Title: A Curious Case of Firmspecific Stock Return Volatility: Implications for Economies in Transition
1A Curious Case of Firm-specific Stock Return
Volatility Implications for Economies in
Transition
Art Durnev
2Based on
- A. Durnev, R. Morck, B. Yeung, 2004, Value
Enhancing Capital Budgeting and Firm-specific
Stock Return Variation, Journal of Finance. - A. Durnev, R. Morck, B. Yeung, P. Zarowin, 2003,
Does Greater Firm-specific Return Variation Mean
More or Less Informed Stock Pricing? Journal of
Accounting Research. - A. Durnev, K. Li, R. Morck, B. Yeung, 2004,
Capital Markets and Capital Allocation
Implication for Economies in Transition, The
Economics of Transition. - A. Durnev, R. Morck, B. Yeung, M. Fox, 2004,
Law, Share Price Accuracy, and Economic
Performance The New Evidence, Michigan Law
Review. - A. Durnev, R. Morck, B. Yeung, 2000, Does
Firm-specific Information in Stock Prices Guide
Capital Allocation?, NBER working paper. - A. Durnev, R. Morck, B. Yeung, 2004,
Firm-specific Return Variation and the Use of
Outside Financing, working paper. - and other work in progress
3 Observed stock price synchronicity in the stock
markets of selected countries,1995
Source Morck, Yeung, Yu (2000)
4Overview
- Decompose
- by running
- High R2 low per capita GDP
- High R2 in emerging economies
- Low R2 within developed economies
5 Country-Level R2, Sorted by GDP, 1995
R2
Source Morck, Yeung, Yu (2000)
6Overview
- What is GDP representing?
- Country size, industrial structure, macroeconomic
stability, synchronous fundamentals - General private property rights, investor
property rights
7Good Government Index plotted against R2
8Overview
- R2 is low in economies that have good property
rights respecting government. - Conditioned on having a good property rights
respecting government, protection for outside
shareholders property rights is associated with
high firm specific stock return variation . - In economies where the institution environment
encourages informed risk arbitrage, firm specific
stock return volatility is higher and R2 is
lower.
9Overview
- Informed risk arbitrage can be a story
- Political factors may be more difficult to
predict in emerging economies - Politicians can confiscate corporate earnings
- Hard for outsiders to gauge corporate earnings
- Even if they can predict corporate earnings the
property rights are uncertain - Firm-specific arbitrage becomes unattractive
- Traders become noise traders
- Poor investor protection
- Potential arbitrageurs would see little point in
gathering and processing information to estimate
firm-specific fundamentals if any surplus is
siphoned off by insiders.
10Overview
- We interpret that low R2 is indicative of higher
information contents in stock prices. - may or may not be right!!
- While our interpretation is unorthodox, their
interpretation if proven right can be very
useful. - The reason is that if we can measure how
informed stock prices are, we can implement
empirical research on the functionality of
capital markets.
11I. R2 inverse of information index
12H. J. Heinz and SP 500 index over the five
trading days prior to Dec. 7 1999
13Wall Street Journal, December 7, 1999 Heinz
Posts Sharply Stronger Net As Restructuring Plan
Progresses Dow Jones Newswires, PITTSBURGH --
H.J. Heinz Co., nine months into a sweeping
restructuring effort, Tuesday posted a sharp
increase in fiscal second-quarter net income and
core results that just topped Wall Street's
target. The maker of Heinz ketchup,
StarKist tuna and Ore-Ida frozen french fries
said net income soared 80 to 415.5 million,or
1.14 a share, from 213.3 million, or 63 cents a
share, a year ago. Figures for the quarter ended
Oct. 27 include a pretax gain of 464.6 million,
or 72 cents a share, on the sale of the company's
Weight Watchers low-calorie foods business. The
bottom line also reflects pretax costs of 43.7
million, or eight cents a share, and charges of
33.3 million, or nine cents a share, booked in
connection with "Operation Excel," the company's
restructuring program. Excluding items, Heinz
said earning increased 6.6 to 234.8 million, or
65 cents a share, from 220.2 million, or 60
cents a share, a year ago. Analysts surveyed by
First Call/Thomson Financial expected earnings to
reach 64 cents a share. Revenue inched up 1 to
2.34 billion from 2.32 billion. The
company launched its four-year restructuring bid
in February. At the time, Heinz said it expected
to cut 3,000 jobs and close some 20 factories,
primarily in Europe. The company also said it
expected pretax restructuring charges to top 900
million. Heinz said Tuesday it intends to
accelerate its restructuring bid to ensure that
all charges will be recognized by the end of
fiscal 2000, a year earlier than expected. The
company said savings should total 215 million by
2002.
