The effectiveness of corporate restructuring: an analysis.

Ramaswamy, Vinita M. ; Fernandez, Ramon

INTRODUCTION

In recent years, restructuring charges have been an increasingly common element on earnings statements. A restructuring charge is a company's estimate of the future costs of a major change in business strategy or operations, such as laying off a large percentage of its workforce, or closing an entire division. The financial press frequently reports a positive market reaction to restructuring, because a company's net income often improves, at least in the short run, in the wake of a non-routine charge against earnings. However, critics complain that such write-offs tend to increase immediate earnings without actual improvement in performance, and that repeated restructurings result in a blurred portrayal of the company's earnings trends. A further consequence of persistent restructuring has been complaints of overworked employees, low morale, dissatisfied customers, and excessive use of expensive contract workers. Such frequent and uncontrolled write-offs provided the impetus for the FASB to take steps to clarify and restrict the use of restructuring charges. During 1995, the FASB issued SFAS 121, "Accounting for the Impairment of Long-Lived Assets to be Disposed," and EITF 95-3, "Liability Recognition for Certain Employee Termination Benefits and Other Costs to Exit an Activity (including certain costs incurred in a restructuring)."

Given such conflicting signals from the market and financial analysts, this study examines the financial performance of restructuring firms to determine if restructuring truly resulted in the efficient, streamlined companies that management and shareholders expected. This research has the potential to be of interest to the users of accounting information by providing them with an enhanced awareness of the effectiveness of corporate restructuring. It is also of interest to the promulgators of financial accounting standards (the SEC and the FASB) who have tried to regulate the use of unnecessary restructuring charges.

RESTRUCTURING COSTS DEFINED

What is a restructuring charge? The charge is basically a company's estimate of the future costs of some drastic change in business strategy or operations, undertaken to improve future profitability, such as laying off a large percentage of its workforce or closing an entire division. Companies took these pretax charges as a signal to investors that prior earnings were overstated and that these charges should be considered a subtraction from shareholders' equity. Write-offs were usually taken by companies with severe problems. Elliot and Shaw (1988) found that firms making the restructuring charges (1) tended to be larger and carry more debt than the rest of the industry, (2) experienced declining ROA's and ROCE's during the three years preceding the write-offs, and (3) experienced lower security returns for three years before, coincident with, and eighteen months after the write-offs. The fast pace of change in technology and world competitiveness also required discretionary write-offs to get rid of unnecessary capitalizations.

During the early 1990's, many companies reported restructuring charges that encompassed several broad categories of cost. These costs included employee termination benefits, costs associated with the elimination and reduction of product lines, costs related to the consolidation or relocation of plant facilities, new systems development or acquisition costs, and losses on asset impairments and disposal of assets. These pervasive write-offs have also been in response to a variety of changes in the global economic environment, including the NAFTA and the GATT. By reducing asset values or cutting costs, corporate management intended the resultant, streamlined companies to be well situated for future growth in the world market. Table 1 provides a sample list of companies taking restructuring charges, and the reasons for such charges.

PERCEIVED PROBLEMS WITH RESTRUCTURING CHARGES

Some investors and critics are skeptical of such restructuring charges. According to them, write-offs can boost a company's future earnings without an actual improvement in performance. Companies were able to pack all negative charges into one year, leaving the other years free of the charges that actually occurred during the year. A variation of this technique was the use of write-offs to mask an earnings slowdown. A non-recurring write-off during the current year would make reported earnings look good the following year.

Critics also protest that repeated restructurings tend to confound a company's earnings picture. One or maybe even two nonrecurring charges were understandable, because they cleared the books for future earnings gains, but repeated write-offs could neither be nonrecurring nor unusual. It is difficult for investors to believe in the face value of reported earnings--such earnings might be in danger of being canceled out by subsequent write-offs.

Another problem with restructuring charges is that the initial charges are, by definition, management's estimates, and the auditors can never be certain how close these estimates will come to the actual expenses involved in selling off assets or retrenchment. When companies took write-offs, they could later reinject some of the reserves into operating income, thus inflating profits. Also, since restructuring charges were liberally defined under APBO 30, firms were charging off costs which would benefit continuing operations, such as expenditure for equipment, costs associated with relocating and training employees, advertising and legal costs, etc. In some cases, restructuring charges even included ordinary operating items such as environmental liabilities, inventory impairments and write-offs of long obsolete, intangible assets. (WSJ, March 1996).

MARKET REACTION TO RESTRUCTURING CHARGES

According to the financial press, initially, investors responded enthusiastically to announcements of restructuring charges. Wall Street's positive response has been attributed to two reasons (Galarza and Ozanian, 1996). First of all, the charges seem to indicate that management is aggressively rectifying past mistakes, thus clearing the way for future earnings growth. Second, reductions in earnings due to these non-recurring charges do not affect the company's ability to pay its bills or dividends since they are simply accounting adjustments that generally do not involve cash outlays.

Table 1 shows that most restructuring announcements have been met with increases in raw stock returns. However, research studies by Elliot and Shaw (1988) and Brickley and Van Drunen (1990) found that restructuring charges were accompanied by negative stock returns. A recent study by Ramaswamy, Wells, and Loudder (1996) found that the market responded negatively on the day of the restructuring announcement, and positively on the day following the announcement.

