ICNL logo

The International Journal
of Not-for-Profit Law

Volume 5, Issue 1, September 2002

A publication of the International Center for Not-for-Profit Law

Table of Contents


The Economics of Non Profit Accounting and Auditing: Suggestions for a Research Agenda
Marc Jegers

Australian Charity Law Reform Proposals
Prof. Myles McGregor-Lowndes

Charities and Terrorism: The Charity Commission Response
Debra Morris

Charity Law Review in Ireland and the Challenges for the State/Third Sector Partnership
Kerry J. O'Halloran

The Kamehameha Schools Admissions Policy Controversy
Randall W. Roth

Case Notes

Asia Pacific:
Fiji | New Zealand

Central and Eastern Europe: Croatia

Middle East and North Africa:

North America:
Canada | United States

Sub-Saharan Africa:
South Africa

Western Europe:
European Union | The Netherlands

Country Reports

Financial Action Task Force (FATF) on Money Laundering

Asia Pacific:
Regional | Burma | China | Japan | New Zealand | Singapore

Central and Eastern Europe: Estonia | Kosovo | Latvia | Romania

Latin America and the Caribbean: Regional

Middle East and North Africa: Regional

Newly Independent States: Belarus

North America:
Canada | Mexico | United States

South Asia:
Afghanistan | India | Sri Lanka

Sub-Saharan Africa:
Congo | Malawi | Nigeria | South Africa | Sudan | Tanzania | Togo | Zimbabwe

Western Europe:
Regional | Belgium | France | Germany | Italy | The Netherlands | Portugal | Spain | United Kingdom

- - - - - - - - - -

Editorial Board

The Economics of Non Profit Accounting and Auditing: Suggestions for a Research Agenda*

By Marc Jegers

Marc Jegers
Free University Brussels (VUB)
Micro economics of the profit and non-profit sectors
Pleinlaan 2
1050 Brussels
Tel. 02/629.21.13
Fax 02/629.20.60
e-mail: Marc.Jegers@vub.ac.be

An agency based economic theory of non profit accounting and auditing is presented, resulting in thirteen empirically testable hypotheses on their presence, nature and quality. The central agency relation analysed is the board-management relation.

JEL: L31, M41

Keywords: non profit accounting, non profit auditing, positive accounty theory.

* Insightful remarks by Willem Buijink (Tilburg University), R. Scott Gassler (Vesalius College, VUB), Bruno Heyndels (Free University Brussels, VUB), Irvine Lapsley (Edinburgh University) and two anonymous referees helped me to substantially improve the text. For all remaining vaguenesses and errors, I am the only one to blame.

1. Introduction

Non profit organizations (NPOs), as opposed to firms and governmental bodies, play a significant role in developed and developing societies. As an example, Salamon (1999, 6) established that some 1.2 million public serving organizations (including health care and education) in the USA expensed in 1996 an amount of money equivalent to more than 7% of the nation’s GNP. As far as employment is concerned, a comparable percentage in total employment is found for 1990 (Salamon, Anheier (1998, 218)). The average employment percentage in their small sample (Hungary, Italy, Sweden, Japan, Germany, UK, France and the US) reached 3.3% (ibid). Given the importance of NPOs their financial health is an important issue. As in the case of profit organizations (POs), financial statements of NPOs are essential tools in assessing their financial situation. Such statements are mostly drawn up yearly, and contain a balance sheet, an income statement and possibly some explanatory notes. The process of producing these documents is accounting: each of the organization’s transactions affecting its financial situation is registered (bookkeeping) and yearly summarized in the financial statement (financial reporting). The truthfulness of the financial statement can be assessed by an independent expert, called ‘auditor’, who will publish a statement expressing his opinion on its quality. In most countries auditing is mandatory for large POs, however defined, and firms quoted on a stock exchange.

