of differences in time
– Time-Series studies refer to a sequence of data points and look at how the data changed through time
• To test market predictability:
– Cross-Sectional studies look at whether some factors can explain the stock price changes, potentially in contradiction to the random walk hypothesis
– Time-Series studies look at the existence of time-related patterns (e.g. trends, seasonality) or event-specific behavior that would invalidate the random walk hypothesis
FIN501 Asset PricingLecture 01 Intro: Empirical Regularities (7)
Dividend/Price Ratio and Stock Prices
A regression of the S&P 500 index price growth on the dividend/price ratio shows that the D/P ratio is a good predictor of future price growth
Campbell and Shiller (1986)
FIN501 Asset PricingLecture 01 Intro: Empirical Regularities (8)
Stocks are more volatile than consumption growth
-50.0%
-40.0%
-30.0%
-20.0%
-10.0%
0.0%
10.0%
20.0%
30.0%
40.0%
50.0%
1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010
Nominal GDP
S&P500 (w/ Dividends)
FIN501 Asset PricingLecture 01 Intro: Empirical Regularities (9)
In the cross section: return & risk
Source: Ang, Hodrick, Xing, Zhang 2008
FIN501 Asset PricingLecture 01 Intro: Empirical Regularities (10)
Adding stocks in alphabetic order
0.00%
5.00%
10.00%
15.00%
20.00%
25.00%
30.00%
35.00%
40.00%
45.00%
50.00%
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40
S&P 100 Stocks from 1990-2011
FIN501 Asset PricingLecture 01 Intro: Empirical Regularities (11)
Measuring Risk differently
• When combining assets to a portfolio
– Idiosyncratic component diversifies away
– Covariance captures contribution of asset to portfolio’s risk
• Use Covariance as risk measure
FIN501 Asset PricingLecture 01 Intro: Empirical Regularities (12)
Covariance with market
0.00%
0.50%
1.00%
1.50%
2.00%
2.50%
3.00%
3.50%
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07
AAPL ABT
AEP AMGN
APC AXP
BA BAX
BHI BK
BMY CAT
CL CPB
DD DELL
DIS DOW
EMC F
FDX GD
GE HAL
HD HNZ
IBM INTC
JNJ KO
LLY LOW
MCD MDT
MMM MO
MRK MSFT
NSC ORCL
monthly return
FIN501 Asset PricingLecture 01 Intro: Empirical Regularities (13)
Covariance with consumption
0.0%
10.0%
20.0%
30.0%
40.0%
50.0%
60.0%
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07
AAPL ABT
AEP AMGN
APC AXP
BA BAX
BHI BK
BMY CAT
CL CPB
DD DELL
DIS DOW
EMC F
FDX GD
GE HAL
HD HNZ
IBM INTC
JNJ KO
LLY LOW
MCD MDT
MMM MO
MRK MSFT
NSC ORCL
annual return
FIN501 Asset PricingLecture 01 Intro: Empirical Regularities (14)
Size and value effect of stock returns
FIN501 Asset PricingLecture 01 Intro: Empirical Regularities (15)
Value vs. Growth Stocks
-60
-40
-20
0
20
40
60
19
80
19
81
19
82
19
83
19
84
19
85
19
86
19
87
19
88
19
89
19
90
19
91
19
92
19
93
19
94
19
95
19
96
19
97
19
98
19
99
20
00
20
01
20
02
20
03
20
04
20
05
20
06
20
07
20
08
20
09
20
10
Large Value-Growth (%)
Mid Value-Growth (%)
Small Value-Growth (%)
FIN501 Asset PricingLecture 01 Intro: Empirical Regularities (16)
Winner vs. Losers
• All shares are ranked each yearon basis of five-year holdingreturns
– Includes companies that de-list
• Assigned to portfolio by decileof performance
– Graph shows average performanceof equally weighted portfolios adjusted by equally weighted market portfolio
• Momentum or mean-reversion?– Short-run
– Intermediate
– Long-term
• UK Stocks 1960-2002 – see paper http://papers.ssrn.com/sol3/papers.cfm?abstract_id=998418.