MOTIVATION AND EMPIRICAL IMPLICATIONS OF THE STUDY

Amidst these conflicting signals from analysts, researchers, and the securities market, the main purpose of this paper is to focus on the reasons behind restructuring and the effectiveness of such a corporate move. Write-offs are usually taken by companies with severe problems. Studies by Elliot and Shaw (1988) and Brickley and Van Drunen (1990) show that firms facing major problems with cost control and debt management usually take restructuring charges to cut costs and streamline operations, in the expectation that future profitability will improve. Frequently, profits do look good in the year following the restructuring (see Table 6). However, if the downsizing was really effective, the company should show an increasing trend in profitability in the years following the restructuring. If the write-offs were merely temporary, short term measures to reduce immediate costs and boost stock prices, the performance of the company will not improve noticeably in future years. Dechow et. al (1994) studied the effect of restructuring charges on executive's cash compensation and found that compensation committees adjust earnings based incentive compensation for restructuring charges. This implies that executives are not penalized for taking restructuring charges, and are therefore not discouraged from taking write-offs for short term gains. Also, hasty restructurings may cause other problems like strained relations with customers (due to lack of personnel to meet their service requests), demoralized employees, and expensive use of contract workers to fill the void left by the terminated employees. The results of restructuring may not be as efficient as management's expectations.

This paper examines the financial statements of sample companies for three years before the restructurings, the year immediately after the restructuring, and four years after the restructuring. The objective of this analysis is to investigate corporate performance to observe if restructuring de facto contributed to steady long term growth, and was not a hasty decision made in response to low cost foreign competitors, and by the investors' demand for quick returns.

ANALYSIS OF CORPORATE PERFORMANCE

Most financial statement analysis examines some aspect of a firm's profitability or a firm's risk. Assessments of profitability permit the analyst to study a firm's past operating performance and to project expected returns in the future. Assessments of risk permit the analyst to judge a firm's success in coping with the various dimensions of risk in the past, in order to continue operating as a going concern in the future. Described below are some common measures used to evaluate a company's profitability and risk.

Profitability: Profitability is the ability of a firm to generate earnings. Analysis of profit is of vital concern to the stockholders, since they derive revenue in the form of dividends, which is paid from profits. Further, increased profits can cause a rise in market prices, leading to capital gains. Profits are also important to creditors, because profits are a main source of funds for debt coverage. Management has an interest in profits too it is often used as a performance measure. The following measures are important indicators of profitability.

Earnings: The Statement of Financial Accounting Concepts No. 1 (SFAC) indicates that the primary focus of financial reporting is to provide information about a company's performance with measures of earnings. Income reporting has great value as a measure of future cash flows, as a measure of management efficiency, and as a guide to the accomplishment of managerial objectives (Natarajan, 1996). Numerous studies have documented the market reaction to earnings announcements (see Lev, 1989). One of the most robust findings in financial statement research area is that the release of interim and annual earnings is associated with both increased trading volume and increased security return variability (Bamber, 1986, Bamber and Cheon, 1995). Earnings, therefore, seem to provide significant information to the users of accounting information.

Return on Assets ROA: The rate of return on assets measures a firm's success in using assets to generate earnings independent of the financing of those assets. ROA takes the particular set of environmental factors and strategic choices that a firm makes and focuses on the profitability of its operations relative to the investments (assets) in place. ROA is measured as follows:

Net income before Minority Share of Earnings and Non-recurring Items Average Total Assets

Return on Common Shareholders' Equity (ROCE): This ratio measures the return to common shareholders after subtracting from revenues, not only operating expenses, but also costs of financing debt and equity securities that are senior to common stock. Thus, ROCE incorporates the results of a firm's operating, investing and financing decisions. The analyst calculates ROCE as follows:

Net Income--Preferred Stock Dividends Average Common Shareholders' Equity

Research by Walsh (1984) and Bryne (1991) showed that profitability ratios were frequently used by investors, financial analysts, management and creditors, to investigate the performance of a firm.

Risk: A corporation faces numerous and often interrelated types of risk. Such risks could be at the international level (exchange rate fluctuations), at the domestic level (inflation, interest rate changes), at the industry level (changes in technology, competition), and at the firm specific level (lawsuits, management direction). Each of these types of risk ultimately affects net income and cash flows. Risk analysis typically examines (1) the near term ability to generate cash to service working capital needs and debt service requirements, and (2) the longer term ability to generate cash internally or from external sources to satisfy plant capacity and debt repayment needs.

Short Term Risk--the Current Ratio: The ability of an entity to maintain its short-term debt paying ability is important. If the entity cannot maintain a short term debt paying ability, its long term prognosis is apt to be discouraging. An important measure of short term risk is the current ratio, measured as Current Assets/Current liabilities. This ratio indicates the amount of cash available at the balance sheet date, plus the amount of current assets the firm expects to turn into cash within one year of the balance sheet, relative to obligations coming due during that period. Empirical studies of bond default, bankruptcy, and other conditions of financial distress have found the current ratio to have strong predictive power.

Long Term Risk--the Debt/Equity Ratio: Analysts use measures of long-term liquidity to examine a firm's ability to meet interest and principal repayments on long term debt as they come due. The debt equity ratio provides a measure of the proportion of long term debt in financing a firm's capital structure. The higher this proportion, the higher the long term solvency risk. The debt equity ratio is measured as: Long term debt/Shareholders' equity.

Growth: Investors generally use the Price/Earnings (P/E) ratio as a gauge of the future earning power of the firm. It is a measure of the market's assessment of the external economic factors, as well as of the growth prospects, financial strength, capital structure, and other risk factors associated with the enterprise. Companies with high P/E ratios generally have high growth opportunities and vice versa. The P/E ratio is measured as: Market price per share/Earnings per share.

CEO Compensation: An interesting secondary issue in this study was the relationship between CEO compensation and the value added to the firm. Dechow (1994) and Sloan (1994) show that there is a significant statistical association between top-executive cash compensation and reported earnings. If the restructuring positively added value to the firm, it follows that there should be a positive correlation between increase in CEO compensation and increase in the value of the firm. The value of the firm, in this study was measured as the value of $ 100 invested in the company three years earlier. The increase in investment is measured as the result of share price appreciation and dividends, both of which are related to the earnings of a firm.