Although there exists a small literature on the technicalities of non profit accounting and auditing (AA), the development of an underlying economic theory explaining its existence and appearance is still in its infancy, contrary to the situation for POs. The basics and early developments of PO accounting theory are described in the seminal work of Watts and Zimmerman (1986, chapter 1). In essence, the literature [still] contains many studies using the finance-based theory and/or the theory of regulation to explain accounting and auditing practice (Watts, Zimmerman (1986, 7)). Needless to say that nowadays the principal-agent framework is an important component of financial theory. The argument that there is no need for a specific NPO AA theory as the theory developed in Watts and Zimmerman (1986) encompasses all kinds of organizations, does not seem very valid a priori: the economic rationale for the existence of NPOs (market failures such as contract failure (Hansman (1987)), government failure (Wolf (1988)), consumerism (Ben-Ner (1986)),… ) markedly differs from the raison d’être of PO. Therefore, the economic role of AA also differs, given the different categories of stakeholders and their objectives, and idiosyncratic governance configurations present in NPOs. It is unfortunate for the present analysis that the economicsbased literature has been fairly silent on nonprofit governance (Eldenburg et al. (2000,5). See also Dyl et al.(1996,1)).
In the next paragraphs, an economic foundation of AA in NPO is developed, in the tradition of a ‘positive accounting theory’, defined as a theory that tries to explain and predict AA practice (Watts, Zimmerman (1986, 2); Belkaoui (1992, 56)). Up to now no such theory has as yet been developed, let alone empirically verified.

The central theme will be the board-manager relation, modelled in a principal-agent framework with goal incongruence between boards and managers (Fama, Jensen (1983), Herman, Heimovics (1991), Steinberg (1990)). As NPO managers are not a homogeneous group of people, three specific situations will be dealt with: an NPO in which the manager is also the founder of the organization, and NPOs in which the manager is an agent of the board, which can be effective or ineffective. Managers are assumed to trade off the NPO’s objectives with discretionary behavior (defined as managerial decisions not contributing to the realization of the NPO’s objectives), entailing the necessity of more or less AA to contain as much as possible agency costs. The analysis will be presented both for NPOs working in a world without AA regulation and NPOs that are subjected to AA regulations. Additionally, the effect of AA knowledge by NPO board members and managers is explicitly taken into consideration, this being a very relevant feature in the functioning of NPOs, contrary to the situation in firms.

All this leads to thirteen testable hypotheses on the presence, nature and quality of NPO AA, in the hope researchers all over the world will try to corroborate or falsify them empirically. It is not claimed that the framework presented here is the only one capable of generating some of these hypotheses. The only purpose of the author is to show that standard economic theory can be applied in an integrated way to a specific NPO topic.

2. The economics of non profit accounting and auditing in an unregulated environment
2.1. The building blocks

The foundation of the theory to be presented here is the same as that of the AA theory for firms: We assume that all… parties… act as to maximize… their expected utility (Watts, Zimmerman (1986, 3)). In the non profit context one can assume, generalizing Hewitt, Brown (2000, 167), that the utility function of managers contains two arguments: the realization of the organisation’s objectives (P), and managerial discretionary behavior (D) in a Williamsonian (1963) sense. In order to prevent managers from leaving the organization, they require a minimum utility level uo (reservation utility).

For board members (B) (considered to be volunteers) only the realisation of the NPO’s objectives is assumed to influence utility. This is clearly a simplification of the real situation, as considerations about wealth, reputation (as a more general concept than P) (Handy (1995)) or time cost potentially play an important role in their decision to join a board. In the context of the present analysis, where P is the focus of interest, this simplification seems justified.
Two categories of managers can be distinguished in the case of a NPO. In the first we find managers who are also founders of the NPO (MF). They are expected to require a larger increase in their discretionary prerogatives to compensate, in utility terms, for a decrease in NPO performance, as compared to a manager who acts purely as an agent for the founders and for the board of the organization (MA). It is fair to state that on average there is still a difference between a manager-agent in a NPO, and a business firm manager: the people attracted to managerial positions in the nonprofit sector are those who care relatively little about financial gain and relatively highly about putting their own ideals into practice (Rose-Ackerman (1987, 812))i.

Graphically, the utility functions of B, MF and MA (UB, UMF and UMA) can be represented as follows:
Figure 1: Indifference curves of voluntary board members (B), managers-founders (MF), managers-agents (MA).

