Investor Confidence Initiatives: A Cost-Benefit Analysis (Summary Document) The Investor Confidence Initiatives are comprised of three separate Proposed Instruments: Proposed Multilateral Instrument 52-108 Auditor Oversight, Proposed Multilateral Instrument 52-109, Certification Of Disclosure In Companies’ Annual And Interim Filings and Proposed Multilateral Instrument 52-110 Audit Committees. While each of the Proposed Instruments are being published separately, the benefits of these initiatives are wide ranging and, to a significant extent, overlapping. As a result, the analysis of these initiatives has been combined into one document. Consultants Charles River Associates and LECG Economics Finance performed analysis on Certification and the Multi-Jurisdictional Disclosure System (MJDS), respectively. The expected net benefits of the Investor Confidence Initiatives (ICI) are expected to be greater than the sum of the parts. The Cost-Benefit Analysis (CBA) was conducted by a combination of external consultants, the Office of the Chief Economist and the Office of the Chief Accountant. Costs have been as rigorously determined as possible and, for the most part, represent high-end estimates. As is always the case, quantifiable benefits represent a greater challenge. In each case, the benefits presented are incomplete and the probable benefits realized should be substantially greater than the numbers presented here. A positive overall impact on investor confidence could be assumed based on the very evident result of the loss of confidence over the past three years. There is, however, no degree of certainty on how investors will respond to the initiatives or any possible guarantee that financial misstatements or restatements will be eliminated from the capital markets through the implementation of these measures. Instead, we have taken an empirical approach to measuring the benefits accruing to issuers, and their investors, from the experience of firms who have already adopted a governance regime matching the requirements of the Proposed Instruments. We have also relied on survey data of the issuers and market participants to value the benefits in other areas. The parts contributing to the CBA sum include an analysis of CEO/CFO certification, the audit committee requirements and the requirement for auditor oversight by the Canadian Public Accountability Board. We have also considered, but not included in the overall aggregate, the benefits of MJDS as well as the potential benefits if MJDS users were exempted from Sarbanes-Oxley provisions by the SEC where similar Canadian provisions are implemented. While we have employed techniques to quantify costs and benefits, it is very important to recognize several important caveats to our analysis. Our estimates and techniques, while sufficiently rigorous as to be preferable to boilerplate statements or minimalist estimates of costs, should not be interpreted as precise or exact. The primary goal is to determine whether the benefits likely exceed the costs, and not to determine the exact value of any net benefit. The aggregate benefits of the Proposed Instruments for Certification, Audit Committees and Auditor Oversight are in the range of $1.0-10.1 billion. All figures are expressed as a net present value (NPV) over ten years discounted at a rate of 7%. The wide range represents the normal uncertainties associated with estimating the benefits of new policies and the length of the term over which the NPV is calculated. In addition, the significant benefits in a
number of areas not easily quantifiable would suggest that the bias is toward the top end. Normally, efforts would have been made to further narrow this range. However, the lower end of the estimated benefits range significantly outweighs the high end of the aggregate cost estimates, which totalled $163-308 million 1 . With a very low probability of overlap between the low end of the benefits and the high end on costs, no further effort was needed to narrow the ranges based on established CBA methodology. The potential impact of the ICI on MJDS 2 has not been included in these results given that there is no way to assess the probability that the future of MJDS will be affected by the introduction (or lack thereof) of the Proposed Instruments. We find that the main savings of MJDS remain intact despite Canadian MJDS Eligible Issuers being required to comply with SOX. If there is a negative impact from not introducing provisions in line with the Sarbanes-Oxley Act (SOX), then the estimated benefits of MJDS of US$1.6-3.0 billion could become a cost to MJDS issuers. If MJDS issuers were permitted to comply with the requirements set out in the Proposed Instruments instead of the comparable provisions of SOX, there would be a potential benefit of approximately US$273 million 3 . CEO/CFO Certification 45 One of the regulatory changes that is being considered in Canada is the requirement that the chief executive officer (CEO) and chief financial officer (CFO) certify the material accuracy of financial information, including the management discussion and analysis (MD&A) related to the financial statements. Section 302 is the relevant section of SOX containing these certification requirement. Apart from some minor differences related to how information is disclosed in Canada, the proposed national instrument for Canada is very similar to s.302 of SOX, both in form and implementation. 