SEC REVIEW UNDER SECTION 408 OF THE SARBANES OXLEY ACT OF 2002: DOES IT MAKE A DIFFERENCE

1
SEC REVIEW UNDER SECTION 408 OF THE SARBANES
OXLEY ACT OF 2002 DOES IT MAKE A DIFFERENCE?

• Sharad Asthana Jeff Boone
• ACCOUNTING, COB, UTSA

April 2009
2
Motivation

• SEC. 408. ENHANCED REVIEW OF PERIODIC DISCLOSURES
  BY ISSUERS.
• (a) REGULAR AND SYSTEMATIC REVIEW. - The
  Commission shall review disclosures made by
  issuers reporting under section 13(a) of the
  Securities Exchange Act of 1934 (including
  reports filed on Form 10K), and which have a
  class of securities listed on a national
  securities exchange or traded on an automated
  quotation facility of a national securities
  association, on a regular and systematic basis
  for the protection of investors. Such review
  shall include a review of an issuers financial
  statement.
• (b) REVIEW CRITERIA. - For purposes of scheduling
  the reviews required by subsection (a), the
  Commission shall consider, among other factors -
• (1) issuers that have issued material
  restatements of financial results
• (2) issuers that experience significant
  volatility in their stock price as compared to
  other issuers
• (3) issuers with the largest market
  capitalization
• (4) emerging companies with disparities in price
  to earning ratios
• (5) issuers whose operations significantly affect
  any material sector of the economy and
• (6) any other factors that the Commission may
  consider relevant.
• (c) MINIMUM REVIEW PERIOD. - In no event shall an
  issuer required to file reports under section
  13(a) or 15(d) of the Securities Exchange Act of
  1934 be reviewed under this section less
  frequently than once every 3 years.

3
MOTIVATION

• PRELIMINARY REVIEW
• (AT LEAST ONCE EVERY 3 YEARS)
• OUTCOME 1 NO ACTION
• OUTCOME 2 DETAILED REVIEW
• FINANCIAL REVIEW (F/S MDA)
• FULL REVIEW (F/S MDA BUS LEGAL INFO)
• TARGETED REVIEW (SPECIFIC ? DERIVATIVES)
• NO FURTHER COMMENT
• OR
• ACCOUNTING AUDITING ENFORCEMENT RELEASE (AAER)

4
MOTIVATION (continued)

• The SEC states that
• Much of the Divisions review involves reviewing
  the disclosure from a potential investors
  perspective and asking questions that an investor
  might ask when reading the document.
• The mission of the U.S. Securities and Exchange
  Commission is to protect investors, maintain
  fair, orderly, and efficient markets, and
  facilitate capital formation.
• The laws and rules that govern the securities
  industry in the United States derive from a
  simple and straightforward concept all
  investors, whether large institutions or private
  individuals, should have access to certain basic
  facts about an investment

5
MOTIVATION (continued)

• DURING 2004-08, THE DIVISION OF CORPORATE
  FINANCE (SEC) SPENT OVER 429 MILLION ON REVIEW
  OF FILINGS. OF THIS, OVER 300 MILLION WAS SPENT
  ON REVIEW OF 10-Ks.
• DID THESE REVIEWS
• MAKE A DIFFERENCE?

RQ
6
OBJECTIVE

• TO INVESTIGATE THE EFFECTIVENESS OF THE SEC
  REVIEWS OF 10-Ks -
• IN IMPROVING THE QUANTITY QUALITY OF
  INFORMATION
• IN EQUALIZING DISTRIBUTION OF INFORMATION AMONG
  INVESTORS
• THE EFFICIENCY OF THE REVIEWS IN PROVIDING
  INFORMATION VALUE IN EXCESS OF REVIEW COSTS.

7
RESULTS / CONTRIBUTION

• (1) SEC reviews result in changes in both the
  length and content of 10-K annual reports and
  trigger financial restatements
• (2) This change in 10-K length and content
  appears to increase the information content of
  the 10-K as measured by return volatility and
  trading activity
• (3) The increased information content is
  primarily concentrated among small investors
• (4) The adverse selection spread component
  declines following release of the post-review
  10-K
• (5) The SEC reviews produce information valued at
  over 84 for every 1 expended in conducting the
  reviews
• (6) Contributes to the recent debate concerning
  ex ante versus ex post regulation of financial
  reporting.

8
THEORY (1)

• Consumers perceive public goods to be "free.
  Investors may demand more frequent/ more intense
  SEC reviews than would be demanded if they
  purchased the reviews at a price determined by
  market Demand/Supply dynamics (Chen and Johnson
  2008)? the marginal cost of the SEC reviews
  exceed the marginal benefit.