• Original article: DeBondt & Thaler 1985
FIN501 Asset PricingLecture 01 Intro: Empirical Regularities (17)
Market Efficiency in Event Studies
T
t
tT AARCAAR
30
-30 -25 -20 -15 -10 -5 0 5 10 15 20 25 30
Over-reaction
Efficient Reaction
Under-reaction
T
Important: Information has to become public at a single moment
FIN501 Asset PricingLecture 01 Intro: Empirical Regularities (18)
Event StudiesObjective: Examine if new (company specific) information is incorporated into the stock price in one single price jump upon public release?
1. Calculate the daily excess returns 𝐴𝑅𝑡 = 𝑅𝑖𝑡 − 𝑅𝑚𝑡relative to the market or benchmark for 30 days prior and after release
𝑡 = −30,−29,… ,−1,0,1,… , 29,30
2. For each relative date 𝑡, calculate average returns and cumulative returns across events
𝐴𝐴𝑅𝑡 =1
𝑁 𝑖=1𝑁 𝐴𝑅𝑖𝑡
𝐶𝐴𝐴𝑅𝑇 = 𝑡=−30𝑇 𝐴𝐴𝑅𝑡
FIN501 Asset PricingLecture 01 Intro: Empirical Regularities (19)
Event Study: Earning Announcements
Event Study byBall and Brown (1968)Pre-announcement drift prior to earnings due to insider trading
! against strong-form
Post-announcement drift
! against semi-strong form
FIN501 Asset PricingLecture 01 Intro: Empirical Regularities (20)
Event Study: Earning Announcement
(MacKinlay 1997)
Cumulative abnormalreturns around earningannouncements
FIN501 Asset PricingLecture 01 Intro: Empirical Regularities (21)
Event Study: Stock SplitsEvent Study on Stock Splits byFama-French-Fischer-Jensen-Roll(1969)
Split is a signal of good profit
Pre-announcement drift can be dueto selection bias (only firms whose price rose) or insider trading.
! inconclusive
No post-announcement drift! for weak form
Selection bias or Insider trading
FIN501 Asset PricingLecture 01 Intro: Empirical Regularities (22)
Event Study: Take-over Announcement
FIN501 Asset PricingLecture 01 Intro: Empirical Regularities (23)
• The Expectations Hypothesis is the proposition that the long-term yield is determined only by the market’s expectations of future short-term yield
• Let 𝑍(𝑠, 𝑡) and 𝑦(𝑠, 𝑡) be the discount factor (zero-coupon price) and the yield for a zero-coupon price bought at 𝑠 that matures at 𝑡.
• The 1-year return on a 1-year bond
𝑟1,1 = 𝐸𝑡 ln1
𝑍(𝑡, 𝑡 + 1)= ln
1
𝑒−𝑦 𝑡,𝑡+1 1= 𝑦(𝑡, 𝑡 + 1)
• The 1-year return on a m-year bond bought at t and sold at 𝑡 + 1.
𝑟1,𝑚 = 𝐸𝑡 ln𝑍(𝑡 + 1, 𝑡 + 𝑚)
𝑍(𝑡, 𝑡 + 𝑚)= 𝐸𝑡 ln
𝑒𝑦(𝑡+1,𝑡+𝑚)(𝑚−1)
𝑒−𝑦 𝑡,𝑡+𝑚 𝑚
= 𝑚𝑦 𝑡, 𝑡 + 𝑚 − 𝑚 − 1 𝐸𝑡[𝑦(𝑡 + 1, 𝑡 + 𝑚)]
Government Bonds
FIN501 Asset PricingLecture 01 Intro: Empirical Regularities (24)
• By the EH, single-period holding on bonds of all maturities are equal in expectation. Therefore setting 𝑟1,1 = 𝑟1,𝑚and rearranging we get
𝑦 𝑡, 𝑡 + 𝑚 − 𝑦 𝑡, 𝑡 + 1 = 𝑚 − 1 𝐸𝑡 𝑦 𝑡 + 1, 𝑡 + 𝑚 − 𝑦 𝑡, 𝑡 + 𝑚
• According to the EH, the yield spread 𝑦 𝑡, 𝑡 + 𝑚 − 𝑦(𝑡, 𝑡 + 1) has a positive relation with short-term changes in the long-term bond yield. However, this does not hold empirically
• Source: Campbell 1995
Government Bonds
FIN501 Asset PricingLecture 01 Intro: Empirical Regularities (25)
Corporate BondsA very popular model to value
corporate debt is the Merton Model,
which postulates that the assets of a
firm follow a geometric Brownian
motion process.