6 This section provides an analysis of the potential costs and benefits of these new certification requirements to Canadians. The potential costs from these requirements are relatively clear – there are added internal costs to the firms and additional costs paid to outside advisors. In 1 For cost methodology on the Certification CBA, please see pages 3-4 of this document. Methodology for the Audit Committee CBA is described on pages 9-10 of the Summary. Complete methodologies are available in the background papers cited at the beginning of each section, available on the OSC web site. 2 LECG Economics Finance, Professor Poonam Puri, Professor Anindya Sen, May 23, 2003 3 The Net Present Value of the external professional fees paid by Canadian issuers listed in the U.S. to comply with SOX is estimated at US$683 million over a ten-year period, using a discount rate of seven percent. If Canadian issuers listed in the U.S. were exempted from compliance with SOX by the SEC and could instead comply with Canadian rules that conform to SOX, this amount would drop to approximately US$410 million, resulting in savings of approximately US$273 million. 4 Multilateral Instrument 52-109, Certification Of Disclosure In Companies’ Annual And Interim Filings 5 The Cost and Benefits of Management Certification of Financial Reports, Charles River Associates Canada Ltd., May 23, 2003 6 Certification of Disclosures in Companies’ Quarterly and Annual Reports, Release no. 33-8124. 2
order to estimate these costs, we interviewed various industry participants, including interlisted Canadian companies that have had to comply with s.302 of SOX, and collected public data. Using the information collected, we estimate the added time that CFOs and CEOs must take to review financial reports and the increased external costs that will be spent on auditors and lawyers. We then use publicly available data to calculate industry-wide costs from these estimates. There are three components to our cost calculations: • An increase in internal hours spent by the CEO and CFO; • A small increase in CFO salaries; and, • Increased expenditures on auditors and lawyers. We value CEO and CFO time based on salaries without bonuses since the opportunity cost of the time spent on disclosure is unlikely the forgone time spent on activities generating high value for the firm that justify the bonuses. As central estimates we assume CEOs and CFOs of small firms (those listed on the Venture Exchange) spend an additional four hours per quarter reviewing quarterly disclosure filings plus an additional four hours to review year-end disclosure filings. In the initial year we assume an expenditure of 10% of current audit fees on additional audit and legal advice. 7 We assume a further 5% more in audit fees per year for ongoing advice. 8 Finally, we assume an increase of 0.5% in CFO salaries. 9 Not all small firms are likely to require expenditures at this level. Our interviews suggested that some CEOs and CFOs of small firms would feel comfortable signing the disclosure documents without any additional internal effort or expenditure on outside experts. For TSX-listed companies we assume that CEOs and CFOs devote the same 20 hours a year of additional time to review filings prior to their release. We assume a smaller salary increase for the CFO of 0.2%. (The representatives of larger firms did not generally feel that salaries would change at all in response to the legislation.) Based on our interviews, most large firms should not face substantial set-up costs or increased ongoing audit costs. We estimate a set-up cost of 5% of existing external audit fees and an ongoing cost of 1% of audit fees. By using shares of existing audit costs, our cost estimates are appropriately scaled for the size and complexity of the firm. This is consistent with the notion that the new regulations are 7 By contrast, to set up auditable internal controls as per s.404 of SOX would require expenditures on the order of 100% to 300% of existing audit fees. 8 As compared with 15 to 100% expected ongoing costs to comply with s.404 of SOX. 9 This increase is very small, but the incidence of reporting fraud we estimate to be only 0.36%. The CFO can mitigate his or her exposure to risk through better internal controls and attention to financial reporting as implied in the other cost assumptions. Thus a 0.5% increase in salary is likely on the high side even for a risk adverse CFO. (The increase in the CFO salary is a real economic cost in the sense that the CFO is exposed to additional regulatory risk that cannot be reduced to zero.) 3
