Academy of Accounting and Financial Studies Journal (Print ISSN: 1096-3685; Online ISSN: 1528-2635)

Research Article: 2018 Vol: 22 Issue: 1

The Impact of Sarbanes-Oxley on the FASB and Accounting Regulation

Devon Baranek, Rider University

Keyword

Sarbanes-Oxley Act, Financial Accounting Standards Board, Accounting Regulation, Accounting Standard Setting.

Introduction

The Sarbanes-Oxley Act (SOX) was signed into law on July 30, 2002, at the time it was labelled “the most far reaching reform of American business since the time of Franklin D. Roosevelt (Bumiller, 2002). SOX were enacted following a period of high-profile corporate scandals that exposed the corrupt accounting practices of some of the largest companies in the United States. These scandals and related regulatory failures contributed to a loss of public faith in the accounting profession and the agencies responsible for regulating it.

The passage of SOX was intended to restore investor confidence and trust in the accounting profession and to improve previously flawed areas of the regulatory process. Some of the major changes included the creation of the Public Company Accounting Oversight Board (PCAOB), Section 404 increases to internal controls, whistle-blower preventions, increases in personal liability for CEOs and CFOs and revisions for the accounting treatment of certain complicated financial entities (Special Purpose Entities) and accounting methods (Mark-to-Market) (SOX, 2002). Previous literature has assessed the economic consequences of the legislation and documented how SOX influenced the behaviors of publicly traded companies, CEOs, Boards of Directors, public accounting firms, users and other constituent groups (Li, Pincus and Rego 2006, Zhang 2005, Berger, Li and Wong 2005, Coates and Srinivasan 2014, etc.) To date, no study has examined the impact of SOX on the accounting regulators or the regulatory process itself. This paper contributes to the literature on the changes that have been realized via SOX by focusing on the Financial Accounting Standards Board (hereafter FASB or the “Board”) and the accounting standard setting process.

Given the extent of the legislative overhaul, it is important to understand how SOX affected all parts of the regulatory process, including the work of the accounting regulators and the regulatory process itself. Improving the accounting and financial regulatory systems continues to be an on-going topic in the US, as evidenced by more current legislation, including the Dodd-Frank Act and the Financial Choice Act (currently in-process). The Dodd-Frank Act, passed in 2010, focuses on financial services and banking, but also includes requirements to improve transparency in financial reporting. Most recently, the Choice Act, passed in the House of Representatives in June 2017, proposes changes to the existing regulatory structure of the financial system, aiming to improve accountability and transparency. In addition to the creation of new regulatory agencies, changes to existing accounting regulators are proposed, including the Securities and Exchange Commission and the Public Company Accounting Oversight Board. Although it is still in the legislative process (and unlikely to be passed in its current state) the Choice Act underscores the desire for more regulatory reform in the areas of accounting, financial reporting and corporate governance.

By analysing the effects of SOX from a previously unexamined perspective, this paper serves to augment the existing regulatory literature as well as enhance our understanding of some of the lesser-studied aspects of the legislation. The analysis provides guidelines and raises additional questions for future research in this area.

Overview Of Sox And Fasb

Sarbanes-Oxley Act

SOX were devised to rehabilitate investor confidence after the highly publicized financial frauds of the early 2000s. These scandals exposed weaknesses in the existing corporate governance systems and self-regulatory mechanisms of the accounting profession and created doubts about the ability of these systems to communicate reliable financial information to the market. As a result of these shortcomings, a number of major objectives were included in the SOX legislation to create and enforce a system of checks and balances, including strengthening the independence of auditors, improving the quality and transparency of financial statements and corporate disclosures, enhancing corporate governance, improving the objectivity of research and strengthening the enforcement of the federal securities laws (SOX, 2002).

In response to these objectives, major changes occurred in the accounting profession. Some of these include creating new accountability standards for auditors, establishing the Public Company Accounting Oversight Board, increases in governance, oversight and compliance procedures, as well as increases in penalties for noncompliance and more prominent roles for CPAs on corporate boards and audit committees.

Financial Accounting Standards Board

In addition to the changes applied to accounting practitioners, the SOX legislation initiated changes in the accounting regulatory process, particularly for the standard setting body, the FASB. The FASB is tasked with establishing accounting standards through “…a comprehensive and independent process that encourages broad participation, objectively considers all stakeholder views and is subject to oversight by the Foundation’s Board of Trustees” (FASB, 2017).

These standards are crucial to the overall economy; they serve as a reference point for organizations generating financial statements and provide guidance on consistency and transparency in their application to various economic events. Better accounting standards lead to enhanced financial reporting. Ultimately, the FASB affects investor confidence and public trust through its various reporting requirements for the capital market.

As the independent, private-sector, non-profit organization tasked with setting accounting standards, the FASB is not obligated to include the public, but the Board encourages constituent participation with invitations to comment, discussion memoranda, public hearings, open meetings and other open operating procedures. For example, the Board recently issued an invitation to comment on in order to solicit feedback for potential financial accounting and reporting topics to add to the FASB agenda (FASB, 2016).

After the large and public embarrassments of the early 2000s (particularly Enron and Worldcom), the former Chairman of the FASB testified before Congress to defend the FASB and review its action following these economic failures. Herz testified that the Board had undertaken a number of changes to increase transparency in the financial reporting process and restore investor confidence. Some of these changes included operational modifications to improve the efficiency and timeliness of standard setting, the creation of the User Advisory Council to improve communication with the investment community and addressing the financial reporting issues previously abused (Subcommittee, 2003).

Hypotheses

In the years preceding SOX, there were considerable macroeconomic changes occurring in the US, including a downturn in the stock market, increases in corporate frauds and advances in shareholder activism (Lahart, 2002; Labaton, 2002; and Hershey 2002). The passage of SOX was part of a comprehensive change in corporate governance, financial reporting and the accounting profession. As a crucial part of the accounting regulatory process, the FASB was not exempt from the effects of this change. Therefore, I predict that various attributes of the FASB and the standard setting process were likely to be impacted by SOX. I summarize my predictions in Table 1 and include a brief discussion below.

