Investors' perceptions of earnings quality, auditor independence, and the usefulness of audited...

SYNOPSIS: In this paper I investigate whether nonprofessional investors' beliefs mirror the Securities and Exchange Commission's (SEC) concerns that earnings quality and auditor independence have declined over time. I also examine whether lower perceptions of earnings quality are associated

with more or less reliance on a firm's audited financial statements and fundamental analysis of those statements when making investment decisions. My results suggest that the SEC's concerns are valid: Perceived earnings quality for all publicly traded firms has declined overtime, as has perceived auditor independence and the perceived reliability of audited financial information. In contrast, the perceived relevance of audited financial information has increased. In addition, results reveal that lower perceptions of earnings quality are associated with greater reliance on a firm's audited financial statements and fundamental analysis of those statements when making investment decisions. This result suggests either (1) lowe r perceptions of earnings quality lead investors to examine more thoroughly a firm's audited financial statements, or (2) more thorough analysis of a firm's financial statements leads investors to lower their assessments of the firm's earnings quality.

Keywords: investor beliefs; earnings quality; auditor independence; reliability; relevance.

Data Availability: The data and complete survey are available upon request.

INTRODUCTION

In this paper I investigate whether investors' beliefs mirror the Securities and Exchange Commission's (SEC) concerns that earnings quality and auditor independence have declined over time. (1) I also examine whether lower perceptions of earnings quality are associated with more or less reliance on a finn's audited financial statements and fundamental analysis of those statements when making investment decisions.

Within the SEC, Commissioners Isaac Hunt, Jr. and Laura Unger have recently, and repeatedly, expressed their concerns about earnings quality and earnings management (Hunt 2001a, 2001b; Unger 2001a). Their concerns mirror those of former chairman of the SEC, Arthur Levitt, who expressed concern over earnings management and its effect on resource allocation in his 1998 speech, "The Numbers Game." In his speech, Levitt notes that management abuses of "big bath" restructuring charges, premature revenue recognition, "cookie jar" reserves, and write-offs of purchased in-process R&D threaten the credibility of financial reporting. Levitt reiterates these concerns in his book, Take on the Street (Levitt 2002).

To conduct my investigation I surveyed 414 individual investors who are members of the National Association of Investors Corporation (NAIC). Survey-based research complements archival-based research and experimental studies. Archival-based research provides statistical power and cross-sectional variation, but is less suited for examining individual behavior and beliefs. Experimental studies are excellent for determining how changes in variables affect individual behaviors and beliefs, but are less well suited for gathering large amounts of descriptive data on multiple behaviors and beliefs. Surveys offer a balance between archival-based and experimental studies by gathering data on a multitude of beliefs and practices for moderately large samples of individuals. Results from survey research can be used for building and testing theories, for attempting to explain what underlies market-level decision outputs (e.g., prices, returns, or trading frequencies), and for identifying trends in beliefs and practices.

As with all types of research, the survey approach has potential drawbacks. Surveys measure beliefs and not necessarily actions, and surveys might be subject to various response biases. I test for potential biases by comparing statistics from my sample to statistics for the national population of NAIC investors, prior archival-based studies, and a market-level performance metric. Comparative statistics indicate that my sample is similar to the national population of over 400,000 NAIC investors, and investors' self-reported returns and trading frequency data are consistent with those reported in Bange (2000) and Barber and Odean (2001). In addition, investors' self-reported returns minor those of the Dow Jones Industrial Average over the period I analyze. These results provide some assurance against response biases and imply that my sample is representative of the greater population of equity investors.

My results suggest that the SEC's concerns about declining earnings quality and auditor independence are valid: perceived earnings quality for all publicly traded firms has declined overtime, as has perceived auditor independence and the perceived reliability of audited financial information. In contrast, the perceived relevance of audited financial information has increased. Interestingly, as investors rely more on audited financial information, they find that information to be less reliable. My data suggests that one possible reason for the decline in the perceived reliability of audited financial information is the perceived decline in auditor independence.

