The effects of the tax reform act of 1986 on business failure momentum.

Choudhury, Askar H. ; Campbell, Steven V.

ABSTRACT

The Tax Reform Act of 1986 discouraged private workout arrangements in favor of corporate bankruptcy reorganization. We hypothesize by channeling failing firms into the more protracted Chapter 11 procedure, the Tax Reform Act of 1986 slowed the "domino effect" and reduced business failure momentum. We divide a sample of 228 continuous monthly observations of large and small business failures into pre- and post-event periods. For each period, we employ maximum likelihood estimation and regress the number of large and small business failures on business failure momentum. We find the Tax Reform Act of 1986 is associated with a significant reduction in business failure momentum for both large and small firms. Our results suggest private workout arrangements impose higher social costs than corporate bankruptcy reorganizations.

INTRODUCTION

In the early and mid 1980's many failing firms sought to avoid Chapter 11 bankruptcy reorganization by privately resolving conflicts among creditors and stockholders. For the period 1980-1986, 91 of the 192 defaulting New York Stock Exchange and American Stock Exchange companies were reorganized privately (Jensen 1999, p.20). In the late 1980's the trend toward private workout arrangements ended abruptly as changes in the Tax Code sought to curb "speculative excesses" in the highly leveraged transactions market. One tax law in particular, The Tax Reform Act of 1986, effectively discouraged private workout arrangements in favor of the Chapter 11 bankruptcy reorganization procedure. Several commentators have criticized such legal barriers for frustrating the normal market adjustment process, while others have argued private workouts should be discouraged due to the negative externalities they produce. The negative externalities of business failure has been describes as a "domino effect" in which the failure of one firm leads to the failure of another firm, and so on, until the memory of the original failure eventually fades (Campbell and Choudhury, 2002).

This paper investigates whether, by channeling failing firms away from private workouts and into bankruptcy reorganization, the Tax Reform Act of 1986 mitigated the negative externalities of business failure. We measure business failure momentum before and after the implementation of the Tax Reform Act using a time-series of 228 continuous monthly observations of the number of large and small business failures. We control for the number of new business incorporations and use maximum likelihood estimation to avoid problems with autocorrelation. With the pre-event period providing a benchmark, we find the Tax Reform Act of 1986 is associated with a significant reduction in business failure momentum for both large and small firms. These results suggest the Chapter 11 bankruptcy reorganization procedure reduces the social cost of business failure by providing an orderly and transparent process of contractual disengagement.

Section two reviews the related literature. Section three describes the research design. Section four presents the results and section five contains some concluding remarks.

LITERATURE REVIEW

One of the more enduring issues in the business failure literature concerns the efficiency of corporate bankruptcy. Many scholars believe bankruptcy, particularly bankruptcy reorganization, is inefficient and should be eliminated in favor of an auction process (e.g. Roe, 1983; Baird, 1986; Jackson, 1986; Wruck, 1990; Bradley and Rosenzweig, 1992). White (1989) concludes, "The U.S. bankruptcy system, rather than helping the economy move toward long-run efficiency, in fact appears to delay the movement of resources to higher value uses" [p.130]. The primary criticisms of the Chapter 11 procedure involve the high costs and time delays imposed on bankrupt firms (Bradley and Rosenzweig, 1992). For large industrial firms, Weiss (1990) found direct Chapter 11 administrative costs averaged 2.8 percent of total asset book value at the fiscal year-end prior to bankruptcy and the average time spent in Chapter 11 was 2.5 years. For small firms, the time spent in Chapter 11 is shorter but the direct bankruptcy costs are proportionally much higher. Campbell (1997) found closely held firms averaged 1.3 years in Chapter 11 and direct bankruptcy reorganization costs averaged 8.5 percent of total asset book value at the start of the bankruptcy proceeding. The available evidence suggests the direct costs of private workout arrangements are about 10 percent of those incurred in Chapter 11 proceedings of comparable size (Gilson et al., 1990).

