Description
Strategic actions and going-concern audit opinions e Liesbeth Bruynseels and Marleen Willekens.
Faculty of Economics and Applied Economics
DEPARTMENT OF ACCOUNTANCY, FINANCE AND INSURANCE (AFI)
Strategic actions and going-concern audit opinions
e
Liesbeth Bruynseels and Marleen Willekens
AFI 0611
Strategic Actions and Going-Concern Audit Opinions
Liesbeth Bruynseels
KULeuven
Marleen Willekens
Tilburg University and KULeuven
ABSTRACT
In this paper we examine the association between various types of strategic management actions
by distressed companies and the likelihood that they receive a going-concern audit opinion. Prior
going-concern studies that focus on the impact of non-financial information investigate particular
operating turnaround initiatives, such as cost reduction strategies (see, Behn et al., 2001; and
Geiger and Rama, 2003). We contribute to this literature by studying the impact of a broader set
of operating turnaround initiatives (i.e. cost reduction, asset disposal, increased marketing and
product upgrading), as well as a set of strategic growth initiatives (i.e. product innovation,
expansion and cooperative strategies). As the assessment of an auditee’s likelihood of survival
within the next twelve months is a critical in the going-concern decision making context, we
further distinguish between strategic growth initiatives that are likely to generate positive cash
flows in the short run (i.e. cooperative agreements) versus long run (i.e. innovation and expansion
strategies). Based on manually collected data on a sample of 114 distressed manufacturing US
firms, we find that operating turnaround initiatives, as well as strategic initiatives that are likely
to generate positive cash flows in the long run are positively associated with the likelihood that a
going-concern opinion is received. This evidence suggests that these two categories of
management initiatives are perceived as additional going-concern risk factors by auditors. On the
contrary, strategic turnaround initiatives that are likely to generate positive cash flows in the short
run, are negatively associated with the likelihood that a going-concern opinion is received, which
is supportive of this category of initiatives being perceived as mitigating factors.
Keywords: strategic management actions, going-concern audit opinion
Data availability: The strategic company information used in the paper is manually
collected from the 10-Ks filed with the SEC; the financial data are from
public sources identified in the paper.
We gratefully acknowledge the helpful comments and suggestions of Robert Knechel, Rashad Abdel-Khalik, Ann
Gaeremynck,, John Christensen, Wayne Landsman, Christian Leuz, Peter Pope, Diane Breesch, Valery Nikolaev and
participants at the 2005 EAA Annual Meeting, the 2005 Doctoral Colloquium, the 2005 ARNN conference, the 2005
EARNet Ph.d. colloquium, the 2006 BAA National Auditing Conference and the accounting workshop at Tilburg
University and the KULeuven.
Corresponding address: [email protected]
2
1. INTRODUCTION
In this study we investigate whether strategic actions taken by management of financially
distressed firms affect the auditors’ going-concern opinion decision. Hypothesizing a relationship
between client strategic actions and going-concern audit decisions is motivated by at least two
features. First, SAS no. 59 explicitly prescribes the consideration of contrary non-financial
(internal and external) matters and mitigating management plans in making going-concern
decisions. Therefore a broad set of events, actions and management plans – including strategic –
are potential determinants of going-concern opinion decisions. Second, changes in auditing
methodology and technology towards business risk auditing approaches in the second half of the
nineties (such as Strategic-Systems Auditing as introduced in Bell et al. 1997 and further
developed in Bell et al. 2005) further motivate the likelihood that strategic management actions
are an integrated part of audit evidence collection. Note that client strategic analysis
1
is one of the
most innovative aspects of business risk auditing methodologies. As the evidence collected from
strategic analyses is likely to have a substantial impact on subsequently planned and executed
audit procedures as well as the assessment of a client’s future financial viability, it is also very
likely to affect the auditor’s going-concern opinion decision.
It is well documented in the literature that auditors make going-concern decisions based
on reported financial results and compliance with financial obligations (see, for example,
Mutchler, 1985 and 1997; Levitan and Knoblett, 1985; Menon and Schwartz, 1987; Dopuch et
al., 1987; Bell and Tabor, 1991; Chen and Church, 1992; Gaeremynck and Willekens, 2003). The
importance of information other than that contained in the financial statements is also emphasized
in SAS no. 59. Besides the presence of negative financial trends and other indications of possible
financial difficulties – like for example default on loan agreements, SAS No. 59 also defines
certain (non-financial) internal and external matters as conditions and events that may indicate
1
or stated alternatively, acquiring evidence of and from the entity business states (EBS) as advocated in Bell et al.
(2005).
3
that there could be substantial doubt about the entity to continue as a going-concern.
Subsequently, SAS no.59 also requires auditors to consider management plans to mitigate the
effects of these adverse conditions or events when assessing their client’s ability to continue as a
going-concern. The impact on the audit opinion of contrary and mitigating factors in publicly
available disclosures such as the financial press, 10-K’s or management discussions and analyses
has also been documented in the going-concern literature (see Mutchler et al., 1997). Although
the importance of strategic management plans is recognized in today’s auditing practice, research
on the impact of forward-looking management plans on going-concern decisions is scant. Behn et
al. (2001) recognized this caveat and provide evidence of the relationship between the likelihood
of going-concern opinions and a company’s ability to obtain new financing and to reduce costs.
After controlling for financial condition, size, default status, and the propensity to voluntarily
disclose information, their results indicate that going-concern opinion decisions are strongly
linked to publicly available mitigating information regarding certain management plans. In
particular, plans to issue equity and to borrow additional funds exert the strongest association
with the issuance of an unqualified opinion. Recently, Geiger and Rama (2003) report that
companies are more likely to receive a modified report if they entered into a cost reduction plan
or sold off significant assets. However, contrary to the findings of Behn et al. (2001), plans for
the issuance of new debt or equity are not significantly associated with the auditor’s opinion type.
The increased relevance of strategic parameters in the audit decision making context in
general, is attributable to changes in the scope and methodology of auditing that have taken place
in (a number of) large accounting firms in the second half of the nineties (see, for example, Bell
et al. 1997; Lemon, Tatum and Turley 2000; Knechel, 2001; Bell et al. 2005; Curtis and Turley
2005). Whereas traditional auditing approaches adopt a bottom-up focus thereby directing
attention to the nature of account balances, classes of transactions, and properties of the client’s
accounting system, business risk auditing develops a top-down holistic perspective of the client’s
4
business and industry. This entails a thorough analysis of the client’s business and strategic
position. Note that a general evolution towards business risk auditing elements is reflected in
some of the new International Audit Risk Standards. In particular, International Standard on
Auditing (ISA) 315 requires the auditor to develop an understanding of client objectives and
strategies, as well as the related business risks that may result in a material misstatement of the
financial statements. These business risks should not only be evaluated in light of their immediate
consequences for the risk of material misstatements, but also with regard to their longer-term
consequences.
We contribute to the going-concern literature by testing the association between the
likelihood of going-concern opinions and a comprehensive set of strategic actions for a sample of
distressed US manufacturing companies. Like other going-concern studies, we rely on
information disclosed by management in the management discussion and analysis (MD&A), and
remainder of the 10-K (Behn et al., 2001; Geiger and Rama, 2003). Consistent with the strategy
literature
2
(see for example, Barker and Duhaime, 1997; Robbins and Pearce II, 1992;
Sudarsanam and Lai, 2001; Bruton et al. 2003) we distinguish between management actions and
plans aimed at a short-term improvement in financial performance (or, operating turnaround
initiatives) and strategic growth initiatives (or, strategic turnaround initiatives). As going-concern
decision making involves the assessment of the likelihood of survival of an auditee within the
next twelve months, we further sub-divide strategic growth initiatives in those that are likely to
generate positive cash flows in the short run (i.e. cooperative agreements) versus long run (i.e.
innovation and expansion strategies).
Consistent with prior going-concern studies which focused on short-term improvement
(or, operating) initiatives (see, Behn et al. 2001; Geiger and Rama, 2003), we find that cost
reduction strategies are positively associated with the likelihood of receiving a going-concern
2
The strategy literature contains an extensive body of research that focuses on how firms reverse firm-threatening
performance declines. See also hypothesis development section.
5
opinion. In addition, we also find that marketing strategies are positively associated with the
likelihood of receiving a going-concern opinion. Our test of the growth initiatives reveals that the
engagement in cooperative agreements is negatively associated with the likelihood of receiving a
going-concern opinion. Our results are consistent with cooperative agreements providing a
positive signal about the going-concern status of the firm and therefore can be interpreted as a
mitigating factor, whereas the presence of cost reduction and marketing strategies are perceived
as additional going-concern risk factors which increase the likelihood to receive a going-concern
opinion.