In addition, my results reveal that lower perceptions of earnings quality are associated with greater reliance on a firm's audited financial statements and fundamental analysis of those statements when making investment decisions. This result suggests either (1) lower perceptions of earnings quality lead investors to more thoroughly examine a firm's audited financial statements, or (2) more thorough analysis of a firm's financial statements leads investors to lower their assessments of the firm's earnings quality.

RESEARCH METHOD

Sample

I drew my sample from the membership base of a regional chapter of the NAIC. The NAIC consists of over 400,000 individual and investment club members with a total portfolio value greater than $125 billion (NAIC 2002). Nine out of ten NAIC club members buy securities for their own accounts in addition to making club investments (O'Hara and Janke 1998). NAIC investors typically assess stocks by making calculations based on information found in a firm's financial statements and from other sources (such as Value Line). NAIC investors are encouraged to utilize a fundamental approach to investment analysis, one that focuses heavily on earnings, and to analyze a stock to determine if it has the right combination of growth, value, and risk properties given its price. Given the NAIC's focus on earnings for investment analysis, NAIC investors are well suited for testing whether perceptions of earnings quality have changed over time.

Design and Response Rate

The survey was three pages long and consisted of 102 questions. I designed the survey around three general categories: investor beliefs, investing practices and performance, and demographic information. In this paper I focus on the survey questions related to earnings quality, auditor independence, and the reliability and relevance of audited financial information. (2)

Academics from several universities, as well as NAIC members and officers, reviewed early drafts of the survey for question ambiguity and overall clarity. The NAIC regional chapter's board of directors approved the final draft of the survey.

I distributed the survey in two phases. In phase one I inserted the survey in the regional chapter's Spring 2001 quarterly newsletter. The newsletter was mailed to 13,250 members. Members were encouraged to complete the survey as soon as possible by filling out the hardcopy survey contained in the newsletter and returning it postage free, or by completing the survey online. In the second phase I placed a reminder in the Summer 2001 newsletter and announced that I would randomly draw one name from the pool of returned surveys to receive a $200 cash payment. In order to be eligible for the random drawing, members had to complete and submit the survey by the July deadline.

Four hundred fourteen investors completed the survey for a response rate of just over 3 percent. Given the length (three pages) and detail (102 questions) of my survey, a low response rate was expected (Rea and Parker 1997, 12). (3) Rea and Parker (1997) suggest that a sample size that allows for 95 percent confidence with a margin of error, or confidence interval, of approximately 5 percent is adequate for most types of survey research. My sample size exceeds these minimums. (4) In the following sections, I perform several tests to investigate whether nonresponse bias might affect my results.

Early versus Late Responders

Consistent with the suggestion of Wallace and Mellor (1988) and Oppenheim (1992), I compare the responses for individuals who returned the survey prior to the $200 random drawing announcement (287 individuals) to those who returned the survey after the announcement (127 individuals). I refer to these two groups as "early" and "late" responders, respectively. The late responders can be thought of as a sample from the nonresponse group, in the sense that they did not return the survey until offered a financial incentive to do so. For all of the statistics reported in Tables 1-5, only two differ between early and late responders at the 5 percent level. Early responders tend to be older than late responders, and early responders perceive all publicly traded firms' earnings in 1990 to be of higher quality than do late responders. (5)

Of the 102 questions investors answered, 75 are subquestions of nine general topic areas. I therefore calculate multivariate test statistics for each set of subquestions within each topic area. These tests examine whether the patterns of means for subquestions within a given topic area differ between early and late responders. Test results reflect that none of the nine [chi square]s comparing the responses of early and late responders significantly differ (all p-values greater than 0.99). These tests suggest that the responses for early and late respondents are similar, implying that nonresponse bias is not likely a major concern (Wallace and Mellor 1988).

