measuring asymmetry in capital marketspremiacap.com/qwafafew/caragata_20051020.pdf · 7 major...
TRANSCRIPT
1
Measuring Asymmetry in Capital Markets
October-November 2005Dr. Patrick Caragata,
Managing Director & CEORapid Ratings
Offices in Brisbane, Sydney, Toronto, New York, London, Wellington, Singapore
© 2005. All rights reserved.
2
Hedge Fund Growth and Returns are in decline on average
“If the amount of money in US equity-focused hedge funds can so quickly close up the inefficiencies in the $15,000 bn US equities market, perhaps there are not enough inefficiencies in the investing universe to sustain the expansion of the hedge fund universe.
Or, as Bernstein puts it: “Ladies and gentlemen! Welcome to the efficient market!” (Economics and Portfolio Newsletter, July 2005)Financial Times 12 July 2005, p 28
© 2004. All rights reserved.© 2005. All rights reserved.
3
Are Capital Markets Getting More Efficient?perhaps
9000
8000 (2003)
Less than 2000
Number of Hedge Funds
1.06% points2005 (1st half)
2.5 % points2004 (to end October
$820 bn(2003)
4.9 % points2001-20036.4 % points1998-200014 % points1995-97
$60 billion1990
Funds under management
Hedge Fund Profitability (% points
higher than cash)
Year
© 2004. All rights reserved.© 2005. All rights reserved. (FT 21/04/05; FT 7/12/04; FT16/05/05; Business Week Aug. 8/8/05)
Or are hedge funds running out of new ideas??
4
Hedge Fund Worries
Hedge Fund Concerns
Percentage Expressing Concern (total 8000)
1 Poor Returns 72 2 Over-capacity 47 3 Mis-selling 30 4 Operational
risk 28
5 Downward pressure
27
6 Systematic risk 27 Source: Financial Times 30 Sept 2005 (KPMG and Create)
© 2004. All rights reserved.© 2005. All rights reserved. (FT 21/04/05; FT 7/12/04; FT16/05/05)Or are hedge funds running out of new ideas??
5
Finding Alpha
To find alpha, you have to find and measure inefficiencies in capital markets
© 2004. All rights reserved.© 2005. All rights reserved.
6
Measuring Asymmetry in Capital Markets• re-examine assumptions of accepted models• integrate debt-side and equity side analysis;
Rapid Ratings has done this since 1998; new trend
• Verify the time-sequencing of early warning signals for upside and downside performance
4. Share price decline (rise)
5. Options pricing models
6. Multivariate discriminantanalysis
7. Traditional Credit Rating Agency (CRA) warning
1. Corporate financial health analysis
2. Credit default swap spreads
3. Bond spreads
© 2004. All rights reserved.© 2005. All rights reserved.
7
Major Inefficiencies in Capital Markets• Traditional ratings lag the share price by 1 to 3 years
• Structural (Merton) models which are becoming embedded in the banking sector track or lag the share price
• Bond markets use matrix pricing (based on brand name and scale) to determine current pricing
• Institutional investors go overweight or underweight specific industry indices which are based on a scale variable (market cap)
© 2004. All rights reserved.© 2005. All rights reserved.
8
COMMON FOOTSTEPS TO DISASTERSTRUCTURAL DEFICIENCIES BEHAVIOURAL DEFICIENCIES
Weak Risk Standards
Inadequate Back-upor Fail Safe System
InadequateAdvance Screening
Absence of Clear Signals
WeakDecision Centre
Tragic Outcomes
Poor Observationand Planning Skills
Inadequate Funding For Early Warning System
Ineffective Electronic Monitoring & Poor Quality Information
Ignoring EarlyWarning Signs
Ignoring RiskThresholds
© Patrick Caragata, 1999, Business Early Warning
Systems:Corporate Governance for the New Millennium, Butterworths
9
Improving Measurement of Asymmetry in Capital Markets
• you can’t make good decisions unless you have good quality information.
• You can’t develop good quality information unless you have the right tools.
• You will create measurement distortions using your tools unless your starting points are well-grounded and you avoid false assumptions
• Let the data speak for itself: Pasteur, using a microscope in 1857, let the data speak for itself and reversed the perception of causation in the link between bacteria and disease (initially for beer, wine, milk, silk and then human disease) and turned medicine into a science. If you retain false assumptions in your analytical model, you will miss the key messages. – His views were radical at the time.
© 2004. All rights reserved.© 2005. All rights reserved.