On the basis of the above discussion, the paper addresses the following questions to explore corporate performance after a restructuring:

 Have earnings (an overall measure of performance) shown a steady
 growth trend during the years after the restructuring, as compared
 to their performance before restructuring?

 Have the restructuring companies improved their profitability (as
 measured by ROA and ROCE) during the years after the
 restructurings?

 Have the companies reduced their risk after the restructuring?

 How is the growth potential of the restructuring companies (as
 measured by the P/E ratio)

 How has CEO compensation changed as compared to the value of the
 firm during the years following the restructuring?

STATISTICAL ANALYSES USED

To examine the trend of corporate performance in the years surrounding the restructuring, the following types of statistical analyses were used:

T-tests: Simple statistical t-tests were used to compare the profitability, risk and growth measures for three periods: Three years before the restructuring and the year before the restructuring; the years before and after the restructuring; and the year after restructuring and four years after the restructuring. These t-tests were used to determine whether the profitability, risk and growth measures declined or increased during the three periods under study. These three groups of measures were also compared to the industry averages for the three periods under study to ascertain the financial position of the restructuring companies as compared to general industry performance.

Discriminant Analysis: A multiple discriminant analysis and a paired case-control methodology was used to assess the differences in the financial characteristics of restructuring and non-restructuring firms as a further test of economic performance. The six factors listed above (earnings, ROA, ROCE, current ratio, debt/equity ratio, and P/E ratio) were used as the distinguishing variables in the discriminant analysis. The discriminant analysis was performed for each of the three periods under study.

Analysis of Variance (ANOVA): An Analysis of Variance was used to compare the CEO compensation and value of the firm during the restructuring year, and the changes in the two variables after four years for the sample group and the control group selected for the discriminant analysis.

A list of companies engaging in restructuring activities were obtained from the Compact Disclosure archives for the year, 1991. The key search terms used were "restructuring charges" and/or "write-offs" if found in the footnotes to the financial information. This search yielded a list of 787 firms. Out of this list, firms were deleted because:

 Restructuring was being done over a period of years. This study was
 interested in the performance of firms before and after a specific
 restructuring year. Multi-year restructuring would not provide a
 clear picture of firm performance prior to or after one year's
 restructuring charge. 153 firms were deleted as a result of
 multi-year restructuring.

The term "restructuring" was used in the footnotes if the firm had taken a charge during the current year, or during any year presented in the financial statements. The sample required firms which had restructured in 1991. Consequently, firms which had restructured during the prior years were deleted from the list. 166 firms were deleted because of prior year restructuring. In some cases, plans of restructuring were discussed in the footnotes, but no actual charges were taken. Such firms were also deleted (47 firms).

The purpose of this study was to examine financial performance to ascertain that restructuring helped companies achieve their objective of cost control and streamlining operations. For that reason, the main criterion of selection was the objective of cost control/increase in efficiency cited as a reason for restructuring. Firms which restructured for reasons other than cost control were deleted from the sample (95 firms). The resultant sample yielded 326 firms (see Table 2).

Earnings and other ratios (ROA, ROCE, current ratio, debt/equity ratio and P/E ratio) were obtained from the Compact Disclosures for the years, 1988 (three years before restructuring), 1990 (the year before restructuring), 1992 (the year after restructuring) and 1995 (four years after the restructuring). Information about CEO salaries and the value of $ 100 invested was obtained from the Businessweek and Fortune magazines.

For the discriminant analysis, each of the 326 restructuring firms was matched by industry with a control firm (non-restructuring firm). The control firms were also selected by size (as measured by total assets) to match the restructuring firms as closely as possible. The resultant sample had two groups of 326 firms: the test group or the restructuring group and the control group or the non-restructuring group.

DESCRIPTIVE ANALYSIS OF THE SAMPLE

The sample contained 279 firms spread over 10 industries (classified according to their two-digit SIC code), and 47 firms spread over 7 industries, for a total of 326 firms each in the test group and the control group. The maximum number of firms taking restructuring charges were in the Electronic and Other Electrical Equipment industry. This industry contained major computer firms like IBM and Apple--a number of companies in this industry took restructuring charges to keep up with international competition and rapid changes in technology. Banking services came next, with 44 companies. The early 1990's saw major changes in the banking industry after the collapse of the Savings and Loan industry. A number of major mergers and restructurings took place at this time. Communications (with firms like AT&T and Ameritech) came next, once again, in response to rapidly changing technology. (See Table 3)

Companies in the Banking Services industry were the largest, with mean revenues of $ 56 million and assets of more than $ 4 billion. (See Table 4) The Communications industry was a close second, with revenues of $ 35 million and assets of more than $ 1 billion. In both these cases, the differences between mean and median revenues was large, because they had one or two companies with extremely high revenues (for example, AT&T in the Communications industry). The mean/median revenues in the other industries ranged from $ 1 million to $ 10 million, with mean/median assets ranging from $0.2 million to $ 3 million.

The average restructuring charge taken by the companies in the sample was $ 967 million and the median charge was $ 118 million. Once again, there was a great disparity between the mean and the median because of a few companies that had taken billions of dollars in restructuring charges. For example, the Communications industry had taken an average restructuring charge of $ 22 billion, with a median charge of $ 3 billion. The next highest was Measuring and Instruments, where at least one company took more than $ 1 billion in restructuring charges.

Out of 326 firms in the sample, 150 (46%) had taken restructuring charges only once during the past five years. An equal number of firms had taken charges twice during the past five years, while 26 firms (8%) had taken write-offs more than twice during that period. In the Communications industry, Banking, and Industrial Machinery industries, more than 70% of the firms had taken restructuring charges twice during the past five years. In fact, 24% of the firms in the Communications industry had taken restructuring charges more than twice during the past five years (See Table 5).