not intended to prescribe what internal controls are needed, only the outcomes that need to be achieved. To translate the above percentage increases into actual dollar amounts, we first estimate salaries and audit costs and then apply the percentage increases. We do this for a sample of firms and then extrapolate to all firms listed on the TSX or Venture Exchange but that are not interlisted. 10 To estimate salaries and audit fees we undertook the following steps: 1. We hand collected data on CEO salaries, audit fees and assets for a random sample of TSX and Venture Exchange companies from their proxy circulars available on SEDAR. With this information, we estimated the relationship between firm size and audit fees and the relationship between firm size and CEO salary. (A log-linear regression is used in both cases. The details of the estimation are provided in the appendix.) 2. We use the regression coefficients to predict salaries and audit fees based on asset values for all firms for which we were able to collect asset data from Bloomberg. The Bloomberg data represents about 50% of the TSX non-interlisted company population and about 10% of the Venture Exchange company population. We tested to see if there was a selection bias in Bloomberg in the companies reported using our random sample of hand-collected data and found that after controlling for exchange there was no bias. 3. We calculated costs based on the assumed percentage increases above, using a real discount rate of 5% to compute the net present value of costs over a 10 year horizon. 11 We grossed up to industry level costs using the ratio of the size to the total population to the Bloomberg size. The potential benefits from the certification requirements are improved investor confidence leading to an improved financial system. In the extreme, a financial market with a reputation for widespread accounting irregularities will reduce the number of investors thereby raising the cost of capital to those firms seeking equity financing. While clear in principle, these benefits are inherently difficult to measure. 12 Given their intangible nature, we are only able 10 We have assumed that all interlisted firms are listed on a U.S. exchange and are therefore subject to SEC regulation and exempt from the proposed Canadian certification requirements. 11 We assume a 7% nominal discount rate based on the average long bond rate over the past decade and a 2% inflation rate (the middle of the Bank of Canada’s target inflation range of 1 to 3%). In terms of benefits, one might argue that the payoffs are proportional to the value of equity and thus the discount rate should be higher than that applied to costs. However, the COSO study of fraudulent reporting found that frauds were more likely to occur when a firm was performing poorly. This would suggest that reducing fraudulent reporting adds value most when the market overall is performing poorly. This in turn implies a low, potentially negative correlation in the payoffs from reduced fraudulent reporting and thus suggests a lower discount rate is appropriate. We thus use a 5% real discount rate for both costs and benefits. 12 The SEC discusses at a very high level possible benefits and costs of the certification requirements in their final rules (Final Rule: Release No33-8124). The SEC maintains that there are likely significant benefits from 4
to quantify some portion of the potential benefits. For this exercise, we estimate the potential reduction in the incidence of financial misstatements and the value that this reduction would have for honest companies from reduced costs of capital. Quantifying costs or benefits of a regulatory policy aimed at reducing the incidence of errors (intentional or otherwise) in financial reporting is difficult for several reasons. First, the proposed instrument is designed to allow firms to choose the appropriate level of controls that the CEO and CFO (and the Board and audit committee) feel is appropriate to provide the new certificates. 13 While we believe such a flexible regulatory approach is very useful for minimizing the regulatory burden 14 , it makes it more difficult to predict the operational steps that companies will take to implement the regulations—and hence the costs are more difficult to quantify. Second, the quantitative analysis of benefits is partially based on Ontario Securities Commission (OSC) data from continuous disclosure reviews. It also implicitly assumes a level of enforcement that engenders the type of response exhibited by firms that must meet U.S. regulations. Section 906 of SOX imposes significant new criminal penalties including up to 20 years in prison. This has motivated CEOs and CFOs to take actions in response to s.302. The level of response to OSC and other Canadian securities regulators will depend on CEOs and CFOs’ expectations of enforcement and the size of penalties. The interviews did not suggest that market participants view enforcement in Canada to be significantly weaker than in the U.S. such that the firm responses to OSC certification requirements would be different from their response to SEC certification requirements. Nevertheless, the response may be more significant in the U.S. than in Canada due to s.906. The effectiveness of the certification requirements in either country will ultimately depend on how the regulations are enforced. Below, we summarize our findings. Interview Findings: • Certification requirements would motivate many firms to undertake additional actions to meet such requirements, including increased attention by the CEO and CFO to financial disclosures, enhancing disclosure controls and procedures, and, especially for smaller firms, increased consultation with external auditors and lawyers. Still, the certification requirements. The apparent difficulty the SEC had in quantifying costs and benefits is not unique to the U.S. situation and we face similar difficulties. On the other hand, we do have the benefit of discussions with Canadian firms interlisted in the U.S. on how they have responded to the SEC regulations (s.302 of SOX) and their perceptions of likely benefits. 