Economic Consequences of Going Concern Audit Opinions in Nonprofit Charitable Organizations

This study investigates the economic consequences of going concern audit reports (GCARs) in nonprofit charitable organizations (NPOs). A well-established literature examines investor reactions to GCARs in the for-profit context, but no similar evidence exists for the nonprofit setting. Understanding associations between GCARs and sources of NPO financing potentially informs NPO manager decisions and actions of officials who regulate and oversee the nonprofit sector.

A GCAR reflects auditors' judgments about whether an NPO can continue operations. Thus, given that NPOs generally have few information intermediaries, I expect that a GCAR contains incremental information about an NPO because auditors possess inside information that is conveyed by issuing GCARs. If a GCAR carries new information about an NPO, subsequent donations and government grants can be potentially affected, depending on the donors' or grantors' motives to contribute. Conversely, if GCARs have no adverse economic impacts, then regulators and NPO oversight agencies can reallocate limited resources to monitor aspects of NPOs other than GCARs.

My findings show that decreases in government grants follow GCARs. Such results indicate either that granting authorities are reluctant to fund activities of NPOs where going concern is at risk, or that NPOs voluntarily withdraw grant applications that cover less than the full cost of providing the service. The evidence also shows that contributions decrease following a GCAR. I find no statistically significant association between GCARs and the NPO's public support in the post-GCAR year.

This paper focuses on NPOs receiving initial GCARs during Single Audits because initial GCARs are generally believed to be more informative than continuing GCARs.¹ Issuing a first-time GCAR is a particularly difficult decision for auditors (Kida 1980; Mutchler 1984). Auditors do not retract GCARs unless the client's financial condition improves significantly (Nogler 1995).

This paper is structured as follows. The next section provides background information about GCARs and external funding sources in the nonprofit sector. The third section reviews the relevant literature and develops hypotheses. The fourth section specifies the research methodology and describes the sample selection procedures. The fifth section presents empirical results for the main analyses. The final section summarizes the study and provides conclusions.

Statement on Auditing Standards (SAS) No. 59 (American Institute of Certified Public Accountants [AICPA] 1988), which guides going concern judgments, requires auditors to include a going concern explanatory paragraph in the audit report if they have substantial doubts about an entity's ability to operate continuously for one year beyond the date of the financial statement audit. GCARs are issued to NPOs infrequently. Keating, Gordon, Fischer, and Greenlee (2005) document that between 1997 and 2000, only 136 of 12,654 NPO audit reports (1.07 percent) under the Single Audit Act contain GCARs.² The low frequency of GCARs may be because the litigation risk in the nonprofit sector is low compared to the for-profit sector. Thus, an auditor's incentive to issue a GCAR is less substantial.³

NPO bankruptcies are also rare. NPOs comprise only 1 percent of bankruptcy cases even though they constitute 30 percent of all public entities. NPOs cannot be forced to file bankruptcy under federal law (11 U.S.C.A. 303(a)), and creditors must follow state law pertaining to insolvent NPOs (O'Neil and Barnett 1980; Oleck 1992). The lower frequency of bankruptcy in NPOs is perhaps due to high resolution costs (such as attorney fees and efforts to distribute assets), or because an NPO's management cannot profit from liquidation or mergers. A further disincentive to bankruptcy is beneficiaries have to find substitutes for the public goods offered by failed NPOs. Finally, bankruptcy has significant adverse effects for NPO managers; for example, reputation loss. Thus, all parties affected by the NPO have incentives to avoid NPO bankruptcy.

NPOs receive external funding from various sources: individual donors, private foundations, and government grants. The funding received from individual donors is direct support, whereas the funding received from private foundations (such as United Way) is indirect support. These two types of support together are called public support. NPOs also apply for grants from the government. Public support and government grants together constitute an NPO's total contributions.

A GCAR is a negative signal for an NPO. Whether such “bad news” contains incremental information content has been an ongoing debate in the for-profit literature. One strand of research reports no negative market reaction to GCARs (Elliott 1982; Dodd, Dopuch, Holthausen, and Leftwish 1984; Blay and Geiger 2001; Herbohn, Ragunathan, and Garsden 2007), while the other strand of research finds significant negative market reaction to GCARs (Fleak and Wilson 1994; Jones 1996). In a more recent study, Menon and Williams (2010) document a significant negative reaction following the announcement of a GCAR. The authors show that the degree of adverse market reaction increases for firms with a debt covenant that restricts the firm from receiving a GCAR and increases with the level of institutional ownership in the firm. Because NPOs do not issue stock, whether donors and grantors react to a GCAR in the nonprofit sector is an unresolved empirical question.

