lecture 7 Flashcards
Advantages and disadvantages of investigating stock price data
Advantages:
Impact of events on firm value can be investigated
Freely available (e.g. yahoo finance)
Data is available at the daily level, which is more rarely available for other metrics and performance
Forward looking unlike sales and profits which are all backward looking
Disadvantages:
Stock price data is only available for traded companies. This could result in a sampling bias
Main focus is on investors. Other interest groups are only considered indirectly
A causal relationship of events on stock price movements is difficult to identify and requires statistical know-how
impact of a strategic event on stock price:
Interpret a single firms stock price reaction to an event
1) calculate the expected return (as if no event occurred)
2) calculate difference between actual return and expected return
Typical events used in event studies
New product introduction
Alliance formation
Channel restructuring
New market entry
Merger and acquisitions
Hostile takeover
Outsourcing
Conversion of non-voting shares into voting shares
Introduction of an option plan for compensationImpact on stock price from layoffs is not clear (up down sideways). Negative impact on consumers but not necessarily on stock price. Investors trade off future cash flows and future costs. Employees are a huge cost.
Efficient market hypothesis
efficient market hypothesis mostly used as an underlying assumption of the event study methodology
The effect of the event is incorporated instantaneously into stock prices. Thus, stock prices reflect all available information
Whether markets are efficient has been extensively researched and remains controversial. Market may be considered as weak, semi-strong or strong (i.e. stock price reflects all public and private information)
Three main premises of efficient markets
a large number of competing profit-maximizing participants analyze and value securities, each independently of the others
New information regarding securities comes to the market in a random fashion
profit-maximizing investors adjust security prices rapidly to reflect the effect of new information
Design of event studies and relevant questions
1) event definition and sampling
2) treatment of confounding effects
3) selection of an appropriate model
4) tests for significance and their power
5) moderating analysis
Event definition and sampling overview
is the event unambiguously defined and visible to investors?
Determine and define the type of event (e.g. layoffs)
Determine the selection criteria e.g. Only consider one firm, selection of competitors of a focal firm (all firms in the same industry)
Search for all companies that fulfil the selection criteria (premise: fimr must be listed on a stock market)
Search for events
E.g. through a media analysis (factiva) or other databanks
Determine the very first date the information was available to investors
T denotes the event time (in days), t= 0 denotes the event date
Treatment of counfounding variables
how are confouding observations being handled; should they be excluded form analysis?
one common concern in event studies is how to handle cases when multiple annoucements by the same entity occur in close proximity
Confounding events are events that may overlap with the effect of the focal event
For instance, a firm may announce the introduction of a new product a day after they announced that they wil not meet their earnings estimates
If the measurement window for the abnormal returns includes both events, the overall change in st ock returns may be caused by both events
whether confounding evnets should be eliminated remains controversial. Eliminating confounding events may be unnecessary for short term events if sample size is large enough
best alternative: show results for analysis with and without eliminating confunded veents
Selection of an appropriate model: determine estimation and event window
determine the event window:
we will search for abnormal returns over this window
E.g. t= o,…+7, often asymmetric
Be sure to determine the first date when the information reached the market
Start the event window before the event took place (anticipation /insider trading)
Determine the event window:
Previous “clean” period used for estimating parameters
Infer the normal performance of the stock
E.g. t= -200… -15
1) calculation of the return
2) calculation of the expected return
3) calculation of the abnormal returns
4) calculation of the cumulative abnormal returns
Calculate the mean of the returns of firm (j) over the estimation period (t)
Disadvantage: aive but still frequently used model. does not consider market movements
Advantage: no additional data (market index) is needed
Tests for significance and their power
are the (cumulative) abnormal returns at or around the time of the event statistically significant
Moderating analysis (=regression analysis)
under which conditions does the financial gain (alternatively loss) increase or decrease
note: moderation analysis is only possible if we have several observations (many events e.g. across different firms)
Determine the regression line:
CARi = b0 +b1x1 + u
u is an error term that describes the influence on CAR of all unobserved variables