13 While we scale our cost estimates for firm size, we cannot account for other differences across firms, such as the sophistication of existing internal controls, which would result in different costs. Thus, our cost estimate range is based on expected average firm costs. 14 Indeed, one of the benefits of a less prescriptive approach is that it allows the firm to determine how to best meet the regulations based on the firm’s particular circumstances. Firms have generally much more information about their individual circumstances than the regulator and therefore have an information advantage. 5
most of the firms and industry representatives we interviewed do not view the certification requirements as unnecessarily onerous. • Large Canadian interlisted firms viewed the certification requirements positively. The increased costs are modest, while firms could realize benefits by having better information for senior executives to make decisions and by passing on any more accurate information to shareholders. • Smaller firms will face larger proportionate costs than large firms, as the CEO and CFO may need to consult outside expertise. However, small firms generally have simpler business models and more compact organizational structures that should allow most CEOs and CFOs to certify financial information without the need to make significant additional expenditures on internal controls—assuming such controls do not have to be auditable. • There is considerable variation between firms of the same size and industry as to the sophistication of internal controls that are in place. Some firms may decide to use the certification requirements as justification to upgrade internal controls, which would likely be at least a marginally profitable investment. The interview findings were based on a relatively small sample primarily due to a low response rate on inquiries. Many of those contacted were unwilling or unable to participate in the survey because they did not have sufficient time to evaluate SOX or the Proposed Instruments in Canada. As a result, the interview findings were not used in estimating the costs and benefits and the views expressed may not be representative of the Canadian capital markets. Academic Literature Findings: • When firms choose to submit to more onerous disclosure requirements they experience an increase in stock prices, reduced bid-ask spreads and greater share turnover. However, when regulations are imposed, some firms may find the costs outweigh the benefits. • Erroneous financial reporting is especially prevalent among smaller firms and the size of misstatements and misappropriations are proportionally larger for smaller firms. However, problems in financial reporting occur at all firm sizes, as the WorldCom and Enron scandals confirm, and the costs are significant for large firms. Better internal controls as well as setting the “tone at the top” are effective at reducing flawed reporting, though they are not a panacea. Cost/Benefit Findings: • Costs are likely to be relatively higher for smaller firms (Venture Exchange-listed) than larger firms (TSX-listed) relative to firm size (measured by assets). This is largely the result of economies of scale in auditing and governance that benefits larger firms. 6
• There may also be modest increases in the salaries of CFOs and CEOs of smaller-firms and some increase in the cost of Directors and Officers (D&O) insurance, to reflect the greater personal risks associated with new regulations. • We estimate the net present value (NPV) of industry-wide costs over a 10 year horizon to be $120 million to $143 million. The upper cost estimate is less than 0.015% of total assets. • Due to the nature of benefits and data limitations there is considerable uncertainty in our benefit estimates. Nonetheless, we estimate the certification requirements could reduce the net present value of the expected amount of misstatements by anywhere from $10 million to $907 million. Given the limited range of the benefits quantified, the expected impact would be, at the very least, at the upper end of this range. • The benefits of reduced financial misstatements are proportionately larger for smaller firms since the size of misstatements are generally proportionately larger. While the cost of misstatements cannot be directly inferred from the size of misstatements, the limited evidence we have suggests that they are of a similar order of magnitude. The costs and reduction in the amount of misstatements are of a similar order of magnitude whether the firm is large or small. We find that reasonable parameter estimates for the probability of financial misstatements, the effect of certification and the size of misstatements (and their cost) put