In addition to the direct costs, bankruptcy reorganization also involves substantial indirect costs. Indirect costs include lost sales, lost profits, the inability to obtain credit from suppliers, and lost investment opportunities (Titman, 1984). The time delays inherent in the Chapter 11 procedure produce higher indirect costs; however, private workouts usually take only a few months to negotiate and cost much less in terms of both direct and indirect costs (Jensen, 1999). Private workouts can be viewed as a natural market response to the inefficiency of corporate bankruptcy. "Such innovation is to be expected when there are such large efficiency gains to be realized from new reorganization and recontracting procedures [Jensen 1999, p.21]." Evidence from market studies suggests private workout agreements enhance firm value. Gilson, John and Lang (1990) provide statistical evidence consistent with stockholders being systematically better off if their firm's debt is restructured privately. Belker, Franks and Torous (1999) find once the result of a workout attempt is known, the returns to shareholders are greater for firms which successfully complete a private workout arrangement.

Although many bankruptcy scholars have criticized the Chapter 11 procedure for the high costs and time delays imposed on the debtor firms, few have acknowledged any benefits to the Chapter 11 procedure, and those that have taken a more positive view (e.g. Belker, Franks and Torous, 1999) typically focus on strategic advantages for certain stakeholders, rather than the social benefits of the procedure itself. Perhaps the most important feature of Chapter 11 is that the parties negotiate new contractual arrangements in full public view with full disclosure. Baird and Picker (1991) argue such a bankruptcy procedure is needed because these negotiations should not be entirely the province of private contracting. "[T]he manager-shareholder and senior creditor cannot be relied on to protect the rights of third parties (Baird and Picker, 1991, p. 312)."

RESEARCH DESIGN

If the negative impact on third party contractual relationships is mitigated by having a public reorganization procedure, it would suggest different recontracting procedures have different social costs. Third parties include contracting parties without valuable claims on the debtor's assets, such as employees, customers, suppliers, and the local community. In this study we examine the social cost of disrupting third party relationships and test the following hypothesis in the alternative:

Hypothesis: Relative to private workout arrangements, bankruptcy
reorganization mitigates the negative externalities of business
failure.

The Tax Reform Act of 1986 is the event of interest. This law altered the economic incentives to enter into private workout arrangements by severely restricting the use of net operating losses (NOLs) for tax purposes when the reorganization involves a "change of ownership." A change of ownership is defined to occur when old equity holders own less than 50% of any new equity issued. The law however provides an exception for firms reorganizing in Chapter 11, and thus by filing Chapter 11 the debtor preserves its NOL carryover tax benefits. The intent and ultimate affect was to direct firms away from private workouts and into the Chapter 11 procedure.

SAMPLE SELECTION

Our sample is a monthly time series of data obtained from Dun and Bradstreet Corporation beginning in October, 1979, with the implementation of the current Bankruptcy Code. The Code made several major changes in bankruptcy procedure. For example, under the former Bankruptcy Act of 1938 (the Chandler Act) there were different reorganization procedures for large and small firms. Chapter 11 of the Bankruptcy Code combines Chapters X, XI, and XII of the old Bankruptcy

Act into a single procedure for business reorganization. Such a major change in bankruptcy reorganization procedures could confound the results of the present study and therefore, we begin the monthly time series at the Code's implementation date. The sample period ends September, 1998, at the time Dun and Bradstreet reorganized and ceased reporting business failure statistics.

Thus, the sample period is a nineteen year window with 228 continuous monthly observations of the number of business failures and new business formations. The event date, January 1, 1987, is the date the Tax Reform Act of 1986 went into effect. We divide the sample observations into a pre-event period, October 1979 through December 1986, and a post-event period, January 1987 through September 1998. We analyze large and small firms separately. Table 1 presents summary statistics for the pre- and post-event periods for both large and small firms. A "failure" is defined as, "a concern that is involved in a court proceeding or voluntary action that is likely to end in a loss to creditors" (Dun and Bradstreet's measures of failures, 1955-1998). All industrial and commercial enterprises petitioned into the Federal Bankruptcy Courts are considered business failures. Also included are: 1) concerns forced out of business through actions in the state courts such as foreclosures, executions, and attachments with insufficient assets to cover all claims; 2) concerns involved in court actions such as receiverships, reorganizations, or arrangements; 3) voluntary discontinuations with a known loss to creditors; and 4) voluntary out of court compromises with creditors. Thus, the number of business failures is broadly defined to include private workout arrangements, state court actions, and federal bankruptcy proceedings. A small business is defined as a concern having less than $100,000 in current liabilities; a large business is defined as a concern having more than $100,000 in current liabilities. Current liabilities include all accounts and notes payable, whether secured or unsecured, known to be held by banks, officers, affiliated companies, suppliers, or the Government. Not included in current liabilities are long-term publicly held obligations (Dun and Bradstreet's measures of failures, 1955-1998).