RESULTS OF THE ANALYSIS--WITH ONE-TAILED T-TESTS

The financial statements of the sample firms were investigated, based on the questions posed earlier. The variables of interest were Earnings, ROA, ROCE, Debt/Equity ratio, Current ratio, and the P/E ratio. A one-tailed t-test was used to test the following hypotheses:

[H.sub.1]: Between 1988 and 1990, the financial performance of the sample group deteriorated significantly.

[H.sub.2]: Between 1990 and 1992 (a year after the restructuring) the financial performance of the sample group showed signs of improvement.

[H.sub.3]: Between 1992 and 1995, the financial performance of the sample group improved significantly.

The t-tests were used to examine the performance of the sample group, based on the earnings and other ratios specified above.

ANALYSIS OF EARNINGS

A study of earnings behavior of restructuring firms during the period, 1988-1990 (the years leading to the restructuring) showed that earnings declined for more than 90% of the sample. (See Table 6, Panel A). An industry wise analysis showed that the Communications and Banking Services industry were the worst hit, with more than 95% of firms reporting earnings decline. A small percentage of restructuring firms (less than 10%) did show increasing earnings, but the earnings increases were very small.

A study of average earnings during 1991 (the restructuring year) showed that the average earnings for all the sample firms were less than the industry average (The mean earnings of all firms with a 2 digit SIC code was taken to be the industry average.) (Table 6, Panel B). In some cases, the restructuring firms' average earnings were negative (most of the companies within that industry showed losses for the restructuring year); e.g., restructuring firms in Apparel (SIC No. 23) and Primary Metals (SIC No. 33) showed average negative earnings, while the industry average was positive. Some firms showed wide divergence from their industry averages, especially in the Communications industry (SIC No. 48).

During 1992 (the year after the restructuring), 93% of firms showed earnings increases (Table 6, Panel B). In some industries, all the firms in the sample showed earnings increases (for example, Industrial Machinery, SIC No. 35). In most cases, more than 90% of the firms within each industry showed earnings increases, except Food and Kindred Manufacturing (SIC No. 20) where 86% of the firms in the sample showed earnings increases. As the result of the earnings increases, the average earnings of the sample firms increased in 1992. Within each industry studied in this paper, the sample firms showed increased earnings (See Table 6, Panel A). In the case of Apparel and Primary Materials (SIC No. 23 and 33), the average earnings went from a negative balance in 1991 to a positive mean in 1992. The industry average also rose--but in numerous cases, the average earnings of the restructuring firms were very close to the industry average itself--for example, Measuring and Analyzing Instruments (SIC No. 38) and Banking Services (SIC No. 67).

Were the restructuring firms able to sustain such growth? Table 6, Panel B shows that 69% of the firms showed declining earnings during the period, 1992-1995. Only 31% of the firms were able to show increasing earnings. In the Electrical Equipment, Measuring and Analyzing Instruments, and Primary Metals industries (SIC No. 36, 38, and 33) more than 70% of the firms showed declining earnings. Table 6, Panel A also shows that the average earnings of the sample firms declined between 1992 to 1995, in spite of increases in the industry averages.

The above results indicate that the sample firms undertook restructuring in response to declining earnings. These results are consistent with prior research. These firms seem to have shown improved earnings during the following year, but a majority of the firms in the sample were unable to keep up with earnings growth in the years following the restructuring. This suggests that the management of the firms may have resorted to accounting manipulations to increase their earnings in the period following the restructuring to justify the charges against earnings (Ramaswamy and Upneja, 1997). Some of the firms continued to reorganize--about 8% of the sample firms had taken restructuring charges more than twice during the period, 1992-1995 (Table 5).

The firm performance for 1988 and 1990 was compared using a one-tailed t-test, the hypothesis being that performance declined during the years leading up to the restructuring. Results of the t-test are in Table 7.

Between 1990 and 1992, the one-tailed t-test was positive at the 0.1 level of significance. Earnings increased during this period, but the change was not very significant. Looking at industry averages, the earnings average of the sample group was still less than the industry average, but the level of significance was only 0.1 as compared to 0.001 during 1988-1990. The sample group's earnings therefore showed an improvement during this period, though it was not good enough as compared to the industry average. During 1992-1995, earnings declined substantially--the t-test showed a level of significance of 0.05. This was in direct contradiction to the hypothesis posed--the hypothesis stated that firm performance would improve in the years following the restructuring.

Analysis of Profitability: Table 8 provides an analysis of the two ratios that are used to measure the profitability of a company. About 90% of all firms in the sample showed declining ROA's for the period leading up to the restructuring, that is, 1988-1990. (Panel A) The Electronic and Other Electrical Equipment (SIC 36) showed the highest number of firms with declining ROA's (94%). Also, 97% of the firms in the sample had ROA's less than the industry average in 1990. In some industries, for example, Banking Services (SIC No. 67) 100% of the firms in the sample had ratios lower than the industry average.

Performance improved slightly during the years, 1992-1995. Only 69% of the firms showed declining ROA's while 72% of the firms still had their ratios below industry average. Electronic and Other Equipment (SIC No. 36) and Banking Services showed good improvement, while Communications (SIC No. 48) and Chemicals (No. 28) showed limited improvement.

ROCE's followed a similar pattern as the ROA's, with 90% of the sample firms showing declining ratios in 1991 and 97% of the firms had ratios lower than the industry averages. During 1992-1995, 66% of the firms shoed declining ROCE's , a slightly better performance than ROA's (69%), while 69% of the firms still had ROCE's under the industry average (once again, slightly better than the ROA with 72%). The slight improvement in ROCE could be because of stock buybacks--the 1990's saw tremendous surge in treasury stock purchases.

