


































UAFS Advances in Business Research 2012 LOWREZ (2).pdf


Durden and Sennetti                                                                                                                                                      Advances in Business Research 

                                                                                                                                                                                               2012, Vol. 3, No. 1, 79-89 

79 

 

Audit Evidence Supporting the Missing Going Concern Opinions: Bankrupt California 

Public School Districts 
 

S. Joseph Durden, CPA, PC 

John Sennetti, Nova Southeastern University 
 

 

Because of asymmetric payoffs to the auditor, bankrupt local governmental units are unlikely to receive going 

concern audit opinions (GCOs). To increase this likelihood¸ we propose a framework to increase the evidence to 

support this opinion. We then apply this to 400 California School District reports, 54 of which were used to declare 

bankruptcy by 2008. Typically, because of the impending bankruptcy, otherwise reliable State financial information 

becomes unreliable. We address this and then recommend a GCO in 2008 for thirteen of nineteen districts which 

eventually went bankrupt. We confirm as available: 1) the missing evidence and provide, and 2) an improved 

information model demanded by auditing research and practice.   
 

 

Local governments (cities and school districts) across the United States are failing to meet their cash obligations, 

but few are receiving going concern audit opinions (GCOs). Fifty-four (13.5 percent) of the over 400 possible school 

district reports in California from 2003 to 2008 actually received bankruptcy designations, and but few of these 

received GCOs (California Department of Education, 2009). Even though the State of California failed to pay its 

obligations when they came due, i.e., failed as a going concern, the state’s financial statements did not receive a 

going concern opinion (California State Controller’s Office, 2009; Luhby, 2009). 

This audit opinion is not intended to predict economic failures, but is relied upon by vendors (teachers, parents 

and more) as a warning as to possible nonpayment to them when their payments become due. The official guidance 

for these audit opinions explicitly states the absence of GCO does not imply a bankruptcy will not occur [American 

Institute of Certified Public Accountants (AICPA) Statement on Auditing Standards  (SAS) No. 59, “The Auditor’s 

Consideration of an Entity’s Ability to Continue as a Going Concern,” AICPA, 1988; see also Gauthier 2005; and 

Governmental Accounting Standards Board (GASB) 2009, Statement No. 56, “Codification of Accounting and 

Financial Reporting Guidance contained in the AICPA Statements on Auditing Standards”, GASB, 2009]. Because 

these opinions report on the presentation of the recent year’s financial results, auditors must use currently available 

evidence, available at most a few months past the year end date of the financial statements, to determine whether the 

governmental unit will be able to meet short-term financial obligations as they become due.   
 

Arguments for and Against Reporting the Opinion 
 

The GASB Statement No.56, instructs auditors to issue these opinions after observing threatening events, such as 

bond payment defaults, major debt and tax crisis events, noncompliance with statutory capital or reserve 

requirements, legal proceedings or regulatory proceedings. All this reduces the likelihood of a GCO unless a 

declared bankruptcy has already occurred, particularly since the GCO may ‘cause’ bankruptcy, signaling just the 

potential of a bankruptcy and frightening creditors, and hence the auditor has greater payoffs not to report this 

opinion, especially when this report may discontinue the relationship with this client. 

Against this likelihood, the national financial crises affecting local governments should increase the evidence 

and demand by creditors, employees’ unions, taxpayers, and all other stakeholders for more going concern opinions.  

Governments missing tax revenues from foreclosed and reduced tax-valued houses have closed police stations in 

California (Sanchez, 2010), and states have assumed school district debts. Going concern opinions for school 

districts alert employees to possible job losses, students to possible non accredited high school degrees, and 

taxpayers to abnormal salaries for officials, as in the cases of Bell, California (Gore, 2009; Roger, 2010).   

This paper responds to this national stakeholder demand for more GCOs.  We consider the general monitoring 

process created by the states, one that permits variables to be modeled as part of the auditor’s an analytical work.  

With this we increase the evidence and hence the likelihood for the going concern opinion for governmental units.  

We then apply this process to the financially-distressed California school districts to find a sufficient framework to 

increase the going concern opinion evidence. The next section and corresponding sections discuss the relevant 

research, hypotheses, methodology, results and their summaries. 
 

 
 

 



Durden and Sennetti                                                                                                                                                      Advances in Business Research 

                                                                                                                                                                                               2012, Vol. 3, No. 1, 79-89 

80 

 

BACKGROUND AND HYPOTHESIS 
 

General Overview 
  

 Like non-governmental audits, governmental audits do not require an analysis to predict the future viability of 

the local governmental unit being audited. GASB 56 requires certain footnote and MD&A disclosures, however 

these requirements are still ex-post in nature and not predictive of an event that has already occurred. Because of 

this, there was no warning or cautionary communication,  to vendors, employees, police, fire departments, insurers 

of homes, local utilities and other vendors of services and products from the local government bankruptcies of the 

City of Prichard, AL; City of Vallejo, CA, City of Dunmore, PA, Emery Unified School District, CA, West Contra 

Costa Unified School District, CA, Compton Unified School District, CA, St Louis School District, MO, and the 

Richmond Unified School District, CA, just to name a few.  