10
Importance of Starting Points
• The importance of starting points: 1989 Nobel prize winner in economics Norwegian economist TrygveHaavelmo (1956), an early inspiration in econometrics and evolutionary economics wrote about the importance of “initial conditions” and “the environment” in shaping the future “path” of economic change. He warned against the use of “rigid models”with “restrictive assumptions”
• Model assumptions, which are often inaccurate depictions of reality, become the foundations formeasurement error in paradigms and their problem-solving tools.
© 2005. All rights reserved.
11
Are Analytical Tools Getting More Efficient?Models Containing an Assumption of Perfect Information
Model Key Assumption
Implications of the Assumption: one or more of the following:
Competitive GE Model
Capital Asset Pricing Model
EDF indicators based on Options Pricing Models
Static and Dynamic Game Theory
Modigliani-Miller theorem
Perfect Information or Common Knowledge, or perfect transparency of information but not information asymmetry.
All agents receive information simultaneously and it is costless
Miscalculation of risks and required compensation·
Adverse asset selection·
Sub-optimal portfolio construction·
Sub-optimal rewards
12
Akerlof on Information Asymmetry “the basic method of economics which is to emphasize some aspects of reality (especially transactors’ attention to price) while putting blinkers on others, can leave major questions unanswered…[e.g.] how asymmetric information affects markets”
• “The Market for ‘Lemons’”: A Personal and Interpretive Essay by George A. Akerlof -2001 Prize Winner in Economics
AKERLOF’S WORK PROVIDES INSIGHTS INTO PARADIGM SHIFTS AND RAISES MAY ISSUES
• Paradigm blindness• Path dependence• The importance of starting points in modelling• Methodology affects perception• Asymmetry creates inefficiencies
© 2004. All rights reserved.© 2005. All rights reserved.
13
Paradigm Characteristics
Decision Criteria ExperimentationMethods(25-7,33)
Problem & Solution• Identification• Inclusion/Exclusion• Assumptions• Examination• Significance• Priority
• Observation• Data Collection• Data Manipulation• Theory Validation• Theory Falsification
R U L E S A N D S T A N D A R D S
Problem & Solution• Legitimacy or exclusion
rules (47, 65)• Incompleteness and
imperfection of existing data-theory fit (145)
• Natural tendency to anomalyand crisis (121)
ParadigmWeakness
Anomalies• Facts don’t fit theories (81)• “Noteworthy puzzle” can’t be
solved by existing paradigm(144)= reformulation of questions,
facts, assumptions and theory
= threat to old paradigm reputation and credibility
= crisis
CONFLICT RESOLUTION CRITERIA PARADIGM COMPETITION8, 23, 96,152, 156-7, 168 87-9, 91, 109, 147
• Application of better quantitative precision• Production of simpler and neater results• Solves noteworthy puzzle• Best explains and predicts facts and events• Best prospects for future problem solving
PROBLEMS
PROBLEMS SOLVED
REPUTATION & CREDIBILITY
PUBLIC CONFIDENCE,COMMUNITY ACCEPTANCE
ALLEGIANCE
CONFLICT RESOLUTION
• Conflict over:• possible methods/solutions• legitimacy of problems• legitimacy of assumptions• appropriate decision criteria
OLD PARADIGM ADJUSTS NEW PARADIGMAd hoc modifications to “eliminate apparentconflict” (78).
Best meets conflict resolution criteria (23).
© Patrick Caragata , All rights reserved, 1992, 1998, Based on Thomas Kuhn, Structure of Scientific Revolutions (University of Chicago Press, 1962)
PARADIGM UTILITY, TESTING & SHIFTS
© 2005. All rights reserved.
14
Factors Driving Asymmetry in Capital Markets
PERFORMANCE
PARADIGMS
STRUCTURES
ETHICAL STANDARDS
TYPES OF ASYMMETRY IN
CAPITAL MARKETSTOOLS
INFORMATIONQuality and
quantity
STARTING POINTS
KNOWLEDGE
GOALS
© 2005. All rights reserved.
15
Focus on Performance and Information Asymmetry
© 2005. All rights reserved.
Performance Asymmetry• Deviation from average practice (like an index) - BARRA• Deviation from best practice (efficiency frontier)• Deviation from worst practice (inefficiency frontier)• Speed and acceleration of (+ or –) performance (learning or
unlearning)Information Asymmetry • True versus false information (binary)• Perfect (full) versus imperfect (partial) information
(continuous/discrete) • Timing of accurate warnings (early versus late warning signals)-
asynchronicity of information (continuous- discrete)
• By improving our measurement of information asymmetry, we can enhance our understanding of performance asymmetry and its future path.