Finally, we also test the impact of three types of aggregated construct variables capturing
the presence short-term initiatives, strategic growth initiatives and vvv. We find that more
operating turnaround initiatives taken by a distressed firm are associated with a higher likelihood
that a going-concern opinion is issued. A similar result is obtained for strategic initiatives not
capable of generating short-term financial impact. On the contrary, the presence of strategic
turnaround initiatives that are likely to generate a financial impact in the short run is negatively
associated with the likelihood that a going-concern opinion is issued. Thus, our evidence suggests
that auditors perceive the engagement in operating initiatives and strategic growth initiatives that
only yield a financial impact in the long term as additional going-concern risk factors, but the
engagement in strategic growth initiatives that generate a short-term financial impact as a
mitigating factor.
The remainder of this paper is organized as follows. In the next section we develop our
hypotheses. Section 3 then is devoted to the development of the going-concern opinion model
that is tested in this paper. Next, in Section 4 we provide an overview of our sample selection
procedure and data collection approach. In Section 5 we discuss the results of our analyses. We
conclude in Section 6.
6
2. HYPOTHESES DEVELOPMENT
SAS no. 59 clearly states that besides financial indicators – such as negative (financial) trends
and other indications of possible financial difficulties – non-financial internal and external
matters are relevant conditions and events to assess the going-concern status of a client firm.
Examples of internal matters that are included in SAS No. 59 are work stoppages or substantial
dependence on the success of a particular project. External matters listed in SAS No. 59 include,
for example, legal proceedings or the loss of a key franchise, license or patent. Furthermore,
when the identified conditions and events in the aggregate lead to substantial doubt about the
continued existence of the entity as a going-concern, the auditor should identify and evaluate
management’s plans to mitigate the effects of these adverse conditions or events. If the auditor
believes that there exist management plans that overcome this substantial doubt, a going-concern
audit report is not required. Examples of such potentially mitigating management plans are
included in SAS no.59, and relate to the sale of assets, the borrowing or restructuring of debt, the
reduction of expenditures and the increase of ownership equity.
A few prior studies provide evidence that auditors are indeed committed to reviewing
management plans that are dealing with adverse conditions or events when assessing a client’s
ability to continue as a going-concern (Behn et al., 2001; Geiger and Rama, 2003). However,
these studies are confined to assessing the impact of examples of management plans and actions
that are explicitly mentioned in SAS no.59. In this paper, we elaborate on this theme and
investigate the impact of a broader set of potentially contrary or mitigating actions and strategies
on the auditor’s going-concern decision. We motivate this broader strategic focus by the
emergence of business risk auditing in the 1990s. With the emergence of business risk auditing,
traditional auditing methodologies have been complemented with new audit processes based on a
top-down, holistic perspective of the client’s business and industry (see, for example, Bell et al.
1997; Knechel 2001; Lemon, Tatum and Turley 2000). The most innovative aspect of business
7
risk auditing is the assessment of client strategic viability, which can have a substantial impact on
the subsequent audit procedures and the assessment of future financial viability.
To predict the impact of a comprehensive set of viable strategic actions on the going-
concern opinion, we categorize management actions into strategic and operating turnaround
approaches. This is a widely used framework introduced by Hofer (1980). Note that an operating
approach focuses on internal, operating problems of firms through – for example – decreasing
costs, increasing efficiency, disposing assets, or improving sales (Hofer, 1980). A strategic
turnaround approach is aimed at long-term profitability by solving external, strategic problems
through for example a change in the strategic direction of the firm, its positioning, alliances and
product lines (Bruton et al., 2003). Strategic repositioning may be done through business
divestments, acquisitions, alliances, new product development, new markets, and increased
market penetration. Firms experiencing financial distress may adopt a variety of strategies to
return to financial health. The strategy literature offers an extensive body of research that focuses
on how firms reverse firm-threatening performance declines to induce successful company
turnaround
3
(see for example, Barker & Duhaime, 1997; Robbins and Pearce II, 1992;
Sudarsanam & Lai, 2001; Bruton et al. 2003).
Hypothesis 1: Operating turnaround approaches and going-concern opinions
An operating approach to company turnaround typically consists of actions related to cost
reduction, revenue generation and operating-asset reduction. The focus is on achieving short-term
financial relief, without considering long-term changes in the organization’s strategy. In order to
achieve short-term profitability improvement, companies have the opportunity to engage in
classic retrenchment activities such as: divestment, product elimination, cost rationalization and
employee layoffs. In addition to these cost-cutting initiatives, revenue generating strategies may
3
Successful turnaround is defined as the reversal of a firm’s pattern of performance decline (Schendel, Patton and
Riggs, 1976).
8
be pursued focusing on existing lines of products, price-cutting, increased marketing expenditure
or increased direct sales efforts (Hofer, 1980).
Prior studies that examine the association between the implementation of operating
approaches and successful company turnaround have focused on retrenchment activities and
provide mixed results. Several studies report that classic retrenchment strategies are significantly
associated with turnaround success (see, for example, Robbins and Pearce II, 1992), whereas
other studies cast doubt on the value of operating approaches as part of a company’s turnaround
approach (Sudarsanam and Lai, 2001; Barker III and Mone, 1994).
The mixed evidence from the strategy literature indicates that operating turnaround
strategies per se may not be capable of curing deficiencies in a declining firm’s strategic
orientation. In other words, if a declining firm’s problems relate to its strategic positioning, these
short-term cures could be inadequate, given that changing a firm’s strategic orientation is a
prerequisite to recovery (see also Schendel et al., 1976; Hofer, 1980; Barker and Duhaime, 1997).
As we investigate the auditor’s perception of the effectiveness of operating turnaround
approaches
4
for distressed firms, a relevant question is which signal such approaches by
themselves send to the auditor regarding the going-concern status of the company. Given the
evidence reported above, it is likely that auditors perceive operating turnaround strategies as
insufficient to induce recovery for distressed firms. This is indeed consistent with the finding
reported by Geiger and Rama (2003), i.e. that cost-cutting or asset disposal activities are
associated with a higher likelihood of receiving a going-concern audit report. This leads to our
first hypothesis:
4
These include cost-cutting activities, disposal of assets, increasing marketing efforts and improving existing
products and operating processes – see also Section Model Specification.
9
H1: For financially distressed companies, the implementation of an operating turnaround
approach is likely to increase (ceteris paribus) the propensity that a going-concern
opinion is issued.
Hypothesis 2: Strategic turnaround approaches and going-concern opinions
Overall, the evidence from the strategy literature suggests that (long-term) strategic turnaround
approaches are successful turnaround vehicles. Barker III and Duhaime (1997) find that when a
company’s decline is firm-based and not caused by industry contraction, recovering firms
implement more extensive strategic changes (which are consistent with reorientation).
Sudarsanam and Lai (2001) provide evidence that firms recovering from financial distress
typically adopt more forward-looking, expansionary and external market focused strategies than
non-recovery firms. More specifically, recovery firms typically adopt growth-oriented and
external-market focused strategies, whereas non-recovery firms continue to engage in operating
restructuring strategies. Given the evidence from the strategy literature about the effectiveness of
strategic approaches for company turnaround and recovery, it is reasonable to expect that such
strategies may also have a mitigating impact on the auditor’s going-concern opinion. However, as
the auditor’s going-concern opinion is an assessment of the client’s ability to survive during the
next 12 months, only those (long-term) strategic approaches that are expected to have a positive
impact on the company’s liquidity status within the next 12 months will be perceived as
mitigating factors. It is therefore necessary to further examine the short-term impact of the
different types of long-term strategic approaches, i.e. cooperative agreements, product innovation
and acquisition strategies.
Barker III and Duhaime (1997) emphasize that cooperative agreements with other firms
are an essential element of a turnaround approach based on strategic change. Examples of
cooperative strategies include long-term contractual agreements with suppliers or buyers,
10
alliances or joint ventures, subcontracting and technology licensing agreements
5
. Prior research
about the consequences of the implementation of a cooperative strategy has shown that strategic
networks such as strategic alliances, joint-ventures and long-term buyer-supplier relationships
often have positive effects on different measures of corporate performance. For example,
Mitchell and Singh (1996) reported evidence of alliances raising organisational survival rates.
Powell et al. (1996) found that companies which had formed many alliances experienced
accelerated growth rates. Gulati et al. (2000) highlight the idea that one of the most important
benefits of strategic networks is the increased access to information, resources, markets and
technologies. In addition to access to resources, Stuart (2000) found that strategic alliances also
affect firm subsequent-period performance through their influence on an organization’s
reputation, particularly if the firm is of ambiguous quality.