My Sample versus the NAIC National Membership Base

I also examine whether my sample is representative of the 400,000 NAIC members nationwide. As suggested by Wallace and Mellor (1988) and Moore and Reichert (1983), I compare characteristics of responding investors to characteristics of the population at large. If the characteristics match, then my sample can be thought of as representing the population. While I do not have data for the population of all individual investors in the marketplace, I do have data on several demographic characteristics for the 400,000 NAIC members nationwide.

The membership base of the NAIC as a whole is 69 percent female, the median age is 53 years, and 70 percent of members have a college education (NAIC 2002). Within my sample, 68 percent are female, the median age is between 50 and 59 years (age was divided into ten-year increments on the survey), and 75 percent of respondents have a college education. My sample contains a higher percent of college-educated investors than does the NAIC membership base as a whole (75 percent versus 70 percent; [chi square] = 6.31, p = 0.01), but mirrors the NAIC membership base's gender composition (68 percent versus 69 percent; [chi square] 0.09, p 0.76) and median age. (6)

My Sample versus a Market Performance Metric and Prior Archival Findings

In this section I compare self-reported one-year returns data for my sample to that of the Dow Jones Industrial Average, and examine whether my results are consistent with prior archival results. The surveys were filled out between the middle of March 2001 and the end of July 2001. For the period March 2000 to the end of July 2001, the Dow's average return was --0.8 percent. (7) The NAIC reports that of the top 100 stocks held by NAIC investors, 80 percent are traded on the New York Stock Exchange (NAIC 2001), making the Dow a good comparative market-level metric. Excluding one outlier with a self-reported return of --800 percent, the average self-reported return for my sample is --0.7 percent, roughly equivalent to the Dow's decline from mid-March 2000 to the end of July 2001. This result is consistent with the results of Bange (2000), showing that individual investors, on average, do not outperform the market.

Finally, Barber and Odean (2001) examine account data for over 35,000 households from a large discount brokerage firm and find that males exhibit significantly lower returns on average than do females and that males trade significantly more often than do females. They attribute this discrepancy to male investors being more overconfident in their stock-picking abilities than female investors, resulting in male investors on average trading 45 percent more often than do female investors. Excluding the --800 percent outlier, my results are consistent with those of Barber and Odean (2001). Female investors in my sample report significantly higher returns than do male investors (1.66 percent versus --4.56 percent; t = 2.53, p = 0.01, two-tailed), and male investors in my sample trade significantly more times per year (47 percent more often) than do female investors (10.42 times per year versus 7.08 times per year; t = 2.62, p < 0.01). My results are consistent with those of Barber and Odean (2001): male investors i n my sample have on average significantly lower reported returns, and report trading significantly more often, than do female investors in my sample.

Other Concerns with Survey Data

Other concerns with survey data are that questions might not be understood, individuals might submit more than one survey, and respondents might not answer truthfully. To address these issues, I first attempted to limit any ambiguity in the questions by submitting early drafts of the questionnaire to academics from several universities, NAIC members, and NAIC officers for review. Second, I screened the data for duplicate responses using responders' names, mailing addresses, and email addresses. I found six duplicate responders; all six submitted an initial survey in the spring and later submitted another survey after the summer reminder. Only the earliest responses of these six investors remain in the data set. I cannot directly assess whether respondents answered truthfully, but conversations with NAIC members who completed the survey suggest that investors would not take the time to fill out the survey if their intent was to be untruthful.

RESULTS

As noted earlier, my sample is predominately female (68 percent), the median age is between 50 and 59 years, and 75 percent have graduated from college. Seventy-eight percent of the investors in my sample are between the ages of 40 and 69 years, and 36 percent have earned a graduate degree. In addition, investors in my sample report having actively bought and sold equity securities for an average (median) of 9.9 (6.6) years and buy or sell stocks on average 8.4 (5.0) times a year.

