The implementation of Statement on Auditing Standards (SAS) No. 47 "Audit Risk and Materiality in Conducting the Audit" has been a difficult and controversial process. The Statement requires that auditors make a "preliminary judgment about materiality levels" in planning the audit, yet provides limited guidance on how this measure should be determined. Furthermore, although the Statement suggests that this judgment "may or may not be quantified," it asserts that the auditor "generally plans the audit primarily to detect errors that he believes could be large enough, individually or in the aggregate, to be quantitatively material to the financial statements." Significant implementation difficulties have arisen as to how auditors are to operationalize planning materiality in the absence of a quantitative definition. In response, several "rules of thumb" have emerged from both practice and academic research. For instance, all of the following have been suggested as ways of arriving at a preliminary judgment about materiality:

* Percentage of pre-tax income or net income;

* Percentage of gross profit;

* Percentage of total assets;

* Percentage of total revenue;

* Percentage of equity;

* Blended methods involving some or all of these definitions (e.g., compute some combination of the above and find the average);

* "Sliding scale" methods which vary with the size of the entity (e.g., KPMG Peat Marwick's "audit gauge").

While all these definitions represent plausible methods for quantifying materiality, it seems possible that the resulting materiality measures could significantly differ. Since "planning materiality" should affect the scope of both tests of controls and substantive tests, such differences might be of importance. Thus, it is hypothetically possible that two auditors auditing otherwise similar entities might generate differing scopes of audit procedures, solely based on the "planning materiality" definition used. Likewise, audit procedures for the same client could differ substantially from year to year depending on how materiality is operationalized by the audit team each year.

The objective of this article is to compare the magnitude and stability of alternative "rule of thumb" planning materiality definitions using actual financial data for 1977 to 1986 from the Standard and Poor's annual Compustat database for the following companies:

1. Concord Fabrics Inc.--Textile Manufacturing.

2. Mott's Supermarkets--Retail Groceries.

3. Golden West Financial Corp.--Savings & Loans.

We selected three companies to represent fairly typical manufacturing, retail and financial companies to keep them within the range of clients of most CPAs and we selected companies somewhat smaller than the average within the industries.

Five separate materiality measures were computed for comparison within and between companies:

1. 5% of average pre-tax income (using a 3-year average).

2. Sliding scale of gross profit depending on firm size (i.e., 5% of gross profit if between $0 and $20,000; 2% if between $20,000 and $1,000,000; 1% if between $1,000,000 and $100,000,000; 1/2% if over $100,000,000).

3. 1/2% of total assets.

4. 1% of total equity.

5. 1/2% of total revenues.

These and other measures of materiality are discussed in D. A. Leslie, Materiality and its Application to Auditing, (Canadian Institute of Chartered Accountants, 1985). These measures are used for convenience and not because they have any authoritative support. In practice, an even wider array of measures are being used today.

Tables 1 and 2 present the absolute and relative size obtained for planning materiality for each company, on average, over the period from 1977 to 1986. To address the relative magnitude of the measures, the last three rows of Table 1 provide comparisons of largest to smallest, largest to second largest, and second smallest to smallest. This was done to identify cases where one measure might have differed extremely from the others. For example, the ratio of the largest to the smallest measures ranged from 3.1 for Concord to 13.3 for Golden West.

In order to address whether our results were specific only to the three companies selected, we computed the same measures for all companies in the same industries as these firms. Our analysis showed that the ratios of largest to smallest measures for these three firms were somewhat conservative when compared to industry averages (i.e., 5.3 for all textile manufacturers, 8.9 for all retail grocers, and 18.8 for all savings and loans). These findings illustrate that substantial differences in quantitative materiality measures are obtained, depending on the definition chosen. Extending the analysis, these differences could translate into sizable differences in audit scopes (and audit costs) as a function of the measure utilized. To the extent that a CPA firm uses rules of thumb such as these, one might speculate that the selection of a given materiality measure could also affect a CPA firm's competitive position for obtaining and retaining audit clients.

By examining the value in the last two rows, it is evident that the largest measure for all three companies tends to swamp even the second largest measure. Conversely, the smallest measures tend to be fairly close in magnitudes.

Table 2 summarizes rankings of the five materiality measures. The numbers of years where the given measure was largest and smallest are shown parenthetically. As seen in the table, the revenue-based measure was largest for the manufacturer and the retailer for all ten years. Likewise, the asset-based measure was largest for the financial institution. The smallest measure results varied between measures much more, with the income-based measure tending to be among the smallest for all three companies.

The results of this "small sample" study provide empirical evidence that, among five commonly employed "rules of thumb" definitions for materiality, sizable differences can occur depending upon the industry of the client and the definition chosen. For example, for the financial company examined, the largest of the materiality measures was, on average, over 13 times bigger than the smallest measures over the period 1977 to 1986. Such large differences in materiality definitions would presumably lead to corresponding large differences in audit scope decisions, depending on which definition is chosen to quantify "planning materiality." The differences may also carry over into the pricing of audit services by different auditors for the same client, and thus affect the competitive market for audits. While more evidence is needed, it appears that additional authoritative guidance would be helpful. As evidence, given the lack of quantitative guidelines from the FASB, at least one Big 8 firm has developed its own materiality guidelines to ensure intra-firm decision consistency. Perhaps it is time for the profession as a whole to follow suit. Tabular Data 1 & 2 Omitted