Table 1 shows the average number of small business failures rose dramatically over the nineteen year sample period. For the pre-event October 1979 through December 1986 period, small business failures averaged 1396 per month, while for the post-event January 1987 through September 1998 period small business failures averaged 4158 per month. The average number of large business failures also increased. For the pre-event period large business failures averaged 1561 per month, while for the post-event period large business failures averaged 1898 per month. Table 1 also presents the summary statistics for the number of new business incorporations. For the pre-event period, new business incorporations averaged 50,588 per month; for the post-event period, the number of new business incorporations averaged 59,393 per month.

EMPIRICAL TESTS OF THE HYPOTHESIS

We use correlation analysis and regression analysis to compare the momentum of business failure over the pre- and post-event periods. Campbell and Choudhury (2002) describe the negative externalities of business failure as a "domino effect" and its momentum varies over time. Campbell and Choudhury also tested the cumulative lagged effects of business failures over time and found the "memory" for business failure can last up to two years from the point of failure. In the present study the number of business failures is regressed on a proxy measure for business failure momentum in both the pre-event and post-event periods. The variable, MOMENTUM, is a constant growth series beginning at one and growing by one each month. If the Tax Reform Act of 1986 is associated with a decrease in business failure momentum, then the coefficient for MOMENTUM should be less influential in the post-event period. To disentangle the effects of expanding business activity, the regression includes a control variable measuring the number new business incorporations.

Durbin-Watson statistics using ordinary least squares (OLS) estimates indicated the presence of positive autocorrelation. One consequence of autocorrelated errors (or residuals) is the formula variance [[[sigma].sup.2] [(X ' X).sup.-1]] of the OLS estimator is seriously underestimated, where X represents the matrix of independent variables and [[sigma].sup.2] is the error variance (see Choudhury, 1994). This can result in misleading test statistics and confidence intervals. We evaluated the autocorrelation function and partial autocorrelation function of the OLS regression residuals using SAS procedure PROC ARIMA (see SAS/ETS User's Guide, 1993). This was necessary because the Durbin-Watson statistic is not valid for error processes other than first order (see Harvey 1981, pp. 209-210). We observed the degree of autocorrelation and identified the order of the model that sufficiently described the autocorrelation. After evaluating the autocorrelation function and partial autocorrelation function, the residuals model was identified as a second order autoregressive model (1 - [[phi].sub.1]B - [[phi].sub.2][B.sup.]) [v.sub.t] = [[epsilom].sub.t] (see Box, Jenkins, & Reinsel, 1994). The final specification of the regression model is of the following form for large (LGFAIL) and small (SMFAIL) failures respectively:

[LGFAIL.sub.t] = [[beta].sub.0] + [B.sub.1][MOMENTUM.sub.t] + [B.sub.2][NEWBUS.sub.t] + [v.sub.t] and [v.sub.t] = [phi].sub.1][v.sub.t-1] = [phi].sub.2][v.sub.t-2] + [[epsilom].sub.t] (1)

[SMFAIL.sub.t] = [[beta].sub.0] + [B.sub.1][MOMENTUM.sub.t] + [B.sub.2][NEWBUS.sub.t] + [v.sub.t] and [v.sub.t] = [phi].sub.1][v.sub.t-1] = [phi].sub.2][v.sub.t-2] + [[epsilom].sub.t]. (2)

Where:

MOMENTUM = a series starting at 1 and growing at a constant amount B=1 each time period;

NEWBUS = the number of new business formations.