estimated benefits at a similar order of magnitude to estimated costs. In light of the fact that there are also other benefits, such as greater liquidity, lower market risk, and better allocation of resources that we are unable to quantify, we find that the benefits likely exceed the costs. Audit Committees 15 Background and Academic Literature Dozens of studies have been published seeking a connection between firm governance and performance. The results have been mixed with some finding a significant relationship and others a small or insignificant connection. Champions of good governance may be surprised at this, but they should also be aware that a board’s primary duties are expected to focus on longer-term vision and the protection of investors rather than on short-term price movements or day-to-day operations. The loss of investor confidence experienced over the past two years, and the regulatory response, has been based on aggressive accounting. More specifically, the Proposed Multilateral Instrument 52-110 Audit Committees (52-110), which requires an independent audit committee, is designed to enhance the quality of financial reporting. 15 Proposed Multilateral Instrument 52-110 Audit Committees, Analysis by The Office of the Chief Economist, Ontario Securities Commission, May, 2003. 7
One of the most significant issues that tends to degrade market efficiency is information asymmetry. Insiders, in both the issuers and the intermediaries, have access to information not available to the retail investor. While some degree of information asymmetry is unavoidable, the damage to the investor, investor confidence and market integrity is most substantial when the investor is provided with misleading information. This can lead to investors making ill-informed investment decisions to their financial detriment and detracting from overall market efficiency. A substantial number of studies internationally have found a link between governance and accounting choices 16 . With audit committee composition, auditor reporting and certification at the forefront of the investor confidence initiatives, we have chosen to focus this part of the CBA on the relationship between the existence of an independent audit committee and evidence of aggressive accounting. Measuring the degree and frequency of aggressive accounting activity is the first challenge. A number of methods have been proposed, depending on the type of behaviour to be estimated. Some firms may seek to avoid reporting negative earnings in a quarter. Others may wish to show consistent growth over a period of a few years or longer. Earnings may also be managed to generate an earnings “surprise” relative to the consensus of the analysts following the stock. This type of behaviour may precede an offering in the market. There is also a demonstrated managerial incentive to understate earnings, or report a loss, in order to set options prices at a favourable level. By shifting earnings forward, managers can price options at a favourable level and move the stock price higher at a later date to improve the profitability of the options granted. A number of methods have been proposed and evaluated to examine the frequency and impact of each of these methods of earnings management. However, all of these approaches would tend to increase the variability of the deviation between cash flow and earnings. As a result, we have chosen to focus on the differences between the two, relative to measures of the quality of governance. While earnings management comes in many forms and each of those forms may have a significant impact on shareholder value, the most common variety appears to be earnings smoothing. Firms may use accruals and other adjustments in order to report a string of unbroken earnings growth. Following the work of studies done in the U.S. market, we have used the average volatility in cash flow over twelve quarters divided by the average volatility in earnings. If no earnings management has taken place, this ratio should be close to one. Methodology Over all firms in the sample, cash flows were more than 2.5 times as volatile as earnings on average, suggesting a significant and widespread practice of earnings smoothing. Similar studies in the U.S. have found a ratio of just under four times. One-quarter of the sample firms had a mean ratio of almost six times while 44% exhibited a mean of over four times. 16 Bowen, Rajgopal and Venkatchalam (2002), Chtourou, Bedard and Couteau (2001), Xie, Davidson and Dadalt (2002), Ching, Firth and Rui (2002), Pincus and Rajgopal (2002), Becker and DeFond (1995), Warfield and Wild (2002), See References in detailed paper, CBA: Audit Committees 8