Looking at the t-tests in Table 7, a similar pattern can be discerned. The period 1988 to 1990 showed a definite decline in ROA and ROCE, with a significance of .001. The profitability ratios were also well below industry average (significant at 0.001 level). During the period, 1990-1992, the ROA and ROCE showed a slight improvement--a positive t-value, with significance at the 0.10 level. During the same period, the ROA and ROCE of the sample group were still below industry average, but now the difference was significant only at the 0.10 level. From 1992-1995, there was a definite decline in the sample group (not the positive direction expected by the hypothesis posed earlier). Both ROA and ROCE declined significantly (level of significance, 0.05) during this period. The ratios of the sample mean were once again lower than the industry average--the t-test showed a significance of 0.05.

Risk Analysis: The firms in the sample performed reasonably well as far as short term risk (as measured by the current ratio) was concerned. As can be seen from Table 9, Panel A about 55% of the firms in the sample showed declining current ratios during the period, 1988-1990. About 57% of the firms had current ratios lower than the industry averages during that period. Measuring and Analyzing Instruments (SIC 38) had only 29% of the firms with declining current ratios. Electronic and Electrical Equipment (SIC 36) and Communications (SIC 48) had the highest number of firms with declining current ratios--more than 65%.

Short term risk showed no noticeable improvement in the years following the restructuring. In fact, during the period, 1992-1995, 57% of the firms showed declining current ratios, as compared to 55% of the firms in the earlier period. 57% of the firms still had their ratios below industry average. In some industries, the ratios did improve--for example, Chemicals and Allied Products (SIC 28) and Industrial Machinery (SIC 35). However, in most industries, short term risk performance remained the same.

Looking at long term risk, as measured by the debt equity ratio, Table 9 Panel B shows that nearly 93% of the firms in the sample had declining ratios during the period, 1988-1990. The decline was essentially uniform through all the industries. All industries had more than 90% of the restructuring firms exhibiting declining debt/equity ratios.

The change in debt/equity ratios after the restructuring was dramatic. Most companies seemed to have used restructuring to pay off their long term debt and improve their long term risk profile. As can be seen from Table 9, Panel B, 52% of the firms showed declining ratios during the period 1992-1995, as compared to 97% during the earlier period. And only 51% of the firms had ratios under the industry average. The improvement was uniformly spread across industries, even the most troubled industries like Electronics (SIC 36) and Communications (SIC 48). As interesting feature noted here was that in the few cases where debt equity ratio had worsened, profit performance improved considerably--there was a negative correlation between debt/equity ratio and profit performance. This seems to imply that companies leveraged themselves to improve profitability.

The current ratio did not provide significant results in the t-test. During 1988-1990, the current ratio did decline, but the level of significance was only 0.15. the current ratios of the sample group was not significantly lower than the industry average. During the next two periods, current ratio declined slightly, but the differences were not significant. The ratios of the sample group did not differ materially from the industry average.

The debt/equity ratio showed a decline between 1988 and 1990 (significance at 0.10 level). The ratio of the sample group was also lower than the industry average during this period (significance, 0.10). However, during the period, 1990--1992, the debt equity ratio showed a marginally significant increase, and this increase was carried on to the period, 1992 1995. During these two periods, the difference between the sample group and the industry average was insignificant.

Price/Earnings Ratio: A study of the P/E ratio showed no significant results. There was no major change in the pattern of P/E ratios of the sample group. About 56% of the firms had P/E ratios less than the industry average during the period, 1988-1990. While earnings declined during the next two time periods, the P/E ratio actually increased for the sample firms, indicating that the market still perceived growth potential in the restructuring firms.

MULTIPLE DISCRIMINANT ANALYSIS (MDA)

To further study the performance of restructuring firms, multiple discriminant analysis (MDA) was used to study the sample, comparing it with a matched sample of control firms. The restructuring firms and the control firms were matched on industry and firm size. MDA is well-suited to many problems where the dependent variable is non-metric. The primary objective of MDA is to classify entries correctly into mutually exclusive groups by maximizing the ratio of inter-group to intra-group variance-covariance from a set of independent variables. In addition, MDA reveals which individual variables have the greatest discriminating power within a multivariate context.

Table 10 summarizes the standardized discriminant function coefficients, and the canonical discriminant function evaluated at the group means. The discriminant analysis was performed for each of the three periods under study. For the years prior to the restructuring, earnings was the most significant factor in the MDA--the earnings of the restructuring firms was materially below the control group. All the other ratios were also significant--the performance of the sample group was clearly inferior to the control group. The results of the MDA for 1988-1990 was significant at the 0.001 level. For the year following the restructuring (1992), the results were still significant, but barely at the 0 .05 level this time. The performance of the sample firms was below par when compared to the control group, but the difference was less obvious as compared to the last period. The price earnings ratio was no longer a significant factor--the average price/earnings ratio for the sample group increased faster than the control group. The final analysis for 1995 showed significance at the 0.01 level. The sample group, once again, performed at a lower level than the control group. Earnings was the most significant factor in this analysis. Debt/equity ratio was no longer significant, while price/earnings ratio showed a definite increase for the sample firm, changing the direction of significance.

CEO PAY AND INVESTMENT VALUE

Table 11 compares average CEO compensation and the value of $ 100 invested in the company three years earlier for the test group and the sample group. The data for this analysis was obtained from Fortune and BusinessWeek. Due to lack of available data, the sample size in both the test and control groups were reduced to 96 each.

As can be seen from Table 11, in 1991, the average compensation for CEO's of restructuring firms was $ 1,358 million. For similar firms without restructuring charges, the average CEO compensation was $ 1,366 million, a difference of 6%. It should be remembered that management compensation for restructuring firms placed less weight on restructuring charges, thus shielding top management from high restructuring charges (DeChow, et. al, 1994). The value of $ 100 invested, referred to hereinafter as IV was $ 155 in 1991 for restructuring firms; for the control group, the average was $ 177, a difference of 14%.