16
C h a n g e s to S e c to r D is t r ib u t io n o f W o r ld E q u it y M a r k e t s in M a r k e t C ra s h o f 2 0 0 0 - 2 0 0 1
F e b . 2 0 0 0 S e p t 2 0 0 1
1 . R e s o u r c e s 5 % 8 %2 . B a s ic in d u s t r ie s 4 % 4 %3 . G e n e r a l I n d u s t r ia ls 8 % 8 %4 . C y c l ic a l C o n s u m e r G o o d s 3 % 3 %5 . N o n -C y c l ic a l C o n s u m e r G o o d s 1 2 % 2 0 %6 . C y c l ic a l S e r v ic e s 1 1 % 1 1 %7 . N o n -C y c l ic a l S e r v ic e s 1 3 % 9 %8 . U t i l i t ie s 3 % 4 %9 . F in a n c ia ls 1 7 % 2 3 %1 0 . I n fo rm a t io n T e c h n o lo g y 2 4 % 1 1 %
file:///C:/Documents%20and%20Settings/paca/Local%20Settings/Temporary%20Internet%20Files/Content.IE5/3BXBZ5WS/04lectequityiofii%5B1%5D.ppt#504,27,Changes to Sector Distribution of World Equity Markets in Market Crash of 2000-2001
Which Portfolio Best Represents the AllocativeEfficiency of Investment?
© 2005. All rights reserved.
17
Rational Expectations and the Bubble NO
Real Return on a Diversified NASDAQ Portfolio Q1 2005 since…• March 1995: 9.3%• September 1995: 7.3%• Dec 1996 (“irrational exuberance” speech): 8.1%• April 1997: below the long run NASDQ average of 6.5 % points• October 1998: below +3% points (opportunity cost of capital in bonds)• November 1998: negative real returns
Re: “the huge bath taken by investors from February 2000 to September 2002” ….:” “It is next to impossible to interpret these events within the context of the rational expectations model, in which stock prices provide the best possible forecasts of future values. Only those who want their colleagues to doubt their rationality even try.” Brad DeLong FT 19 April 2005
© 2004. All rights reserved.© 2005. All rights reserved.
18
Are Capital Markets Getting More Efficient?
Market Dcharacteristics
Foreign Exchange
Overshooting & Undershooting
Rudiger Dornbusch
Commodities Overshooting & Undershooting
Various Jeffrey Frankel (FT 15/04/05)
Equities Overshooting & Undershooting
???
ominant Source
Our evidence consistently demonstrates that equity markets are constantly beset by overshooting and undershooting
© 2005. All rights reserved.
19
Are Capital Markets Getting More Efficient?So, how timely and accurate are traditional rating approaches and current rating models in measuring the financial health of companies which issue debt and equity?
© 2005. All rights reserved.
20
Forms of Bias in Traditional Ratings
© 2005. All rights reserved.
FORM OF BIAS EXPLANATION IMPACT
1. Scale bias bigger is better/ proxy for risk Rating error
2. Brand Name Bias Brand name is wrongly assumed to be a good proxy for risk.
Rating error
3. Lag bias (rating stickiness)
Ratings lag share price and financial performance
Rating error
4. Subjectivity bias Rating analyst may be captured Rating error
5. Flatlining Big 3 ratings may be unchanged for years Rating error
6. Use of ratings floors Bunching at Triple B to avoid market disruption and self-fulfilling prophesy
Rating error
7. Absolute risk measurement bias
Get relativities right, but not how high, how low and gap measurement.
Rating error
8. Assumption that the company has perfect information
All directors and senior executives have perfect knowledge of what is going on inside the company.
Rating error
21
Forms of Bias in Rating Models
© 2005. All rights reserved.