Capon et al. (1992) report that new product development and strategic acquisitions are
often linked to (long-term) performance improvement, and report evidence that suggests that both
strategies act as substitutes in terms of effectiveness vis-à-vis company turnaround (i.e. non-
innovative firms that involve in acquisitions perform nearly as well as those that engage in
product innovations). However, the short-term performance impact of both types of strategic
actions is less apparent. In a recent meta-analytic review of merger and acquisition performance,
King et al. (2004) report that acquisitions are not improving the short-term financial performance
of acquiring firms, on average
6
. With respect to the short-term performance impact of product
innovation, Mishina et al. (2004) even report a negative association with the rate of short-term
sales growth.
5
Note that strategic alliances are a popular financing vehicle for companies in financial distress, as partnering up
with a successful healthy company can provide distressed companies with additional funding to develop or market
products, or with other benefits such as a more extensive customer base (see, for example, Bruton et al. 2003).
Another vehicle to improve financial position is engaging in a licensing strategy with regard to unused or high-risk
technology. Licensing out proprietary technology can substantially improve a company’s financial position as it
periodically receives royalties and/or milestone payments (see, for example, Sudarsanam and Lai 2003). A company
can also safeguard future sales by engaging into long-term contracting with buyers or distributors (Miller, 1992).
6
Instead, this study indicates that acquisitions either have no significant effect or a modest negative effect on an
acquiring firm’s financial performance in the post-announcement period.
11
To summarize, the evidence from the strategy literature suggests that (long-term) strategic
turnaround approaches are successful turnaround vehicles. However, the short-term performance
impact of different types of strategic turnaround approaches varies. Based on prior research it is
reasonable to expect that the introduction of new products and corporate acquisitions are less
likely to have a positive financial impact within the next 12 months, whereas it is more likely that
cooperative agreements generate positive cash flows within the next 12 months. This leads to our
second hypothesis:
H2: For financially distressed companies the implementation of a strategic turnaround
approach is likely to reduce (ceteris paribus) the propensity that a going-concern audit
opinion is issued, given that the strategy is likely to have a mitigating impact within the
next 12 months.
3. MODEL SPECIFICATION AND VARIABLE MEASUREMENT
We use the following logistic model to test our two hypotheses:
REPORT = f (operating turnaround variables, strategic turnaround variables, control variables).
REPORT is an indicator variable that takes a value equal to one if the auditor issues a going-
concern report, and zero otherwise. The turnaround approach variables contain information
regarding turnaround strategies that have been implemented by the company during the year
under audit to overcome adverse conditions affecting corporate performance. This information is
manually collected from corporate disclosures in the annual report and 10-K. We investigate the
impact of two categories of turnaround strategies that can potentially mitigate the adverse
conditions affecting company performance. In the category of operating turnaround strategies,
we consider the impact of cost-cutting and asset disposal activities, product and operating process
12
improvements and increasing marketing efforts. As strategic turnaround approaches, we consider
cooperative agreements with other firms, the introduction of new products, mergers, and
acquisitions. Finally, the control variables in our model encompass factors that have been found
to be associated with going-concern opinion decisions and the propensity to voluntarily disclose
information in prior research.
[ INSERT TABLE 1 ABOUT HERE ]
3.1 Operating turnaround variables
Our classification of operating turnaround variables is based on Hofer (1980) who distinguishes
between four different types of operating turnaround approaches: a) cost-cutting strategies, b)
asset reduction strategies, c) revenue increasing strategies and d) combination strategies.
Accordingly, we include and test variables reflecting a cost reduction strategy (O-COSTRED), an
asset disposal strategy (O-DISPOSE), a commercial strategy (O-COMMERCIAL), and a strategy
aimed at the improvement of existing products and processes (O-UPGRAD).
We define O-COSTRED as an operating turnaround variable that captures significant cost
reduction efforts. In particular, this variable relates to both employee layoffs and other cost
reduction efforts during the year under audit. O-COSTRED is set equal to one if the company
reports cost reduction strategies for the year under audit, and is set equal to zero otherwise. O-
DISPOSE is defined as an operating turnaround variable that indicates whether a company
engages in the sale of significant assets. O-DISPOSE is set equal to one if the company reports
the sale of assets for the year under audit, and is set equal to zero otherwise. As opposed to
strategic actions aimed at reducing expenditures, short-term operating strategies also include a
number of revenue generating activities (Hofer, 1980). Subsequently, we define O-
13
COMMERCIAL as an operating turnaround variable that indicates whether a company increases
its marketing efforts, and O-UPGRAD as a variable that relates to the realisation of
improvements in existing products and production processes. O-COMMERCIAL is set equal to
one if the company reports increased marketing efforts for the year under audit, and is set equal to
zero otherwise. O-UPGRADE is set equal to one if the company reports product and/or process
improvements for the year under audit, and is set equal to zero otherwise. In order to extend and
refine our analyses, we also define a number of strategic construct variables. This will enable us
to test the aggregate impact of strategic turnaround variables that have similar characteristics on
theoretical grounds. OPERATING is defined as the sum of O-COSTRED, O-DISPOSE, O-
COMMERCIAL and O-UPGRADE, scaled by its maximum value in the sample. Note that a
further refinement in short-term versus long-term impact variables would be tautological in the
context of operating initiatives, as they are all expected to have a short-term impact on financial
performance.
3.2 Strategic turnaround variables
Long-term strategic turnaround approaches typically relate to reconfiguration of the assets
and/or the corporate portfolio, and product and/or market refocusing. S-EXPANSION is a
strategic turnaround variable capturing whether a company engages in an acquisition strategy (of
other companies) to accelerate growth. S-EXPANSION is set equal to one if the company
reports acquisitions for the year under audit, and is set equal to zero if such is not the case. We
also include S-COOP, a variable that indicates whether a company enters into strategic alliances,
joint-ventures, licensing agreements and other cooperative arrangements. S-COOP is set equal to
one if a company entered in cooperative arrangements during the year under audit, and is set
equal to zero if such is not the case. Further, we define S-PRODUCT as a strategic turnaround
variable that assesses whether a company has recently introduced new products. S-PRODUCT is
14
set equal to one if the company reports the introduction of new products during the year under
audit, and is set equal to zero if such is not the case.
Finally, we also introduce strategic construct variables to capture the aggregate impact of
several strategic turnaround initiatives engaged in by the audited firm. First, we define a (rough)
strategic construct variable STRATEGIC to capture the aggregate impact of all strategic
turnaround initiatives that have been implemented by the audited company during the past year.
This categorization is based on the strategic literature (see, Hofer, 1980). STRATEGIC is a
discrete variable representing the sum of all strategic turnaround variables defined above, scaled
by its maximum value in the sample. In a more refined categorization of strategic variables, we
distinguish between strategic initiatives that are expected to have a positive impact on firm
financial performance in the short term (STRAT_ST) or in the long term (STRAT_LT). This
categorization can be motivated by recent evidence in the strategy literature (see, King et al.
2004; Mishina et al. 2004) that acquisitions and product innovations are not improving the short-
term financial performance of acquiring and innovating firms, respectively. Such a distinction in
short-term and long-term financial impact is warranted given the going-concern decision context
that we are investigating. STRAT_ST is an indicator variable which happens to coincide with S-
COOP, as the presence of cooperative agreements is the only strategic turnaround variable
defined in this paper that is likely to generate a financial impact within the next twelve months.
STRAT_LT is defined as the sum of S-PRODUCTS and S-ACQUIS, divided by its maximum
value in the sample. We also refer to Table 1, for an overview of definitions of the strategic
variables.
3.3 Control variables
The issuance of a going-concern opinion is obviously conditional upon the auditee’s financial
condition. Therefore a first category of control variables that are included in our model capture
15
the financial condition of the firm. Based on prior audit opinion studies (see, for example,
Mutchler 1985, 1997; Chen and Church, 1992) we include cash flow from operations divided by
total liabilities (CFO/TL), the current ratio (CR), and long-term debt divided by total assets
(LTD/TA), as control variables. Following Menon and Schwartz (1987), we also include a
change variable, namely the change in current ratio (?CR). In line with Bell and Tabor (1991),
we also control for a company’s liquidity performance relative to the industry, by including an
indicator variable (INDCR), taking the value of one if the current ratio of the company exceeds
the industry median current ratio. As in Chen and Church (1992), we also add DEFAULT, an
indicator variable that takes a value equal to one if the company defaults on debt payments or is
in technical default of loan covenants
7
, and zero otherwise. Following prior research, we also
include the log of total assets to control for company size (see, for example, Chen and Church,
1992).