We use maximum likelihood estimation instead of two step generalized least squares to estimate the regression parameters in equations (1) and (2). Maximum likelihood estimation estimates both regression parameters and autoregressive parameters simultaneously and accounts for the determinant of the variance-covariance matrix in its objective function (likelihood function). In general, the likelihood function of a regression model with auto-correlated errors has the following form:

L([beta][theta][[sigma].sup.2]) = - n / 2 1n ([[sigma]].sup.2]) - 1 / 2 ln [absolute value of [OMEGA]] - (Y -X[beta])'[[OMEGA].sup.-1] (Y - X[beta]) / 2[[sigma].sup.2]

where,

Y - vector of response variable (number of failures), X - matrix of independent variables (MOMENTUM, NEWBUS, and Intercept), [beta] - vector of regression parameters, [theta] - vector of autoregressive parameters, [[sigma].sup.2] - error variance, [OMEGA] - variance-covariance matrix of autocorrelated regression errors.

For further discussion on different estimation methods and the likelihood function, see Choudhury et al. (1999); also see SAS/ETS User's Guide, 1993, for expressions of the likelihood function.

RESULTS

In this section we report the results of tests investigating the association between the implementation of the Tax Reform Act of 1986 and business failure momentum. The strong but weakening correlations reflected in Table 2 suggest a strong memory of business failure that gradually weakens over time. The memory of large business failures is longer and stronger in the pre-event period than in the post-event period (the correlation statistic for a one month lag in the pre-event period is .88 while in the post-event period it is .77). Also, the positive correlations remain statistically significant for more than two years in the pre-event period, while in the post-event period the correlation ceases to be statistically significant after about 16 months. The correlation results reported in Table 2 for small business failures are similar to those reported for large. A one month lag in the number of small business failures has a .91 correlation in the pre-event period, compared to a .85 correlation in the post-event period. At 24 months the correlation remains strong at .89 in the pre-event period, but has weakened to .27 in the post-event period. These results suggest the Tax Reform Act shortened the memory of business failure for both large and small firms.

The regression analysis results indicate an association between the implementation of the Tax Reform Act of 1986 and a slowdown in business failure momentum. Table 3 reports the regression results for the October 1979 through December 1986 pre-event period. The estimated coefficient for business failure momentum, MOMENTUM, in the pre-event period is statistically significant and positive for both large and small businesses. Interpreting these results for large businesses, if time is increased by one month, the number of business failures increases by 26 firms. Similarly, if time is increased by one month, the number of business failures increases by 32 firms. The control variable for new business formations, NEWBUS, is not significant in the pre-event period.

The regression results reported in Table 4 for the post-event period, January 1987 through September 1998, indicate a slowdown in business failure momentum. The estimated coefficient for MOMENTUM is not statistically significant in either the large or small firm regressions. The estimated coefficient for MOMENTUM is close to zero for large business failures and less than five for small business failures; however, the estimated coefficient for the control variable NEWBUS is significant in both regressions. Overall, these results suggest the Tax Reform Act of 1986 is associated with a reduction in business failure momentum and the impact is slightly more pronounced for large businesses than for small businesses.

SUMMARY AND CONCLUSIONS

The Tax Reform Act of 1986 gave large and small businesses an economic incentive to restructure under the Chapter 11 procedure, rather than attempt a private workout arrangement. After controlling for increases in new business formations, we find strong evidence suggesting the implementation of the Tax Reform Act of 1986 is associated with a shorter the memory for business failure and a reduction in business failure momentum. Our results contribute to the literature by documenting the negative externalities of business failure and, for the first time, associating alternative recontracting procedures with differences in business failure momentum. The evidence suggests private restructurings impose greater social costs than the Chapter 11 corporate bankruptcy procedure. It is an open question whether the efficiency gains inherent in private workout arrangements can justify the additional social cost of the negative externalities.

REFERENCES

Baird, D. (1986). The uneasy case for corporate reorganization. Journal of Legal Studies, 15, 127-147.

Baird, D. & Picker, R. (1991). A simple noncooperative bargaining model of corporate reorganizations. Journal of Legal Studies, 20, 311-349.

Belker, B., Franks, J. & Torous, W. (1999). Are stockholders better off when debt is restructured privately. In E.I. Altman (Ed.), Bankruptcy and Distressed Restructurings: Analytical Issues and Investment Opportunities (pp. 391-400). Washington, D.C., Beard Books.

Box, G., Jenkins, G. & Reinsel, G. (1994). Time Series Analysis: Forecasting and Control. Englewood Cliffs, NJ: Prentice-Hall.