FO R M O F BIA S EX PLA N A T IO N IM PA C T 1 . S c o p e b ia s Fo c u se s o n n a rro w ra n g e o f
v a ria b le s (4 -1 2 ) R a t in g e rro r
2 . M a rk e t su b je c t iv it y b ia s
U se s e q u it y p ric e s a s in p u t s (O p t io n s p ric in g m o d e ls)
R a t in g e rro r
3 . Pa n in d u st ry b ia s Ig n o re s a n in d u st ry sp e c if ic a p p ro a c h : w e ig h t s m u st b e in d u st ry sp e c if ic
R a t in g e rro r
4 . S p a t ia l sm a ll sa m p le b ia s
M o d e ls u sin g le ss t h a n 1 0 0 , 0 0 0 f irm s h a v e u n st a b le w e ig h t s
R a t in g e rro r
5 . T e m p o ra l sm a ll sa m p le b ia s
M o d e ls u sin g le ss t h a n 2 0 y e a rs h a v e u n st a b le w e ig h t s
R a t in g e rro r
6 . S h o rt t e rm b ia s (1 y e a r)
Pa st t re n d s a re lo st ; s lo p e s o f c u rv e s n o t c a lc u la t e d
R a t in g e rro r
7 . S h o rt f o re c a st h o rizo n b ia s (n e x t 1 2 m o n t h s)
A b se n c e o f m e d iu m a n d lo n g t e rm o u t lo o k t re n d s p re v e n t b e t t e r a n a ly sis o f st re sse s f a c in g t h e c o m p a n y
R a t in g e rro r
8 . A ssu m p t io n t h a t t h e c o m p a n y h a s p e rf e c t in f o rm a t io n
S e e S e rv ig n y a n d R e n a u lt , 2 0 0 4 , M e a su rin g a n d M a n a g in g C re d it R isk , M c G ra w -H ill p . 4
R a t in g e rro r
22
Forms of Bias in Rating Models
© 2005. All rights reserved.
FORM OF BIAS EXPLANATION IMPACT9. Data availability Inadequate scale and scope
of empirical data Rating error
10. National bias Avoids cross-country approach and falls prey e.g. to excessive tolerance of high leverage.
Rating error
11. Homogeneity of firms
All firms are not alike. Rating error
12. Limited firm behaviour
Firms make many moves Rating error
13. One unique determinant of default
liquidity, equity/assets, or debt/assets or volatility in ROR are treated as the best predictors of default.
Rating error
14. Default risk assumed to be constant over time
Default risk changes with the financial health of firms
Rating error
15. Benchmarking is based on “average” not “best-to-worst” practice.
Ignores measuring distance to the frontier
Rating error
23
FORM OF BIAS EXPLANATION IMPACT16. Bondholders can’t force firm into default prior to maturity
Simplifying, unrealistic assumption
Rating error
17. Default occurs if the value of the equity drops below the debt level.
Simplifying, unrealistic assumption of gradual decline (geometric Brownian motion)
Rating error
18. Value of firm is perfectly observable
Simplifying unrealistic assumption of perfect information; market value of the debt is not always apparent; same is true for assets such as goodwill and off-balance sheet items
Rating error
19. Riskless interest rates are constant through time and maturity
Simplifying unrealistic assumption
Rating error
20. No room for renegotiation of debt
Simplifying unrealistic assumption
Rating error
21. Only one type of debt
Most firms have multiple types of debt
Rating error
24
Focus for Measuring AsymmetryFocus on OUTCOMES: gains and losses in debt and equity
markets to provide a benchmark for accuracy
How do you combine an analysis of performance asymmetry and information asymmetry? Compare trends in the financial health of companies with the share price trends
Most likely outcome is an asymmetrical distribution of outcomes not a normal or log normal distribution
© 2005. All rights reserved.
25
R ap id R a tin g s A gen cy A gen cy E D F A 1 9 5 A A A A aa 0 .0 2 A 2 9 0 A A a a2 0 .0 4 A 3 8 5 A A - a a3 0 .0 6 A 4 8 0 A A 2 0 .1 1 B 1 7 5 A - A 3 0 .1 9 B 2 7 0 B B B B aa2 0 .3 1
B 3 6 5 B B B - B a a 3 0 .5 B 4 6 0 B B + B a1 0 .7 8 C 1 5 5 B B B a2 1 .2 C 2 5 0 B B - B a3 1 .7 C 3 4 5 B + B 1 2 .6 C 4 4 0 B B 2 3 .6 D 1 3 5 B - B 3 5 D 2 3 0 C C C + C aa1 6 .7 D 3 2 5 C C C C aa2 8 .8 D 4 2 0 C C C - C aa3 1 1 E 1 1 5 C C C a 1 4 E 2 1 0 C C 1 7 E 3 5 D D 2 0 E 4 0
© 2005. All rights reserved.
26
Investment grade threshold at 65
Higher risk
Speculative buy – vulture capital territory below 45
Worldtex Files Chapter 11, Plan -- March 13, 2001
MP is the earlier name for Rapid Ratings© 2005. All rights reserved.