Increasing the demand for predictive GCOs began in the United States with the rising inflationary crises of the 

1970’s, when costs exceeded revenues (and similar to recent deflationary crises with declining revenues). Practicing 

auditors requested a going concern opinion for those units subject to governmental and public audits (Belluomini, 

1977). Absent prior research of these governmental issues, the auditors adopted for-profit opinions (SAS No. 59) 

and their predictive models, although predictive models for governmental units are now available (Trussel and 

Patrick, 2009). The for-profit research on these opinions consider the bankruptcy probability (Hopwood, McKeown, 

and Mutchler, 1994) from perceived variables from other stressed companies (Mutchler, 1984), the predictability of 

future going concern issues based on prior financial distress (Geiger et al., 2005), the influences of contrary 

information  (Mutchler, Hopwood, and McKeown, 1997),  publicly available information (Mutchler, 1985), and user 

interpretation of the going concern opinions (Ponemon and Raghunandan, 1994).   
 

Whether to Report a Going Concern Opinion: Agency and Other Issues 
 

In addition to the available economic information, the auditor considers whether the effect of the opinion may 

increase the likelihood of these past negative conditions and bankruptcy, and thereby increasing the corresponding 

legal liability (Mutchler, 1985; Raghunandan and Rama, 1995; Akers. Maher, and Giacomino, 2003; Carcello and 

Neal, 2003; Bellovary, Giacomino, and Akers, 2006). A lawsuit from or dismissal by the client may occur even in 

the presence of predictive evidence (Akers et al., 2003; Carcello and Neal, 2003; Raghunandan and Rama, 1995; 

Bellovary et al., 2006). In some cases, these conditions create opinion shopping (Carcello and Neal 2003; Bellovary 

et al., 2006).  
 

Type I and Type II Decision Errors  
 

Two types of misclassifications or auditor errors may occur in reporting. A Type 1 misclassification occurs when 

the client remains viable after the issuance of a going concern opinion. Conversely, a Type II misclassification 

occurs when the client declares bankruptcy after the issuance an unqualified opinion for the period preceding the 

bankruptcy filing (Geiger et al., 2005). Type II misclassifications are common while Type I are not. Less than half 

of businesses filing for bankruptcy received a going concern opinion for the preceding financial statements in the 

1980’s (Hopwood et al. 1989; Raghunandan and Rama, 1995; Geiger et al., 2005).  
 

Asymmetric Payoffs for Type I and Type II Decision Errors  
 

The auditor has a knowledge advantage over the principal (school district taxpayers) because of the “expertise, 

functional indispensability, and intrinsic ambiguity associated with the services they provide” (Sharma, 1997, 768).  

The cost to the auditor of a Type I decision is likely a loss of a client. The cost of a Type II decision to the auditor 

for government auditors, is likely minimal, since their lawsuits for these audits are rare, although the taxpayers may 

try to harm the auditors’ reputations. Opportunism emerges as the auditors decide in their best interests to not issue a 

GCO.  

 When a GCO is not given for bankrupt corporations, lawsuits from taxpayers, debt holders and more also arise, 

but SAS No.59 argues that when one is given it does not mean that the corporation will be solvent, and in this case 

the auditor is less likely dismissed. When a GCO is not given for bankrupt governmental units, even when 

subsequent challenges are made to the financial statements, the former city manager is more likely to be sued than 

the auditors (e.g. Palmeri, 2010). Also, since for the unit, the state may likely assume the debt, the client may 

continue (but not as the same going concern) so arguments for not giving a GCO remain. Next, because of the time 

delays due to state takeovers, creditors too are more likely to wait and less likely to sue the auditor. Specifically for 



Durden and Sennetti                                                                                                                                                      Advances in Business Research 

                                                                                                                                                                                               2012, Vol. 3, No. 1, 79-89 

81 

 

California school districts, there exists an additional funding source (loan) from the State of California after the 

school district declares insolvency, thereby again reducing the likelihood of taxpayer lawsuits and reducing auditors’ 

concerns of client retention (State of California, 2009), and reducing the likelihood of a GCO.  

Next, given the unit’s poor financial condition, the collection of the audit fees themselves or even client retention 

may be in doubt, and since the fees themselves may be so low, or even not sufficiently budgeted, that the 

corresponding quality of audit itself may low and the GCO evidence produced insufficient for this opinion, 

following low auditor-budgeted costs. Finally, the first to feel the bankruptcy consequences are the unit’s 

employees, less likely as a individuals to sue, but more likely to sue as part of a union, if one is permitted. Again 

these lawsuits are more likely first against management than the auditors, reducing once again the likelihood of the 

GCO. 
 

Public Interest Arguments for Increasing the Likelihood of the GCO 
 

Against this, the public interest stakeholders, too numerous to mention and unaware of the consequences of 

bankruptcy or extreme financial stress conditions, and as voters, do demand more warnings from the accountants. 

The consequences can be too severe: closed police stations and less protection for cities in California (Sanchez, 

2010), failed city water and fire systems and higher housing insurance in Harrisburg, Pennsylvania (McNichol, 

2010). In Georgia, bankrupt school districts create non-accredited high school degrees and Las Vegas, Nevada, 

closed city-supported cancer and kidney units for urgent care. These events argue for increasing the auditors’ 

evidence for the supply of GCOs: The “hope is that issuing a going-concern opinion might promote timelier rescue 

activity” (Venuti, 2004: 41). 
 

Audit Evidence for the Bankruptcy of Governmental Units 
 

Local governments are able to file for bankruptcy under the Municipal Bankruptcy Act of 1937 and 579 

municipalities have filed for bankruptcy since the passage of the legislation (Landry and Deal, 2008).  Park (2004) 

considers the causes of municipal cash shortages and bankruptcies. Research on audit opinions associated with these 

begins with Belluomini (1977) who provides the variables in Table 1 to be considered before providing a going 

concern opinion for governmental units. 
 