14R2 Inverse of Information IndexAccounting
Evidence
- Regress current stock return on future changes in
earnings (Durnev, Morck, Yeung Zarowin (2003))
Informativeness
15R2 Inverse of Information IndexAccounting
Evidence
- In regression analyses we control for industry
characteristics (match-pair design,
diversification, size), factors having an
intrinsic effect on informativeness (earnings
volatility, institutional ownership), effects of
earnings timeliness (stock return, RD). - Future earnings explain more of current share
price with low R2 firms than with high R2 firms. - Suggests that low R2 firms indeed have more
accurate share prices.
16II. Informed equity prices and resources
allocation
17Stock Market Efficiency
- Malinvestment (poor microeconomic capital
allocation) - Hayek (1941) The Pure Theory of Capital
- Malinvestment can exist even with the optimal
aggregate quantity of capital goods if those
goods are not allocated to their best uses. - Major impediment to macroeconomic growth.
- The Functional Form of the Efficient Markets
Hypothesis - Tobin (1982)
- EMH gt microeconomic efficiency of capital
allocation. - A stock market exhibits functional market
efficiency if stock prices direct capital to its
highest value uses with an acceptably low error.
18Greater Functional Efficiency
- Investors trust markets more
- Investors lacking special information -
effectively, the great majority - can buy
securities confident that the prices they pay
roughly reflect the value of the attached future
income stream. - This increases the overall amount of capital that
savers provide via financial markets by
mitigating the well-known lemons effect
associated with external financing. - Managers trust markets more
- Firms weighted average costs of capital better
reflect each firms risks and opportunities.
This should reduce malinvestment. - Managers can use stock prices as feedback on
investors collective assessments of the uses to
which they are putting the firms capital. - Such feedback plausibly also reduces
malinvestment problems, or (if ignored) invites
the market for corporate control to reduce them.
19Greater Functional Efficiency
- More informed investment community ? firms have
grater access to outside financing. - More informed investment community ? firms
managers conduct more value maximizing
investment. - More informed investment community ? higher
economic growth (especially productivity growth).
201996-1999 GDP per capita growth rate plotted
against R2
CORR -0.278 PVAL 0.10
Source Durnev, Li, Morck, Yeung (2004)
211996-1999 Average growth rate in productivity
plotted against R2
CORR -0.326 PVAL 0.05
Source Durnev, Li, Morck, Yeung (2004)
22Informed Equity Prices and Growth Cross-Country
Regression Results
- Results are statistically and economically
significant - Controls initial GDP, schooling, inflation,
black market premium, government size, openness,
rule of law, bank credit, stock market size
23Capital Allocation Precision Measured by the
Elasticity of Capital Expenditure with Respect to
Value Added and R2
Germany
1
Hong Kon
France
New Zeal
Spain
Denmark
Sweden
Austria
Japan
United K
Belgium
United S
Australi
Ireland
Italy
Peru
Korea, R
Greece
Norway
Netherla
Finland
Canada
Portugal
Capital Allocation Precision
.5
Singapor
Mexico
Philippi
Chile
Malaysia
Pakistan
Turkey
Indonesi
Colombia
India
0
0
.2
.4
R2
CORR -0.500 PVAL 0.00
Source Wurgler (2000)
24Cross-Industry Results
- In using the US data, we assume that each
industry is a country, there is perfect capital
mobility, all countries have identical monetary
and fiscal policies. - Run firm-level regressions on 1990-1992 daily
data - Define where
- where
-
where
25Capital Investment Opportunity Schedule
Informed Equity Prices and Capital Budgeting
Quality
26Informed Equity Prices and Quality of Capital
Budgeting
- Marginal q estimation
- PV of the cash flows of a marginal dollar of
capital - Operationalize for 1993-1997 sample
- Cross-multiply and simplify
- Firm-time random effects regression on firm-level
data by industry
27Reality Check
- Estimate industry-level marginal q by running
- Average coefficients across all industry-level
regressions
28The Deviation of Marginal Tobins q from One with
Industries Grouped by Industry-Average Firm-Level
Market Model R2.