A second category of control variables constitute mitigating factors identified in prior
audit opinion research. Behn et al. (2001) find that plans to use existing bank lines of credit and
other approved lines of credit are negatively associated with the likelihood of a going-concern
opinion. Consistent with Behn et al. (2001) we include BORROW, a variable that is set equal to
one if the auditee plans to borrow funds through existing bank lines of credit or other approved
debt instruments; and STOCK, a variable that is set equal to one if the auditee plans to issue
equity through existing or committed arrangements.
4. SAMPLE AND DATA
4.1 Sample selection
Consistent with prior going-concern studies (see, for example, Mutchler, 1985; Chen and Church,
1992; and Behn et al., 2001) we select a sample adopting a matched pair design. Note that a
7
A company’s default status was determined by reading the MD&A and debt footnotes in the financial statements.
16
matched pair design is most often used when the research design necessitates manual data
collection, as is the case in this study. In particular, we first selected a sample of companies that
received a first-time going-concern opinion and then a matched sample of distressed companies
that did not receive a going-concern opinion.
Selection of going-concern firms
We identified all firms from the Worldscope database that are listed on AMEX, NASDAQ and
NYSE and received a going-concern audit opinion in the period 1998-2001. Consistent with
prior studies (Mutchler and Williams 1990; Behn et al. 2001; Blay and Geiger 2001), we
restricted our sample to companies in the manufacturing industries (SIC 20 to 39) to eliminate
confounding industry effects. This resulted in an initial sample of 276 manufacturing companies
that received a going-concern audit opinion in fiscal years from 1998-2001. From this initial
sample, we then eliminated companies that received a going-concern report in the previous year
to control for potential confounding effects from prior going-concern opinions (see also,
Mutchler, 1985, Blay & Geiger, 2001; Behn et al., 2001). In addition, we also excluded firms that
faced bankruptcy proceedings as the decision to issue a going-concern opinion is trivial for such
firms, and firms for which no match could be identified. This resulted in 57 firms with going-
concern opinions: 8 firms in 1998, 8 firms in 1999, 15 firms in 2000, and 26 firms in 2001.
Selection of control firms
To test our going-concern model, we matched the going-concern sample with a sample of
distressed companies that did not receive a going-concern report. We searched the Worldscope
database from 1998 to 2001 to identify all manufacturing companies listed on NASDAQ, NYSE
or AMEX that received a clean audit opinion. Consistent with prior studies, we further restricted
the control sample to firms in financial distress (see, for example, Mutchler, 1985; Chen &
17
Church, 1992; Behn et al., 2001; McKeown et al., 1991). Based on Chen and Church (1992) we
adopt the following criteria to identify distressed companies: 1) negative retained earnings, 2)
negative operating income, 3) negative net income , 4) negative working capital, 5) negative net
worth, and 6) negative operating cash flows. Note that Chen and Church (1992) classified a
company as stressed if it meets at least one of above criteria. We use a more stringent rule for
financial distress, by classifying a company to be stressed if it meets at least two of these stress
criteria. This procedure yielded 2929 “distressed” companies that received a clean opinion during
the period 1998-2001.
As we use a matched-pair design, we matched control-sample companies to the going-
concern firms based on year, size (proxied by total assets) and two-digit SIC classifications. This
procedure ensures that we include similar companies with respect to size and industry in both
sub-samples. A matched group design has been used previously by Mutchler (1985), Chen and
Church (1992), Behn et al. (2001) and Geiger and Rama (2003). A limitation of this approach is
that it overstates the issuance of a going-concern opinion in this experimental setting. Note that
we address this issue by our statistical approach, as we adopt logistic regression analysis (see
further).
[ INSERT TABLE 2 ABOUT HERE ]
4.2 Data collection and strategic scorecard
As strategic company information is not publicly available, we manually collected this
information from the relevant 10-Ks filed with the SEC, by reading these documents back-to-
cover and completing a strategic scorecard. The strategic scorecard used to test our hypotheses is
included in Appendix A. For each of the defined turnaround approaches we assessed whether the
company has engaged in actions related to that specific turnaround initiative during the year
18
under audit. Per initiative, a score equal to one is assigned if a firm discloses such action, and
zero otherwise.
The suitability of 10-K filings for strategic information collection is supported by prior
studies that investigated the association between disclosures in the MD&A and future corporate
financial performance. These studies report evidence that indicates that the information content of
narrative disclosures in the annual report is significantly associated with future viability of
distressed firms (see, for example, Tennyson et al., 1990; Boo and Simnett, 2002). Boo and
Simnett (2002) examine the reliability of management’s prospective comments for a sample of
140 Australian public companies that had experienced significant losses. Their results indicate
that management’s prospective comments have significant information content with respect to the
company’s future viability. Note also that SAS No. 8 requires auditors to ensure that the ‘other
information’ attached to financial statements is not materially inconsistent with the financial
statements, and does not contain any material misstatement or fact. Furthermore, the costs of
potential litigation and loss of reputation are important factors to prevent management from
disclosing misleading information.
Note that the dependent variable and (most of) the control variables in the model are
collected from the WORLDSCOPE data base. The information regarding management's plans to
engage in additional borrowings and equity issues is also retrieved from 10K’s.
5. RESULTS
5.1 Descriptive statistics and univariate results
Tables 3 and 4 contain the descriptive statistics for the test and control variables. The descriptive
statistics in Table 3 relate to the full sample (of distressed companies), whereas the descriptive
19
statistics in Table 4 are given for the going-concern and non-going concern samples separately.
Table 4 also reports the results of a t-test of differences between the going-concern and non-going
concern samples. Inspection of Table 3 reveals that the most common turnaround approaches in
our (full) sample of distressed firms are the cost reduction strategy (O-COSTRED, appearing in
60% of the sample firms) and the cooperative strategy (S-COOP, appearing in 61% of the sample
firms). All other approaches only occur in between 15 and 34 percent of the sample firms.
The results in Table 4 indicate that the companies that received a going-concern audit
report have a significantly lower current ratio (CR, t-statistic = 2.84), are more likely to have a
lower current ratio than the industry average value (INDCR, t-statistic = 4.83), are less likely to
be in default (DEFAULT, t-statistic = 3.71) and engage less frequently in additional borrowings
(BORROWING, t-statistic = 5.38). These results are consistent with going-concern opinions
being issued for distressed companies that face short term liquidity problems. Note that the non-
significant differences between the two samples with respect to the other financial distress
variables (other than liquidity measures) and total assets are supporting the efficacy of our
matching procedures.
Only a few turnaround activities appear to be significantly different for going-concern and
non-going concern companies. The strongest result is found for cost-cutting activities which are
significantly more common in the sample of going-concern firms (O-COSTRED, t-statistic =
1.92). The aggregate OPERATING variable is also (weakly) significantly higher for the going-
concern sample (OPERATING, t-statistic = 1.98). The occurrence of (long-term) strategic
activities is not significantly different between going-concern and non-going concern firms.
[ INSERT TABLE 3 ABOUT HERE ]
[ INSERT TABLE 4 ABOUT HERE ]
20
5.2 Multivariate logistic analysis
To test our hypotheses and assess which (type) strategic and operating variables are significantly
different between going-concern and other firms we estimate four logistic regression models. Our
choice of logistic regression analysis (instead of a probit analysis) is inspired by the matched
sampling approach we adopt. The use of logistic regression analysis neutralizes potential
problems resulting from oversampling going-concern companies relative to the population
proportion. In logistic regression analysis the coefficients of the independent variables will not
be affected by disproportionate sampling, only the intercept term is affected. However, since we
are not obtaining parameter estimates for the purpose of developing a predictive model, the bias
in the intercept term has no effect on our analysis (Maddala, 1991). We further also tested
whether there are multicollinearity problems between the independent variables that may affect
our results. Inspection of the correlation matrix (see Appendix B) indicates that most correlations
between the independent variables are below 30 percent. As there are some larger correlations,
we also calculated VIF factors, but all VIF scores are below 4.11.
We report the results from our multivariate logistic analyses in Tables 5 and 6. We
estimate four models. Model 1 is estimated to establish a base model for going-concern opinions
based on prior audit opinion studies, and thus mainly includes financial health variables and
variables that capture the ability to engage in additional borrowings and stock issues. Models 2, 3
and 4 are estimated to test our hypotheses and assess which (types of) strategic and operating
variables have incremental explanatory power beyond the control variables.
[ INSERT TABLE 5 ABOUT HERE ]
[INSERT TABLE 6 ABOUT HERE]
21
Model 1 has good explanatory power with a chi-square statistic equal to 66.44, a pseudo
R
2
equal to 57% and a McFadden R² equal to 42%. Only the current ratio (CR, p < 0.05), the
company’s current ratio relative to the industry (IND_CR, p < 0.05) operating cash flow over
total liabilities (CFOTL, p < 0.01), DEFAULT (p
Strategic actions and going-concern audit opinions e Liesbeth Bruynseels and Marleen Willekens.