A GCAR becomes a potentially valuable source of information for contributors because it contains auditors' insights about an NPO obtained through field examination.⁴ Thus, a GCAR reflects an auditor's view about the NPO's ability to continue operations, which can discourage donors and grantors from contributing or giving grants to the NPO. Donors and grantors make contributions to increase an NPO's service or provision level. For instance, donors or grantors contribute to NPOs to increase the frequency or quality of art exhibits, increase the number of children fed or educated in developing countries, or help low-income people to earn more (Vesterlund 2006). When an NPO receives a GCAR, donors or grantors may discontinue their support because they are concerned about whether the NPO can continue to carry out its mission or whether contributed resources will be consumed in a bankruptcy process. In turn, these donors or grantors may choose to reallocate their resources to another NPO that carries out a similar mission. Plus, some donors are reluctant to make contributions just to pay the debt of a financially distressed NPO without directly supporting a mission (O'Neil and Barnett 1980).

The government requires audit reports from NPOs that meet either the revenue threshold imposed by state legislations or the federal expenditure threshold under the Single Audit Act. Often, state auditors themselves audit NPOs and prepare audit reports. Some government agencies use GCARs as one of the screening criteria when they make grant decisions. For example, the New York State Homeless Housing and Assistance Corporation (HHAC), a subsidiary of the New York State Housing Finance Agency (HFA), considers going concern as one of the criteria to qualify an NPO for funding under the Homeless Housing and Assistance Program (HHAP).

The above discussion leads to the proposition that donors and grantors may react negatively to a GCAR. Given that there are only few GCARs for NPOs, the GCARs that are issued perhaps have more impact on donors and grantors. Furthermore, an NPO may choose not to apply grants after receiving a GCAR, since most grants require an NPO to have sufficient funds to cover a portion of the cost of providing the services the grant covers.

On the other hand, although prospective donors can obtain A-133 audit reports under the Freedom of Information Act (FOIA) or through publicly available databases, most donors (especially individual donors) typically do not know an NPO's audit reports unless the NPO has a high enough profile to have the GCAR reported in the media. Other donors may increase contributions to help the struggling organization survive. For example, government may offer special grants to support a financially distressed NPO that receives a GCAR simply because the organization serves a crucial mission in the communities, such as a homeless shelter in an impoverished region or a clinic in a remote rural area. In addition, those donors who seek a warm glow (Andreoni 1990; Ribar and Wilhelm 2002) may not be affected by a GCAR when they make donation decisions.

Therefore, the net effect of a GCAR on an NPO's funding sources is unclear. The preceding arguments lead to the following hypothesis:

• H: 

  An NPO's funding sources are independent of whether the NPO receives a GCAR.

An empirical challenge when investigating going concerns is disentangling the impact of receiving a GCAR from general financial distress that precedes the GCAR. To control for financial distress, going concern studies in the context of periodically tested firms (Mutchler 1984, 1985; Chen and Church 1992; Feng 2010; Menon and Williams 2010) typically identify a control firm for each sample firm from a pool of financially distressed firms in the same year and industry. I adopt the same approach to select the control group. I also include the distress indicators, as developed for NPOs by Greenlee and Trussel (2000) and Trussel (2002), to disentangle the impact of financial distress on subsequent contributions, public support, and grants.

I use the specification below to investigate associations between GCARs and NPO funding sources:

where LnFundSource is a generic dependent variable, indicating specific dependent variables LnContributions, LnPublicSupport, and LnGovernmentGrants. In all specifications, the variable of interest is GCAR, which is equal to 1 for the year following the GCAR. A negative coefficient for GCAR implies that receiving a GCAR is associated with decreases in the funding that an NPO receives in the post-GCAR year. A positive coefficient for GCAR suggests an increase in the funding that an NPO receives in the post-GCAR year.