P: NPO performance (realization of objectives)
D: managerial discretionary behavior

Given the available resources, one can imagine higher levels of managerial discretion to cost some of the realization of the NPO’s objectives. The transformation curve between P and D is assumed to be concave as already suggested by Newhouse (1970, 68): at low levels of discretionary behavior performance P will be marginally lower than PMAX, the maximum performance that can be obtained. At higher levels of D the decrease in P becomes more and more severe (fig. 2). A convex transformation curve would be rather inappropriate, implying even the slightest level of managerial discretion heavily affecting the organization’s performance.

Figure 2: Transformation curve between the NPO’s performance (P) and discretionary behavior (D).

In the following subsections, the interactions between utility and transformation will be analysed in order to derive hypotheses on the presence, nature and quality of AA in NPOs in the different circumstances described in the Introduction. In a first step, AA will be treated as a given system with a given cost. Afterwards, the AA system will be decomposed in three elements with their own costs and effects: the conceptual accounting basis chosen (cash accounting versus accrual accounting), the disclosure decision, and the decision to seek external auditing. Clearly, further refinements are conceivable: e.g. the scale of the audit can range from modest to very comprehensive. As these can be dealt with in a way similar to the one applied to take account of the three aforementioned elements of the AA system, they will not be included in the present analysis.

2.2. The simplest case: a manager-founder, no board, no donors

The utility maximizing situation for this case is depicted in figure 3:

Figure 3: Utility maximizing manager-founder

Depending on the exact shape of u*MF (the indifference curve reflecting the maximum utility that can be obtained by the manager-founder), the difference between PMAX and P*MF can be small or large. Given the fact that we assume no regulation (or accreditation or other standards to qualify for legal or tax privileges), there is no higher authority to impose managerial behavior resulting in a higher level of P (P*MF<P<PMAX) which would correspond to the reservation utility uo MF of the manager-founder. Therefore, there is no agency problem, as there is no distinct principal, and AA cannot be considered to be instruments to attenuate agency costs.

This does not preclude an NPO run by a manager-founder to implement an accounting system and even having an external audit if there are sound managerial economic reasons for doing so, i.e. if their costs are at least compensated by increases in managerial efficiency (Anthony (1989, 27)). This will be more likely in larger NPOs who perform complex or diversified activities (hypothesis 1). Furthermore, diversification can necessitate the development of a cost accounting system. A particular example of this can be found in hospitals, where efficiency enhancement has been documented also to be achieved in an indirect way by the mere provision of cost accounting information to physicians (Conrad et al. (1996), Eldenburg (1994)).

2.3. The case of an NPO run by a manager-agent, in the presence of a board, and no donors
2.3.1. The ineffective board

A theoretically rather uninteresting case is that of the ineffective board. Unfortunately this situation is too frequently observed in NPOs. Indeed, the prescriptive literature advances a ‘heroic’ model of the non profit board ( Herman, Heimovics (1990, 168)), but in reality many board members and managers alike contend that boards often function poorly (Middleton (1987, 141)). Given the shape of uMA as compared to uMF, utility optimization of the manager will result in P*MA < P*MF < PMAX (=u*B) and D*MA > D*MF, for a given organization (fig. 4). The manager will shirk and enjoy perquisites at the expense of the maximal attainment of the NPO’s goals.

Figure 4: Manager-agent with ineffective board


In the Jensen and Meckling (1976) terminology (agency costs= monitoring costs + bonding costs + residual loss) the agency costs (in performance units) consist only of the residual loss, reflected by PMAX-P*MA. If there is accounting and possibly external auditing, the economic rationale for it is comparable to that elucidated in the previous section.

2.3.2. The effective board

When the board is effective and managerial behavior is perfectly observable, the manager can be forced to function at his reservation utility level u°MA (fig. 5). This will result in a higher performance level P°MA>P*MA, reducing the residual loss to PMAX-P°MA. It is more realistic, though, to assume imperfect observability of managerial behavior. AA additional to the AA level implemented for the aforementioned internal efficiency reasons can be understood as monitoring devices. Due to their additional cost they will shift inwards the P-D transformation curve.