Perceptions of Earnings Quality

Consistent with Pratt (2000, 750), I define earnings quality as the "extent to which net income reported on the income statement differs from true earnings." To examine whether investors' perceptions of earnings quality have declined over time, I asked investors how often do reported earnings reflect "true" (unbiased and accurate) earnings for all publicly traded firms in 1990 and for all publicly traded firms in 2000. Investors responded on five-point scales with endpoints labeled 1 = seldom and 5 = often. Given the SEC's concerns about declining earnings quality, I expect investors to judge earnings quality to be lower in 2000 than in 1990.

The number and frequency of responses are shown in Panel A of Table 1. To test whether perceptions of earnings quality have decreased over time I compare each investor's perception of earnings quality in 1990 to his/her perception of earnings quality in 2000 using a paired-samples t-test. Results shown in Panel B of Table 1 show that perceived earnings quality for all publicly traded firms in 1990 is greater than perceived earnings quality for all publicly traded firms in 2000 (t = 3.16, p < 0.0 l). (8) This finding is consistent with the SEC's concern that the quality of publicly traded firms' earnings has declined overtime. (9)

In conjunction with collecting information on perceptions of earnings quality overtime, I asked investors for their perceptions of earnings quality of several different industries. Descriptive data shown in Panel C of Table I reflects that investors ranked Dow Jones Industrial Average firms as having the highest quality of earnings among the industries listed and dot-coin firms as having the lowest quality of earnings.

Investors also answered the following question related to earnings quality: "In your opinion, how frequently do managers of publicly traded firms manage earnings (i.e., undertake actions that cause reported earnings to differ from 'true' (unbiased and accurate) earnings)?" Investors responded on an 11-point scale with endpoints labeled "0 percent of the time" and "100 percent of the time." The number and frequency of responses are shown in Table 2. Results show that investors on average believe that managers manage earnings approximately 50 percent of the time (mean = 48 percent, median = 50 percent).

Responses (not tabulated in a table) to a related question, "In your opinion, how often do managers of publicly traded firms undertake each of the following actions to affect reported earnings?" reveal that boosting earnings with one-time gains (e.g., by realizing investment gains or reversing unused restructuring charges) was viewed as the most prominent way managers manage earnings, followed by inappropriate expense recognition practices. (10) Using evidence from auditors about managers' attempts to manage earnings, Nelson et al. (2003) report that managers most often use reserves (general and restructuring) to manage earnings. It appears investors' beliefs about how managers manage earnings are in line with auditors' experiences.

Perceptions of Auditor Independence and Usefulness of Audited Financial Information

The SEC has suggested that auditor independence issues are threatening the usefulness of audited financial information (Hunt 2001b). A report by the SEC's chief accountant covering more than half of the Fortune 1000 companies shows that, on average, for every dollar of audit fees audit clients paid to their auditors, they paid $2.69 for nonaudit services (Unger 2001b). The SEC contends that such statistics suggest a possible conflict of interest for auditors. Some empirical research supports this contention (e.g., Frankel et al. 2003), while other empirical research does not (e.g., Ashbaugh et al. 2002; DeFond et al. 2002).

I examine investors' perceptions of auditor independence and the usefulness of audited accounting information by asking investors the following three questions: (11)

1. In your opinion, has auditor independence changed over the last five years? (Response options were "auditors are less independent," "independence has not changed," and "auditors are more independent.") (12)

2. In your opinion, has the reliability of audited financial information changed over the last five years?

(Response options were "reliability has decreased," "reliability has not changed," and "reliability has increased.")

3. In your opinion, has the relevance of audited financial information changed over the last five years?

(Response options were "relevance has decreased," "relevance has not changed," and "relevance has increased.")

Given the SEC's concerns, I expect investors' beliefs to reflect a decline in auditor independence and the reliability of audited financial information. I am uncertain about whether investors will perceive a change in the relevance of audited financial information. On one hand, if investors believe that managers are managing earnings, then they might perceive audited financial information as less relevant. On the other hand, if investors distrust the earnings number reported by management, then they might decide to more thoroughly analyze the audited financial statements to determine the validity of the reported numbers, potentially making the statements more relevant.