9
THEORY (2)

• SEC reviews are an extra layer of financial
  monitoring on top of that done by the independent
  auditor. The auditor has significantly more
  economic incentives than the SEC to detect
  material misstatements in financial statements ?
  SEC reviews are unlikely to result in meaningful
  improvement in 10-K annual reports.

10
THEORY (3)

• Audited financials can comply with GAAP and yet
  lack an appropriate level of transparency if GAAP
  is slow to respond to innovations in business
  practices ? the SEC reviews can improve 10-K
  annual reports by correcting accounting and
  disclosure deficiencies arising from unsettled,
  emerging accounting and disclosure issues
  ineffectively addressed by extant GAAP

11
PRIOR RESEARCH

• Ettredge et al. (2008) ? Examine comment letters
  addressing non-compliance with auditor change
  disclosures in Item 4 of Form 8-K ? the
  probability of non-compliance is positively
  associated with the 8-K containing bad news,
  negatively associated with corporate governance
  quality, and not associated with firm size.

12
PRIOR RESEARCH

• Ertimur and Nondorf (2006) ? Examine comment
  letters issued on registration statements of
  initial public offerings ? Higher managerial
  expertise is associated with fewer comments and a
  faster review ? Little association between
  corporate governance and number of
  comments/review time ?. Hypothesize that firms
  with fewer comments and faster review will have
  lower under-pricing, greater depth, and narrower
  bid-ask spreads.

13
PRIOR RESEARCH

• Chen and Johnston (2008)? Examine the change in
  earnings announcement for a sample of firms
  receiving comment letters on a 10-K or 10-Q
• SEC REVIEW ? PRECISION ? EARNINGS REACTION
• ___________________________________________(KIM
  VERRECCHIA 1997)______
• UNANSWERED QUESTIONS
• ?10-K POST-REVIEW?
• ?QUANTITY OF INFORMATION?
• ?QUALITY OF INFORMATION?
• ?DISTRIBUTION OF INFORMATION
• ? LARGE V/S SMALL?
• COSTS V/S BENEFITS OF REVIEW?
• Ex Ante V/S Ex Post?
• DO RESULTS HOLD FOR LARGE SAMPLES (gt206)?

14
SAMPLE

• 36,771 CL (MAY 1ST, 2005 JUNE 30TH, 2008)
• 16,987 (10-K RELATED)
• 10,414 (DATA ON COMPUSTAT CRSP)
• 3,572 CL FIRMS 3,754 NO CL FIRMS

15
RESEARCH DESIGN

• POST-REVIEW 10-K CHANGES
• LENGTH ANALYSIS

16
RESEARCH DESIGN

• POST-REVIEW 10-K CHANGES
• CONTENT ANALYSIS
• LEVENSHTEIN_DIFF
• LEVENSHTEIN (CL) - LEVENSHTEIN (NO-CL)
• ALSO USE LEVENSHTEIN_CHAR

17
RESEARCH DESIGN

• POST-REVIEW 10-K CHANGES
• RESTATEMENT ANALYSIS
• H3A The number of SEC-triggered restatements is
  greater than expected in the absence of an SEC
  review
• OR P(Rgtr) gt 0.10

18
RESEARCH DESIGN

• POST-REVIEW 10-K CHANGES
• INFORMATION CONTENT ANALYSIS
• ?CAR CARPOST- CARPRE
• ESTIMATION PERIOD (-250, -5)
• TEST PERIOD (-1, 1)
• Fama and Frenchs (1993) model with the Carharts
  (1997) momentum modification
• H4A the mean/median value of ?CARgt0

19
RESEARCH DESIGN

• POST-REVIEW 10-K CHANGES
• INFORMATION CONTENT ANALYSIS
• CAT S( )
• TRist Number of trades on day t for size
  stratum s in firm i
• µ(TRist) Mean (TRist) in the period (-49, -5)
• s (TRist) Standard Deviation (TRist) in the
  period (-49, -5)
• Large ? gt50,000
• Small ? lt 5,000 (or lt 100share price if gt 50)
• ?CATsmall CATsmall,post CATsmall,pre
• H5A the mean/median value of ?CATsmall gt0
• H6A the mean/median value of ?CATsmall ?
  ?CATlarge
• Repeat with ?CAS (Cumulative Abnormal Shares)
  H7,8