Monitoring too can be either perfect or imperfect. When perfect, we obtain P°MA,PM (assuming the board has enough power to impose that intersection of the transformation curve with u° with the highest P-value), for which it can easily be seen that P°MA,PM<P°MA<PMAX. It is also clear that P°MA,PM > P*MA, P°MA,PM = P*MA, P°MA,PM < P*MA depending on the shapes of the U*MA curves, the level of U °MA, the transformation curve and the height of the monitoring costs which determines the shift of the transformation curve. It is obvious that if P°MA,PM<P*MA, the imposition of the monitoring system is not effective.

Figure 5: Manager-agent with effective monitoring.

In reality, as both practitioners and theoreticians know, monitoring with AA techniques cannot completely eliminate managerial discretion. Nevertheless it has some effect in the sense that it will mitigate the manager-agent’s inclination to maximize utility, without being able to reduce it to U°MA. In figure 6 the utility level reached by the manager is UMA,IM, which is higher than his reservation utility U°MA, but lower than the maximum utility attainable (U*MA,IM) given the transformation curve with monitoring (starting at PMAX,M).

Figure 6: Manager-agent, with imperfect monitoring.

The agency cost here is PMAX-PMA,IM, where PMAX-PMAX,M can be defined as the monitoring cost, and PMAX,M-PMA,IM as residual loss. The implementation of a monitoring system is rational only if it can be expected that PMA,IM>P*MA, as is the case in figure 6. Reasoning on the potential influences on the probability that PMA,IM>P*MA(p(PMA,IM>P*MA)) gives rise to a number of testable hypotheses.

A first series of hypotheses explores, in a ceteris paribus context, the effect of (PMAX-PMAX,M), the monitoring cost. It is clear that the larger this difference turns out to be, the lower p(PMA,IM>P*MA) will be, and consequently the lower the probability that the monitoring system will be effective, and therefore implemented. It is fair to assume that the monitoring cost is (strongly) positively correlated with the cost of the AA system. Remember that we deal with an AA system that is additional to the one possibly already in place for reasons of internal efficiency. Theoretically, the latter may generate enough information to an effective board to make managerial behaviour perfectly observable, a case dealt with in the previous section. As the fixed cost of devising and implementing an AA system is considerable, the relative cost of this is smaller in larger NPOs, leading to the hypothesis that in larger NPO, ceteris paribus, more AA systems will be found as monitoring devices (hypothesis 2).

Read in combination, hypotheses 1 and 2 both predict a positive relationship between the organization’s size and the presence of AA. Empirically, this poses the problem of disentangling efficiency determinants and monitoring effects. In some circumstances it is possible that accounting is introduced in the NPO for efficiency reasons, whereas auditing is added as an instrument for monitoring (see section

Increasing complexity and diversity of activities increases the cost of an AA system apropopriate for monitoring purposes, resulting in a lower probability, again ceteris paribus, of occurrence (hypothesis 3). This effect is opposite to the one postulated in hypothesis 1, where efficiency reasons are put forward to explain the existence of AA. One could argue that the efficiency argument is dominant: not running a large hospital, for example, with a performing accounting system (probably including a sophisticated cost accounting system) would make the transformation curve shrink in such a way that the ensuing utility level for the manager-agent would be even lower than UMA,IM.

2.3.3. A refinement Introduction

Up to now AA was presented as a monolithic take-it-or-leave-it system, with a given cost. Reality is of course more subtle. At least three components with their own costs and effects seem relevant in an NPO context: the choice between no accounting, cash accounting or accrual accounting, disclosure and auditing.

The board’s choice will depend on the eventual level of P to be reached. Consider, as an example, figure 7.

Figure 7: Decomposition of accounting and auditing effects.

Five transformation curves are depicted based on figure 6: no accounting, cash accounting (CA), accrual accounting (AC), accrual accounting and disclosure (AC+DI), and the same situation together with auditing (AC+DI+AU). In this example cash accounting is optimal from the board’s point of view. The combination accrual accounting + disclosure + auditing is not even feasible, as it never leads to at least the reservation utility for the manager.

It is clear from this example that it is the interaction between costs and utility shifts towards u° that determines the final outcome.