Table 1: Summary Of Predictions
Attribute Predicted
Effect of SOX:
Empirical Proxies Predicted Empirical Observations
FASB workload Increase Total amount of guidance issued Increase
FASB speed Increase Amount of time from the issuance of an Exposure Draft to a final Standard issuance Decrease
    Length of guidance Decrease
Constituent participation Increase Amount of public hearings held Increase
    Amount of comment letters received Increase
Cost of regulation ? FASB operating revenue and
expenses
?
    Salaries and wages ?
    Cost per document issued ?

FASB Workload

SOX charges public companies with greater financial reporting responsibilities, particularly in the areas of internal control and corporate governance. Additionally, management is tasked with exercising a greater amount of judgment (and incurring a greater amount of personal risk) in determining the appropriate methods and procedures addressing the likelihood and magnitude of financial misstatements. The increased reporting responsibilities likely increased the demand for accounting regulatory guidance following the passage of SOX, effectively increasing the workload for the FASB.

On the other hand, SOX was a fairly unpopular and controversial piece of corporate legislation with expensive implications. The accounting profession has a long history of self-regulation and SOX is in direct conflict of that. It’s possible that FASB constituents were feeling a sense of regulatory “fatigue” from the initial outlays associated with compliance of the many provisions. Preparers and auditors, in particular, faced significant changes and increased responsibilities from the legislation. In this case, it’s possible that constituents would want to limit or reduce any additional accounting guidance in the years following SOX, decreasing the workload for the FASB. Overall, I anticipate an increase in the workload of the FASB in the post-SOX period.

FASB Speed

The FASB has long been criticized for being too slow to act on current accounting issues or too slow to issue guidance on controversial topics (SEC, 2002; Mundstock, 2003). If changes from SOX increased the demand for accounting guidance, then the FASB would face pressure to issue GAAP and other associated documents more quickly.

However, the accounting profession suffered a considerable reputational hit after the financial scandals leading up to the passage of SOX. It’s possible that in response to the public criticism and bad press, the FASB was more cautious and deliberate in issuing GAAP (for example, increasing the time spent on research, increasing discussions and negotiations, increasing requests for feedback from constituents, etc.), leading to a decrease in the pace of standard setting. Additionally, there is an ample amount of constituent feedback requesting the
FASB slow the pace of standard setting, particularly when large amounts of guidance are issued, so that constituents can absorb the changes and allocate their resources effectively. If the Board is responsive to these requests, than a shift in the speed of the FASB would be expected after the initial (mandatory) changes from SOX are implemented. Overall, I predict the speed of the FASB increased in response to the new legislation.

Constituent Participation

The rigorous due process of the FASB during standard setting provides constituents with numerous opportunities for participation, including public Board meetings, the issuance of exposure drafts for proposed standards, solicitations of constituent comments and roundtable discussions (FASB, 2017). Timely participation from knowledgeable and experienced constituents is considered a vital part of the FASB’s efforts to create standards that are useful, meaningful and present the best solutions for current accounting and reporting issues. Despite the importance of constituent participation to the FASB, actual participation levels are usually very low. Tandy and Wilburn (2004) estimate that for the first 100 standards issued by the FASB, only 0.002% of US corporations and 0.06% of public accounting firms submitted a comment letter to the FASB. Recent empirical evidence confirms this lacklustre level of participation has continued in more recent years and even suggests on a relative basis it has declined (Lysak, 2017).

After the demise of Enron, many placed blame on accounting regulators, for creating standards that were limited in scope and full of loopholes, allowing companies to take advantage of Special Purpose Entities (SPEs) and Mark-to-Market accounting. “When the tool to defraud is handed down by a regulator…companies will not only abuse their privileges to the fullest extent, but will feel they are doing so lawfully” (Prashad, 2006). Others were quick to blame public accounting, accusing Arthur Andersen of, among other things “…being in cahoots with company crooks” (Press, 2002). SOX authorized new and increased amounts of oversight for the accounting profession and permanently altered the peer review environment that had previously existed. This was not popular among a profession that has always been well respected and self-regulated.

The idea that oversight and enforcement mechanisms are tied to levels of constituent participation in accounting standard setting is supported by empirical evidence on an international level. Jorissen et al. (2013) find that non-preparers originating from countries with strong ex ante enforcement mechanisms are significantly more present in the public consultation stage of standard setting for the International Accounting Standards Board and that preparers from countries with strong ex post enforcement mechanisms are significantly more present in the public consultation stage. In light of the increased public scrutiny of financial reporting, subpar accounting standards blamed for Enron, defamation to the accounting profession, as well as the more onerous regulatory oversight and enforcement, I anticipate a greater number of constituents were motivated to participate in the standard setting process in the years following the passage of SOX. One caveat to this prediction is the simultaneous implementation of various technologies (the FASB website, email and online comment letter submission, etc.) around the same time SOX was passed. These technological advances enable constituents to more easily stay informed and participate in the standard setting process and likely augment the level of constituent participation, but it is difficult to tease out the effects of technology from my observations.

Cost of Regulation

The FASB is funded primarily by accounting support fees levied on publicly traded companies in the US (based on market capitalization) under Section 109 of SOX. These support fees are used to cover the annual recoverable expenses (roughly equivalent to operating expenses) of the Board and are reviewed by the SEC each year (FAF, 2017). The collection of accounting support fees allowed the FASB to nearly double its operating revenue in the pre- and post-SOX periods. It’s possible that with additional resources, economies of scale could be reached and the cost of regulation (on a per document basis) may decrease.

However, it’s also likely the seven-man board is limited in its capacity to generate and issue accounting guidance, particularly on complicated and controversial financial reporting issues and faces diminishing returns regardless of the size of the budget or support staff. For these reasons, I make no directional prediction for the effect of SOX on the cost of accounting regulation.