Results for the above three questions are reported in Panels A-C of Table 3. (13) To test whether investors' opinions have changed over the last five years, I use a Chi-squared test to compare the proportion of investors who responded that independence, relevance, and reliability have declined to the proportion who responded that independence, relevance, and reliability have increased.

Results reported in Panel D of Table 3 show that more respondents believe that auditor independence has decreased than increased over the last five years (34 percent versus 11 percent; [chi square] = 194.38, p < 0.01). Similarly, more respondents believe that the reliability of audited financial information has decreased than increased over the last five years (27 percent versus 18 percent; [chi square] = 18.63, p < 0.01). In contrast to the above results, more respondents believe that the relevance of audited financial information has increased than decreased over the last five years (32 percent versus 17 percent; [chi square] 40.69, p < 0.01). Parametric t-tests that examine whether the average response for each variable is significantly different from a "no change" response yield inferentially identical results (all p-values less than 0.01).

Overall, my results are consistent with the SEC's concerns and suggest that investors perceive today's audited financial information as being more relevant, but less reliable, than audited information was five years ago. My findings with respect to auditor independence suggest that a possible reason for the decrease in the reliability of financial information is the perceived decrease in auditor independence over the last five years. Results are consistent with this conjecture: the correlation between reliability and auditor independence is positive and significant (r = 0.54, p < 0.01, two-tailed).

Earnings Quality Beliefs and the Use of Financial Statements for Investing Purposes

The final question I address is whether investors' earnings quality beliefs are associated with their use of audited financial information. This analysis is exploratory as I am unable to predict or conclude whether investors' beliefs about earnings quality drive their use of audited financial information, or whether their use of audited financial information drives their beliefs about earnings quality. This analysis, however, will show whether lower perceptions of earnings quality are associated with investors using audited financial information more or less when making investment decisions. I conduct my analysis by correlating beliefs about how often managers of publicly traded firms undertake actions that cause reported earnings to differ from "true" (accurate and unbiased) earnings (i.e., actions that reduce earnings quality), with how often investors report using audited financial information to make investment decisions.

Table 4 reports that the correlations between how frequently investors believe managers manage earnings and how frequently investors report using audited financial statements are positive and significant for both a financial statements composite measure and each individual financial statement (all p-values less than 0.03, two-tailed).'4 In addition, the correlation between how frequently investors believe managers manage earnings and how frequently they report using financial statement! ratio analysis is positive and significant (r = 0.13, p = .02, two-tailed). Though I can conclude that lower assessments of earnings quality are positively associated with greater use and analysis of a firm's audited financial statements, I cannot conclude that lower assessments lead investors to more thoroughly use and analyze a firm's audited financial information. It is possible that greater use and more thorough analysis of a firm's audited financial information lead investors to lower their assessments of earnings quality.

LIMITATIONS AND CONCLUSIONS

In this paper I use the results of a survey to investigate whether investors' beliefs mirror the Securities and Exchange Commission's (SEC) concerns that earnings quality and auditor independence have declined over time. I also examine whether lower perceptions of earnings quality are associated with more or less reliance on a firm's audited financial statements and fundamental analysis of those statements when making investment decisions.

Results suggest that the SEC's concerns about declining earnings quality and auditor independence are valid: perceived earnings quality for all publicly traded firms has declined over time, as has perceived auditor independence and the perceived reliability of audited financial information. In contrast, the perceived relevance of audited financial information has increased. In addition, results reveal that lower perceptions of earnings quality are associated with greater reliance on a firm's audited financial statement and fundamental analysis of those statements when making investment decisions. The latter result suggests either (1) lower perceptions of earnings quality lead investors to more thoroughly examine a firm's audited financial statements, or (2) more thorough analysis of a firm's financial statements leads investors to lower their assessments of the firm's earnings quality. Hopefully, future research will disentangle this relationship.