Following Petrovits, Shakespeare, and Shih (2011), I include the price of donating (LnPrice) and fundraising expense (LnFundraisingExpense). The natural log of the variables in the model mitigates potential heterogeneity. The price of donating (price, hereafter) is measured as how many dollars of after-tax income a donor must give up in order to provide one additional dollar to an NPO's final output. Following Weisbrod and Dominguez (1986) and Marudas and Jacobs (2004), I consider price a function of the program service ratio and the tax benefit of donating:

Weisbrod and Dominguez (1986) and Petrovits et al. (2011) find a significant negative relationship between price and level of contributions.⁵ Weisbrod and Dominguez (1986), who use fundraising expenses to control for the organization's efforts to lower information asymmetry, report that fundraising expenses have a positive impact on contributions. In contrast, Tinkelman (2004) documents a negative correlation between fundraising expenses and donations, because donors prefer an NPO to have lower fundraising expenses so that more of their donations are allocated to program spending.

I also include distress indicators developed by Greenlee and Trussel (2000) and Trussel (2002): debt ratio (DebtRatio), revenue concentration ratio (RevenueConcentration), administrative ratio (AdminRatio), and surplus margin (Margin). Financial distress reflected by these measurements may affect resources received. For instance, a low debt ratio can be attractive to potential contributors. On the other hand, a low debt ratio can also suggest that an NPO is financially self-sufficient and has less need to solicit contributions. Therefore, I do not specify expected signs for these distress indicators.

Petrovits et al. (2011) document that internal control deficiencies adversely affect contributions; therefore, I include internal control weakness (ICWeakness) in the model and expect a negative correlation between ICWeakness and subsequent contributions. The nonprofit literature presents mixed results regarding the correlation between auditor size and audit quality. Krishnan and Schauer (2000) find that nonprofit reporting quality is positively related to auditor size. In contrast, Tate (2007) does not find consistent evidence that audit quality and Big 5 auditors vary directly. In a recent study, Kitching (2009) finds that NPOs with Big 5 auditors receive more contributions than NPOs with smaller auditors. Thus, I include an indicator variable, BigN, to consider auditor reputation, and I expect a positive coefficient for BigN.

To control for size, I include total assets (LnSize). Organization age and LnSize are highly correlated and, therefore, LnSize also controls for NPO reputation (Tinkelman 2004). I also include the lagged contributions to consider organization characteristics not specifically considered.⁶ Contributions do not immediately respond to influencing factors. Therefore, I expect that contributions are sensitive to the indicators of organizational performance in the preceding accounting period. To control for unspecified factors correlated with industry or year, the specification also includes the industry and year fixed effects. Specifically, I distinguish six industries: Arts, Culture, and Humanities; Education; Health; Human Services; Public and Societal Benefits; and Other, and the six years 1998 to 2003 that encompass the sample period.

I use the Federal Audit Clearinghouse (FAC) database and the National Center for Charitable Statistics (NCCS) database. The FAC database contains the Single Audit data for NPOs, including auditors' opinions on the financial reporting. The NCCS database contains NPOs' financial data obtained from Form 990. Nonprofit researchers use the NCCS data in lieu of data from consolidated financial statements, because the two sets of data are highly correlated and it is difficult to obtain NPOs' financial statements (Krishnan, M. Yetman, and R. Yetman 2006; Keating, Parsons, and Roberts 2008; Kitching 2009; Petrovits et al. 2011).

The GCAR sample consists of NPOs that received an initial GCAR between 1998 and 2003 in the FAC database. I select this period because digitized files in the NCCS database contain necessary financial data on public charities.⁷ Table 1, Panel A shows the sample selection process. The final sample includes 405 NPOs that received initial GCARs within the study period. Table 1, Panel B displays the frequency of NPOs that received their first-time GCARs between 1998 and 2003, and shows an even distribution across the study period. Panel C shows that observations are more concentrated in the Human Services (47 percent) and Health sectors (29 percent). The Education (10 percent) and Public and Societal Benefits (10 percent) sectors are the next two largest groups. The Arts, Culture, and Humanities (1 percent) and Other sectors (2 percent) are the two smallest groups. This sector distribution is generally consistent with that of the NPO population.

Chen and Church (1992) and Mutchler (1984, 1985) note that in the for-profit setting, auditors first identify problem firms and then decide whether to issue GCARs. To determine whether their selection criteria apply to NPOs, I obtained 172 audit reports for 172 sample NPOs through the FOIA procedure. The three most frequently cited reasons for issuing GCARs, as calculated by the frequency with which auditors cited each justification, are the following: (1) negative net assets or insolvency (i.e., total liabilities are greater than total assets), (2) losses, or (3) negative working capital.⁸ For each sample NPO with a GCAR, I identify a control NPO that meets one of the three criteria. More specifically, I identify an NPO without a GCAR that has the closest total assets within the same National Taxonomy of Exempt Entity (NTEE) and fiscal year combination, and that has either negative net assets, operating losses, or negative working capital. Thus, the control group consists of 405 NPOs through this matching process. In order to mitigate the impact of outliers, I winsorize the main financial data at the 1 percent and 99 percent levels after selecting control organizations.