However, this model
implies a credit spread to
Treasuries that is
consistently lower than
the observed credit
spreads.
FIN501 Asset PricingLecture 01 Intro: Empirical Regularities (26)
Fixed income: Commercial paper
FIN501 Asset PricingLecture 01 Intro: Empirical Regularities (27)
Rollover risk: Composition of ABCP
0
50
100
150
200
250
300
350
Ma
r-00
Se
p-0
0
Ma
r-01
Se
p-0
1
Ma
r-02
Se
p-0
2
Ma
r-03
Se
p-0
3
Ma
r-04
Se
p-0
4
Ma
r-05
Se
p-0
5
Ma
r-06
Se
p-0
6
Ma
r-07
Se
p-0
7
Ma
r-08
Se
p-0
8
$ B
illi
on
s
Other
Non-U.S. ResidentialMortgages
Student Loans
Credit Cards
Autos
Commercial RealEstate
Home Equity(Subprime)
ABS issuance
Source: JPMorgan
FIN501 Asset PricingLecture 01 Intro: Empirical Regularities (28)
Important spreads
FIN501 Asset PricingLecture 01 Intro: Empirical Regularities (29)
Derivatives
What’s a Credit Default Swap?
FIN501 Asset PricingLecture 01 Intro: Empirical Regularities (30)
ABX index
FIN501 Asset PricingLecture 01 Intro: Empirical Regularities (31)
DerivativesIn the Black-Scholes model for options pricing, a European
call option on a stock worth S with strike K and maturity T
today is worth:
back out the volatility for
different strikes with observed
option prices.
• Implied volatility
• Model ⇒ constant
• Reality ⇒ smile
Strike
Imp
lied
Vola
tilit
y
FIN501 Asset PricingLecture 01 Intro: Empirical Regularities (32)
The VIX
0.00
10.00
20.00
30.00
40.00
50.00
60.00
70.00
80.00
90.00
12/26/03 12/26/04 12/26/05 12/26/06 12/26/07 12/26/08 12/26/09 12/26/10 12/26/11
FIN501 Asset PricingLecture 01 Intro: Empirical Regularities (33)
Frictions
• In Physics
– Drop of a stone = drop of a feather
– Aviation wouldn’t work without frictions
• In Finance
– Walrasian auctioneer
– Asymmetric information
• Margins/haircuts/
• Limited risk bearing capacity, limited (risky) borrowing, …
FIN501 Asset PricingLecture 01 Intro: Empirical Regularities (34)
Beyond prices: Margins/Haircuts
• Margin/haircut/LTV determines max leverage ratio
Gorton-Metrick (2011)
“Repo Run”
But …
FIN501 Asset PricingLecture 01 Intro: Empirical Regularities (35)
Copeland,Martin, Walker (2011)
Margins very stable in tri-party repo market
– contrasts with Gorton&Metrick (2011)
– no general run on certain types of collateral
• http://www.ny.frb.org/research/staff_reports/sr477.pdf
Run (non-renewed financing) only on select counterparties
– Bear Stearns (anecdotally)
– Lehman (in the data)
Like 100% haircut…(counterparty specific!)
FIN501 Asset PricingLecture 01 Intro: Empirical Regularities (36)
Liquidity mismatch: sensitivity of wealth shifts
• Macro: real projects techno + market liquidity– Total supply in economy irreversibility + specificity
• Finance: financial claims market liquidity– How liquidity is distributed in economy – intermediation chain
• Shifts in wealth distributions affects costs of liquidity (amplification)
Technological liquidity Reversibility of investment
Market liquidity Specificity of capital
Price impact of capital sale
A L
Maturity mismatch
Funding liquidity Maturity structure of debt
Can’t roll over short term debt
Sensitivity of margins
Margin-funding is recalled
FIN501 Asset PricingLecture 01 Intro: Empirical Regularities (37)
Conclusion
• Risk & Return
– What is risk?
• Fixed income – risky borrowing/lending
• Derivatives (CDS)
• Market efficiency
– Asymmetric information
• Friction Finance
– Margins/haircuts/LTV
FIN501 Asset PricingLecture 01 Intro: Empirical Regularities (38)
Overview
• Frictionless Finance
One period model Part I
Multi-period model Part II
• Friction Finance Part III