Faculty of Economics and Applied Economics
DEPARTMENT OF ACCOUNTANCY, FINANCE AND INSURANCE (AFI)
Strategic actions and going-concern audit opinions
e
Liesbeth Bruynseels and Marleen Willekens
AFI 0611
Strategic Actions and Going-Concern Audit Opinions
Liesbeth Bruynseels
KULeuven
Marleen Willekens
Tilburg University and KULeuven
ABSTRACT
In this paper we examine the association between various types of strategic management actions
by distressed companies and the likelihood that they receive a going-concern audit opinion. Prior
going-concern studies that focus on the impact of non-financial information investigate particular
operating turnaround initiatives, such as cost reduction strategies (see, Behn et al., 2001; and
Geiger and Rama, 2003). We contribute to this literature by studying the impact of a broader set
of operating turnaround initiatives (i.e. cost reduction, asset disposal, increased marketing and
product upgrading), as well as a set of strategic growth initiatives (i.e. product innovation,
expansion and cooperative strategies). As the assessment of an auditee’s likelihood of survival
within the next twelve months is a critical in the going-concern decision making context, we
further distinguish between strategic growth initiatives that are likely to generate positive cash
flows in the short run (i.e. cooperative agreements) versus long run (i.e. innovation and expansion
strategies). Based on manually collected data on a sample of 114 distressed manufacturing US
firms, we find that operating turnaround initiatives, as well as strategic initiatives that are likely
to generate positive cash flows in the long run are positively associated with the likelihood that a
going-concern opinion is received. This evidence suggests that these two categories of
management initiatives are perceived as additional going-concern risk factors by auditors. On the
contrary, strategic turnaround initiatives that are likely to generate positive cash flows in the short
run, are negatively associated with the likelihood that a going-concern opinion is received, which
is supportive of this category of initiatives being perceived as mitigating factors.
Keywords: strategic management actions, going-concern audit opinion
Data availability: The strategic company information used in the paper is manually
collected from the 10-Ks filed with the SEC; the financial data are from
public sources identified in the paper.
We gratefully acknowledge the helpful comments and suggestions of Robert Knechel, Rashad Abdel-Khalik, Ann
Gaeremynck,, John Christensen, Wayne Landsman, Christian Leuz, Peter Pope, Diane Breesch, Valery Nikolaev and
participants at the 2005 EAA Annual Meeting, the 2005 Doctoral Colloquium, the 2005 ARNN conference, the 2005
EARNet Ph.d. colloquium, the 2006 BAA National Auditing Conference and the accounting workshop at Tilburg
University and the KULeuven.
Corresponding address: [email protected]
2
1. INTRODUCTION
In this study we investigate whether strategic actions taken by management of financially
distressed firms affect the auditors’ going-concern opinion decision. Hypothesizing a relationship
between client strategic actions and going-concern audit decisions is motivated by at least two
features. First, SAS no. 59 explicitly prescribes the consideration of contrary non-financial
(internal and external) matters and mitigating management plans in making going-concern
decisions. Therefore a broad set of events, actions and management plans – including strategic –
are potential determinants of going-concern opinion decisions. Second, changes in auditing
methodology and technology towards business risk auditing approaches in the second half of the
nineties (such as Strategic-Systems Auditing as introduced in Bell et al. 1997 and further
developed in Bell et al. 2005) further motivate the likelihood that strategic management actions
are an integrated part of audit evidence collection. Note that client strategic analysis
1
is one of the
most innovative aspects of business risk auditing methodologies. As the evidence collected from
strategic analyses is likely to have a substantial impact on subsequently planned and executed
audit procedures as well as the assessment of a client’s future financial viability, it is also very
likely to affect the auditor’s going-concern opinion decision.
It is well documented in the literature that auditors make going-concern decisions based
on reported financial results and compliance with financial obligations (see, for example,
Mutchler, 1985 and 1997; Levitan and Knoblett, 1985; Menon and Schwartz, 1987; Dopuch et
al., 1987; Bell and Tabor, 1991; Chen and Church, 1992; Gaeremynck and Willekens, 2003). The
importance of information other than that contained in the financial statements is also emphasized
in SAS no. 59. Besides the presence of negative financial trends and other indications of possible
financial difficulties – like for example default on loan agreements, SAS No. 59 also defines
certain (non-financial) internal and external matters as conditions and events that may indicate
1
or stated alternatively, acquiring evidence of and from the entity business states (EBS) as advocated in Bell et al.
(2005).
3
that there could be substantial doubt about the entity to continue as a going-concern.
Subsequently, SAS no.59 also requires auditors to consider management plans to mitigate the
effects of these adverse conditions or events when assessing their client’s ability to continue as a
going-concern. The impact on the audit opinion of contrary and mitigating factors in publicly
available disclosures such as the financial press, 10-K’s or management discussions and analyses
has also been documented in the going-concern literature (see Mutchler et al., 1997). Although
the importance of strategic management plans is recognized in today’s auditing practice, research
on the impact of forward-looking management plans on going-concern decisions is scant. Behn et
al. (2001) recognized this caveat and provide evidence of the relationship between the likelihood
of going-concern opinions and a company’s ability to obtain new financing and to reduce costs.
After controlling for financial condition, size, default status, and the propensity to voluntarily
disclose information, their results indicate that going-concern opinion decisions are strongly
linked to publicly available mitigating information regarding certain management plans. In
particular, plans to issue equity and to borrow additional funds exert the strongest association
with the issuance of an unqualified opinion. Recently, Geiger and Rama (2003) report that
companies are more likely to receive a modified report if they entered into a cost reduction plan
or sold off significant assets. However, contrary to the findings of Behn et al. (2001), plans for
the issuance of new debt or equity are not significantly associated with the auditor’s opinion type.
The increased relevance of strategic parameters in the audit decision making context in
general, is attributable to changes in the scope and methodology of auditing that have taken place
in (a number of) large accounting firms in the second half of the nineties (see, for example, Bell
et al. 1997; Lemon, Tatum and Turley 2000; Knechel, 2001; Bell et al. 2005; Curtis and Turley
2005). Whereas traditional auditing approaches adopt a bottom-up focus thereby directing
attention to the nature of account balances, classes of transactions, and properties of the client’s
accounting system, business risk auditing develops a top-down holistic perspective of the client’s
4
business and industry. This entails a thorough analysis of the client’s business and strategic
position. Note that a general evolution towards business risk auditing elements is reflected in
some of the new International Audit Risk Standards. In particular, International Standard on
Auditing (ISA) 315 requires the auditor to develop an understanding of client objectives and
strategies, as well as the related business risks that may result in a material misstatement of the
financial statements. These business risks should not only be evaluated in light of their immediate
consequences for the risk of material misstatements, but also with regard to their longer-term
consequences.
We contribute to the going-concern literature by testing the association between the
likelihood of going-concern opinions and a comprehensive set of strategic actions for a sample of
distressed US manufacturing companies. Like other going-concern studies, we rely on
information disclosed by management in the management discussion and analysis (MD&A), and
remainder of the 10-K (Behn et al., 2001; Geiger and Rama, 2003). Consistent with the strategy
literature
2
(see for example, Barker and Duhaime, 1997; Robbins and Pearce II, 1992;
Sudarsanam and Lai, 2001; Bruton et al. 2003) we distinguish between management actions and
plans aimed at a short-term improvement in financial performance (or, operating turnaround
initiatives) and strategic growth initiatives (or, strategic turnaround initiatives). As going-concern
decision making involves the assessment of the likelihood of survival of an auditee within the
next twelve months, we further sub-divide strategic growth initiatives in those that are likely to
generate positive cash flows in the short run (i.e. cooperative agreements) versus long run (i.e.
innovation and expansion strategies).
Consistent with prior going-concern studies which focused on short-term improvement
(or, operating) initiatives (see, Behn et al. 2001; Geiger and Rama, 2003), we find that cost
reduction strategies are positively associated with the likelihood of receiving a going-concern
2
The strategy literature contains an extensive body of research that focuses on how firms reverse firm-threatening
performance declines. See also hypothesis development section.
5
opinion. In addition, we also find that marketing strategies are positively associated with the
likelihood of receiving a going-concern opinion. Our test of the growth initiatives reveals that the
engagement in cooperative agreements is negatively associated with the likelihood of receiving a
going-concern opinion. Our results are consistent with cooperative agreements providing a
positive signal about the going-concern status of the firm and therefore can be interpreted as a
mitigating factor, whereas the presence of cost reduction and marketing strategies are perceived
as additional going-concern risk factors which increase the likelihood to receive a going-concern
opinion.