While in any given quarter, there may be a justifiable and legitimate reason for a deviation between cash flow and earnings, persistent differences in volatility of four to six times is highly indicative of earnings management. The high percentage of the sample showing this persistence confirms our choice of this variable as a focus. With the very widespread nature of this activity, efforts to reduce it should show the greatest benefit for the overall market. The proxies chosen for the quality of governance are based on the measurable components of 52-110, an audit committee composed solely of independent directors with the auditors reporting directly to the audit committee. Our first hypothesis was that firms with a better governance regime would show a lower incidence of earnings smoothing (a volatility ratio closer to one) and that the governance variables would be significant. Assuming that governance has an impact on the decision to manipulate earnings, we then looked for a connection to shareholder value in order to estimate the benefits of improved governance. This link is also well supported in the studies noted above among others. There are a number of possible measures of shareholder value including equity price movements, market value-added, return on capital, return on equity and total return. Based on recent studies, we chose to focus on economic value added (EVA) 17 . EVA is defined as the rate of return less the cost of capital multiplied by the capital employed. In other words, is the company generating a sufficient return to cover the cost of obtaining capital and, for the total value, how much capital has been employed? In addition to the governance factors, other variables found to have a significant impact on EVA were added in order to ensure a robust and fully specified model. These variables included net income, total assets and the weighted average cost of capital. A sample of 306 publicly listed firms on the TSX was used, approximately one-quarter of the total number of firms listed on the TSX. This is almost double the normal sample size expected to show statistical significance. Governance data for these firms, analogous to requirements of 52-110, was compiled in conjunction with the Rotman School of Business at the University of Toronto through publicly available documents and, where public documents were incomplete for the purposes of the analysis, direct interviews with firm representatives. Costs and Benefits An independent audit committee was found to have a very significant impact on the incidence of earnings smoothing. The Chair-CEO split and the auditor reporting to the audit committee were not found to be significant. Other studies have found that any management influence in the auditor-audit committee relationship negates the impact of independence. In our study, the lack of significance in the auditor report variable may be related to data problems. This variable was based on verbal reports from the issuers and may not conform to the requirements of 52-110. More specifically, while the auditor may be reporting to the audit committee, there may also be significant management influence in the relationship. 17 Hall (2002), Davidson (2001) 9
In turn, the earnings smoothing variable was found to have a substantial and robust impact on EVA. Given that half of the sample already has an independent audit committee, the net impact of independence was applied to half of the firms in the TSX. Given that having the auditor report to the audit committee has been found to be significant in other studies, not to mention the other requirements of 52-110, this is very likely to represent an understatement of the total benefits from the implementation of 52-110. In addition, there are other forms of earnings manipulation that would not be accurately captured in our measure. There may be a significant benefit related to having a financial expert on the audit committee deriving from improved results and investor perception. Conversely, firms without a financial expert may experience lower investor confidence and a higher cost of capital. With no requirement to have a financial expert in 52-110, these costs and benefits were not built into the analysis. Through the impact of reduced earnings smoothing and other manipulation as measured by this variable, we would expect benefits in the range of $1.0-9.2 billion or 0.05-0.4% of total assets, discounted over ten years at a 7% discount rate. From a cost perspective, the sample of 306 TSX-listed companies was broken down into firms that currently meet the criteria, firms that could meet the criteria with independent directors currently serving on the board but not on the audit committee, and firms that would need to hire additional independent directors. 154 companies already meet the requirements of audit committee independence. 102 companies could fulfil the requirements by replacing inside directors on the audit committee with independent directors already on the board. In terms of additional cost as a result of 52-110, we focused on the 50 remaining companies that would have to hire 71 additional independent directors. Cost information was based on a report by Patrick O’Callaghan & Associates 18 . Using low and high end estimates for the additional cost of search fees, meeting fees, committee retainers and director fees, the additional cost range was estimated at $43-165 million on a present discounted value basis over ten years. It was assumed that the new directors would be covered under the current directors and officers (D&O) insurance policies and that there would be no increase in costs resulting from the introduction of 52-110. A survey of the major insurance companies in Canada confirmed this assumption. A substantial increase in D&O costs in the U.S. based on a survey by Foley Lardner 19 has been cited on this topic. That study makes the erroneous assumption that increasing D&O costs are a function of the implementation of the SOX Act. The costs of D&O insurance were on the rise well before SOX was introduced and are a result of a number of factors. The losses associated with the problems at Enron, Worldcom and others would have figured prominently among these factors. Improved governance, with the possible exception of certification as noted above, should reduce the cost of insurance if there is any positive impact at all resulting from these initiatives. 