This study is subject to several limitations. First, surveys capture perceptions that might not accurately match factual base rates. This limitation is particularly relevant to regulators as they consider my results. For example, it is possible that the perceived decrease in auditor independence reported previously is a result of the SEC's recent emphasis on this issue and not investors' actual experiences. If this is the case, then using perception data to justify the passage of stricter independence rules is circular and amounts to self-justification.

Second, self-reported survey data might not fully reflect the actual beliefs and practices of investors. Though comparative statistics from my sample are similar to statistics for the national population of over 400,000 NAIC investors, prior archival-based studies, and a market-level performance metric, it is important to keep this limitation in mind when interpreting the results. Despite these limitations, this study provides a snapshot of the beliefs of active equity investors on issues at the forefront of the SEC's agenda: earnings quality, auditor independence, and the reliability and relevance of audited financial information.

TABLE 1

Perceived Changes in Earnings Quality over Time

Panel A

How Often Do Reported Earnings Reflect "True" (Accurate And
Unbiased) Earnings

                All Publicly Traded             All Publicly Traded
                   Firms in 1990                   Firms in 2000

          Number      Frequency (%) (a)  Number     Frequency (%) (a)

Seldom 1    15               4.8           12              3.7
       2    43              13.7           47             14.6
       3   112              35.6          146             45.2
       4   107              34.0           91             28.2
 Often 5    38              12.1           27              8.4
           315             100.0          323            100.0

Mean       3.4                            3.2

Panel B

Paired-Samples Test of Differences in Perceptions of Earnings
Quality in 1990 versus 2000

All Publicly Traded  All Publicly Traded
   Firms in 1990        Firms in 2000     df   t-statistic  p-value (b)

        3.4                  3.2          313     3.16         0.01

versus

Panel C

Quality-of-Earnings Perceptions of Industries (c)


Dot-Com firms                       2.3
Foreign firms                       2.6
Technology firms                    2.8
NASDAQ firms                        3.1
Retail firms                        3.3
Financial services firms            3.3
Manufacturing firms                 3.4
Dow Jones Industries Average firms  3.6

(a)Frequencies do not sum to 100.0% due to rounding.

(b)p-value is one-tailed.

(c)Investors answered the following question: "In your opinion, how
often do reported earnings reflect 'true' (unbiased and accurate)
earnings for each of the following groups of firms?" Investors responded
on five-point scales with endpoints labeled 1 = seldom and 5 = often.

Note: Table made from bar graph

TABLE 2

How Frequently Do Managers Manage Earnings? (a,b)


100% of the time   1.1%
             90%   4.6%
             80%   7.7%
             70%  10.6%
             60%  14.9%
             50%  19.5%
             40%   7.7%
             30%  14.6%
             20%  13.8%
             10%   3.7%
  0% of the time   1.7%

Mean 48.2% of the time

Median 50.0% of the time

(a)Investors answered the following question: "In your opinion, how
frequently do managers of publicly traded firms manage earnings (i.e.,
undertake actions that cause reported earnings to differ from 'true'
(unbiased and accurate) earnings)?"

(b)n = 349.

Note: Table made from bar graph

TABLE 3

Perceptions of Auditor Independence and the Reliability and Relevance of
Audited Financial Information over Time

Panel A

Auditor Independence (n = 331) (a)


Auditors are less independent  34.1%
Independence has not changed   55.3%
Auditors are more independent  10.6%

Note: Table made from bar graph

Panel B

The Reliability of Audited Financial Information (n = 348) (b)


Reliability has decreased    27.0%
Reliability has not changed  54.9%
Reliability has increased    18.1%

Note: Table made from bar graph

Panel C

The Relevance of Audited Financial Information (a = 352) (c)


Relevance has decreased    16.5%
Relevance has not changed  51.1%
Relevance has increased    32.4%

Note: Table made from bar graph

Panel D

Chi-Squared Test of Proportions

                      Decreased          Increased  df

Auditor independence    34.1%    versus    10.6%    1
Reliability             27.0%    versus    18.1%    1
Relevance               16.5%    versus    32.4%    1

                      [chi square]-statistic  p-value

Auditor independence          194.38           0.00
Reliability                    18.63           0.00
Relevance                      40.69           0.00

(a)Investors answered the following question: "In your opinion, has
auditor independence changed over the last five years?"