Data Description

I utilize an extensive hand collected data sample to explore the impact of SOX on the FASB, sourcing Statements of Financial Accounting Standards (SFAS) and Exposure Drafts of standards (available on line in the FASB Reference Library), Financial Accounting Foundation Annual Reports (requested for each year from the FASB) and other resources available on the FASB website. The sample period extends from 1973, the inception of the FASB, to 2008, the final year before the new FASB Accounting Standards Codification began, which overhauled the standard and pronouncement classification system. This period includes 166 SFAS. Since SOX was passed in 2002 and my sample period ends in 2008, I focus on a subsample of six years pre-SOX (1996-2001) and six years post-SOX (2003-2008) to improve comparability. This subsample period includes 38 SFAS.

Table 2A: Summary Of Sfas Panel A: Standards Issued By Year
Year SFAS Year SFAS
n Percent n Percent
1973 1 0.6 1991 2 1.2
1974 2 1.2 1992 5 3.01
1975 9 5.42 1993 4 2.41
1976 2 1.2 1994 2 1.2
1977 6 3.61 1995 5 3.01
1978 5 3.01 1996 3 1.81
1979 9 5.42 1997 4 2.41
1980 10 6.02 1998 3 1.81
1981 9 5.42 1999 3 1.81
1982 18 10.84 2000 3 1.81
1983 7 4.22 2001 4 2.41
1984 4 2.41 2002 4 2.41
1985 6 3.61 2003 3 1.81
1986 3 1.81 2004 4 2.41
1987 6 3.61 2005 1 0.6
1988 4 2.41 2006 5 3.01
1989 3 1.81 2007 2 1.2
1990 2 1.2 2008 3 1.81
Total 166 100

Table 2 summarizes this sample of standards. Panel A shows the number and percentage of SFAS issued each year over the total sample period, with the pre-SOX and post-SOX subsample in bold. To help illustrate content, the standards are classified by theme in Panel B, using the theme classifications as defined by Wallace (2001) and extended to include more recent years.

Table 2B: Summary Of Sfas Panel B: Standard Themes
Theme Standard # of SFAS in Theme Category
Revenue Recognition 45, 48, 66, 97 4
Research and Development 2, 7, 44, 68, 86, 142 6
Income Taxes 9, 31, 37, 96, 109 5
EPS and Comprehensive Income 55, 85, 128, 130 4
Cash Flow 95, 104 2
Marketable Securities, Derivatives and Hedging 12, 20, 80, 107, 115, 119, 133, 138, 149, 150, 155, 157, 159, 161 14
Impairment of Long-lived Assets 121, 144 2
Capitalization of Interest 34, 42, 58, 62 4
Leases 13, 17, 22, 23, 26, 27, 28, 29, 91, 98 10
Debt and Securitizations 4, 6, 15, 47, 49, 64, 76, 77, 78, 84, 105, 114, 118, 125, 129, 134, 140, 156 18
Contingencies 5, 11, 38 3
Pensions and Other Compensation 35, 36, 43, 74, 81, 87, 88, 106, 110, 112, 123, 132, 123R, 132R, 148, 158 16
Accounting Changes/Prior Period Adjustments 3,16,56,83, 154 5
Business Combinations 10, 72, 94, 141, 141R, 147, 160 7
Segments and Major 14, 18, 24, 30, 131 5
Customers Foreign Currency 1, 8, 52 3
Changing Prices 33, 39, 40, 41, 46, 54, 70, 82, 89 9
Specialized Practices/Industry Guidance 19, 32, 50, 51, 53, 60, 61, 63, 65, 67, 69, 71, 73, 90, 92, 101, 113, 120, 122, 143, 146, 151, 152, 153, 162, 163 26
Related Party Disclosures 57 1
Not-for-Profits 93, 116, 117, 124, 136 5
Exclusion of Certain Groups from Reporting 21, 25, 79, 102, 126 5
Deferrals 59, 75, 99, 100, 103, 108, 127, 137 8
Rescission/Elimination 111, 135, 139, 145 4
    166

Variable Measurement And Summary Statistics

Since the attributes of the FASB and the standard setting process I am examining can be difficult to identify and/or quantity, I employ the following empirical proxies:

FASB Speed

To measure the speed of the FASB in standard setting, I inspect the number of days from when the Exposure Draft of a standard is issued to the final standard issuance for the 166 SFAS issued from 1973-2008. Many of the projects tackled by the FASB extend over several years and vary in their level of complexity, so a great deal of time and effort is spent from the time an item is added to the Agenda and extending to the day an Exposure Draft is issued. This period would usually cover when the majority of research, writing and debating around a standard occur. By examining only the period between when an Exposure Draft is issued and a standard is finalized, I’m isolating a unique portion of the standard setting process that is applicable to all standards, which would allow for a comparison of the speed in which the FASB is able to finalize guidance once an agreed-upon draft has already been established. This data is presented for the pre-SOX post-SOX subsample in Table 3 and for all 166 standards in Appendix A.