18 “Corporate Board Governance and Director Compensation in Canada: A Review of 2001”, by Patrick O’Callaghan and Associates, December 2001 19 Foley Lardner, The Increased Financial And Non-Financial Cost Of Staying Public, April, 2003 10
One area that presented a significant problem in the cost estimation process was the option of naming a financial expert on the audit committee. While financial expertise does not necessarily imply an accounting or other financial designation (and vice versa), we used these designations as a proxy for additional costs that firms may incur. Since it will be left up to the board to determine the need for a financial expert and how that expert would be qualified, these additional costs may be zero. However, using the professional designations (C.A., C.M.A, C.G.A., C.F.A.) as a proxy, we estimated potential additional costs of $37-143 million. Given that this is an option, not a requirement, these costs have not been included in the totals. It was also assumed that additional administrative costs for an additional director or two would be minimal compared to the other costs noted above. The Canadian Public Accounting Oversight Board 20 The benefits provided by Proposed Multilateral Instrument 52-108 Auditor Oversight include improvements in the quality of audits and reliability of the financial statements filed by reporting issuers. This will improve investors’ confidence in our market and, as a result, help reduce the cost of capital for reporting issuers. These advantages will also harmonize our regulatory regime with the U.S. system. The CPAB will be self-funding and operating costs will be recovered through fees levied on the participating public accounting firms that are inspected by the CPAB. The exact fee structure and amounts have not yet been determined but fees are likely to include three elements: (i) start-up cost recovery fees, (ii) initial registration fees, and (iii) annual/recurring fees. Start-up Cost Recovery Fees - It is expected that the largest four or six public accounting firms will pay the bulk of the start-up cost recovery fees over a two or three year period. The amount to be paid by each firm may vary to reflect relative size of each firm. Initial Registration Fees - This fee will cover the administrative costs of maintaining a register of public accounting firms that have decided to participate in the CPAB Oversight Program. This fee is likely to reflect each firm’s ability to pay and may vary depending upon the number of reporting issuers that each firm audits. Annual/Recurring Fees - It is estimated that the CPAB will have an annual budget in the range of $3 to $5 million, on top of the approximately $3 million that is currently spent on practice inspection by the accounting profession through provincial CA Institutes/Ordre. Indirect costs related to requirements placed on auditors by the CPAB can not be estimated at this time, as the requirements have not been proposed or determined. 20 Proposed Multilateral Instrument 52-108 Auditor Oversight, Analysis by The Office of the Chief Accountant, Ontario Securities Commission, May 29, 2003. 11
Summary For the most part, the CBA Summary focuses on the quantifiable aspects of the ICI. There will almost certainly be qualitative improvements in investor confidence that would translate into higher firm valuations and reduced cost of capital. If we had assumed that the loss of market value over the last few years was solely related to a loss of investor confidence, then even a partial reversal of that decline would cover the costs of the ICI many times over. However, there were many other factors involved in the market sell-off including an overvaluation of equities going into this period, the terrorist attack on the World Trade Center, the war in Iraq and other issues. It is not possible to separate out the impact of the market frauds and conflicts which have damaged investor confidence. Instead, we relied on demonstrable improvements in firm valuations, reduced capital costs and other cost reductions that could be linked to improved firm governance regimes. These estimates contain a degree of uncertainty that is reflected in the wide ranges for both costs and benefits reported. In aggregate, though, even the lower end of the range of quantifiable benefits outweighs the high end of estimated costs by a substantial margin. The currently unquantifiable benefits could extend that margin dramatically. 12
You are being directed to the most recent version of the statute which may not be the version considered at the time of the judgment.