Table 1: Going Concern Considerations for Governmental Units 
 

1. Significant interfund receivables or receivables from other governmental units may not be realizable in full; as a result, the unit will 
have significant cash flow problems. 

2. Taxable property values may be decreasing and/or delinquent taxes as a percentage of the total tax levy may be increasing and the 

current tax rate may be approaching the maximum legal limit. As a result, if there are no other revenue sources, the unit may not be 
able to provide adequate levels of service and meet debt obligations. 

3. The unit's actual expenditures may have been in excess of budgeted expenditures for one or more years and/or actual revenues may 

have fallen short of budgeted revenues for a year of more. 

4. On a per capita basis, bonded debt or expenditures may have increased at a rate in excess of inflation. 

5. Salaries and fringe benefits may have increased as a percentage of total expenditures without an offsetting increase in productivity. 

6. Substantial unfunded pension liabilities requiring significant funding in future years may exist. 

7. The unit may use long term debt to finance current operations or to meet debt service requirements. 

8. There may be increasing amounts of short term borrowings that are not liquidated on a timely basis. 

9. The unit may be unable to issue long term debt to pay off bond anticipation notes. 

10. Enterprise fund activities may not be able to continue as a result of a lack of operating funds, because of continuing operating 
deficits and/or the failure to obtain grants-in-aid from other governmental units to subsidize operations. 

11. The unit's bonds may be downgraded by a bond rating agency. 

12. The governing body may fail to recognize the seriousness of the unit's financial difficulties and as a result may fail to provide 

appropriate remedies. 

 

Examples from California School Districts 
  
 Consider two “independent” school districts in California, the West Coast Costa and Chico districts. West Coast 

Costa district received an unqualified opinion and Chico a going concern opinion (a GCO). Both districts have 

almost identical financial characteristics and should have received a bankrupt rating by the State of California for the 

same year (June 30, 2008), given by the State and found in the auditor report each given in December 2008, and 

when the “preliminary negative certification,” one signal that bankruptcy was imminent but not yet official for both 

school districts. Yet, West Coast Costa did not officially receive this final negative certification rating or a GCO, 

even though the preliminary certification was mentioned in the West Coast Costa Management and Discussion 



Durden and Sennetti                                                                                                                                                      Advances in Business Research 

                                                                                                                                                                                               2012, Vol. 3, No. 1, 79-89 

82 

 

Analysis footnote for that year (Chico Unified School District 2008; West Costa Contra Unified School District 

2008). Apparently the available evidence was insufficient to sustain a GCO. We hypothesize this evidence was 

available but the framework for finding it was not.   

Table 1 begins a framework of signals under which the bankruptcy of governmental units should become more 

explicit. California adds to this framework by providing the information on the attributes listed in Table 2 of two 

types of financially stressed districts, those listed as qualified (stressed) and as negative (bankrupt). From the latter 

we should be able to predict bankrupt districts in August, the time of the audit, for the past school year ending in 

June, before the list of officially bankrupt districts becomes available from the State of California. However, as the 

district becomes more stressed, the inputs to these variables may be changing, e.g. the State may be intervening into 

local districts and making efforts to prevent financial default, by adding resources or closing buildings and 

consolidating classes. Also, the auditors are not supposed to predict defaults with future information but with current 

information, which for California schools may be constantly changing and therefore statistically not relevant. In 

question is whether the State will declare a school negative (bankrupt) after the auditor’s opinion has been rendered.  

So the question remains open as to whether the auditor may unknowingly have sufficient evidence for the GCO, our 

hypothesis, H.  
 

H: California school district financial reports not given going concern opinions but associated with attributes 

of bankrupt district reports are likely to contain evidence sufficient to merit going concern opinions.  
 

METHODOLOGY AND RESULTS 
 

Data 
  

Table 2 presents the California Department of Education’s Annual Reports covering the fiscal years 2004 

through 2008. These reports contained a listing of the school districts that received a (a) positive, (b) qualified, or (c) 

negative certification. A positive certification indicates a school district or county office of education that will meet 

its financial obligations for the current fiscal year and subsequent two fiscal years. A qualified certification indicates 

the school district may not be able to meet its financial obligations during the year, but is not one of substantial 

doubt, and can be considered similar to the SAS No. 59 condition where notes are given but no reference is made in 

the opinion, and one that is not a GCO. A negative certification signifies the school district will not meet its 

financial obligations for the current fiscal year or subsequent fiscal year.  We use this qualification as the basis for 

designating a school district bankrupt.  
 

Design and Framework 
 

 Using publicly held financial and non-financial information on variables suggested by for-profit and not-for-

profit research, we create in-sample statistical models to identify from troubled school districts, those variables 

related to the changing conditions from fiscal stability to government intervention. This approach may require 

different models for different times periods as the conditions of the governmental unit changes from (1) fiscal 

stability to (2) fiscal instability and finally to (3) bankruptcy. In each case the auditor has an advantage of a 

preliminary certification (preliminary level 3) by the State of California, six months after the financial statements 

and about the time of the auditor’s report.  Often this announcement is designated as a subsequent event. 
 