Table 3: Sfas Details Of Issuance, Length And Participation For Pre-Sox And Post-Sox Subsample
Standard Days from ED to Issue Word Count Paragraphs Public Hearing Comment Letters Issue Date Classification
SFAS 125 245 1,538 19 No 86 Pre-SOX
SFAS 126 244 2,882 21 Yes 112 Pre-SOX
SFAS 127 86 516 6 No 76 Pre-SOX
SFAS 128 30 502 6 No 29 Pre-SOX
SFAS 129 397 4,823 43 No 104 Pre-SOX
SFAS 130 397 697 11 No 104 Pre-SOX
SFAS 131 365 2,793 34 Yes 281 Pre-SOX
SFAS 132 517 4,507 40 No 221 Pre-SOX
SFAS 133 245 2,371 15 No 90 Pre-SOX
SFAS 134 94 4,282 20 No 97 Pre-SOX
SFAS 135 730 14,007 56 Yes 300 Pre-SOX
SFAS 136 183 491 6 No 25 Pre-SOX
SFAS 137 125 3,712 6 No 9 Pre-SOX
SFAS 138 335 2,430 22 No 450 Pre-SOX
SFAS 139 26 408 4 No 77 Pre-SOX
SFAS 140 92 7,441 6 No 82 Pre-SOX
SFAS 141 608 643 8 No 28 Pre-SOX
SFAS 142 458 4,211 25 Yes 40 Pre-SOX
SFAS 143 120 7,572 62 Yes 211 Pre-SOX
SFAS 144 913 9,029 77 Yes 280 Pre-SOX
SFAS 145 120 8,308 61 Yes 211 2002
SFAS 146 486 3,547 28 No 50 2002
SFAS 147 426 6,222 51 Yes 53 2002
SFAS 148 59 2,891 12 No 10 2002
SFAS 132R 730 2,609 21 Yes 53 Post-SOX
SFAS 149 153 1,403 17 No 24 Post-SOX
SFAS 150 61 1,609 5 No 70 Post-SOX
SFAS 123R 335 7,929 40 No 40 Post-SOX
SFAS 151 930 3,383 31 Yes 71 Post-SOX
SFAS 152 336 976 5 No 26 Post-SOX
SFAS 153 669 716 5 No 233 Post-SOX
SFAS 154 366 2,604 7 No 30 Post-SOX
SFAS 155 517 4,161 27 No 66 Post-SOX
SFAS 156 184 2,210 7 No 24 Post-SOX
SFAS 157 212 11,225 12 No 26 Post-SOX
SFAS 158 822 5,686 39 Yes 100 Post-SOX
SFAS 159 184 4,334 22 No 245 Post-SOX
SFAS 141R 396 4,994 30 No 80 Post-SOX
SFAS 160 898 529 6 Yes 49 Post-SOX
SFAS 161 456 4,307 8 No 63 Post-SOX
SFAS 162 1,126 965 8 No 32 Post-SOX
SFAS 163 396 4,067 35 Yes 87 Post-SOX
Table 4: Fasb Revenues, Expenses And Cost Per Document
      FASB            
Year Operating Revenue % Change Salaries and Wages % Change FASB Total Expenses % Change Expenses/Revenue #Docs Cost per Document
1973 3,880,627   1,381,945 - 2,047,112   0.53 2 1,940,314
1974 4,374,844 0.13 1,952,714 0.41 3,095,659 0.51 0.71 18 243,047
1975 5,255,011 0.20 2,165,956 0.11 3,415,437 0.10 0.65 23 228,479
1976 5,464,533 0.04 2,398,037 0.11 3,970,703 0.16 0.73 18 303,585
1977 6,384,448 0.17 2,604,533 0.09 4,218,230 0.06 0.66 25 255,378
1978 6,535,339 0.02 3,438,904 0.32 5,642,380 0.34 0.86 32 204,229
1979 7,036,289 0.08 3,800,757 0.11 6,262,685 0.11 0.89 26 270,627
1980 7,803,431 0.11 4,244,215 0.12 7,210,417 0.15 0.92 34 229,513
1981 8,224,333 0.05 4,612,372 0.09 7,417,163 0.03 0.90 32 257,010
1982 9,519,000 0.16 5,092,000 0.10 8,048,000 0.09 0.85 29 328,241
1983 9,558,000 0.00 5,310,000 0.04 8,448,000 0.05 0.88 23 415,565
1984 9,675,000 0.01 5,543,000 0.04 8,898,000 0.05 0.92 18 537,500
1985 9,845,000 0.02 5,581,000 0.01 8,890,000 0.00 0.90 15 656,333
1986 10,445,000 0.06 6,112,000 0.10 9,845,000 0.11 0.94 13 803,462
1987 11,139,000 0.07 6,804,000 0.11 11,049,000 0.12 0.99 14 795,643
1988 11,788,000 0.06 7,845,000 0.15 13,637,000 0.23 1.16 5 2,357,600
1989 12,111,000 0.03 8,179,000 0.04 13,754,000 0.01 1.14 8 1,513,875
1990 11,953,000 -0.01 8,950,000 0.09 14,325,000 0.04 1.20 8 1,494,125
1991 12,992,000 0.09 8,612,000 -0.04 13,709,000 -0.04 1.06 11 1,181,091
1992 15,105,000 0.16 9,174,000 0.07 14,368,000 0.05 0.95 12 1,258,750
1993 15,536,000 0.03 8,106,000 -0.12 14,701,000 0.02 0.95 8 1,942,000
1994 15,876,000 0.02 8,264,000 0.02 15,255,000 0.04 0.96 13 1,221,231
1995 15,701,000 -0.01 8,454,000 0.02 15,734,000 0.03 1.00 15 1,046,733
1996 15,969,000 0.02 8,267,000 -0.02 15,817,000 0.01 0.99 14 1,140,643
1997 16,042,000 0.00 8,661,000 0.05 16,318,000 0.03 1.02 10 1,604,200
1998 16,932,000 0.06 9,464,000 0.09 17,253,000 0.06 1.02 10 1,693,200
1999 16,786,000 -0.01 8,583,000 -0.09 17,711,000 0.03 1.06 17 987,412
2000 17,113,000 0.02 9,225,000 0.07 18,435,000 0.04 1.08 14 1,222,357
2001 18,345,000 0.07 9,296,000 0.01 19,386,000 0.05 1.06 8 2,293,125
2002 15,815,000 -0.14 9,327,000 0.00 20,015,000 0.03 1.27 8 1,976,875
2003 31,307,000 0.98 10,284,000 0.10 20,528,000 0.03 0.66 19 1,647,737
2004 36,895,000 0.18 12,379,000 0.20 23,680,000 0.15 0.64 19 1,941,842
2005 31,459,000 -0.15 12,942,000 0.05 24,695,000 0.04 0.78 34 925,265
2006 33,086,000 0.05 13,546,000 0.05 25,929,000 0.05 0.78 38 870,684
2007 31,969,000 -0.03 14,264,000 0.05 25,939,000 0.00 0.81 30 1,065,633
2008 34,315,000 0.07 14,299,000 0.00 27,514,000 0.06 0.80 40 857,875
TABLE 5: Summary Statistics
Panel A: Entire Sample Period, 1973-2008
Variable Obs. Mean Median Std. Dev. Min Max
daysedissue 166 257.60 201.00 206.69 26.00 1,126.00
totalnumdocs 166 21.10 23.00 9.52 2.00 40.00
numpars 166 20.23 13.50 18.39 3.00 115.00
wrdcnt 166 2,689 1,782 2,730 179 14,034
PubHear 166 0.53 0.00 0.65 0.00 3.00
numedcmtltr 166 232.50 78.50 1,117.21 0.00 14,239.00
oprev 166 13,100,000 9,845,000 8,085,084 3,880,627 36,900,000
fasbsalwage 166 6,629,844 5,581,000 3,156,642 1,381,945 14,300,000
totexpenses 166 11,800,000 8,890,000 7,034,896 2,047,112 31,500,000
costperdoc 166 1,047,533 1,017,073 650,518 204,229 2,357,600
Panel B: Pre-SOX, 1996-2001
Variable Obs. Mean Median Std. Dev. Min Max
daysedissue 20 299.50 290.00 202.47 26.00 730.00
totalnumdocs 20 11.85 10.00 3.17 8.00 17.00
numpars 20 25.55 21.50 20.25 4.00 62.00
wrdcnt 20 3,904 3,215 3,471 408 14,007
PubHear 20 1.25 1.00 0.97 0.00 3.00
numedcmtltr 20 127.65 86.00 115.48 9.00 450.00
oprev 20 16,900,000 16,900,000 860,693 16,000,000 18,300,000
fasbsalwage 20 8,922,250 8,943,000 436,526 8,267,000 9,464,000
totexpenses 20 17,500,000 17,500,000 1,282,318 15,800,000 19,400,000
costperdoc 20 1,536,007 1,604,200 461,837 987,412 2,293,125
Panel C: Post-SOX, 2003-2008
Variable Obs. Mean Median Std. Dev. Min Max
daysedissue 18 505.17 396.00 310.50 94.00 1,126.00
totalnumdocs 18 29.83 32.00 9.34 19.00 40.00
numpars 18 25.78 21.00 23.60 5.00 85.00
wrdcnt 18 4,669 4,222 3,518 529 12,638
PubHear 18 1.06 1.00 0.42 0.00 2.00
numedcmtltr 18 877.11 68.50 3,335.62 24.00 14,239.00
oprev 18 33,600,000 33,100,000 2,057,241 31,300,000 36,900,000
fasbsalwage 18 12,900,000 13,500,000 1,399,436 10,300,000 14,300,000
totexpenses 18 27,200,000 28,100,000 3,209,519 22,100,000 31,500,000
costperdoc 18 1,260,786 995,449 469,117 857,875 1,941,842