In the next subsections the role of the accounting system choice, disclosure and auditing in NPOs are more thoroughly discussed. Cash accounting versus accrual accounting

Jones and Pendlebury (1996, 139) distinguish five important accounting techniques for the public and non profit sectors: budgetary accounting, commitment accounting, fund accounting, cash accounting and accrual accounting.
Budgetary accounting seems more useful in the public sector than in the private non profit sector (Herzlinger, Sherman (1980), Razek, Hosch (1985)). When commitment accounting or fund accounting are applied, the choice between cash accounting and accrual accounting still has to be made, as the former only affect the way the accounts are grouped and do not define the recognition principles (Jones, Pendlebury (1996, 156 and 158)). Therefore we confine ourselves to the choice between cash accounting and accrual accounting. This is not a choice between ‘right’ or ‘wrong’: Both are ... important sources of information about the financial results of an entity but both are incomplete. Income under the accrual basis is a better measure… for relating accomplishments to efforts… . The cash basis focuses on the important issue of an entity’s ability to generate cash from its current operations (Horngren et al. (1993, 44); see also Jones, Pendlebury (1996, 143)). Given the lower emphasis on income in NPO, one can expect cash accounting to be more useful in NPO than in profit organizations.

An underlying assumption when speculating on the NPO’s choice between cash accounting and accrual accounting is that accrual accounting can be applied in NPO. Sunder (1997, 194) is sceptical, at least for public goods: How can the accrual method be used in absence of direct exchange transactions when such transactions form the very basis of the accrual method? His fear is, in my opinion, excessive. In a large number of NPOs, transactions can be observed: this is evident in the so called commercial NPOs and mutual NPOs, which frequently are comparable to insurance companies, but also in more service-like NPOs such as hospitals or schools. This leaves us essentially with some public goods and advocacy organizations: political parties, unions, environmental organizations, religious organizations,… Here, the problem is not the absence of any visible activity, which frequently takes the form of making available (alleged) information, but the difficulty of measuring or even unambigously defining this ‘transaction’. Therefore, recognition of costs and (financial) revenues poses no conceptual problem, and consequently the choice between cash accounting and accrual accounting is a genuine choice to be made in the case of NPOs. It goes without saying that PMAX,CA>PMAX,AC, as for comparable organizations the cost of implementing an accrual accounting system is higher than for a cash accounting system (Jones, Pendlebury (1996, 143)). The question is to know whether the higher cost will have an impact as to result in UMA, IM,AC far enough to the left to result in a higher value for P than under cash accounting. Such a shift can be expected only when the board consists of persons with enough accounting knowledge, as otherwise accounting information produced cannot be correctly interpreted, let alone be of any help in monitoring. This leads us to hypothesis 4: the more accounting professionals (by education and/or experience) in the NPO board, the more an accrual accounting system will be preferred compared to a cash accounting system (see also hypothesis 7). This is completely in line with Sunder’s statement pertaining to all kinds of organizations (Sunder (1997, 29)): Each organization develops accounting suitable to its own unique characteristics so that it may serve as an effective instrument of control.

2.3.3. Disclosure

Both cash accounting and accrual accounting periodically result in a financial statement that is supposed to be evaluated and discussed during a board meeting. It should be clear that a financial statement is a report on the financial situation of the organization, even though financial well-being is not the primary (or even any) objective of the organization. Other documents are to be used to report on the achievement of the NPO’s objectives. Therefore, in my opinion, statements such as The financial reports of nonprofit organizations need to reflect the service story of the entity instead of the net income or net loss (Trigg, Nabangi (1995, 259)) are based on an unfortunate confusion of financial and non financial performance variables.

The question now is whether this statement should be disclosed to a wider public in order to reduce agency problems between the board and management. Remember that we assume for the time being that the NPO is not substantially financed by donors. The effect of the presence of donors is discussed in a later paragraph.