Finally, we also test the impact of three types of aggregated construct variables capturing
the presence short-term initiatives, strategic growth initiatives and vvv. We find that more
operating turnaround initiatives taken by a distressed firm are associated with a higher likelihood
that a going-concern opinion is issued. A similar result is obtained for strategic initiatives not
capable of generating short-term financial impact. On the contrary, the presence of strategic
turnaround initiatives that are likely to generate a financial impact in the short run is negatively
associated with the likelihood that a going-concern opinion is issued. Thus, our evidence suggests
that auditors perceive the engagement in operating initiatives and strategic growth initiatives that
only yield a financial impact in the long term as additional going-concern risk factors, but the
engagement in strategic growth initiatives that generate a short-term financial impact as a
mitigating factor.
The remainder of this paper is organized as follows. In the next section we develop our
hypotheses. Section 3 then is devoted to the development of the going-concern opinion model
that is tested in this paper. Next, in Section 4 we provide an overview of our sample selection
procedure and data collection approach. In Section 5 we discuss the results of our analyses. We
conclude in Section 6.
6
2. HYPOTHESES DEVELOPMENT
SAS no. 59 clearly states that besides financial indicators – such as negative (financial) trends
and other indications of possible financial difficulties – non-financial internal and external
matters are relevant conditions and events to assess the going-concern status of a client firm.
Examples of internal matters that are included in SAS No. 59 are work stoppages or substantial
dependence on the success of a particular project. External matters listed in SAS No. 59 include,
for example, legal proceedings or the loss of a key franchise, license or patent. Furthermore,
when the identified conditions and events in the aggregate lead to substantial doubt about the
continued existence of the entity as a going-concern, the auditor should identify and evaluate
management’s plans to mitigate the effects of these adverse conditions or events. If the auditor
believes that there exist management plans that overcome this substantial doubt, a going-concern
audit report is not required. Examples of such potentially mitigating management plans are
included in SAS no.59, and relate to the sale of assets, the borrowing or restructuring of debt, the
reduction of expenditures and the increase of ownership equity.
A few prior studies provide evidence that auditors are indeed committed to reviewing
management plans that are dealing with adverse conditions or events when assessing a client’s
ability to continue as a going-concern (Behn et al., 2001; Geiger and Rama, 2003). However,
these studies are confined to assessing the impact of examples of management plans and actions
that are explicitly mentioned in SAS no.59. In this paper, we elaborate on this theme and
investigate the impact of a broader set of potentially contrary or mitigating actions and strategies
on the auditor’s going-concern decision. We motivate this broader strategic focus by the
emergence of business risk auditing in the 1990s. With the emergence of business risk auditing,
traditional auditing methodologies have been complemented with new audit processes based on a
top-down, holistic perspective of the client’s business and industry (see, for example, Bell et al.
1997; Knechel 2001; Lemon, Tatum and Turley 2000). The most innovative aspect of business
7
risk auditing is the assessment of client strategic viability, which can have a substantial impact on
the subsequent audit procedures and the assessment of future financial viability.
To predict the impact of a comprehensive set of viable strategic actions on the going-
concern opinion, we categorize management actions into strategic and operating turnaround
approaches. This is a widely used framework introduced by Hofer (1980). Note that an operating
approach focuses on internal, operating problems of firms through – for example – decreasing
costs, increasing efficiency, disposing assets, or improving sales (Hofer, 1980). A strategic
turnaround approach is aimed at long-term profitability by solving external, strategic problems
through for example a change in the strategic direction of the firm, its positioning, alliances and
product lines (Bruton et al., 2003). Strategic repositioning may be done through business
divestments, acquisitions, alliances, new product development, new markets, and increased
market penetration. Firms experiencing financial distress may adopt a variety of strategies to
return to financial health. The strategy literature offers an extensive body of research that focuses
on how firms reverse firm-threatening performance declines to induce successful company
turnaround
3
(see for example, Barker & Duhaime, 1997; Robbins and Pearce II, 1992;
Sudarsanam & Lai, 2001; Bruton et al. 2003).
Hypothesis 1: Operating turnaround approaches and going-concern opinions
An operating approach to company turnaround typically consists of actions related to cost
reduction, revenue generation and operating-asset reduction. The focus is on achieving short-term
financial relief, without considering long-term changes in the organization’s strategy. In order to
achieve short-term profitability improvement, companies have the opportunity to engage in
classic retrenchment activities such as: divestment, product elimination, cost rationalization and
employee layoffs. In addition to these cost-cutting initiatives, revenue generating strategies may
3
Successful turnaround is defined as the reversal of a firm’s pattern of performance decline (Schendel, Patton and
Riggs, 1976).
8
be pursued focusing on existing lines of products, price-cutting, increased marketing expenditure
or increased direct sales efforts (Hofer, 1980).
Prior studies that examine the association between the implementation of operating
approaches and successful company turnaround have focused on retrenchment activities and
provide mixed results. Several studies report that classic retrenchment strategies are significantly
associated with turnaround success (see, for example, Robbins and Pearce II, 1992), whereas
other studies cast doubt on the value of operating approaches as part of a company’s turnaround
approach (Sudarsanam and Lai, 2001; Barker III and Mone, 1994).
The mixed evidence from the strategy literature indicates that operating turnaround
strategies per se may not be capable of curing deficiencies in a declining firm’s strategic
orientation. In other words, if a declining firm’s problems relate to its strategic positioning, these
short-term cures could be inadequate, given that changing a firm’s strategic orientation is a
prerequisite to recovery (see also Schendel et al., 1976; Hofer, 1980; Barker and Duhaime, 1997).
As we investigate the auditor’s perception of the effectiveness of operating turnaround
approaches
4
for distressed firms, a relevant question is which signal such approaches by
themselves send to the auditor regarding the going-concern status of the company. Given the
evidence reported above, it is likely that auditors perceive operating turnaround strategies as
insufficient to induce recovery for distressed firms. This is indeed consistent with the finding
reported by Geiger and Rama (2003), i.e. that cost-cutting or asset disposal activities are
associated with a higher likelihood of receiving a going-concern audit report. This leads to our
first hypothesis:
4
These include cost-cutting activities, disposal of assets, increasing marketing efforts and improving existing
products and operating processes – see also Section Model Specification.
9
H1: For financially distressed companies, the implementation of an operating turnaround
approach is likely to increase (ceteris paribus) the propensity that a going-concern
opinion is issued.
Hypothesis 2: Strategic turnaround approaches and going-concern opinions
Overall, the evidence from the strategy literature suggests that (long-term) strategic turnaround
approaches are successful turnaround vehicles. Barker III and Duhaime (1997) find that when a
company’s decline is firm-based and not caused by industry contraction, recovering firms
implement more extensive strategic changes (which are consistent with reorientation).
Sudarsanam and Lai (2001) provide evidence that firms recovering from financial distress
typically adopt more forward-looking, expansionary and external market focused strategies than
non-recovery firms. More specifically, recovery firms typically adopt growth-oriented and
external-market focused strategies, whereas non-recovery firms continue to engage in operating
restructuring strategies. Given the evidence from the strategy literature about the effectiveness of
strategic approaches for company turnaround and recovery, it is reasonable to expect that such
strategies may also have a mitigating impact on the auditor’s going-concern opinion. However, as
the auditor’s going-concern opinion is an assessment of the client’s ability to survive during the
next 12 months, only those (long-term) strategic approaches that are expected to have a positive
impact on the company’s liquidity status within the next 12 months will be perceived as
mitigating factors. It is therefore necessary to further examine the short-term impact of the
different types of long-term strategic approaches, i.e. cooperative agreements, product innovation
and acquisition strategies.
Barker III and Duhaime (1997) emphasize that cooperative agreements with other firms
are an essential element of a turnaround approach based on strategic change. Examples of
cooperative strategies include long-term contractual agreements with suppliers or buyers,
10
alliances or joint ventures, subcontracting and technology licensing agreements
5
. Prior research
about the consequences of the implementation of a cooperative strategy has shown that strategic
networks such as strategic alliances, joint-ventures and long-term buyer-supplier relationships
often have positive effects on different measures of corporate performance. For example,
Mitchell and Singh (1996) reported evidence of alliances raising organisational survival rates.
Powell et al. (1996) found that companies which had formed many alliances experienced
accelerated growth rates. Gulati et al. (2000) highlight the idea that one of the most important
benefits of strategic networks is the increased access to information, resources, markets and
technologies. In addition to access to resources, Stuart (2000) found that strategic alliances also
affect firm subsequent-period performance through their influence on an organization’s
reputation, particularly if the firm is of ambiguous quality.