Summary statistics for FASB speed in standard setting are shown in Table 5. The average number of days from Exposure Draft to final standard issuance for the entire sample period is 257.13 days. For the pre-SOX period 1996-2001, this average is 299.50 days and in the post- SOX period 2003-2008, this average is much higher, at 505.17 days. The longest period from Exposure Draft to issuance is for SFAS 162, “The Hierarchy of Generally Accepted Accounting Principles”, at 1,126 days and the shortest is for SFAS 137, “Accounting for Derivative Instruments and Hedging Activities” at 26 days.

FASB Workload

To measure the FASB workload, I examine both the total amount and length of guidance issued. “Guidance” includes SFAS, Statements of Financial Accounting Concepts (SFAC), Interpretations of existing literature (FIN), Exposure Drafts (ED), Discussion Memorandums (DM) and Other Documents. To capture the length of guidance, I collect both word and paragraph counts for each SFAS. This data is presented for the pre-SOX post-SOX subsample in Table 3 and for all 166 standards in Appendix A. The length and amount of content being generated by the FASB is only one measure of FASB output. Setting financial reporting standards is a complicated process and these proxies do not capture all of the work completed by the Board, however they do provide some insight about the content and amount of detail included in standards.

Summary statistics in Table 5 show that the FASB has issued as many as 40 pieces of guidance in a single year (in 2008) and as few as 2 (in 1973), with the average around 21. For the entire sample period, the average length of a standard is about 20 paragraphs and contains 2700 words. For the pre-SOX subsample, the average standard is longer, about 26 paragraphs and 3,904 words and for the post-SOX subsample, the standards appear to be similar in the number of paragraphs (26) but a bit wordier (4,669 words).

TABLE 6: Univariate Tests Panel A: Sub-Sample, 1996-2008 (Excluding 2002)
Variable Pre-SOX Mean Post-SOX Mean Difference T-stat Two-tailed p-value
daysedissue 299.50 505.17 205.67 2.44 0.02
totalnumdocs 11.85 29.83 17.98 8.12 0.00
numpars 25.55 25.78 0.23 0.03 0.97
wrdcnt 3,904 4,669 764 0.67 0.50
PubHear 1.25 1.06 -0.19 -0.79 0.44
numedcmtltr 127.65 877.11 749.46 1.01 0.32
oprev 16,900,000 33,600,000 16,700,000 33.31 0.00
fasbsalwage 8,922,250 12,900,000 3,977,750 12.14 0.00
totexpenses 17,500,000 27,200,000 9,700,000 12.43 0.00
costperdoc 1,536,007 1,260,786 -275,221 -1.82 0.08
Panel B: Entire period, 1973-2008 (excluding 2002)
Variable Pre-SOX Mean Post-SOX Mean Difference T-stat Two-tailed p-value
daysedissue 226.66 505.17 278.51 5.94 0.00
totalnumdocs 20.37 29.83 9.47 4.23 0.00
numpars 19.72 25.78 6.06 1.31 0.19
wrdcnt 2,457 4,669 2,211.55 3.30 0.00
PubHear 0.47 1.06 0.58 3.72 0.00
numedcmtltr 157.29 877.11 719.82 2.59 0.01
oprev 10,500,000 33,600,000 23,100,000 25.00 0.00
fasbsalwage 5,769,313 12,900,000 7,130,687 12.80 0.00
totexpenses 9,906,874 27,200,000 17,293,126 15.50 0.00
costperdoc 753,869 1,260,786 506,917 3.30 0.00

Constituent Participation

To measure the level of constituent participation, I include the number of public hearings held and number of comment letters received for each SFAS during the standard setting process (Table 3 and Appendix A). Public hearings are held for about one third of the standards issued (57 out of 166). The number of comment letters varies greatly, depending on the content of the standard. One caveat to note when using comment letters to proxy for constituent participation is the use of “form letters”, where participants simply add their name and organization name to a pre-written template and submit letters of identical content. While these constituents may not have participated in the FASB due process exactly as intended, they are still taking the time to complete and submit a letter, therefore their participation is counted.’