Theoretically, scrutiny by a larger public should make monitoring more effective, but it is realistic to say that in most cases disclosure will not be a viable option for an NPO trying to induce more appropriate managerial behavior (from the board’s point of view). In terms of figure 7: the cost of disclosure will mostly not be compensated by a commensurate shift of UMA,IM. Auditing

A last question to be dealt with in this section (where we are in a situation without donors, without regulation, and with an effective board) is the question to know whether an external auditor will be hired to certify the financial statement of the NPO, be it disclosed or not. In general the economic role of auditing is to reduce agency costs (Sunder (1997, 115)). It is well known that audit prices are differentiated (e.g. between Big 5 and non Big 5 firms), affecting the shift of the transformation curve, and that audit services are not of a homogeneous quality, with consequently different effects on UMA,IM. It can be expected that in larger NPOs, where supervision by board members is more difficult, the effect on UMA,IM as compared to the effect on the transformation curve, will be larger, even if they engage expensive auditors. This leads to the hypothesis that larger NPOs will be more frequently audited than smaller NPO’s (hypothesis 5). This is exactly the result found by Buijink (1992, 92) for Dutch profit organizations in 1926, a year chosen because at that time there was no audit regulation in the Netherlands.

Finally, it should be noticed that internal control systems in NPO are generally very weak (Ortman, Schlesinger (1997)) exacerbating the role of auditing in eliminating agency (and sometimes even efficiency) problems.

2.4. Accounting and auditing knowledge

A consideration with respect to AA costs that is typical for (smaller) NPO and has, to my knowledge, never been documented is the following: the employees (and mostly volunteers) in smaller NPOs are mostly very professional as far as their input in the realization of the NPO’s objective is concerned, but frequently lack any sound knowledge of AA concepts or techniques (and sometimes even do not want to acquire this knowledge, as it is seen, in a nostalgic 1968 spirit, as an emanation of objectionable capitalism, a faith also sometimes reserved for marketing or budgeting). A board demanding a full fledged AA system will have to realize that this will be implemented at a higher cost than in organizations employing professional accountants, reducing the probability that it will result in PMA,IM>P*MA (hypothesis 6).

Hypothesis 6 can also easily be understood as follows: the lower accounting professionality of the NPO staff makes the AA system less effective, reducing monitoring and constraints on managerial behavior.

An analogous reasoning applies to the boards. In a lot of cases, especially in smaller NPO, board members are highly involved in the realization of the organization’s main objectives, but less interested in accounting technicalities, again reducing (U*MA-uMA,IM), and therefore p(PMA,IM>P*MA). The presence of board members with an appropriate AA knowledge will therefore enhance the implementation of AA monitoring systems (hypothesis 7).

2.5 The influence of debt

From the beginning of accounting theory for profit organizations, a major role has been played by debt (see e.g. Watts (1977)). Debt can also be cast in an important role in the theory presented here. Creditors of course know they are lending money to an NPO,11 accepting that the organization pursues its objectives, but within a financial context that will enable the organization to pay interest and repay the debt in due time. Two agency situations can be discerned here: the creditors versus the board (now as agent), and the creditors versus management. AA, and more specifically auditing, will naturally arise as creditor’s instrument to monitor (professionally) the organization (hypotheses 8, 9). One effect is that managers will feel more constrained, increasing ((U*MA-uMA,IM), which increases the probability that the board (in its own principal-agent situation vis-à-vis management), will find itself in the favorable situation in which PMA,IM>P*MA. Notice that in the opposite case the creditor will still be inclined to impose AA.

When survival is at stake, either management or the board can be inclined to present the existence of an AA system as a bonding strategy. This does not influence the overall agency cost in terms of P, but only the distribution of non-residual costs between monitoring costs and bonding costs.

2.6. The effect of donor financing

There has been some literature on the effects of substantial NPO financing by gifts (individual and corporate) on the NPO’s financial structure and financial vulnerability (Chang, Tuckman (1994), Salamon, Anheier (1998)), but none, to my knowledge, on its interaction with financial disclosure or auditing. The theory developed in the previous sections can easily be adapted to incorporate this dimension, under ceteris paribus conditions. In figure 8 the effective board, imperfect monitoring case is illustrated. Other situations are analyzed analogously.