During 1995, average CEO compensation for restructuring companies was $ 2,018 million--an increase of 48% from 1991. The average compensation for the control group was $ 2,498 million, an increase of 82%. The disparity between the two groups was 23% in 1995, as compared to 6% in 1991. If management compensation is an indicator of corporate performance, obviously the restructuring firms are not doing as well as similar firms in the industry.

This can also be seen by comparing IV in 1995. For the sample group, IV declined from $ 155 in 1991 to $ 125 in 1995, a decline of 19%. The control group showed an average IV of $ 175--decline of 1% from 1991. The disparity between the two groups also increased--from 14% in 1991 to 40% in 1995. This seems to indicate that restructuring corporations have been performing poorly over the past few years.

An Analysis of Variance performed on the average CEO pay and average IV for the sample group and the control group was significant at the 0.01 level. Pairwise comparisons showed that CEO pay was considerably different for the sample group between 1991 and 1995 (significance, 0.05). There was also considerable difference in CEO pay between the sample group and the control group. IV showed significant differences for the sample group between 1991 and 1995 (significance, 0.05). The control group did not show an extensive difference in IV during the two periods (significance, 0.20). However, IV of the sample group and the IV of the control group showed marginally significant difference in 1991 (at 0.10) and notably increased significance for 1995 (0.01 level). The results of the ANOVA also show that IV decreased considerably for the restructuring firms, while CEO compensation (which is usually tied to performance) increased during this period.

SUMMARY AND CONCLUSIONS

Numerous corporations have taken restructuring charges during the past decade, ostensibly to reduce costs, streamline operations, and therefore to improve profitability. This paper analyzes the financial performance of a sample of restructuring firms to observe the effectiveness of such a strategy. Diverse techniques such as t-tests, discriminant analysis and ANOVA were used to examine the performance of restructuring firms as compared to a control group of nonrestructuring firms.

The results indicate that restructuring did not contribute extensively to improved financial performance. In fact, earnings and profitability of restructuring firms have shown a declining trend in the years following restructuring for a majority of companies. As a result, companies are now modifying their strategy to "growth" rather than "downsizing" as a approach to increased profitability.

REFERENCES

Accounting Principles Board Opinion No. 30. 1973. Reporting the results of operations Reporting the effects of disposal of a segment of business, and extraordinary, unusual and infrequently occurring events and transactions. New York: American Institute of Certified Public Accountants, Inc.

Bamber, L. 1986. The information content of annual earnings releases: A trading volume approach. Journal of Accounting Research, 24 (Spring): 40-56.

--and Y.S. Cheon. 1995. Differential price and volume reactions to accounting earnings announcements. The Accounting Review, 70 (July): 417-441.

Brickley, J.A., and L.D. Van Drunen. 1990. Internal corporate restructurings: An empirical analysis. Journal of Accounting and Economics, 2(January): 251-280.

Dechow,P, M. Huson, and R.Sloan. 1994. The effect of restructuring charges on executives' cash compensation. The Accounting Review (January): 138-156

Elliot, J.A. and W. Shaw. 1988. Write-offs as accounting procedures to manage perceptions. Journal of Accounting Research (Supplement): 91-119.

Galarza, P. and M.K.Ozanian. 1996. Forgive nothing: Here's how to deal with all those huge "nonrecurring" charges. Financial World (March): 18-20.

Natarajan, R. 1996. Stewardship value of earnings components: Additional evidence of the determinants of executive compensation. The Accounting Review (January): 1-22.

Ramaswamy, V., D.Wells, and M. Loudder. 1996. Market reaction to restructuring Announcements--a re-examination. Presented at the Academy of Accounting and Finance Conference, New Orleans.

Sloan, R. 1993. Accounting earnings and top executive compensation. Journal of Accounting and Economics (January/April/July): 55-100.

Vinita M. Ramaswamy, University of St. Thomas

Ramon Fernandez, University of St. Thomas

TABLE 1
EXAMPLES OF COMPANIES TAKING RESTRUCTURING CHARGES

NAME AMOUNT OF CHARGE REASONS FOR CHARGE

Unisys $ 225 million Employee layoffs
1. Westinghouse $ 906 million Sales of business
 Electric
Ameritech $ 335 million Write-down of equipment
Kerr McGee $ 5 million Severance
1. Eli Lilly $ 1.2 million Streamline operations
2. Nynex $ 1.3 billion Severance
3. Borden $ 642 million Selling off snack business
K Mart $ 535 million Store closings

NAME STOCK PRICE
 MOVEMENT

Unisys Up 12.5 cents
1. Westinghouse Down 12.5 cents
 Electric
Ameritech Down 37.5 cents
Kerr McGee Down 12.5 cents
1. Eli Lilly Up 12.5 cents
2. Nynex Up 75 cents
3. Borden Down $ 2.375
K Mart Down 50 cents

TABLE 2: SAMPLE OF FIRMS TAKING RESTRUCTURING CHARGES

DESCRIPTION NUMBER OF FIRMS

Firms initially selected (with the term
 "restructuring" in the footnotes) 787
Firms deleted due to:
 Restructuring being done over a period of years 153
 Restructuring done within the past three years 166
 Plans being announced but no charges taken 47
 Cost control or efficiency not stated as the
 main reason for the restructuring 95
Total firms selected in the sample 326

TABLE 3
SAMPLE FIRMS BY INDUSTRY

SIC Code Industry name Number
(2 digit) of firms

20 Food and Kindred Manufacturing 15
23 Apparel and Other Finished Products 12
28 Chemicals and Allied Products 32
33 Primary Metals 21
35 Industrial Machinery 29
36 Electronic and Other Electrical Equipment 52
38 Measuring and Analyzing Instruments 17
48 Communications 34
67 Banking Services 44
73 Computer Services 23
Others Miscellaneous 47
 Total 326