Summary statistics in Table 5 show that the average number of letters received for a SFAS over the entire sample period is 233, with some standards receiving ten or less (SFAS 59, SFAS 75, SFAS 135 and SFAS 145). The standard generating the largest number of comment letters is SFAS 123R, which provided guidance on share-based payment to employees.

In the pre-SOX period, it appears that constituent participation may have been dwindling, as the average number of comment letters received per standard during that time was 128, far lower than the overall average of the entire sample period (Table 5). The number of public hearings was consistent with the past years, at about 30% (7 out of 20). Data on how many or which constituents attend FASB public hearings, roundtables and other meetings is not publicly available, so I am unable to determine if the attendance in these meetings varies in the pre- and post-SOX periods.

Cost of Regulation

To proxy for the cost of regulation, I examine the FASB operating revenue and expenses per year, as well as the salaries and wages of Board members and compute the total cost per document as issued by the FASB (Operating revenue/total number of documents issued). Many standards take more than one year to issue, so this measure is imperfect, but it gives some idea of the “bang for the buck” of the FASB. This information is presented in Table 4 (additional financial details are available in Appendix B) and summary statistics are presented in Table 5.

Results and Discussion

In this section, I present the result of univariate tests of the means for the variables of interest to determine if and how the FASB and the accounting regulatory process changed after the passage of SOX. Table 6 Panel A presents results for the pre-SOX and post-SOX subsample and Panel B present results for the entire sample period. FASB speed, workload, constituent participation and the overall cost of regulation are analysed.

FASB speed

I find a significant increase in the average number of days from Exposure Draft to final standard issuance in both the pre-SOX post-SOX subsample and the entire sample period. It appears that the FASB is taking longer to generate standards over time, even as constituents demand more responsive and timely guidance.

These findings are inconsistent with my prediction that the speed of the FASB in standard setting would increase after the passage of SOX. This decrease may be caused by one or more of the following factors: 1) The FASB is tackling more complex financial reporting issues which require a greater amount of time to generate, 2) the FASB is Increasing the number of projects undertaken simultaneously in response to an increase in demand or 3) the Board is increasing the amount of research, preparation and diligence applied in standard setting as a result of the negative publicity from previous corporate frauds.

FASB Workload

I find that the total amount of guidance provided by the FASB has significantly increased over both the sub-sample and entire sample period. Interesting, while the number of paragraphs in FASB standards has remained about the same over time, the length and wordiness of those paragraphs has gradually increased, leading to longer standards, especially in the subsample period. Consistent with my prediction, more guidance is being issued after the passage of SOX, but it appears the length and content of the guidance has remained constant.

Constituent Participation

I find an increase in constituent participation as measured by comment letters in the sub-sample period, from 128 per standard pre-SOX to 877 per standard post-SOX as well as the number of public hearings per standard. In the subsample these increases are directionally as predicted, but not significant. This is inconsistent with my prediction – the data shows that constituents simply failed to alter their behavior and voluntarily participate in the FASB’s due process after SOX was passed. I find this particularly surprising since the FASB has accepted comment letters online in more recent years, which simplifies and streamlines the submission process. When the entire sample period is taken into consideration, the pre-SOX/post-SOX means are significant. Overall, it appears that constituent participation has increased over time, but not significantly in the post-SOX period.

Cost of Regulation

I find significant results in the subsample period for all of the financial metrics relating to the cost of regulation. Operating revenue, salaries and wages and total expenses all increased significantly in the post-SOX period. The change in operating revenue is expected, since a provision of SOX provided financing for the FASB. The increase in salaries and wages may have been in response to the increased workload, stress and public scrutiny faced by the Board during the regulatory process. Alternatively, FASB members may have been underpaid before this time and the bump in salary served to re-align the FASB more closely with the market.

I find the cost per document issued decreased significantly, from $1,536,007 to $1,260,786 in the post-SOX period. Cost per document includes many different types of FASB guidance, so this effect may be due to a number of different factors. For example, the increased operating revenue of the Board may have allowed more qualified staff to be hired, who was able to improve productivity, thus reducing the cost per document issued. Alternatively, the Board could be issuing shorter, directed guidance in the form of Staff Positions or Other Documents, the catchall for FASB direction that doesn’t fall into a pre-existing category.

Conclusion

Overall, it appears that, analogous to the rest of the accounting profession, SOX left indelible marks on the FASB and the accounting standard setting process. I find evidence of a post-SOX increase in the workload of the FASB, along with a concurrent decrease in the speed of standard setting. Counter to the public outcry against the profession, I do not find evidence of an increase in constituent interest or participation in the period after SOX. Finally, I find mixed results on the impact of SOX and the cost of standard setting. The cost of running and operating the FASB increased in the post-SOX period, but the cost per document issued by the Board decreased, suggesting that some previously untapped efficiencies or economies of scale are now being utilized.

Although SOX was passed 15 years ago, more research is necessary to fully comprehend the impact of the legislation on all areas of the accounting profession. Evidence from this study enhances our understanding of how the legislation influenced the FASB and the accounting standard setting process and provides some guidelines for future research. For example, have the mandates of SOX so significantly altered the FASB’s workload/speed that changes to the nature of the Board are necessary to maintain consistency in standard setting? If the FASB is operating with an increasingly large budget and issuing increasing amounts of guidance, should its oversight also be increased? As new legislation looks to reduce some of the mandates of SOX should we also expect to see changes in the standard setting process? These questions are raised for future research.