Capon et al. (1992) report that new product development and strategic acquisitions are
often linked to (long-term) performance improvement, and report evidence that suggests that both
strategies act as substitutes in terms of effectiveness vis-à-vis company turnaround (i.e. non-
innovative firms that involve in acquisitions perform nearly as well as those that engage in
product innovations). However, the short-term performance impact of both types of strategic
actions is less apparent. In a recent meta-analytic review of merger and acquisition performance,
King et al. (2004) report that acquisitions are not improving the short-term financial performance
of acquiring firms, on average
6
. With respect to the short-term performance impact of product
innovation, Mishina et al. (2004) even report a negative association with the rate of short-term
sales growth.
5
Note that strategic alliances are a popular financing vehicle for companies in financial distress, as partnering up
with a successful healthy company can provide distressed companies with additional funding to develop or market
products, or with other benefits such as a more extensive customer base (see, for example, Bruton et al. 2003).
Another vehicle to improve financial position is engaging in a licensing strategy with regard to unused or high-risk
technology. Licensing out proprietary technology can substantially improve a company’s financial position as it
periodically receives royalties and/or milestone payments (see, for example, Sudarsanam and Lai 2003). A company
can also safeguard future sales by engaging into long-term contracting with buyers or distributors (Miller, 1992).
6
Instead, this study indicates that acquisitions either have no significant effect or a modest negative effect on an
acquiring firm’s financial performance in the post-announcement period.
11
To summarize, the evidence from the strategy literature suggests that (long-term) strategic
turnaround approaches are successful turnaround vehicles. However, the short-term performance
impact of different types of strategic turnaround approaches varies. Based on prior research it is
reasonable to expect that the introduction of new products and corporate acquisitions are less
likely to have a positive financial impact within the next 12 months, whereas it is more likely that
cooperative agreements generate positive cash flows within the next 12 months. This leads to our
second hypothesis:
H2: For financially distressed companies the implementation of a strategic turnaround
approach is likely to reduce (ceteris paribus) the propensity that a going-concern audit
opinion is issued, given that the strategy is likely to have a mitigating impact within the
next 12 months.
3. MODEL SPECIFICATION AND VARIABLE MEASUREMENT
We use the following logistic model to test our two hypotheses:
REPORT = f (operating turnaround variables, strategic turnaround variables, control variables).
REPORT is an indicator variable that takes a value equal to one if the auditor issues a going-
concern report, and zero otherwise. The turnaround approach variables contain information
regarding turnaround strategies that have been implemented by the company during the year
under audit to overcome adverse conditions affecting corporate performance. This information is
manually collected from corporate disclosures in the annual report and 10-K. We investigate the
impact of two categories of turnaround strategies that can potentially mitigate the adverse
conditions affecting company performance. In the category of operating turnaround strategies,
we consider the impact of cost-cutting and asset disposal activities, product and operating process
12
improvements and increasing marketing efforts. As strategic turnaround approaches, we consider
cooperative agreements with other firms, the introduction of new products, mergers, and
acquisitions. Finally, the control variables in our model encompass factors that have been found
to be associated with going-concern opinion decisions and the propensity to voluntarily disclose
information in prior research.
[ INSERT TABLE 1 ABOUT HERE ]
3.1 Operating turnaround variables
Our classification of operating turnaround variables is based on Hofer (1980) who distinguishes
between four different types of operating turnaround approaches: a) cost-cutting strategies, b)
asset reduction strategies, c) revenue increasing strategies and d) combination strategies.
Accordingly, we include and test variables reflecting a cost reduction strategy (O-COSTRED), an
asset disposal strategy (O-DISPOSE), a commercial strategy (O-COMMERCIAL), and a strategy
aimed at the improvement of existing products and processes (O-UPGRAD).
We define O-COSTRED as an operating turnaround variable that captures significant cost
reduction efforts. In particular, this variable relates to both employee layoffs and other cost
reduction efforts during the year under audit. O-COSTRED is set equal to one if the company
reports cost reduction strategies for the year under audit, and is set equal to zero otherwise. O-
DISPOSE is defined as an operating turnaround variable that indicates whether a company
engages in the sale of significant assets. O-DISPOSE is set equal to one if the company reports
the sale of assets for the year under audit, and is set equal to zero otherwise. As opposed to
strategic actions aimed at reducing expenditures, short-term operating strategies also include a
number of revenue generating activities (Hofer, 1980). Subsequently, we define O-
13
COMMERCIAL as an operating turnaround variable that indicates whether a company increases
its marketing efforts, and O-UPGRAD as a variable that relates to the realisation of
improvements in existing products and production processes. O-COMMERCIAL is set equal to
one if the company reports increased marketing efforts for the year under audit, and is set equal to
zero otherwise. O-UPGRADE is set equal to one if the company reports product and/or process
improvements for the year under audit, and is set equal to zero otherwise. In order to extend and
refine our analyses, we also define a number of strategic construct variables. This will enable us
to test the aggregate impact of strategic turnaround variables that have similar characteristics on
theoretical grounds. OPERATING is defined as the sum of O-COSTRED, O-DISPOSE, O-
COMMERCIAL and O-UPGRADE, scaled by its maximum value in the sample. Note that a
further refinement in short-term versus long-term impact variables would be tautological in the
context of operating initiatives, as they are all expected to have a short-term impact on financial
performance.
3.2 Strategic turnaround variables
Long-term strategic turnaround approaches typically relate to reconfiguration of the assets
and/or the corporate portfolio, and product and/or market refocusing. S-EXPANSION is a
strategic turnaround variable capturing whether a company engages in an acquisition strategy (of
other companies) to accelerate growth. S-EXPANSION is set equal to one if the company
reports acquisitions for the year under audit, and is set equal to zero if such is not the case. We
also include S-COOP, a variable that indicates whether a company enters into strategic alliances,
joint-ventures, licensing agreements and other cooperative arrangements. S-COOP is set equal to
one if a company entered in cooperative arrangements during the year under audit, and is set
equal to zero if such is not the case. Further, we define S-PRODUCT as a strategic turnaround
variable that assesses whether a company has recently introduced new products. S-PRODUCT is
14
set equal to one if the company reports the introduction of new products during the year under
audit, and is set equal to zero if such is not the case.
Finally, we also introduce strategic construct variables to capture the aggregate impact of
several strategic turnaround initiatives engaged in by the audited firm. First, we define a (rough)
strategic construct variable STRATEGIC to capture the aggregate impact of all strategic
turnaround initiatives that have been implemented by the audited company during the past year.
This categorization is based on the strategic literature (see, Hofer, 1980). STRATEGIC is a
discrete variable representing the sum of all strategic turnaround variables defined above, scaled
by its maximum value in the sample. In a more refined categorization of strategic variables, we
distinguish between strategic initiatives that are expected to have a positive impact on firm
financial performance in the short term (STRAT_ST) or in the long term (STRAT_LT). This
categorization can be motivated by recent evidence in the strategy literature (see, King et al.
2004; Mishina et al. 2004) that acquisitions and product innovations are not improving the short-
term financial performance of acquiring and innovating firms, respectively. Such a distinction in
short-term and long-term financial impact is warranted given the going-concern decision context
that we are investigating. STRAT_ST is an indicator variable which happens to coincide with S-
COOP, as the presence of cooperative agreements is the only strategic turnaround variable
defined in this paper that is likely to generate a financial impact within the next twelve months.
STRAT_LT is defined as the sum of S-PRODUCTS and S-ACQUIS, divided by its maximum
value in the sample. We also refer to Table 1, for an overview of definitions of the strategic
variables.
3.3 Control variables
The issuance of a going-concern opinion is obviously conditional upon the auditee’s financial
condition. Therefore a first category of control variables that are included in our model capture
15
the financial condition of the firm. Based on prior audit opinion studies (see, for example,
Mutchler 1985, 1997; Chen and Church, 1992) we include cash flow from operations divided by
total liabilities (CFO/TL), the current ratio (CR), and long-term debt divided by total assets
(LTD/TA), as control variables. Following Menon and Schwartz (1987), we also include a
change variable, namely the change in current ratio (?CR). In line with Bell and Tabor (1991),
we also control for a company’s liquidity performance relative to the industry, by including an
indicator variable (INDCR), taking the value of one if the current ratio of the company exceeds
the industry median current ratio. As in Chen and Church (1992), we also add DEFAULT, an
indicator variable that takes a value equal to one if the company defaults on debt payments or is
in technical default of loan covenants
7
, and zero otherwise. Following prior research, we also
include the log of total assets to control for company size (see, for example, Chen and Church,
1992).