Table 2 provides descriptive statistics for the initial GCAR sample and the control sample in the year when an NPO received its initial GCAR. Paired t-tests for the differences in means reveal that, on average, NPOs with GCARs have a higher donation price (t = 2.95, p < 0.01), debt ratio (t = 8.21, p < 0.01), admin ratio (t = 3.87, p < 0.01), and internal control weakness (t = 9.54, p < 0.01). NPOs with GCARs also have lower public support (t = −1.74, p < 0.10) and fundraising expenses (t = −1.63, p < 0.10) than peer NPOs. In sum, Table 2 suggests that the GCAR sample, in general, may have higher financing risk and weaker internal control than the control group.

Finally, sample NPOs receive, on average, 51 percent of their revenues from government grants and 8 percent of their revenues from public support (untabulated). Therefore, total contributions (i.e., the sum of government grants and public support) constitute 59 percent of total revenues.

Pearson correlations between major variables are in Table 3. As expected, the correlations between GCAR and total contributions and between GCAR and public support are negative, although statistically insignificant. The correlation between GCAR and government grants in the post-GCAR year is unexpectedly positive. Without controlling for other variables in a multivariate context, however, definitive conclusions about these associations are premature.

Table 4 presents OLS regression estimates for the impacts of GCARs on NPOs' funding sources (i.e., total contributions, public support, and government grants).¹⁰ Public support and government grants are subsets of total contributions.

In the specification LnContributions (Column 1 of Table 4), GCAR has a negative coefficient (−0.13). The result rejects the null hypothesis. The evidence suggests that, all else equal, receiving a GCAR is associated with fewer contributions in the post-GCAR year and that GCARs in NPOs contain substantial incremental information content. As expected, fundraising expenses are significantly and positively correlated with total contributions. Finally, contributions in the GCAR year are significantly and positively correlated with the contributions of the subsequent post-GCAR year.

In the specification LnPublicSupport (Column 2 of Table 4), the coefficient for the variable GCAR is insignificant. As previously discussed, there are many potential explanations as to why GCARs may not affect public support. Some donors perhaps contribute for self-fulfillment and, thus, do not care about GCARs. Other donors may contribute to help the NPO survive when a GCAR is issued. Finally, individual donors may not be aware of the GCAR because they lack incentives to obtain and review financial information.

Column 3 of Table 4 shows a negative and statistically significant coefficient on GCAR (−0.179) in the specification LnGovernmentGrants, which implies that a GCAR decreases subsequent government grants. This decrease could be imposed on the NPO by government funding agents following a GCAR. A government agency that funds NPOs is more likely to be aware of the going concern opinion than an individual donor. The reduction in the subsequent grants can also be a choice by the NPO. Government grants often require NPOs to use other funds in conjunction with government grants, as many grants do not fully compensate the cost of providing services. If an NPO is operating under financial distress, it may no longer provide the service. Thus, the NPO may decide not to pursue or accept further grant funds.

The variable LnGovernmentGrants_t−1, the natural log of grants in the prior year, has a significantly positive coefficient (0.904), suggesting a high correlation with the government grants in the GCAR year. Coefficients for other control variables are not statistically significant.

The purpose of this study is to investigate some of the economic consequences of GCARs in the context of nonprofit charitable organizations. More specifically, I examine contributions that an NPO receives following an initial GCAR. The results show declines in government grants, indicating that either government agencies react negatively to GCARs or the affected NPOs voluntarily withdraw their grant applications. An NPO's subsequent contributions are also reduced in the post-GCAR year. There is no significant correlation between GCARs and the subsequent public support that NPOs receive following GCARs.

The study offers the first empirical investigation of the economic consequences of GCARs in the nonprofit sector. The analysis documents the negative correlations between GCARs and contributions and government grants that an NPO receives subsequent to a GCAR. The decline in contributions is attributable to a decline in government grants rather than a decline in donations. Examining the economic consequences of GCARs helps us to understand the value of GCARs. These findings are relevant for assessing the extent to which GCARs in NPOs have information content, that is, how GCARs influence NPOs and their stakeholders.