A low R2 indicates high firm-specific return
variation relative to market and industry-related
variation. The height of each bar is the group
average deviation of marginal q below and above
one.
29Informed Equity Prices and Capital Budgeting
Quality
- Measure malinvestment by the distance between
marginal q and one - ( -1)2 and -1
- Simple Correlations
- Marginal q is closer to one in industries where
stock returns are more asynchronous - Multiple regressions
- Control for fundamentals asynchronicity,
diversification, size, industry concentration,
liquidity, leverage, investment in intangibles. - Marginal q is closer to one in industries where
stock returns are more asynchronous.
30Mean Marginal q for Industries Subsamples with
Marginal q Above One and Below One, Grouped by
Industry-Average Firm-Level Market Model R2.
A low R2 indicates high firm-specific return
variation relative to market and industry-related
variation. The height of each bar Is the group
mean marginal q. The number of observations in
each group is listed at the top of each bar. The
sample sizes for 0 to 10, 10 to 20, 20 to
30 and 30 to 40 are 3, 34, 26, and 11
industries with marginal q above one and 9, 48,
48, and 7 industries with marginal q below one.
31Informed Equity Prices and Capital Budgeting
Quality
- Let the data choose such that the LHS
variable is - Best fit is at 0.78-0.83
- Consider over- and under-investment separately
- Partition sample into and
subsamples - R2 is associated with proximity to 1 only in the
subsample - Partition sample into and
subsamples - R2 is associated with proximity to 0.8 in both
subsamples
32Informed Equity Prices and Value Creation
- Measure firm value by average q
- Simple Correlations
- Average q is higher in industries where stock
returns are more asynchronous - Multiple regressions
- Control for fundamentals asynchronicity,
intangible assets, etc. - Average q is higher in industries where stock
returns are more asynchronous
33Informed Equity Prices and Use of Outside
Financing
- New long-term debt coverage of capital spending
- New equity coverage of capital spending
34 Industry-Level External Financing Grouped by
R2
CORR(Dld,R2) -0.160 PVAL 0.03 CORR(De,R2)
-0.151 PVAL 0.05
Source Durnev, Morck, Yeung (2001)
35Greater Functional Efficiency
- More informed investment community ? firms have
grater access to outside financing. - More informed investment community ? firms
managers conduct more value maximizing
investment. - More informed investment community ? higher
economic growth (especially productivity growth).
36Conclusions Our Spin
- 1-R2 is a measure of the capitalization of
firm-specific information into stock prices - The extent to which firm-specific information is
incorporated into stock prices varies across
countries, firms, and industries - This has real effects, viz. 1-R2
- Growth
- More use of external financing
- More efficient capital allocation
- More value creation
37- III. Yet to be answered
- Why do stock prices of some firms, industries,
and countries have more information content
(firm-specific variation) than those of others? - What explains variation in firm-specific
volatility through time?
38The Information Content of Stock Prices
- Incomplete risk arbitrage
- The absence of firm-specific arbitrage
- Agency problems and firm-specific arbitrage
- Noise traders and firm-specific arbitrage
39The Information Content of Stock Prices
- A lot of stock price variation is firm-specific
and not related to obvious public information - French and Roll (1986), Roll (1988)
- Risk arbitrage is harder than introductory
textbooks would have us believe - Shleifer and Vishny (1997)
40The Information Content of Stock Prices
- Exogenous and endogenous firm/industry
characteristics - Firms/industries that attract analysts are easy
to value (?) - Larger firms are easier to value (?)