Bradley, M. & Rosenzweig, M. (1992). The untenable case for Chapter 11. Yale Law Journal, 101, 1043-1095.

Campbell, S. (1997). An Investigation of the direct costs of bankruptcy reorganization for closely held firms. Journal of Small Business Management, 35(3), 21-29.

Campbell, S. & Choudhury, A. (2002). An empirical analysis of the business failure process for large and small firms. Academy of Entrepreneurship Journal, 8, 107-120.

Choudhury, A. (1994). Untransformed first observation problem in regression model with moving average process. Communications in Statistics: Theory and Methods, 23(10), 2927?2937.

Choudhury, A., Hubata, R. & St. Louis, R. (1999). Understanding Time-Series Regression Estimators. The American Statistician, 53(4), 342-348.

Dunn and Bradstreet. (1998). Business Failure Record. New York, NY: Dunn and Bradstreet.

Gilson, S., Kose, J. & Lang, L. (1990). Troubled debt restructurings: An empirical study of private reorganization of companies in default. Journal of Financial Economics, 27, 315-354.

Harvey, A.C. (1981). The Econometric Analysis of Time Series. London: Philip Allan.

Jackson, T. (1986). The Logic and Limits of Bankruptcy Law. Cambridge, MA. Harvard University Press.

Jensen, M. (1999). Corporate control and the politics of finance. In E.I. Altman (Ed.), Bankruptcy and Distressed Restructurings: Analytical Issues and Investment Opportunities (pp. 3-43). Washington, D.C., Beard Books.

Roe, M. (1983). Bankruptcy and debt: A new model for corporate reorganization. Columbia Law Review, 83, 528-602.

SAS/ETS User's Guide. (1993). SAS Institute, Inc, Cary, North Carolina.

Titman, S. (1984). The effect of capital structure on a firm's liquidation decision. Journal of Financial Economics, 13, 137-152.

Weiss, L.A. (1990). Bankruptcy resolution: Direct costs and violation of priority of claims. Journal of Financial Economics, 27, 285-314.

White, M. (1989). The corporate bankruptcy decision. Journal of Economic Perspectives, 3, 129-151.

Wruck, K. (1990). Financial distress, reorganization, and organizational efficiency. Journal of Financial Economics, 27(2), 419-444.

Askar H. Choudhury, Illinois State University

Steven V. Campbell, University of Idaho

Table 1: Summary Statistics for Large and Small Firm Failures
for the Periods October 1979 - December 1986 and January
1987 - September 1998 (Monthly Data) (a)

Variables Period Monthly Standard Minimums Maximums
(b) 19-- Means Deviations

SMFAIL 79-86 1396.23 912.62 242.00 3952.00
 87-98 4158.55 942.75 2476.00 6365.00
LGFAIL 79-86 1561.63 971.72 259.00 4145.00
 87-98 1898.54 363.22 1223.00 2778.00
NEWBUS 79-86 50588.47 5730.41 27234.00 68087.00
 87-98 59393.38 5439.17 48688.00 73060.00

(a) Small firms have less than $100,000 in current liabilities;
large firms have more than $100,000 in current liabilities. A
failure is defined as, "a concern that is involved in a court
proceeding or voluntary action that is likely to end in a loss to
creditors." Source: Dun & Bradstreet, Inc

(b) Variable Definitions:
SMFAIL = number of small firm failures;
LGFAIL = number of large firm failures;
NEWBUS = number of new business incorporations.

Table 2: Correlation between Number of Failures
and Their Monthly Lags for the Periods October 1979
December 1986 and January 1987 - September 1998