VARIABLES 
 

First design: Suggested predictive variables 
 

For the California school districts with nine independent (predictive) variables suggested by the CDE, we are 

interested in predicting bankruptcy, classification 3 in Table 2, the negative certification, our first dependent variable 

(as Y = 1). [We add to the predictive variables the “Other” category, and now identified as a “size” variable, as size 

is important to all previous research on organizations and in particular for the California School Districts, bringing 

the total for Table 4 to ten predictive variables].  

The modeling is a two-step process. First in Logistic Regression Model 1  we compare the dependent variable, 

classification 2 (Y = 1), the qualified certification, Table 2, to the positive certification, classification 1, Table 2 (for 

Y = 0), giving the 1-2 comparison, in order to determine which of the CDE suggested variables serve to predict 

financial stress. We would like to identify predictive variables before a stable school district becomes unstable. In 

Model 2, we then compare the negative certification or bankruptcy, classification 3 (Y = 1), to the positive 

certification, classification 1, Table 2 (again, as Y = 0) for the 1-3 comparison, using only those variables of Model 



Durden and Sennetti                                                                                                                                                      Advances in Business Research 
                                                                                                                                                                                               2012, Vol. 3, No. 1, 79-89 

83 

 

1 found useful in some way for predicting bankruptcy, 1-3. Preliminary negative certifications do not become final 
until after the end of the school year and are available to auditors during the audit period and therefore an in-sample, 
same year sample analysis and the GCO can be given before the declaration of bankruptcy is declared. 
 

Second design: All available predictive variables 
 

We repeat the first design. We add one more available predictive variable suggested by prior for-profit going 
concern opinion research (Geiger et al., 2005), the evidence of prior bankruptcy (1, 0), bringing the total number to 
eleven independent variables. In creating the variables for the logistic modeling process from Table 2, all predictive 
variables are coded as 1 if present, 0 otherwise. For the final variables, the evidence of prior bankruptcy (the school 
district had a previous negative certification), is coded as 1, otherwise 0. For average daily enrollment variable, 
identified in the model as SIZE AVG, if  the school district has an average enrollment greater than 7,916 (which 
represents the average daily enrollment for all school districts within the data set), the value is set to 1, otherwise 0. 

 

Table 2: California School District Variables 
 

These financial and non-financial indicators were utilized by the California State Controller to determine financial 
distress experienced by a school district. Fiscal years are July 1 to June 30. (California Department of Education 
July 2009, July, 2008, October 2007, October 2006, June 2005). 
  

FY 2007-2008 FY 2006-2007 FY 2005-2006 FY 2004-2005 

Declining Enrollment1 Declining Enrollment Declining Enrollment Declining Enrollment 

Deficit Spending2 Deficit Spending Deficit Spending Deficit Spending 

Revenue Limit3 Inadequate Reserves Inadequate Reserves Inadequate Reserves 

Other Expenditures4 Mid-Year Budget Adj.  Encroachment Issues Special Ed Cont.5 

Salary and Benefit Issues Salary and Benefit Issues Salary and Benefit Issues Reserves6 

Management Turnover Health Welfare Benefit Issues Prior Audit Adjustments Other Revenues7 

Negative Fund Balance Forest Reserve Funds Loss Benefit Related Costs Labor Agreements8 

Charter School Issues Charter School Issues Other Fund Encroach. Indept. Position Control9 

Other Issues Other Issues District Man. Issues Litigation/Labor Claims 

 

Based on a review of these variables, the school district receives one of three following classifications from the 
California State Controller: 
 

  Positive: A school district or county office of education that will meet its financial obligations for the current 
fiscal year and subsequent two fiscal years. 

  Qualified: A school district or county office of education that may not meet its financial obligations for the 
current fiscal year or subsequent two fiscal years. 

  Negative: A school district or county office of education that will not be able to meet its financial obligations for 
the current fiscal year or subsequent fiscal year. 

  
1Declining Enrollment: Enrollment decreased in both the prior and current fiscal years. 

  
2Deficit Spending: Unrestricted deficit spending exceeded one-third of available reserves in any of the current or 
two subsequent fiscal years.  

  
3Revenue Limit: Projected revenue limit for any of the current or two subsequent fiscal years changed by more 
than 2% since budget adoption of first interim.  

  
4Other Expenditures: Projected operating expenditures (e.g., books and supplies) for the current and two 
subsequent fiscal years changed by more than 5% since budget adoption or first interim. 

  
5Contributions: Contributions from unrestricted to restricted resources, or transfers to or from the general fund to 
cover operating deficits changed by more than $20,000  and more than  5% since budget adoption for any of the 
current or two subsequent fiscal years.  

  
6Reserves: Available reserves (e.g., designated for economic uncertainties, undesignated amounts) did not meet 
minimum requirements for the current and two   subsequent fiscal years.  

  
7Other Revenues: Projected operating revenues (e.g., federal, other state) for the current and two subsequent 
fiscal years changed by more than 5% since budget adoption or first interim.  

  
8Status of Labor Agreements:  Salaries and benefit negotiations are unsettled for certificated employees, 
classified employees, and/or management/supervisor/confidential employees as of budget adoption or second 
interim projections.  

  
9Independent Position Control: Personnel position control is independent from the payroll system. 



Durden and Sennetti                                                                                                                                                      Advances in Business Research 

                                                                                                                                                                                               2012, Vol. 3, No. 1, 79-89 

84 

 

  
10

Average Daily Attendance: The mean daily attendance was calculated for each year with daily attendance for 

each school district greater than the mean receives a 1 and less than the mean scores a 0. (SIZE AVG) 

  
11

Prior Bankruptcy: A school district receiving a negative classification in a prior reporting period scores a 1 and 

no negative rating receives a 0.  
 