Figure 8: The presence of donors

If (actual and potential) donors could perfectly observe management’s discretionary behavior, there would be no additional problem for the board. But generally, the general public’s perception of managerial discretion is imperfect, often leading to an overestimation of D. Call TPP the transformation curve under perfect perception, and TIP under imperfect perception. Its construction is justified as follows: given a real level D1 of managerial discretion, A will be on the transformation curve, but if under imperfect perception the public perceives discretion as being at a level D2, gifts will be reduced (point C under perfect perception) so as to allow performance PB. In the figure it is assumed that there is no error in perception when D=0, but it can easily be generalized for more general perceptional patterns. Clearly, the equilibrium obtained given TIP must be worse than that on TPP. Voluntary disclosure and auditing can partly remedy this situation, as illustrated in figure 9.

Figure 9: The presence of donors: the effect of disclosure and auditing.

We can assume that (UMA,IM-uMA,IM,DI+AU) (and thus the difference between P1 and P2) will be larger, the more donors the organization has, and/or the larger it is. Therefore, we can hypothesize that we will observe more disclosure and auditing in large, donative NPOs (hypothesis 10). This might explain why large donative organizations with a substantial number of donors (such as Amnesty International or Foster Parents Plan) send their certified financial statements to almost everybody whose address they can find.

2.7. Accounting choices

One of the core topics of accounting theory for profit organizations is the study of accounting choices. Their potential influence on management remuneration through reported financial performance is one of the reasons for this attention. As such remuneration mechanisms are not frequently present in NPOs, one can wonder whether accounting choice has its place in an accounting theory of NPO. As witnessed by Chase and Coffman (1994), the answer is definitely yes. They put forward two lines of reasoning that are relevant here (ibid., 234 sqq.): a political cost argument and a traditional debt covenant argument. Political cost is defined as the negative impact on subsidies (and also gifts) of showing too much wealth. But subsidies can also enter the scene another way, viz. in situations where subsidies are determined by reported costs, such as in the retrospective hospital financing systems. In that case it is in the mutual interest of board and managers to make allowable cost increasing accounting choices, as this will induce a shift of the transformation curve to the right, by generating a higher level of funding.

Therefore, in general, we can conclude that in some circumstances, which have more to do with cost recovery and financing than with managerial behavior, accounting choices do matter for NPOs (hypothesis 11). Empirical corroboration should then take into consideration the specificities of e.g. subsidy rules or existing debt convenants.

3. Nonprofit accounting in a regulated environment

Numerous NPOs are subject to AA regulations. Applying the AA rules is frequently a necessary condition to be eligible for subsidization. These rules are clearly a monitoring device for the subsidizing authorities, though other reasons to enact AA regulations may exist (Maijoor (1991, chapter 3)). Whatever the reason might be, subsidies are a persuasive instrument, so we can expect to observe more AA in subsidized NPOs than in other NPOs (hypothesis 12).
From the point of view of the organization, the question of compliance can be examined. A decision on the degree of compliance can be analysed as an economic decision comparing costs and benefits. Potential benefits of reducing compliance are a decrease in collection costs, processing costs and auditing costs. Potential costs are possible fines or other penalties (such as reduced subsidies), additional administrative costs when corrected figures have to be produced, loss of reputation of the organization and its managers. There are some studies on compliance of firms with AA regulations (DeFond, Jiambalvo (1991, 1993), Ham et al. (1985), Hylas, Ashton (1982), Jegers, Buijink (1987), Kinney , McDaniel (1989), Nelson (1993), Willingham, Wright (1985) are good examples), and almost none on NPO (exceptions are Jegers, Houtman (1993), Krishnan, Schauer (2000)).

In the context of the theory presented here, compliance has an inward shifting effect on the location of the transformation curve, and therefore on the eventual position of UMA. As the expected financial benefits of not complying for large NPOs are relatively smaller than for smaller NPO, we can expect more compliance (and therefore financial statements of higher quality and higher quality audits) by large NPOs (hypothesis 13). The two mentioned empirical studies on NPOs do not contradict this hypothesis.