Appendix A: Additional Sfas Details
Standard # Days ED to Issue Word Count Paragraphs Public Hearing Comment Letters Standard # Days ED to Issue Word Count Paragraphs Public Hearing Comment Letters
SFAS 1 57 848 7 No 74 SFAS 39 177 1,998 16 Yes 124
SFAS 2 132 1,622 16 Yes 168 SFAS 40 208 786 8 Yes 124
SFAS 3 34 2,130 16 No 55 SFAS 41 208 867 8 Yes 124
SFAS 4 43 1,231 12 No 120 SFAS 42 207 454 5 No 63
SFAS 5 145 2,252 20 Yes 212 SFAS 43 334 778 9 No 217
SFAS 6 185 1,913 17 No 92 SFAS 44 52 666 8 No 41
SFAS 7 331 1,667 16 Yes 138 SFAS 45 104 2,661 25 No 25
SFAS 8 288 3,673 37 Yes 190 SFAS 46 34 646 9 No 18
SFAS 9 173 1,914 18 Yes 98 SFAS 47 349 1,173 11 No 102
SFAS 10 37 616 8 No 22 SFAS 48 34 1,100 12 No 36
SFAS 11 45 2,024 11 No 45 SFAS 49 126 1,266 11 No 34
SFAS 12 39 3,204 23 Yes 272 SFAS 50 156 1,316 17 No 12
SFAS 13 116 12,208 51 Yes 282 SFAS 51 156 1,217 16 No 23
SFAS 14 442 5,833 41 Yes 233 SFAS 52 168 4,129 38 Yes 260
SFAS 15 167 5,582 45 Yes 96 SFAS 53 186 2,171 25 No 23
SFAS 16 321 2,353 17 Yes 162 SFAS 54 60 304 4 No 27
SFAS 17 99 1,102 9 No 42 SFAS 55 101 825 8 No 68
SFAS 18 56 711 9 No 65 SFAS 56 101 998 12 No 25
SFAS 19 153 9,441 64 Yes 195 SFAS 57 129 579 5 No 66
SFAS 20 38 1,834 15 No 30 SFAS 58 197 871 9 No 72
SFAS 21 47 1,860 16 No 126 SFAS 59 52 235 4 No 0
SFAS 22 178 2,364 16 No 26 SFAS 60 209 6,559 65 No 56
SFAS 23 239 1,286 11 No 30 SFAS 61 209 926 11 No 14
SFAS 24 149 1,023 6 No 35 SFAS 62 175 820 9 No 94
SFAS 25 100 964 10 No 27 SFAS 63 368 1,159 13 No 45
SFAS 26 114 1,200 8 No 33 SFAS 64 204 593 5 No 90
SFAS 27 91 1,017 9 No 25 SFAS 65 224 3,197 33 No 42
SFAS 28 145 1,147 4 No 37 SFAS 66 304 5,928 52 No 47
SFAS 29 176 1,223 14 No 37 SFAS 67 304 2,291 27 No 37
SFAS 30 139 735 7 No 36 SFAS 68 171 1,576 15 No 37
SFAS 31 47 881 9 No 49 SFAS 69 214 3,977 39 Yes 113
SFAS 32 106 1,260 12 No 53 SFAS 70 358 2,342 21 No 69
SFAS 33 261 7,548 69 Yes 450 SFAS 71 286 2,430 24 Yes 172
SFAS 34 304 2,358 23 Yes 269 SFAS 72 131 1,729 15 Yes 80
SFAS 35 1,066 3,810 30 Yes 700 SFAS 73 125 181 3 No 28
SFAS 36 308 1,200 11 No 228 SFAS 74 230 418 5 No 160
SFAS 37 123 505 5 No 67 SFAS 75 161 221 4 No 10
SFAS 38 264 1,320 10 No 59 SFAS 76 124 1,221 12 No 75
Appendix A2:Cont. Additional sfas details
Standard # Days ED to Issue Word Count Paragraphs Public Hearing Comment Letters Standard # Days ED to Issue Word Count Paragraphs Public Hearing Comment Letters
SFAS 77 471 1,021 11 No 72 SFAS 122 334 1,961 6 No 89
SFAS 78 503 985 6 No 85 SFAS 123 852 7,095 54 Yes 1786
SFAS 79 134 411 7 No 46 SFAS 124 259 12,638 85 Yes 14239
SFAS 80 398 2,316 14 No 153 SFAS 125 245 1,538 19 No 86
SFAS 81 135 594 9 Yes 162 SFAS 126 244 2,882 21 Yes 112
SFAS 82 36 323 5 No 120 SFAS 127 86 516 6 No 76
SFAS 83 99 637 9 No 24 SFAS 128 30 502 6 No 29
SFAS 84 99 720 6 No 59 SFAS 129 397 4,823 43 No 104
SFAS 85 99 797 5 No 53 SFAS 130 397 697 11 No 104
SFAS 86 349 1,507 17 Yes 210 SFAS 131 365 2,793 34 Yes 281
SFAS 87 275 8,030 77 Yes 400 SFAS 132 517 4,507 40 No 221
SFAS 88 184 2,139 21 Yes 110 SFAS 133 245 2,371 15 No 90
SFAS 89 76 338 4 No 200 SFAS 134 94 4,282 20 No 97
SFAS 90 365 2,026 13 Yes 1400 SFAS 135 730 14,007 56 Yes 300
SFAS 91 365 4,073 28 Yes 822 SFAS 136 183 491 6 No 25
SFAS 92 608 2,007 18 Yes 1400 SFAS 137 125 3,712 6 No 9
SFAS 93 235 737 8 Yes 193 SFAS 138 335 2,430 22 No 450
SFAS 94 303 2,021 17 Yes 232 SFAS 139 26 408 4 No 77
SFAS 95 488 3,123 34 Yes 450 SFAS 140 92 7,441 6 No 82
SFAS 96 456 5,716 36 Yes 400 SFAS 141 608 643 8 No 28
SFAS 97 365 3,077 33 Yes 111 SFAS 142 458 4,211 25 Yes 40
SFAS 98 258 4,957 25 No 72 SFAS 143 120 7,572 62 Yes 211
SFAS 99 101 359 3 No 132 SFAS 144 913 9,029 77 Yes 280
SFAS 100 63 179 4 No 270 SFAS 145 120 8,308 61 Yes 211