27
Delphi Bankruptcy October 2005
28
Enron Bankruptcy Dec 2001Enron Corporation (Ends 31st December)
0
10
20
30
40
50
60
70
80
90
100
96 97Jun-97
Sep-97
Dec-97
Mar-98
Jun-98
Sep-98
Dec-98
Mar-99
Jun-99
Sep-99
Dec-99
Mar-00
Jun-00
Sep-00
Dec-00
Mar-01
Jun-01
Sep-01
Dec-01
Mar-02
Years
Cre
dit S
core
0.00
8.93
17.85
26.78
35.70
44.63
53.55
62.48
71.40
80.33
89.25
Ave
rage
Sha
re P
rice
$(U
SD)
Rapid Ratings Short Term RatingRapid Ratings Medium Term RatingRapid Ratings Long Term RatingInvestment Grade ThresholdAverage Share Price
Source: Rapid Ratings Pty Ltd. ©2005
E3
D3
D3
C2
E1
E1
UndershootingOvershooting
Convergence
D2
D1
C3
A1
Share prices are adjusted to take into account dividends and share splits. The perceived relationship between share price and credit score may vary if a large number of new shares are issued.
Higher risk, VERY speculative – vulture capital territory below 45
“Corporate quality BOND spreads widened dramatically in early Feb 2001”
KMV downgrade MAY Nov 28 Agency Downgrades
Agency ratingZ Score
Dec- Mar-
“BUY”: investment banks© 2005. All rights reserved.
29
Stelco bankruptcy January 2004
0
10
20
30
40
50
60
70
80
90
100
Cre
dit S
core
0.00
1.10
2.20
3.30
4.40
5.50
6.60
7.70
8.80
9.90
11.00
Ave
rage
Sha
re P
rice
$(C
AD
)
Rapid Ratings Short Term RatingRapid Ratings Medium Term RatingRapid Ratings Long Term RatingInvestment Grade ThresholdTraditional Agency RatingAverage Share Price
E3 E3
E3E2
E1D4
E2
C4
E2
D2
D3D2
C4
C3
C1C2
Overshooting
BB-
BB+B4
E3E3E2
BB-
Share price lags RR by 4 years for junk status and demise
Higher risk, speculative – vulture capital territory below 45Stelco Inc (End 31st December)
1996 1997 1998 1999 2000 2001 2002 2003 2004
YearsThe perceived relationship between the share prices and credit ratings may vary if the number of shares issued are changed significantly. Source: Rapid Ratings Pty Ltd. ©2004
30
General Motors Downgrade 2005General Motors Corp (End 31st December)
2005 Share price average is week of 4-11 April
0
10
20
30
40
50
60
70
80
90
100
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Years
Cre
dit S
core
0.00
5.89
11.79
17.68
23.57
29.47
35.36
41.26
47.15
53.04
Ave
rage
Sha
re P
rice
$(U
SD)
Rapid Ratings Short Term RatingRapid Ratings Medium Term RatingRapid Ratings Long Term RatingInvestment Grade ThresholdAGENCY RATINGAverage Share Price
Source: Rapid Ratings Pty Ltd. ©2005
D1
BBB+
A
C4
C4
D3
C3
D1
C3
E1
C1
C4
B4
C4
C1
C4C4
B4
B1
Overshooting
Share prices are adjusted to take into account dividends and share splits. The perceived relationship between share price and credit score may vary if a large number of new shares are issued.
BBB
BBB
C4
C4C4
B3
Overshooting
58.94
31
Anticipation of Share Price RisesNextel Communications (End 31st December)
2005 Share price average is week of 4-11 April
0
10
20
30
40
50
60
70
80
90
100
1998 1999 2000 2001 2002 2003 2004 2005 2006
Years
Cre
dit S
core
0.00
5.47
10.94
16.41
21.89
27.36
32.83
38.30
43.77
49.24
Ave
rage
Sha
re P
rice
$(U
SD)
Rapid Ratings Short Term RatingRapid Ratings Medium Term RatingRapid Ratings Long Term RatingInvestment Grade ThresholdAGENCY RATINGAverage Share Price
Source: Rapid Ratings Pty Ltd. ©2005
E3
B1
A1
D4
E2
E1E1
D2
E1
Undershooting
Convergence
Overshooting
B+ B+BB-
B1
A4
A1
Share prices are adjusted to take into account dividends and share splits. The perceived relationship between share price and credit score may vary if a large number of new shares are issued.