Table 3: Predictive Variables of School District Bankruptcy Based Upon the CDE 
 

Panel A: Top Three Variables-Begin with 2003 Variables 
 

Year    Dependent Variablesi    Predictive Variableii      (Wald Z-value)       Pseudo-R2    

2007-2008   1-2        Declining Enrollment     (1.638)**     0.67485 
              Deficit Spending      (0.592)* 

              SIZE AVG       (2.036)*  

           1-3        Declining Enrollment     (1.662)**     0.62814 
              Deficit Spending      (0.261)* 

              SIZE AVG       (0.084)* 

2006-2007   1-2        Declining Enrollment     (2.754)**     0.69749 
              Deficit Spending      (1.495)* 

              SIZE AVG       (0.952)*  
           1-3        Declining Enrollment     (2.092)**     0.65707 

              Deficit Spending      (0.783)* 

              SIZE AVG       (0.058)* 
2003-2004**   1-2        Declining Enrollment     (1.453)**     0.48046 

              Deficit Spending      (0.649)* 

              SIZE AVG       (0.151)*  
           1-3        Declining Enrollment     (2.275)**     0.77779 

              Deficit Spending      (0.158)* 

              SIZE AVG       (1.877)* 
 

Panel B: Based Upon All Available Predictive Variables, the Top Three in 2008 
     

Year           Dependent Variablesi     Predictive Variableii      (Wald Z-value)       Pseudo-R2    
   2007-2008   1-2        Prior Bankruptcy      (-1.082)**     0.87475 

              Revenue Limit      (2.422)* 

              SIZE AVG       (1.592)* 
           1-3        Prior Bankruptcy      (3.500)**     0.99267 

              Revenue Limit      (-0.126)* 

              SIZE AVG       (1.112)* 
_________________________________ 
 

*  statistically significant, Z score exceeds 1.645 (one sided p-value<0.05) 
** data insufficient for logistic modeling for years 2006-2005 and 2005-2004. 

i      Refer to Table 4: Positive certification (1), Qualified certification (2), Negative certification (3). 

ii     Refer to Table 4: Footnotes 1-11.          
 

Table 4: Successful Predictions of Classifications 
 

Panel A. State of California classifications by type and year 
 

Classification 2007-2008 2006-2007 2005-2006 2004-2005 2003-2004 Total 

Positive (‘1’) 27 9 2 13 16  
Qualified (‘2’) 85 108 19 29 38  

Negative (bankrupt) (‘3’) 19 13 5 3 14 (13.5%)54 

Total 131 130 26 45 68 400 
 

Panel B: Classification by Most Recent Year: Predictive Model for (2007-2008) 
 

Prediction of Positive (1) v Qualified (2) Classification  
 

Classification Count Probability Exceeds % Successfully Predicted (Count) 

Positive (1) 27  22.222   (6) 

Qualified (2) 85 0.50000 97.647  (84) 

Total 112 0.50000 79.464  (90) 

 

 

 

 



Durden and Sennetti                                                                                                                                                      Advances in Business Research 

                                                                                                                                                                                               2012, Vol. 3, No. 1, 79-89 

85 

 

Prediction of Positive (1) v Bankruptcy (3) Classification  
 

Classification Count Probability Exceeds % Successfully Predicted (Count) 

Positive (1) 27 0.50000 96.296   (26) 

Negative (bankruptcy declared) (3) 19 0.50000 68.421   (13) 

Total 46  84.783   (39) 

 

School districts in California are required to file interim reports twice a year confirming their status to CDE 

authorities. During the fiscal year 2007-2008, there were 131 school districts that filed either a qualified or negative 

certification during either the 1
st
 or 2

nd
 certification filing. If a school district filed either a qualified or negative 

certification for the 1
st
 certification filing, then if for the 2

nd
 certification the financial condition was improved they 

received a positive certification for that 2
nd

 certification period. The 131 school districts are thereby classified into 

three divisions (dependent variables) based on the 2
nd

 certification, there are 19 classified as negative, 85 classified 

as qualified and 27 classified as positive. The same classification procedure is utilized for the previous years under 

study. There is insufficient published data on bankrupt schools for the years 2004 to 2006, as the bankruptcies range 

from 68 in 2004 to 26 in 2006, suggesting improving U.S. and California economic school district conditions as 

shown in Tables 3 and 4. 
 

Results-necessary and sufficient conditions 
 

Necessary conditions: Tables 1 and 2 and local laws provide the necessary conditions for predicting (in-sample) 

bankruptcy for governmental units. Tables 3 and 4 present the sufficient conditions following the unique modeling 

process described earlier.  In the first design for the governmental variables given by California, we create a logistic 

regression model to predict bankruptcy for the year 2003. But we find no one set of variables to serve for every year. 

To resolve this issue, Table 3 (Panel A) presents the best (maximum likelihood, pseudo-R-square) three variable 

logistic model for predicting 2003 to 2004 bankruptcies (the 1-3 Model). We find Declining Enrollment, Size AVG 

(the school district exceeds average enrollment) as statistically significant and Deficit Spending as the third variable 

found by the model predictive of bankruptcy (1-3) for the year 2003 to 2004. These two variables, although not 

statistically significant also appear in the three variable model found for the predictions of the at risk (qualified) 

condition (1-2) for the year 2003 to 2004. These two statistically significant variables also appear in 2007 to 2008 

(1-2) predictions. Although only one of the two (Declining Enrollment) is significant for predicting bankruptcy in 

that year, both are significant for predicting the qualified condition (1-2 model). Therefore, only Declining 

Enrollment and the prior bankruptcy of the school districts should be considered (is necessary) in evaluating the 

bankruptcy of school districts. 