 Large Firm Failures (b)
Monthly
Lags (a) Oct.79-Dec.86 Jan.87-Sep.98

FAILLAG1 0.87823 0.76594
 (<0.0001) (<0.0001)
FAILLAG2 0.87129 0.73143
 (<0.0001) (<0.0001)
FAILLAG3 0.81817 0.64610
 (<0.0001) (<0.0001)
FAILLAG4 0.75185 0.50392
 (<0.0001) (<0.0001)
FAILLAG5 0.75779 0.54871
 (<0.0001) (<0.0001)
FAILLAG6 0.71762 0.48062
 (<0.0001) (<0.0001)
FAILLAG7 0.71184 0.46910
 (<0.0001) (<0.0001)
FAILLAG8 0.68256 0.41579
 (<0.0001) (<0.0001)
FAILLAG9 0.68120 0.43216
 (<0.0001) (<0.0001)
FAILLAG10 0.66324 0.39391
 (<0.0001) (<0.0001)
FAILLAG11 0.68155 0.36121
 (<0.0001) (<0.0001)
FAILLAG12 0.70954 0.43185
 (<0.0001) (<0.0001)
FAILLAG13 0.67994 0.27081
 (<0.0001) (<0.0012)
FAILLAG14 0.71860 0.29078
 (<0.0001) (<0.0005)
FAILLAG15 0.63883 0.22843
 (<0.0001) (<0.0064)
FAILLAG16 0.62056 0.15580
 (<0.0001) (<0.0651)
FAILAG17 0.61673 0.20843
 (<0.0001) (<0.0131)
FAILLAG18 0.56125 0.07768
 (<0.0001) (<0.3599)
FAILLAG19 0.57335 0.06572
 (<0.0001) (<0.4388)
FAILLAG20 0.5541 0.00324
 (<0.0001) (<0.9696)
FAILLAG21 0.55311 -0.05888
 (<0.0001) (<0.4880)
FAILLAG22 0.55511 -0.03818
 (<0.0001) (<0.6531)
FAILAG23 0.53799 -0.07297
 (<0.0001) (<0.3899)
FAILLAG24 0.53625 -0.09331
 (<0.0001) (<0.2711)

 Small Firm Failures (b)
Monthly
Lags (a) Oct.79-Dec.86 Jan.87-Sep.98

FAILLAG1 0.91336 0.85561(<0.0001)
 (<0.0001)
FAILLAG2 0.92202 0.83484(<0.0001)
 (<0.0001)
FAILLAG3 0.92119 0.80535(<0.0001)
 (<0.0001)
FAILLAG4 0.88898 0.75113
 (<0.0001) (<0.0001)
FAILLAG5 0.88991 0.79412
 (<0.0001) (<0.0001)
FAILLAG6 0.87171 0.74067
 (<0.0001) (<0.0001)
FAILLAG7 0.85140 0.72654
 (<0.0001) (<0.0001)
FAILLAG8 0.83895 0.67937
 (<0.0001) (<0.0001)
FAILLAG9 0.83825 0.66902
 (<0.0001) (<0.0001)
FAILLAG10 0.79460 0.64135
 (<0.0001) (<0.0001)
FAILLAG11 0.82508 0.60979
 (<0.0001) (<0.0001)
FAILLAG12 0.85642 0.65161
 (<0.0001) (<0.0001)
FAILLAG13 0.81353 0.53713
 (<0.0001) (<0.0001)
FAILLAG14 0.85533 0.54068
 (<0.0001) (<0.0001)
FAILLAG15 0.83955 0.48146
 (<0.0001) (<0.0001)
FAILLAG16 0.82220 0.41738
 (<0.0001) (<0.0001)
FAILAG17 0.84967 0.43814
 (<0.0001) (<0.0001)
FAILLAG18 0.83383 0.33767
 (<0.0001) (<0.0001)
FAILLAG19 0.86171 0.33488
 (<0.0001) (<0.0001)
FAILLAG20 0.84768 0.31011
 (<0.0001) (<0.0002)
FAILLAG21 0.84938 0.28239
 (<0.0001) (<0.0007)
FAILLAG22 0.8412 0.27236
 (<0.0001) (<0.0011)
FAILAG23 0.8806 0.2642
 (<0.0001) (<0.0015)
FAILLAG24 0.89472 0.27123
 (<0.0001) (<0.0011)

() p-values

(a) Variable Definitions:
FAILLAG(J) = number of firm failures, large or small,
lagged J months back in time

(b) Small firms have less than $100,000 in current
liabilities; large firms have more than $100,000
in current liabilities. A failure is defined as, "a
concern that is involved in a court proceeding or
voluntary action that is likely to end in a loss to
creditors." Source: Dun & Bradstreet, Inc.