SFAS 101 160 1,274 12 No 81 SFAS 146 486 3,547 28 No 50
SFAS 102 77 1,265 11 No 69 SFAS 147 426 6,222 51 Yes 53
SFAS 103 57 258 6 No 120 SFAS 148 59 2,891 12 No 10
SFAS 104 143 1,021 8 No 112 132R 730 2,609 21 Yes 53
SFAS 105 237 2,478 22 No 188 SFAS 149 153 1,403 17 No 24
SFAS 106 668 14,034 115 Yes 475 SFAS 150 61 1,609 5 No 70
SFAS 107 349 1,666 17 Yes 204 123R 335 7,929 40 No 40
SFAS 108 181 385 7 No 50 SFAS 151 930 3,383 31 Yes 71
SFAS 109 245 8,883 59 Yes 250 SFAS 152 336 976 5 No 26
SFAS 110 148 1,143 9 No 48 SFAS 153 669 716 5 No 233
SFAS 111 138 3,960 11 No 21 SFAS 154 366 2,604 7 No 30
SFAS 112 187 1,524 12 No 59 SFAS 155 517 4,161 27 No 66
SFAS 113 275 3,250 33 Yes 53 SFAS 156 184 2,210 7 No 24
SFAS 114 334 3,169 27 Yes 160 SFAS 157 212 11,225 12 No 26
SFAS 115 242 3,025 25 Yes 600 SFAS 158 822 5,686 39 Yes 100
SFAS 116 212 2,777 30 Yes 280 SFAS 159 184 4,334 22 No 245
SFAS 117 243 3,661 31 Yes 280 141R 396 4,994 30 No 80
SFAS 118 198 1,470 7 No 57 SFAS 160 898 529 6 Yes 49
SFAS 119 183 2,344 17 No 144 SFAS 161 456 4,307 8 No 63
SFAS 120 297 956 11 No 31 SFAS 162 1,126 965 8 No 32
SFAS 121 485 4,037 35 Yes 147 SFAS 163 396 4,067 35 Yes 87
APPENDIX B: Fasb Operating Revenue Panel A: Pre-Sox And Post-Sox Subsample Period
Year Net Contributions + Accounting Support Fees + Sales & Royalties - Direct Cost of Sales = Operating Revenue
    % Change   % Change   % Change   % Change   % Change
1996 51,09,000 0 - 1,32,15,000 23,55,000 1,59,69,000
1997 51,83,000 0.01 0 - 1,26,82,000 -0.04 18,23,000 -0.23 1,60,42,000 0
1998 49,62,000 -0.04 0 - 1,37,24,000 0.08 17,54,000 -0.04 1,69,32,000 0.06
1999 48,31,000 -0.03 0 - 1,37,19,000 -0.13 17,64,000 0.01 1,67,86,000 -0.01
2000 48,03,000 -0.01 0 - 1,40,13,000 0.17 17,03,000 -0.03 1,71,13,000 0.02
2001 51,13,000 0.06 0 - 1,48,18,000 0.06 15,86,000 -0.07 1,83,45,000 0.07
2002 38,95,000 -0.24 0 - 1,33,48,000 -0.1 14,28,000 -0.1 1,58,15,000 -0.14
2003 2,63,000 -0.93 1,96,97,000 - 1,26,02,000 -0.06 12,55,000 -0.12 3,13,07,000 0.98
2004 1,23,000 -0.53 2,53,55,000 0.29 1,26,57,000 0 12,40,000 -0.01 3,68,95,000 0.18
2005 48,000 -0.61 2,02,25,000 -0.2 1,40,70,000 0.11 28,84,000 1.33 3,14,59,000 -0.15
2006 56,000 0.17 2,24,36,000 0.11 1,48,65,000 0.06 42,71,000 0.48 3,30,86,000 0.05
2007 1,01,000 0.8 2,25,14,000 0 1,45,57,000 -0.02 52,03,000 0.22 3,19,69,000 -0.03
2008 1,10,000 0.09 2,37,10,000 0.05 1,41,02,000 -0.03 36,07,000 -0.31 3,43,15,000 0.07
APPENDIX B:Fasb Operating Revenue Panel B: Fasb Total Expenses In Pre-Sox And Post-Sox Period
Year Salaries & Wages + Employee Benefits + Occupancy & Equipment + Other Operating Expenses = FASB Program Expenses + Support Expenses = Total Expenses
1996 82,67,000   14,78,000   9,04,000   11,87,000   1,18,36,000   39,81,000   1,58,17,000
1997 86,61,000 13,99,000 8,00,000 11,08,000 1,19,68,000 43,50,000 1,63,18,000
1998 94,64,000 14,48,000 7,94,000 11,42,000 1,28,48,000 44,05,000 1,72,53,000
1999 85,83,000 19,72,000 10,34,000 15,21,000 1,31,10,000 46,01,000 1,77,11,000
2000 92,25,000 17,50,000 10,18,000 16,42,000 1,36,35,000 48,00,000 1,84,35,000
2001 92,96,000 22,60,000 10,31,000 19,16,000 1,45,03,000 48,83,000 1,93,86,000
2002 93,27,000 25,97,000 10,72,000 17,80,000 1,47,76,000 52,39,000 2,00,15,000
2003 1,02,84,000 32,58,000 10,77,000 16,58,000 1,62,77,000 58,53,000 2,21,30,000
2004 1,23,79,000 36,12,000 11,44,000 19,39,000 1,90,74,000 60,95,000 2,51,69,000
2005 1,29,42,000 39,11,000 10,55,000 13,84,000 1,92,92,000 70,53,000 2,63,45,000
2006 1,35,46,000 41,29,000 9,33,000 14,13,000 2,00,21,000 81,20,000 2,81,41,000
2007 1,42,64,000 39,21,000 10,47,000 17,11,000 2,09,43,000 95,72,000 3,05,15,000
2008 1,42,99,000 40,10,000 10,82,000 21,21,000 2,15,12,000 99,45,000 3,14,57,000

References