Table 3 (Panel B) presents the best (maximum likelihood, pseudo-R-square) three variable logistic model for 

predicting the most recent year, the 2007 to 2008 bankruptcies (the 1-3 Model) with the most available data and 

adds one more variable found from adding the non-governmental bankruptcy research variable, Prior Bankruptcy 

(1,0). The best three variable model does confirm prior bankruptcy as a predictor, as one of three variables in the 

three variable (1-3) model. It is also statistically significant, but the variables from Table 4 found in Table 5(a) have 

changed. Prior Bankruptcy and Revenue Limit have replaced Declining Enrollment and Deficit Spending in the (1-

2) model. Only Prior Bankruptcy is significant in the bankruptcy (1-3 model). As shown in both Table 3 (a), the 

partial model, and Table 3 (B), the full model, different variables for different years are predictive of bankruptcy.  

This can be expected. As the State takes over school districts, it changes them and provides improved funding levels, 

causing different variables to be predictive each year. It also seems that schools which had prior bankruptcies and 

exceeded their revenue limits also had declining enrollments and deficit spending, as these variables are collinear. 
 

Sufficient Conditions 
 

 Table 4 demonstrates the ability of the full model of variables. In Table 3B, the 1-2 model only agrees with the 

1-3 model in the variable, prior bankruptcy, for the most recent year, 2007 to 2008. We see in Table 4 that the 1-3 

model has a success rate of around 85%, predicting 13 of the 19 bankruptcies and 26 of the 27 stable school districts, 

using the preliminary negative rating for the dependent variable. Therefore 13 of the 19 bankruptcies could have 

been predicted using this full model, two-step approach, and the preliminary information available on bankruptcies 

from the State of California (California Department of Education, July 2009).   

 The 19 bankrupt school districts for 2008 we identify as GCO candidates as (see bold): Chico Unified, Orange 

Center Elementary, El Rancho Unified, Wilsona Elementary, King City Joint Union High, King City Union 

Elementary, Val Verde Unified, Aromas-San Juan Unified, Julian Union High, La Honda-Pescadero Unified, Pajaro 

Valley Unified, Santa Cruz City Schools, Travis Unified, Vallejo City Unified, Healdsburg Unified, Piner-Olivet 



Durden and Sennetti                                                                                                                                                      Advances in Business Research 

                                                                                                                                                                                               2012, Vol. 3, No. 1, 79-89 

86 

 

Union Elementary, Chinese Camp Elementary, Dixon Unified and Westwood Unified. These school districts meet 

the definition of a going concern opinion under SAS No. 59 and GASB No. 54. The two school districts, West Coast 

Costa and Chico that seemed very similar, differed in their final certifications. West Coast Costa, although similar to 

Chico in financial characteristics, was not identified as a bankrupt school by the State, and it also would not qualify 

in our process as sufficient for a GCO since it has no prior negative certification before its preliminary negative 

certification. Table 6 shows one other district (Willows Unified) is identified when the model is applied to 27 the 

positive certification districts, creating a Type I error of 1/27 or 3.7%, and bringing the total to fourteen districts 

recommended for a GCO.  
 

Summary and Conclusion 
 

 This research presents a process to increase the evidence for the going concern opinion for local governmental 

units and applies it to the California school districts. This process first identifies relevant variables from a database 

constructed by the State of California, the stakeholder must likely to assume the district’s debt. It then adds relevant 

unique variables from general bankruptcy research. Statistical analysis presents only one of these, whether or not a 

prior bankruptcy exists. These changing models are unique to local government bankruptcies because as stress 

occurs, changes in the values and in the importance of each variable follow as consequences. For example, the 

variable “revenue changes,” may be an important stress variable prior to bankruptcy, but is not a consistent predictor 

as is the variable, “prior bankruptcy.” We see in Table 2 and in Table 3 that the State of California predictor 

variables vary by year, and then must be confirmed by comparing the (1-2) analysis models with the (1-3) predictive 

models.   
 

Contributions to Practice 
 

  We respond to a specific request: “Additional research into the development and application of such financial 

predictor models for governments would enhance the models and, in the long run, serve the interests of the 

taxpayers.” (Ghnay 2009, 40). Under our process it is likely that some local governments now exist as undeclared 

going concern candidates. Current economic downtown and numerous media reports of government financial 

distress (the closing of police stations, fire departments and schools) underscore the demand for more GCOs.     
 

Contributions to Prior Research 
 

 This research adds to Ghany (2009), Landry and Deal (2008), Kloha et al. (2005), Gauthier (2005), Venuti 

(2004), Brown (1993), Belluomini (1977) and Trussel and Patrick (2009), by adding a framework to find predictive 

models, as no one model is or will ever be sufficient by itself, given changing a) state and local laws and  

regulations, b) economic conditions and c) forms of governmental units. 
 