- More diversified units are harder to value
(confirmed) - more focused behavior of managers leads to more
risk arbitrage - More concentrated units are easier to value (?)
- Younger units are harder to value (?)
- Companies with more intangibles (RD) are harder
to value (?)
41The Information Content of Stock Prices
- Internet companies (?)
- Foreign ownership (?)
- Type of exchange (NYSE, NASDAQ) (?)
- Business groups (?)
- Anomalies (momentum, etc) (?)
42The Information Content of Stock Prices
- Corporate policy variables such as dividend,
debt, institutional ownership can signal
insiders pre-commitment to deliver a
considerable portion of cash flow to outside
shareholders. - the action of sending back cash to outside
shareholders and to creditors itself limits the
insiders ability to abuse outside stakeholders
rights. Thus, there will be more informed risk
arbitrage.
43The Information Content of Stock Prices
- Higher institutional ownership causes R2
(confirmed) - monitoring from institutions improves
shareholders property rights and induces more
informed risk arbitrage - institutions may be subject to sentiment shocks
which become reflected in their trading strategy - But, institutions may prefer more liquid heavily
traded stocks which may tend to be more volatile - More leverage causes R2 (confirmed, dual
causality) - But, when leverage increases, shareholders bear
greater share of the total cash flow risk - Companies that have higher dividends have lower
R2 (?) - But, dividend payout can increase information
sensitivity of equity
44The Information Content of Stock Prices
- Cost of arbitrage, Pontiff (1983)
- transaction cost
- More expensive, larger firms are easier to trade
- holding cost variables
- Dividend payments reduce holding costs by
reducing the amount of capital that must be
provided to the arbitrage positions in future
periods.
45The Information Content of Stock Prices
- Examples of high firm-specific variation
industries - Commercial sports
- Knitting mills
- Crude petroleum natural gas
- Tobacco
- Examples of low firm-specific variation
industries - Engines and turbines
- General building constructors
- Department stores
- Operative builders
- Drug stores
46Average R2 Across Stocks Based on Monthly
Returns, 1926 to 1995
Source Durnev, Morck, Yeung (2001)
47Systematic and Firm-specific Variation in the US
Source Durnev, Morck, Yeung (2001)
48R2
49Systematic Variation
50Firm-specific Variation
51Changing Returns Synchronicity in Transition
Economies China
Source Li, Morck, Yeung (2004)
52Changing Returns Synchronicity in Transition
Economies Czech Republic
Source Li, Morck, Yeung (2004)
53Changing Returns Synchronicity in Transition
Economies Hungary
Source Li, Morck, Yeung (2004)
54Changing Returns Synchronicity in Transition
Economies Poland
Source Li, Morck, Yeung (2004)
55Changing Returns Synchronicity in Transition
Economies Romania
Source Li, Morck, Yeung (2004)
56Changing Returns Synchronicity in Transition
Economies Russia
Source Li, Morck, Yeung (2004)
57Variation of R2 through time
- Business cycles
- Change in laws
- Liberalization
- The relationship between the firm-specific stock
return variation and explanatory variables is
stronger in later days. - The quality of the stock market to process
information evolves through time and it is better
in the 90s than in the 60s, 70s, or 80s. - lower intensity of noise traders
- better shareholders protection rights
- better institutional framework
58Mandatory Disclosure and R2
- Mandatory disclosure is not needed (Macey, Choi
Guzman, Romano) - issuers have market incentives to provide the
optimal level of disclosure - much of what is required is not useful, the
information impounded in share prices comes via
routes other than mandatory disclosure - Mandatory disclosure is needed (Coffee,
Easterbrook Fischel, Bebchuk, Fox) - existence of externalities will result in a
market failure in the absence of regulation
59Mandatory Disclosure and R2
- We study imposition on December 15, 1980 of
enhanced (MDA) - Requires issuer managers to disclose any material
information that suggests the issuers recent
results are not necessarily indicative of future
operating results. - We observe a statistically significant reduction
in R2 after the new requirements. - Suggests the enhanced requirements did result in
revelation of useful information and hence
improved share price accuracy.