A second category of control variables constitute mitigating factors identified in prior
audit opinion research. Behn et al. (2001) find that plans to use existing bank lines of credit and
other approved lines of credit are negatively associated with the likelihood of a going-concern
opinion. Consistent with Behn et al. (2001) we include BORROW, a variable that is set equal to
one if the auditee plans to borrow funds through existing bank lines of credit or other approved
debt instruments; and STOCK, a variable that is set equal to one if the auditee plans to issue
equity through existing or committed arrangements.
4. SAMPLE AND DATA
4.1 Sample selection
Consistent with prior going-concern studies (see, for example, Mutchler, 1985; Chen and Church,
1992; and Behn et al., 2001) we select a sample adopting a matched pair design. Note that a
7
A company’s default status was determined by reading the MD&A and debt footnotes in the financial statements.
16
matched pair design is most often used when the research design necessitates manual data
collection, as is the case in this study. In particular, we first selected a sample of companies that
received a first-time going-concern opinion and then a matched sample of distressed companies
that did not receive a going-concern opinion.
Selection of going-concern firms
We identified all firms from the Worldscope database that are listed on AMEX, NASDAQ and
NYSE and received a going-concern audit opinion in the period 1998-2001. Consistent with
prior studies (Mutchler and Williams 1990; Behn et al. 2001; Blay and Geiger 2001), we
restricted our sample to companies in the manufacturing industries (SIC 20 to 39) to eliminate
confounding industry effects. This resulted in an initial sample of 276 manufacturing companies
that received a going-concern audit opinion in fiscal years from 1998-2001. From this initial
sample, we then eliminated companies that received a going-concern report in the previous year
to control for potential confounding effects from prior going-concern opinions (see also,
Mutchler, 1985, Blay & Geiger, 2001; Behn et al., 2001). In addition, we also excluded firms that
faced bankruptcy proceedings as the decision to issue a going-concern opinion is trivial for such
firms, and firms for which no match could be identified. This resulted in 57 firms with going-
concern opinions: 8 firms in 1998, 8 firms in 1999, 15 firms in 2000, and 26 firms in 2001.
Selection of control firms
To test our going-concern model, we matched the going-concern sample with a sample of
distressed companies that did not receive a going-concern report. We searched the Worldscope
database from 1998 to 2001 to identify all manufacturing companies listed on NASDAQ, NYSE
or AMEX that received a clean audit opinion. Consistent with prior studies, we further restricted
the control sample to firms in financial distress (see, for example, Mutchler, 1985; Chen &
17
Church, 1992; Behn et al., 2001; McKeown et al., 1991). Based on Chen and Church (1992) we
adopt the following criteria to identify distressed companies: 1) negative retained earnings, 2)
negative operating income, 3) negative net income , 4) negative working capital, 5) negative net
worth, and 6) negative operating cash flows. Note that Chen and Church (1992) classified a
company as stressed if it meets at least one of above criteria. We use a more stringent rule for
financial distress, by classifying a company to be stressed if it meets at least two of these stress
criteria. This procedure yielded 2929 “distressed” companies that received a clean opinion during
the period 1998-2001.
As we use a matched-pair design, we matched control-sample companies to the going-
concern firms based on year, size (proxied by total assets) and two-digit SIC classifications. This
procedure ensures that we include similar companies with respect to size and industry in both
sub-samples. A matched group design has been used previously by Mutchler (1985), Chen and
Church (1992), Behn et al. (2001) and Geiger and Rama (2003). A limitation of this approach is
that it overstates the issuance of a going-concern opinion in this experimental setting. Note that
we address this issue by our statistical approach, as we adopt logistic regression analysis (see
further).
[ INSERT TABLE 2 ABOUT HERE ]
4.2 Data collection and strategic scorecard
As strategic company information is not publicly available, we manually collected this
information from the relevant 10-Ks filed with the SEC, by reading these documents back-to-
cover and completing a strategic scorecard. The strategic scorecard used to test our hypotheses is
included in Appendix A. For each of the defined turnaround approaches we assessed whether the
company has engaged in actions related to that specific turnaround initiative during the year
18
under audit. Per initiative, a score equal to one is assigned if a firm discloses such action, and
zero otherwise.
The suitability of 10-K filings for strategic information collection is supported by prior
studies that investigated the association between disclosures in the MD&A and future corporate
financial performance. These studies report evidence that indicates that the information content of
narrative disclosures in the annual report is significantly associated with future viability of
distressed firms (see, for example, Tennyson et al., 1990; Boo and Simnett, 2002). Boo and
Simnett (2002) examine the reliability of management’s prospective comments for a sample of
140 Australian public companies that had experienced significant losses. Their results indicate
that management’s prospective comments have significant information content with respect to the
company’s future viability. Note also that SAS No. 8 requires auditors to ensure that the ‘other
information’ attached to financial statements is not materially inconsistent with the financial
statements, and does not contain any material misstatement or fact. Furthermore, the costs of
potential litigation and loss of reputation are important factors to prevent management from
disclosing misleading information.
Note that the dependent variable and (most of) the control variables in the model are
collected from the WORLDSCOPE data base. The information regarding management's plans to
engage in additional borrowings and equity issues is also retrieved from 10K’s.
5. RESULTS
5.1 Descriptive statistics and univariate results
Tables 3 and 4 contain the descriptive statistics for the test and control variables. The descriptive
statistics in Table 3 relate to the full sample (of distressed companies), whereas the descriptive
19
statistics in Table 4 are given for the going-concern and non-going concern samples separately.
Table 4 also reports the results of a t-test of differences between the going-concern and non-going
concern samples. Inspection of Table 3 reveals that the most common turnaround approaches in
our (full) sample of distressed firms are the cost reduction strategy (O-COSTRED, appearing in
60% of the sample firms) and the cooperative strategy (S-COOP, appearing in 61% of the sample
firms). All other approaches only occur in between 15 and 34 percent of the sample firms.
The results in Table 4 indicate that the companies that received a going-concern audit
report have a significantly lower current ratio (CR, t-statistic = 2.84), are more likely to have a
lower current ratio than the industry average value (INDCR, t-statistic = 4.83), are less likely to
be in default (DEFAULT, t-statistic = 3.71) and engage less frequently in additional borrowings
(BORROWING, t-statistic = 5.38). These results are consistent with going-concern opinions
being issued for distressed companies that face short term liquidity problems. Note that the non-
significant differences between the two samples with respect to the other financial distress
variables (other than liquidity measures) and total assets are supporting the efficacy of our
matching procedures.
Only a few turnaround activities appear to be significantly different for going-concern and
non-going concern companies. The strongest result is found for cost-cutting activities which are
significantly more common in the sample of going-concern firms (O-COSTRED, t-statistic =
1.92). The aggregate OPERATING variable is also (weakly) significantly higher for the going-
concern sample (OPERATING, t-statistic = 1.98). The occurrence of (long-term) strategic
activities is not significantly different between going-concern and non-going concern firms.
[ INSERT TABLE 3 ABOUT HERE ]
[ INSERT TABLE 4 ABOUT HERE ]
20
5.2 Multivariate logistic analysis
To test our hypotheses and assess which (type) strategic and operating variables are significantly
different between going-concern and other firms we estimate four logistic regression models. Our
choice of logistic regression analysis (instead of a probit analysis) is inspired by the matched
sampling approach we adopt. The use of logistic regression analysis neutralizes potential
problems resulting from oversampling going-concern companies relative to the population
proportion. In logistic regression analysis the coefficients of the independent variables will not
be affected by disproportionate sampling, only the intercept term is affected. However, since we
are not obtaining parameter estimates for the purpose of developing a predictive model, the bias
in the intercept term has no effect on our analysis (Maddala, 1991). We further also tested
whether there are multicollinearity problems between the independent variables that may affect
our results. Inspection of the correlation matrix (see Appendix B) indicates that most correlations
between the independent variables are below 30 percent. As there are some larger correlations,
we also calculated VIF factors, but all VIF scores are below 4.11.
We report the results from our multivariate logistic analyses in Tables 5 and 6. We
estimate four models. Model 1 is estimated to establish a base model for going-concern opinions
based on prior audit opinion studies, and thus mainly includes financial health variables and
variables that capture the ability to engage in additional borrowings and stock issues. Models 2, 3
and 4 are estimated to test our hypotheses and assess which (types of) strategic and operating
variables have incremental explanatory power beyond the control variables.
[ INSERT TABLE 5 ABOUT HERE ]
[INSERT TABLE 6 ABOUT HERE]
21
Model 1 has good explanatory power with a chi-square statistic equal to 66.44, a pseudo
R
2
equal to 57% and a McFadden R² equal to 42%. Only the current ratio (CR, p < 0.05), the
company’s current ratio relative to the industry (IND_CR, p < 0.05) operating cash flow over
total liabilities (CFOTL, p < 0.01), DEFAULT (p