(b)Investors answered the following question: "In your opinion, has th
reliability of audited financial information changed over the last five
years?"

(c)Investors answered the following question: "In your opinion, has th
relevance of audited financial information changed over the last five
years?"

TABLE 4

Earnings Quality Beliefs and the Use of Financial Statements for
Investing Purposes

Correlation between the                      Pearson Correlation
Perceived Frequency of
Earnings Management and               n            r        p-value (a)

Use of four financial statements     319         0.19        0.01
 for investing purposes (b)
Use of individual financial
 statements for investing
 purposes (c)
 Balance sheet                       334         0.15        0.01
 Income statement                    330         0.13        0.02
Cash flow statement                  328         0.17        0.00
Changes in owners' equity statement  322         0.15        0.01
Use of financial statement/ratio     323         0.13        0.02
 analysis to analyze stocks (d)

(a)p-value is two-tailed.

(b)Four financial statements is a composite measure calculated by adding
the mean response for each individual financial statement and dividing
by four.

(c)Investors answered the following question: "Please indicate how often
you use the following information for investing purposes."

(d)Inestors answered the following question: "How often do you use the
following techniques when analyzing stocks?"

(1.) Consistent with Pratt (2000, 750), I define earnings quality as the extent to which net income reported on the income statement differs from "true" (unbiased and accurate) earnings.

(2.) The complete survey is available upon request.

(3.) I decided in conjunction with the NAIC regional chapter board that, despite the likeliness of a low response rate, a more detailed questionnaire would provide the most useful data for research purposes, and the most useful feedback to the regional chapter board for decision-making purposes.

(4.) According to Rea and Parker (1997, 119, Table 7.1), the minimum sample size required for 95 percent confidence with a confidence interval (margin of error) of 5 percent is 385 respondents.

(5.) I report the statistical test using this variable in Panel B of Table 2. Rerunning this test using only early or late responders yields inferentially identical results.

(6.) I am unable to statistically compare the average age of my sample to that of the NAIC because age was divided into tenyear increments on my survey, whereas the NAIC discloses age as a single number.

(7.) For the 12-month periods ending March 1, April 1, May 1, June 1, July 1, and August 1, 2001, the Dow changed 0.8 percent, --9.5 percent, 3.0 percent, 1.8 percent, 0.5 percent, and --2.3 percent, respectively.

(8.) A nonparametric Wilcoxon signed ranks test and a nonparametric Chi-squared test of proportions yield inferentially identical results (all p-values less than 0.01).

(9.) I received all responses by July 2001, well in advance of the collapse of Enron and Arthur Andersen. It is likely that investors' perceptions of earnings quality have declined even further since these events.

(10.) I created the response list for this question using Schilit (1993, 2).

(11.) To examine "usefulness," I ask investors to assess the reliability and relevance of audited financial information. I ask about relevance and reliability because within the Financial Accounting Standards Board's (FASB) Conceptual Framework "decision usefulness" is defined in terms of the reliability and relevance of accounting information (FASB 1980).

(12.) I elected to ask about changes in auditor independence and the reliability and relevance of audited financial information over the last five years because former SEC Chairman Arthur Levitt initially expressed his concerns about earnings management in a speech entitled "The Numbers Game" in 1998, approximately five years prior to my survey.

(13.) Correlations between these three variables are all positive and significant (all p-values less than 0.01, two-tailed).

(14.) The financial statement composite measure treats the four financial statements as one document by adding the mean response for each individual financial statement and dividing the total by four.

REFERENCES

Ashbaugh, H., R. Lafond, and B. Mayhew. 2002. Do non-audit services compromise auditor independence? Further evidence. Working paper, University of Wisconsin.