B1A4A4
54.72
32
Independent Backtesting Results of Rapid Ratings
Re: % of times fall/rise in share price anticipated by RR credit ratingPeriod 1998-2004 Over 1500 companies
Falls in Share Prices•share price drop is at least 10% and RR credit rating is either: (1) always less than investment grade or fell below investment grade during the period or where fall in credit rating was at least 50% of the fall in the share price, but did not fall below investment grade.Success rate: 80%, of which 89% have a lead of one year or more.
Rises in Share PricesSimilar methodologySuccess rate: 80%, of which 85% have a lead of one year or more.
More accurate ratings help to alleviate effects of asymmetry© 2005. All rights reserved.
33
Does Independent Research Make a Difference?
Sells in 2004 as Percentage of total recommendations: 9.1%
Brokers……………………………………9.1%
Independent Research Providers………… 8.7%
© 2005. All rights reserved.
34
Sources1. Akerlof, George A., Writing the “The Market for ‘Lemons”: A Personal and Interpretive Essay,
http://nobelprize.org/economics/articles/akerlof/
2. Akerlof, George A., "The Market for ’Lemons:’ Quality Uncertainty and the Market Mechanism,“ Quarterly Journal of Economics, August 1970, 84, 488-500.
3. Burton, Jonathan ,1998, Revisiting The Capital Asset Pricing Model, Dow Jones Asset Manager May/June 1998, 20-28
4. Creighton, Adam, Luke Gower and Anthony Richards , 2004, The Impact Of Rating Changes In Australian Financial Markets, Research Discussion Paper 2004-02 , March 2004, System Stability Department, Economic Research Department, Sydney, Reserve Bank of Australia
5. DEUTSCHE BUNDESBANK Monthly Report, December 2004, Credit default swaps – functions, importance and information content,
6. Doerpinghaus, Helen I. and William T. Moore, 1994, Insurance Contract Valuation, Experience Rating, and Asymmetric Information, Journal Of Financial And Strategic Decisions, Volume 7 Number 2 Summer 1994
7. Ederington, L., and J. Goh, 1998, Bond rating agencies and stock analysts: who knows what when?”, The Journal of Finance and Quantitative Analysis, 33 (4), pp569-585
8. Haavelmo, T. (1956) A Study of the Theory of Economic Evolution, North Holland, Publishing Company, Amsterdam.
9. Hull, John, Mirela Predescu, and Alan White, 2004, The Relationship Between Credit Default Swap, Spreads, Bond Yields, and Credit Rating Announcements, Joseph L. Rotman School of Management, University of Toronto
10. Lakatos, I. (1970) “Falsification and the Methodology of Scientific Research Programmes”, in Lakatos, I and A. Musgrave (ed) Criticism and the Growth of Knowledge, Cambridge University Press, Cambridge.
11. Macey, Jonathan R. 2002, J. DuPratt White Professor of Law, Cornell Law School,Testimony before the United States Committee on Governmental Affairs March 20, 2002 , “Nationally Recognized Statistical Ratings Organizations and Investor Protection”
35
Sources12. Micu, Marian, Eli M Remolona, and Philip D Wooldridge, 2003, The price impact of rating announcements: evidence
from the credit default swap market, BIS Quarterly Review, June 2004
13. Norden, Lars and Weber, Martin, 2004, Informational Efficiency of Credit Default Swap and Stock Markets: The Impact of Credit Rating Announcements, Centre for Economic Policy Research, 90--98 Goswell Road, London EC1V 7RR
14. Odders-White, Elizabeth R. and Mark J. Ready, 2003, Credit Ratings and Liquidity, Department of Finance, University of Wisconsin
15. Oderda , G., Michel M. Dacorogna, Tobias Jung, 2002, Credit Risk Models Do they Deliver their Promises? A Quantitative Assessment, Submitted for publication in Economic Notes, November 13, 2002
16. Pasteur: http://www.historylearningsite.co.uk/louis_pasteur.htm
17. Per Botolf Maurseth , 2000, Growth Theory and Philosophy of Science - A Comparison of Neo-Classical and Evolutionary Perspectives. The Norwegian Institute of International Affairs. http://www.druid.dk/conferences/winter2000/maurseth.pdf
18. Pinches, George E. & J. Clay Singleton, 1978, The Adjustment of Stock Prices to Bond Rating Changes, vol. 33 Journal of Finance pp, 29-55 at 39.
19. http://www.wealtheffect.com/stocks/b13.asp comments in the 2001 10-Ks for Intel Corp. and Symantec Corp. (note the identical wording in the two documents)
20. http://www.freddiemac.com/debt/pdf/refpoint200204.pdf