Limitations 
 

The generalization of our results and the extension of their external validity to all local governmental units is 

limited: 1) to sample analysis which in this case consists of only the school districts in California,  2) to the number 

of bankruptcies for any year, as only 19 were given for 2008 (and only 26 in Kansas City (Associated Press, 2010),  

3) to the state’s extensive database of variables which varies by state, and 4) to the availability of early and 

preliminary negative confirmations (potential bankruptcies, which although not certain give weight to the auditor 

who files a GCO for that same year) and those states in similar financial conditions which do give preliminary 

signals, such as Kansas City, (Missouri) Detroit (Michigan), Germantown (Pennsylvania), Harrisburg 

(Pennsylvania) and St. Mary Parish (Louisiana.), just to name a few (Gundle-Krieg 2009; Kersey and Van Beek 

2009; Eisele-Dyrli 2010; Hepp 2010; Hollingsworth, 2009). 
 

REFERENCES 
 

Akers, M., Maher, M., & Giacomino, D. 2003. Going-concern opinions: Broadening the expectations gap. CPA 

Journal, 73(10), 38-42. 
  

American Institute of Certified Public Accountants (AICPA). 1981 . The auditor’s considerations when a 

question arises about an entity’s continued existence. Statement on Auditing Standards No. 34. New York, 

NY: AICPA. 



Durden and Sennetti                                                                                                                                                      Advances in Business Research 

                                                                                                                                                                                               2012, Vol. 3, No. 1, 79-89 

87 

 

American Institute of Certified Public Accountants (AICPA). 1988. The auditor’s consideration of an entity’s 

ability to continue as a going concern. Statement on Auditing Standards No. 59. New York, NY: AICPA. 
 

Associated Press. 2010. Kansas City closing 26 public schools. http://www.cnbc.com/id/35812823/Kansas_City_ 

Closing_26_Public_Schools_UPDATED. 
 

Brown, K. 1993. The 10-point test of financial condition: Toward an easy-to-use assessment tool for smaller cities. 

Government Finance Review, 9(6), 21-26.  
 

Bellovary, J., Giacomino, D., & Akers, M. 2006. Weighing the public interest. CPA Journal, 76(1), 16-21.   
 

Belluomini, F. 1977. Local governmental units: The going concern question. Journal of Accountancy, 143(6), 60-

64.   
 

California Department of Education. 2009. Annual financial report of California K-12 schools, report to the 

superintendent of public instruction, for the period of July 1, 2007, through June 30, 2008. July. California 

State Controller.  http://www.sco.ca.gov/aud_pubs_auditrpt_k12.html. 
 

California Department of Education. 2008. Annual financial report of California K-12 schools, report to the 

superintendent of public instruction, for the period of July 1, 2006, through June 30, 2007. July. California 

State Controller. http://www.sco.ca.gov/aud_pubs_auditrpt_k12.html. 
 

California Department of Education. 2007. Annual financial report of California K-12 schools, report to the 

superintendent of public instruction, for the period of July 1, 2005, through June 30, 2006. October. 

California State Controller. http://www.sco.ca.gov/aud_pubs_auditrpt_k12.html.  
 

California Department of Education. 2006. Annual financial report of California K-12 schools, report to the 

superintendent of public instruction, for the period of July 1, 2004, through June 30, 2005. October. 

California State Controller. http://www.sco.ca.gov/aud_pubs_auditrpt_k12.html. 
 

California Department of Education. 2005. Annual financial report of California K-12 schools, report to the 

superintendent of public instruction, for the period of July 1, 2003, through June 30, 2004. June. California 

State Controller.  http://www.sco.ca.gov/aud_pubs_auditrpt_k12.html. 
 

California State Controller’s Office. 2009. State of California comprehensive annual financial report for the 

fiscal year ended June 30, 2009. http://www.sco.ca.gov/ard_state_cafr.html. 
 

Carcello, J., & Neal, T. 2003. Audit committee characteristics and auditor dismissals following "new" going-

concern reports. Accounting Review, 78(1), 95-117. 
  
Chico Unified School District. 2008. Financial statements and supplementary information for the year ended 

June 30, 2008. Author. http://www.chicousd.org/__documents/0708%20Audit.pdf. 
 

Eisele-Dyrli, K. 2010, January. Will pensions bankrupt your district? District Administration: Solutions for 

school district management (January). http://www.districtadministration.com/viewarticle.aspx?articleid=2253. 
 

Governmental Accounting Standards Board (GASB). 2009. Statement No. 56: Codification of accounting and 

financial reporting guidance contained in the AICPA Statements on Auditing Standards. Norwalk, CT: 

GASB. 
 

Gauthier, S. 2005. Governmental accounting, auditing, and financial reporting. Chicago, IL: Government 

Finance Officers Association. 
 

Geiger, M., & Rama, D. 2003. Audit fees, nonaudit fees, and auditor reporting on stressed companies. Auditing: A 

Journal of Practice & Theory, 22(2), 53-69. 
 

Geiger, M., Raghunandan, K., & Rama, D. 2005. Recent changes in the association between bankruptcies and prior 

audit opinions. Auditing: A Journal of Practice & Theory, 24(1), 21-35. 
 



Durden and Sennetti                                                                                                                                                      Advances in Business Research 

                                                                                                                                                                                               2012, Vol. 3, No. 1, 79-89 

88 

 

Ghany, K. 2009. Measuring financial stress on state and local governments: A proposal for early detection. CPA 

Journal, 79(10), 38-40.  
 

Gore, A. 2009. Why do cities hoard cash? Determinants and implications of municipal cash holdings. Accounting 

Review, 84(1), 183-207. 
 