Bange, M. 2000. Do the portfolios of small investors reflect positive feedback trading? Journal of Financial and Quantitative Analysis (June): 239-255.

Barber, B., and T. Odean. 2001. Boys will be boys: Gender, overconfidence, and common stock investment. Quarterly Journal of Economics (February): 261-292.

DeFond, M., K. Raghunandan, and K. Subramanyam. 2002. Do non-audit service fees impair auditor independence? Evidence from going concern audit opinions. Journal of Accounting Research (September): 1247-1274.

Financial Accounting Standards Board (FASB). 1980. Qualitative Characteristics of Accounting Information. Statement of Financial Accounting Concepts No. 2. Stamford, CT: FASB.

Frankel, R., M. Johnson, and K. Nelson. 2002. The relation between auditor's fees for nonaudit services and earnings quality. The Accounting Review 77 (Supplement): 71-105.

Hunt, I. 2001a. Financial reporting challenges. Remarks delivered at the Financial Executives International Benchmarking the Quality of Earnings Program, New York, NY, April 26. Washington, D.C.: SEC.

-----. 2001b. Emerging issues under the federal securities laws. Remarks delivered to the DC Bar Corporation, Finance and Securities Section, Washington, D.C., May 15. Washington, D.C.: SEC.

Levitt, A. 1998. The numbers game. Remarks delivered at the NYU Center for Law and Business, New York, NY, September 28. Washington, D.C.: SEC.

-----. 2002. Take on the Street. New York, NY: Pantheon Books. Moore, J., and A. Reichert. 1983. An analysis of the financial management techniques currently employed by large U.S. corporations. Journal of Business Finance and Accounting 10: 623-645.

National Association of Investors Corporation. 2001. Better investing top 100 index. Available at: http:// old.better-investing.org/ content/top 100index.html.

-----. 2002. NAIC fact sheet: Statistical highlights. Available at: http://new.better-investing.org/about/fact.html. Nelson, M., J. Elliott, and R. Tarpley. 2002. Evidence from auditors about managers' and auditors' earnings-management decisions. The Accounting Review 77 (Supplement): 175-202.

O'Hara, T., and K. Janke. 1998. Starting and Running a Profitable Investment Club. New York, NY: Random House.

Oppenheim, A. N. 1992. Questionnaire Design, Interviewing, and Attitude Measurement. New York, NY: Pinter Publishers.

Pratt, J. 2000. Financial Accounting in an Economic Context. 4th edition. Cincinnati, OH: South-Western College Publishing.

Rea, L., and R. Parker. 1997. Designing and Conducting Survey Research. 2nd edition. San Francisco, CA: Jossey Bass.

Schilit, H. 1993. Financial Shenanigans: How to Detect Accounting Gimmicks and Fraud in Financial Reports. New York, NY: McGraw-Hill.

Unger, L. 2001a. Protecting the integrity of financial information in today's marketplace. Remarks delivered to the Philadelphia Bar Association, Philadelphia, PA, March 30. Washington, D.C.: SEC.

-----, 2001b. This year's proxy season: Sunlight shines on auditor independence and executive compensation. Remarks delivered to the Center for Professional Education, Inc., Washington, D.C., June 25. Washington, D.C.: SEC.

Wallace, R., and C. Mellor. 1988. Nonresponse bias in mail accounting surveys: A pedagogical note. British Accounting Review 20:131-139.

Frank D. Hodge is an Assistant Professor at the University of Washington.

I thank the individual investors who unselfishly gave their time and efforts to this project and to the regional chapter of the National Association of Investors Corporation for access to their members. I also thank Patricia Dechow (editor), Jim Jiambalvo, Jane Kennedy, Dee Ann Lommers-Johnson, Jamie Pratt, Terry Shevlin, and two anonymous referees for helpful comments, as well as Brooke Elliott for her research assistance. I acknowledge financial support of the William R. Gregory Fellowship and the Accounting Department at the University of Washington.