20
RESEARCH DESIGN

• CHANGES IN INFORMATION ASYMMETRY
• ASC ADVERSE SELECTION COMPONENT
• ?ASC ASC(0 to 59) - ASC(-60 to -1)
• H9A The mean/median value of
• ?ASCPOST ?ASCPRE ?0

21
Jeff to discuss results
22
RQ Does SEC review of an annual 10-K report lead
to 10-K changes?

• Test 1 examine change in length of 10-K
• Test 2 change in content of 10-K
• Test 3 restatements

23
RQ Does SEC review of an annual 10-K report lead
to 10-K changes?

• Test 1 examine change in length of 10-K
• Test result
• Mean 10-K length increased by about 6 (Table 5
  Panel A)

24
RQ Does SEC review of an annual 10-K report lead
to 10-K changes?

• Test 2 change in content of 10-K
• Test result
• Greater change in 10-K content (post-review
  versus pre-review 10-K) among CL firms as
  compared to industry/size matched No CL control
  group (Table 5 Panel B).

25
RQ Does SEC review of an annual 10-K report lead
to 10-K changes?

• Test 3 restatements
• Test result (Table 5 Panel C)
• 4.5 restatement rate among CL firms
• 4.5 162 restatements out of 3,572 CL firms
• 1.5 restatement rate among No CL firms
• Probability of observing 162 or more restatements
  given a 1.5 base restatement rate (absent SEC
  review) is very small (prob lt 0.001).

26
RQ Does SEC review of an annual 10-K report lead
to 10-K changes?

• Conclusion
• Significant increase in length of 10-K
• Significant change in content of 10-K
• More restatements than expected by chance.
• Answer to RQ Yes.

27
RQ Does new information in 10-K

• Add to the information set used by investors?
• Equalize the distribution of information among
  investor classes?

28
RQ Does new information in 10-K add to the
information set used by investors?

• Test 1
• Compare return volatility during post-review and
  pre-review 10-K filing event (-1,1)
• Compare trading activity during post-review and
  pre-review 10-K filing event (-1,1)
• Test result (Table 6 Panels A-C).
• Increase in return volatility (Table 6 Panel A)
• Increase in number of trades (Table 6 Panel B)
• Increase in number shares traded (Table 6 Panel
  C)
• Answer to RQ Yes

29
RQ Does new information in 10-K equalize the
distribution of information among investor
classes?

• Test 1
• Compare trading activity associated with new
  information between large and small investor
  classes.
• Test result (Table 6 Panels B and C)
• Small investors showed greater increase in
  trading activity than large investors.
• Suggests that new information in 10-K had
  greater information content for small investors,
  potentially because larger investors already had
  obtained at least some pre-disclosure knowledge
  of the information through private information
  search.
• Could also mean that the new information was
  commonly interpreted by large investors but was
  subject to varying interpretation by small
  investors.
• Spread analysis will help distinguish between
  these two competing explanations..

30
RQ Does new information in 10-K equalize the
distribution of information among investor
classes?

• Test 2
• Use adverse selection spread component as
  observable indicator of information asymmetry
• For both post-review and pre-review 10-K filing
  events
• Measure mean adverse selection spread component
  during 60 days following the 10-K filing event.
• Subtract mean adverse selection spread component
  during 60 days preceding the 10-K filing event to
  control for across-time changes in spread levels.
  
• i.e., 60 day mean less -60 day mean).
• Mean spread during 60 post-filing period
  (controlling for mean spread preceding the
  filing) should be lower than means spread during
  corresponding pre-review 10-K period if new
  10-K information equalized the distribution of
  information among investors.
• (slide continues)

31
RQ Does new information in 10-K equalize the
distribution of information among investor
classes?

• Test 2 (continued)
• Test result
• Mean spread during 60 post-filing period
  (controlling for mean spread preceding the
  filing) was lower than mean spread during
  corresponding pre-review 10-K period.
• Spread analysis suggests that difference in
  trading activity among investor classes was due
  to equalizing distribution of information among
  investors.
• Answer to RQ Yes

32
RQ Is the information value of SEC review
greater than review cost?

• Test
• Determine net change in equity values around
  comment letter issuance dates (two dates
  delivery of letter to registrant and uploading of
  letters to EDGAR) using (-1,1) window.
• Measure net information value of review as summed
  net change in equity values.
• Compare net information value of review to review
  cost per SEC budget.
• Test result
• 7 billion decline in equity values during
  comment letter issuance event.
• 87 million cost in conducting reviews
• 84 information per 1 spent on review.

33
Takeaway

• SEC reviews are substantive.
• Provides new information
• Mitigates information asymmetry among investors.
• Probably not an instance of overregulation.
• All of the above surprises me.