Table 3 Regression Results for Number of Large and Small
Firm Failures for the Period October 1979 - December 1986
(Monthly Data) (a) : Maximum Likelihood Estimates

Independent Large Firm Failures Small Firm Failures
Variables (b) (corrected for (corrected for
 autocorrelationd) autocorrelatione)

Intercept 8073.00C (-2.58) ** -10760.00 (-6.35)***
MOMENTUM 25.78 (2.95) *** 32.16 (6.55) ***
NEWBUS -0.003 (-0.20) 0.001 (0.12)
R-Squared 0.82 0.89
Durbin-Watson 1.96 2.18

(a) Small firms have less than $100,000 in current liabilities;
large firms have more than $100,000 in current liabilities. A
failure is defined as, "a concern that is involved in a court
proceeding or voluntary action that is likely to end in a loss to
creditors." Source: Dun & Bradstreet, Inc.

(b) Variable Definitions: MOMENTUM = a series starting at 1 and
growing at a constant amount B=1 each time period; NEWBUS = the
number of new business formations

(c) The t-statistics reported in parenthesis are significant at ten
(*), five (**), and one (***) percent levels.

(d) The regression residuals model was identified as, (1-
[[PHI].sub.1] [beta] - [[PHI].sub.2] [[beta].sup.2]) [v.sub.t] =
[[member of].sub.t] and the estimated first and second order
autoregressive (AR) parameters from SAS were, (1 + 0.45[beta] +
0.37[[beta].sup.2])[v.sub.t] = [[member of].sub.t]. Where
t-statistics for autoregressive parameters are reported in
parentheses and they are both significant at the one (***) percent
level.

(e) The regression residuals model was identified as,
(1 - [[PHI].sub.1][beta] - [[PHI].sub.2][[beta].sub.2])
[v.sub.t] = [[member of].sub.t] and the estimated first and
second order autoregressive (AR) parameters from SAS were,
(1 + 0.32 [beta] + 0.42 [[beta].sup.2])[v.sup.t] = [[member of].sup.t].
(3.22) *** (4.16) ***

Where t-statistics for autoregressive parameters are reported in
parentheses and they are both significant at the one (***) percent
level.

Table 4: Regression Results for Number of Large and
Small Firm Failures for the Period January 1987 -
September 1998 (Monthly Data) (a):

 Large Firm Small Firm
 Failures Failures
Independent (corrected for (corrected for
Variables (b) autocorrelation (d)) autocorrelation (e))

Intercept 847.09c -607.56
 (0.67) (-0.16)

MOMENTUM 0.6326 4.89
 (0.24) (0.64)

NEWBUS 0.0127 0.04
 (2.57) ** (3.91) ***

R-Squared 0.66 0.79

Durbin-Watson 1.96 2.05

(a) Small firms have less than $100,000 in current liabilities;
large firms have more than $100,000 in current liabilities. A
failure is defined as, "a concern that is involved in a court
proceeding or voluntary action that is likely to end in a loss to
creditors." Source: Dun & Bradstreet, Inc

(b) Variable Definitions:

MOMENTUM = a series starting at 1 and growing at a constant
amount B=1 each time period;

NEWBUS = the number of new business formations;

(c) The t-statistics reported in parenthesis are significant
at ten (*), five (**), and one (***) percent levels

(d) The regression residuals model was identified as, (1 -
[[PHI].sub.1] [beta] - [[PHI].sub.2] [[beta].sup.2])[v.sub.t] =
[[member of].sub.t] and the estimated first and second order
autoregressive (AR) parameters from SAS were, (1 + 0.51 [beta] +
0.35 [[beta].sup.2] [v.sub.t] = [[member of].sub.t] 6.26 *** 4.31
*** Where t-statistics for autoregressive parameters are reported
in parentheses and they are both significant at the one (***)
percent level.

(e) The regression residuals model was identified as, (1 -
[[PHI].sub.1] [beta] - [[PHI].sub.2] [[beta].sup.2])[v.sub.t] =
[[member of].sub.t] and the estimated first and second order
autoregressive (AR) parameters from SAS were, ((1 + 0.53 [beta] +
0.38 [[beta].sup.2] [v.sub.t] = [[member of].sub.t] 6.67 *** 4.72
***.

Where t-statistics for autoregressive parameters are reported in
parentheses and they are both significant at the one (***) percent
level.