TABLE 4
DESCRIPTIVE CHARACTERISTICS OF RESTRUCTURING FIRMS
(figures in 000's)

SIC CODE REVENUES SENIOR CAPITAL

 MEAN MEDIAN MEAN MEDIAN

20 9,097 6,365 1,477 1,203
23 1,021 788 187 206
28 3,997 8,272 2,727 1,405
33 4,261 4,819 568 682
35 8,302 10,291 3,982 1,010
36 3,238 2,380 462 263
38 7,699 6,057 673 832
48 34,035 13,307 1,569 * 237 *
67 56,355 48,427 3,086 * 1,121 *
73 1,027 956 283 731
Others 4,952 3,639 2,995 3,179

SIC CODE TOTAL ASSETS

 MEAN MEDIAN

20 1,875 1,985
23 268 325
28 3,572 2,895
33 1,279 1,367
35 5,635 2,983
36 738 432
38 1,010 1,167
48 1,789 * 836 *
67 4,036 * 2,871 *
73 681 1,039
Others 4,295 3,327

* figures in billions

TABLE 5: Pattern of Restructuring Charges Taken

Sic Average Charge
Code (In Millions of $)

 Mean Median

All 967 118
20 27.1 21.1
23 15.3 7.9
28 44.2 29.5
33 86 10.7
35 321 120
36 115 149.3
38 786 347
48 22 * 3.57 *
Others 44.2 35.7

Sic Write-offs Taken Writeoffs Taken
Code Once in past 5 Yrs Twice in past 5 Yrs

 No. Of Firms No. Of Firms

All 150 (46%) 150 (46%)
20 11 (73%) 3 (25%)
23 10 (83%) 2 (17%)
28 17 (53%) 14 (43%)
33 7 (33%) 11 (52%)
35 5 (17%) 21 (72%)
36 35 (67%) 14 (26%)
38 10 (58%) 7 (42%)
48 2 (6%) 24 (70%)
Others 35 (75%) 9 (19%)

Sic Writeoffs
Code More than Twice

 No. Of Firms

All 26 (8%)
20 1 (2%)
23 0
28 1 (3%)
33 3 (15%)
35 3 (11%)
36 3 (7%)
38 0
48 8 (24%)
Others 3 (6%)

* figures in billions

TABLE 6 - PANEL A
EARNINGS PERFORMANCE OF RESTRUCTURING FIRMS

SIC During 1988 - 1991,
CODE % of firms whose earnings

 Declined
 Increased

All 93 7
20 93 7
23 93 7
28 90 10
33 90 10
35 93 7
36 94 6
38 94 6
48 96 4
67 95 5
73 91 9
Others 93 7

SIC During 1991 -1992
CODE % of firms whose earnings

 Increased Declined

All 7 93
20 14 86
23 0 100
28 9 91
33 10 90
35 0 100
36 4 96
38 3 97
48 5 95
67 6 94
73 3 97
Others 3 97

SIC During 1992 - 1995
CODE % of firms whose earnings

 Increased Declined

All 69 31
20 59 41
23 58 42
28 81 19
33 71 29
35 65 35
36 75 25
38 70 30
48 62 38
67 63 37
73 60 40
Others 53 47

TABLE 6 - PANEL B
EARNINGS PERFORMANCE OF RESTRUCTURING FIRMS

SIC Average Industry Avg. Average
 I ndustry Avg.
Code Earnings Earnings Earnings
 (1991) * (1991) (1992) **
 (sample firms) (sample firms)

All 5321.5 6781.5 6236.8
20 351.6 539.5 521.5
23 NM 550.8 221.2
28 311.5 427.6 401.8
33 NM 1221.3 826.4
35 434.9 636.4 651.2
36 691.4 821.5 762.5
38 1125.9 1420.9 1391.2
48 9097.2 13058.2 11986.2
67 21087.6 22331.2 23082.5
73 421.5 546.2 499.2
Others 2128.1 2322.2 2319.2

SIC Industry Avg. Average

Code Earnings Earnings Earnings
 (1992) (1995) *** 1995
 (sample firms)

All 6619.5 5678.9 6521.8
20 576.2 321.5 589.4
23 599.2 199.3 607.9
28 456.2 345.2 525.9
33 1026.5 629.3 1321.2
35 725.5 399.5 742.5
36 835.2 534.9 899.6
38 1455.9 1196.3 1489.2
48 13051.2 10891.6 13782.1
67 23149.8 20597.4 23854.7
73 577.6 435.0 584.3
Others 2592.6 2198.7 2651.3

* The restructuring year, 1991
** One year after restructuring, 1992
*** Four NM: Not Meaningful, due to numerous negatives in the data
for averaging.

TABLE 7
Results of the T-test Comparative Firm Performance For The
Three Periods under Study

Description 1988 vs. 1990 1990 vs. 1992
 t-values P>t t-values P>t

Earnings (5.69) 0.00 1.06 0.15
ROA (4.31) 0.00 1.12 0.15
ROCE (4.99) 0.00 1.21 0.15
Debt/Equity (1.31) 0.10 1.78 0.05
Current ratio (1.11) 0.15 (0.72) 0.20
P/E ratio (0.51) 0.25 0.42 0.25

Description 1992 vs. 1995
 t-values P>t

Earnings (1.70) 0.05
ROA (1.83) 0.05
ROCE (1.67) 0.05
Debt/Equity 1.59 0.10
Current ratio (0.11) 0.30
P/E ratio 0.61 0.20

TABLE 8 - PANEL A
ANALYSIS OF PROFITABILITY - ROA

SIC Firms with declining ROA less than
Code ROA in 1988 vs. 1991 industry average