Greenlee, J., & Trussel, J. 2000. Predicting the financial vulnerability in charitable organizations. Nonprofit 

Management & Leadership, 11(2), 199-210. 
 

Gundle-Krieg, D. 2009. Will Detroit public schools declare bankruptcy? Education Examiner (July 7). 

http://www.examiner.com/education-in-national/will-detroit-public-schools-declare-bankruptcy. 
 

Hepp, C. 2010. Germantown Settlement subsidiary going bankrupt. The Philadelphia Inquirer (April 1). 

http://www.philly.com/inquirer/business/20100401_Germantown_Settlement_subsidiary_going_bankrupt.html. 
 

Hollingsworth, H. 2009. Kansas City public schools: Bankruptcy may force 50% Of Missouri City's schools to 

shut doors. Huffington Post (July 24). http://www.huffingtonpost.com/2010/03/07/kansas-city-public-school_n_ 

489145.html. 
 

Hopwood, W., McKeown, J., & Mutchler, J. 1989. A test of the incremental explanatory power of opinions qualified 

for consistency and uncertainty. Accounting Review, 64(1), 28-48.  
 

Hopwood, W., McKeown, J., & Mutchler, J. 1994. A reexamination of auditor versus model accuracy within the 

context of the going-concern opinion decision. Contemporary Accounting Research, 10(2), 409-430.  
 

Jones, F. 1987. Current techniques in bankruptcy prediction. Journal of Accounting Literature,  6, 131-164. 
 

Kersey, P., & Van Beek, M. 2009. Detroit’s schools are going bankrupt, too: Now’s the time to cast off collective 

bargaining agreements and introduce school choice. Wall Street Journal, (July 24).  
 

Kloha, P., Weissert, C., & Kleine, R. 2005. Developing and testing a composite model to predict local fiscal distress. 

Public Administrative Review, 65(3), 313-323. 
  

Landry, R., & Deal, K. 2008. More municipalities likely to face Chapter 9: Is a perfect storm brewing? American 

Bankruptcy Institute Journal, 27(6), 18.  
 

Luhby, T. 2009. Cash-poor California turns to IOU’s. CNNMoney. (July 2).  
 

McNichol, D. 2010. Pennsylvania state capital mulls bankruptcy as a budget option. Bloomberg Businessweek. 

(February 4).  
 

Mutchler, J. 1984. Auditors’ perceptions of the going-concern opinion decision. Auditing: A Journal of Practice & 

Theory, 3(2), 17-30. 
  
Mutchler, J. 1985. A multivariate analysis of the auditor’s going-concern opinion decision. Journal of Accounting 

Research, 23(2), 668-682. 
 

Mutchler, J. 1986. Empirical evidence regarding the auditor’s going-concern opinion decision. Auditing: A Journal 

of Practice & Theory, 6(1), 148-163. 
  

Mutchler, J. Hopwood, W., & McKeown, J. 1997. The influence of contrary information and mitigating factors on 

audit opinion decisions on bankrupt companies. Journal of Accounting Research, 35(2), 295-310. 
  

Park, K.  2004. To file or not to file: The causes of municipal bankruptcy in the United States. Journal of Public 

Budgeting, Accounting & Financial Management, 16(2), 228-256.   
 

Ponemon, L., & Raghunandan, K. 1994. What is “substantial doubt”? Accounting Horizons, 8(2), 44-54. 
  

Raghunandan, K., & Rama, D. 1995. Audit reports for companies in financial distress: Before and after SAS No. 59. 

Auditing, 14(1),  50-63. 



Durden and Sennetti                                                                                                                                                      Advances in Business Research 

                                                                                                                                                                                               2012, Vol. 3, No. 1, 79-89 

89 

 

Roger, J. 2010. Mayor of California City to stop taking high salary. Yahoo News. http://news.yahoo.com/s/ap/ 

20100726/ap_on_re_us/us_bell_salaries. 
 

Sanchez, R. 2010. American cities teeter on brink of bankruptcy. ABC News / Money (June 28). http:// 

abcnews.go.com/Business/maywood-california-top-cities-overwhelmed-recession/story?id=11032953. 
 

Sharma, A. 1997. Professional as agent: Knowledge asymmetry in agency exchange. Academy of Management 

Review, 22(3), 758-798. 
 

State of California. 2009. Education code. Section 42122-42129, 41320-41322. http://www.leginfo.ca.gov/cgi-

bin/calawquery?codesection=edc&codebody=&hits=20. 
 

Trussel, J., & Patrick, P. 2009. A predictive model of fiscal distress in local governments. Journal of Public 

Budgeting, Accounting & Financial Management,  21(4), 578-616. 
  

Venuti, E. 2004. The going-concern assumption revisited: Assessing a company's future viability. CPA Journal, 

74(5), 40-43. 
 

West Contra Costa Unified School District. 2008. Financial statements and supplementary information for the 

year ended June 30, 2008. http://www.wccusd.net. 
 

 

S. Joseph Durden is a CPA, PC. He has given going concern audit opinions to some governmental units. His 

publications include Issues in Accounting Education of the American Accounting Association. 
 

John Sennetti is a professor of graduate statistics and accounting at Nova Southeastern University. He received his 

Ph.D. from Virginia Tech. He supports the research of his graduate students, and has also published in Auditing: A 

Journal of Practice and Theory, Journal of Finance, Journal of Business Ethics, Advances in Accounting Behavioral 

Research, and other. 

 