60Mandatory Disclosure and R2
- Issuer managers will be more likely, other things
being equal, to reveal in a timely fashion good
news than bad news. - To the extent that disclosure is voluntary, firms
with good news will have a higher level of
disclosure on average than firms with bad news.
61Mandatory Disclosure and R2
- Firms with good news in a given year should have
returns better than the industry average. - Firms with bad news in a given year should have
returns - Below the industry average where the bad news is
revealed during the year. - Equal to the industry average where the bad news
is not revealed during the year (when the bad
news does eventually become revealed in a
subsequent year, share price is marked down
then).
62Mandatory Disclosure and R2
- Following the MDA enhancement, we would expect
- Firms with current or past unrevealed bad news
that would, but for the MDA enhancement, have
been in the medium return group are instead in
the bad return group. - The medium return group is now less mixed,
the firms remaining in the group having more
informed prices. - The goodreturn group does not experience
significantly improved transparency.
63Mandatory Disclosure and R2
- If the regulations increase share price accuracy
- H Decrease in the R2 of the medium return
group should exceed that of the good return
group.
64Mandatory Disclosure and R2
- The results suggest that mandatory issuer
disclosure does result in the revelation of
useful information and thus adds to share price
accuracy.
651982-1980 Change in Firm-specific Variation
(SSE), Systematic Variation (SSM), and 1-R2 for
Good, Medium, and Bad Return Groups of
Firms
MEDIUM
SSE
BAD
1-R2
1-R2
SSE
SSM
SSE
SSM
1-R2
SSM
GOOD
66R2 Inverse of Information Index
- While supportive evidence is piling up, more of
theoretical and empirical work needs to be done
before we can firmly believe that more firm
specific stock return variation more informed
stock prices.
67Additional Evidence
- Beny (2000) finds that a countrys stock returns
are less synchronous where insider trading is
more restricted.
68Number of Analysts per Firm and
R2
Germany
United S
30
Netherla
Hong Kon
France
Spain
Italy
Singapor
Sweden
United K
Malaysia
20
Mexico
Canada
Number of Analysts per Firm
Brazil
Belgium
Finland
Japan
Denmark
Norway
Australi
India
Philippi
China
Korea, R
10
Thailand
Indonesi
New Zeal
Austria
Peru
Turkey
Poland
South Af
Taiwan
Greece
Chile
Ireland
Portugal
Czech
Pakistan
Colombia
0
0
.2
.4
.6
R2
CORR -0.271 PVAL 0.09
69U.S. Home Bias and R2
1
Taiwan
China
India
South Af
Greece
Japan
Chile
Malaysia
Pakistan
Belgium
Korea, R
Thailand
Philippi
Denmark
Turkey
Indonesi
Germany
Austria
Hong Kon
Singapor
Poland
Brazil
France
.8
Spain
Australi
United K
Italy
Sweden
Portugal
Norway
Canada
Finland
U.S. Home Bias
New Zeal
.6
Netherla
.4
Mexico
Ireland
.2
0
.2
.4
.6
R2
CORR 0.347 PVAL 0.04
70Earnings Management Measure and R2
.4
Turkey
India
Indonesi
China
.3
Greece
Pakistan
Negative Earnings Avoidance
Mexico
Poland
Chile
Thailand
South Af
Brazil
Malaysia
Ireland
Italy
Philippi
.2
Singapor
United K
United S
Netherla
Taiwan
Sweden
Austria
Spain
Australi
Germany
Finland
Hong Kon
Denmark
Canada
Portugal
Belgium
New Zeal
France
Japan
Norway
.1
0
.2
.4
.6
R2
CORR 0.464 PVAL 0.00