All 296 (90%) 318 (97%)
20 13 (86%) 15 (100%)
23 11 (97%) 12 (100%)
28 30 (93%) 30 (93%)
33 19 (90%) 20 (94%)
35 26 (89%) 29 (100%)
36 49 (94%) 50 (97%)
38 15 (88%) 15 (88%)
48 31 (87%) 33 (96%)
67 39 (88%) 44 (100%)
73 22 (95%) 23 (100%)
Others 43 (91%) 45 (96%)

SIC Firms with declining ROA less than
Code ROA in 1992 vs. 1995 industry average

All 225 (69%) 233 (72%)
20 9 (60%) 10 (67%)
23 7 (58%) 9 (75%)
28 26 (81%) 23 (90%)
33 15 (71%) 17 (81%)
35 19 (65%) 23 (79%)
36 39 (75%) 42 (80%)
38 12 (70%) 11 (67%)
48 27 (80%) 25 (73%)
67 27 (63%) 30 (68%)
73 14 (63%) 14 (63%)
Others 25 (53%) 29 (62%)

TABLE 9 - PANEL B

 ANALYSIS OF PROFITABILITY--ROCE

 Firms with Firms with
 declining ROCE declining ROCE
 ROCE in less than ROCE in less than
SIC 1989 vs. industry 1992 vs. industry
Code 1991 average 1995 average

All 296 (90%) 318 (97%) 220 (66%) 229 (69%)
20 13 (86%) 15 (100%) 9 (60%) 10 (67%)
23 12 (100%) 12 (100%) 7 (58%) 9 (75%)
28 30 (93%) 30 (93%) 25 (78%) 21 (86%)
33 19 (90%) 20 (94%) 15 (71%) 17 (81%)
35 26 (89%) 29 (100%) 18 (61%) 23 (79%)
36 49 (94%) 50 (97%) 38 (70%) 41 (78%)
38 15 (88%) 15 (88%) 12 (70%) 11 (67%)
48 31 (87%) 33 (96%) 25 (74%) 24 (70%)
67 39 (88%) 44 (100%) 27 (63%) 30 (68%)
73 22 (95%) 23 (100%) 14 (63%) 14 (63%)
Others 43 (91%) 45 (96%) 25 (53%) 29 (62%)

TABLE 9 - PANEL A

 ANALYSIS OF RISK--CURRENT RATIO

 Firms with Current Firms with Current
 declining ratio declining ratio
 Current less Current less
 ratio in than ratio in than
SIC 1989 vs. industry 1992 vs. industry
Code 1,991 avg. 1,995 avg.

All 182 (55%) 187 (57%) 180 (57%) 184 (57%)
20 6 (40%) 7 (43%) 9 (60%) 9 (60%)
23 5 (41%) 5 (41%) 5 (41%) 5 (41%)
28 19 (59%) 21 (67%) 17 (53%) 18 (57%)
33 11 (52%) 11 (52%) 8 (38%) 9 (41%)
35 16 (55%) 16 (55%) 14 (48%) 14 (48%)
36 35 (67%) 35 (67%) 37 (71%) 38 (73%)
38 5 (29%) 5 (29%) 4 (23%) 4 (23%)
48 22 (64%) 24 (70%) 27 (79%) 28 (82%)
67 29 (65%) 29 (65%) 30 (69%) 30 (69%)
73 9 (39%) 9 (39%) 8 (35%) 8 (35%)
Others 25 (53%) 25 (53%) 21 (45%) 21 (45%)

TABLE 9 - PANEL B

 ANALYSIS OF RISK--DEBT/EQUITY RATIO (D/E)

 Firms with D/E less Firms with D/E less
 declining than declining than
SIC D/E in 1989 industry D/E in 1992 industry
Code vs. 1991 avg. vs. 1995 avg.

All 303 (93%) 305 (95%) 170 (52%) 169 (51%)
20 14 (93%) 14 (93%) 5 (38%) 5 (38%)
23 11 (93%) 11 (93%) 5 (41%) 5 (41%)
28 29 (90%) 29 (90%) 16 (50%) 16 (50%)
33 19 (91%) 19 (91%) 13 (58%) 11 (52%)
35 27 (93%) 27 (93%) 17 (57%) 17 (57%)
36 49 (94%) 50 (97%) 33 (63%) 33 (63%)
38 16 (94%) 16 (94%) 5 (29%) 5 (29%)
48 32 (94%) 33 (97%) 19 (55%) 20 (58%)
67 42 (95%) 42 (95%) 25 (56%) 25 (56%)
73 21 (91%) 21 (91%) 11 (47%) 11 (47%)
Others 43 (93%) 43 (91%) 21 (44%) 21 (44%)

TABLE 10: Results of the Discriminant Analysis

 Standard Canonical Discriminant Function Coefficient for

Variable 1988 vs. 1990 1990 vs. 1992 1992 vs. 1995

Earnings (0.592) (0.472) (0.561)
ROA (0.337) (0.246) (0.298)
ROCE (0.362) (0.271) (0.312)
Current ratio (0.113) (0.102) (0.121)
Debt/Equity (0.233) (0.216) (0.169)
P/E ratio (0.121) 0.105 0.192
Canonical Correlation 0.6265 0.4972 0.5779
Chi-square 27.82 15.73 18.97
p level 0.001 0.05 0.01

TABLE 11: CEO Pay and Value of Money Invested ('000)

Description 1991 1995 % Difference

Average CEO compensation 1358 2018 48%
 for restructuring companies
Average CEO compensation 1366 2498 82%
 for similar companies
% Difference 6% 23%
Value of $ 100 invested 3 years 155 125 -19%
 earlier in restructuring companies
Value of $ 100 invested 3 years 177 175 -1%
 earlier in similar companies
% Difference 14% 40%