An additional argument is that the reputation cost in case of fraud is more than proportionally higher. Also, due to lower visibility, management of smaller NPOs can be supposed to have more possibilities to shift UMA to be right by non complying, meaning not reporting correct figures on costs directly related to the achievement of the NPO’s objectives and on costs of managerial discretion.

Regulation per se does not automatically render the discussion in section 2 pointless. Whether this is the case or not hinges upon the difference between the AA level required by the regulator, and the AA level necessary to cope with agency problems in the NPO. If the former is the higher, the mechanisms described in hypotheses 2-11 will be suppressed. In the opposite case, additional AA efforts will be necessary to attenuate agency problems.

4. Summary and conclusions

A positive theory of accounting and auditing in NPO is presented in an agency framework. Taking into consideration genuine characteristics of managerial utility functions, thirteen empirically testable hypotheses, mostly of the ceteris paribus type, are derived. These pertain to the presence, nature and quality of accounting and auditing in NPO, and consider different contexts in which the NPO are active: regulation or not, presence of donor financing, effective or ineffective boards, effective or ineffective monitoring. They are reproduced in Table 1.

Table 1: Overview of hypotheses


For efficiency reasons, larger NPOs with more complex activities will have more AA.

2 In larger NPOs, more AA will be present as a monitoring device*
3 In NPOs with complex activities, less AA will be present as a monitoring device*

The more AA knowledge by board members, the higher the probability to have an accrual accounting system*.

5 Larger NPOs will be more frequently audited than smaller NPOs*.

The more AA knowledge by NPO employees, the higher the probability to have on AA system*.


The more AA knowledge by board members, the higher the probability to have on AA system*.

8 The more debt, the more the NPO will be audited.
9 The more debt, the more AA.

There will be more auditing and disclosure in large donative NPOs than in other NPOs*.

11 In specific circumstances (subsidies, debt covenants,… ) accounting choices matter*.
12 There will be more AA in subsidized NPOs*.
13 Larger NPOs will comply more to AA regulations


AA: accounting and auditing
*: if the AA level required by efficiency reasons and/or regulation is not sufficient to maximally reduce agency costs

The validity of these hypotheses can only be assessed by empirical research in different countries, for different kinds of NPOs, in different circumstances. As they are all formulated at the organizational level, this seems a priori a feasible endeavour, even if the operationalization of some of the concepts may require the development of some new measurement tools. Furthermore, it is evident that further theoretical refinements should result in new hypotheses. Nevertheless, it is hoped that the present paper is a first step towards a comprehensive positive theory of accounting and auditing in NPOs. Given their importance in modern society, they deserve it.


Anthony R.N., 1989, Should business and nonbusiness accounting be different?, Boston, HBS Press, Boston.

Belkaoui A.R., 1992, Accounting theory, London: Academic Press.

Ben-Ner A., 1986, “Nonprofit organizations: why do they exist in market economies?”, in: Rose-Ackerman S., 1986, The economics of nonprofit organizations, New York:Wiley.

Buijink W.F., 1992, “Some evidence on the demand for external auditing in an unregulated environment: the case of the Netherlands”, in: Buijink W.F., 1992, Empirical financial accounting research: compliance with regulations, distributional proporties of financial ratios and demand for external auditing, Maastrich, UPM.

Chang C.F., Tuckman H.P., 1994, “Revenue diversification among non-profits”, Voluntas, 273-290.

Chase B.W., Coffman E.N., 1994, ”Choice of accounting method by not-for-profit institutions: accounting for investments by colleges and universities”, Journal of Accounting and Economics, 233-243.

Conrad D., Wickizer T., Maynard C. Klastorin T., Lessler D., Ross A., Soderstrom N., Sullivan S., Alexander J., Travis K., 1996, “Managing care, incentives, and information: an exploratory look inside the ‘Black Box’ of hospital efficiency”, Health Services Research, 235-259.

DeFond M.L., Jiambalvo J., 1991, “Incidence and circumstances of accounting errors”, Accounting Review, 643-655.

DeFond M.L., Jiambalvo J., 1994, “Debt covenant violation and manipulation of accruals”, Journal of Accounting and Economics, 146-176.

Dyl E.A., Frant H.L., Stephenson C.A., 1996, Governance structure and performance of nonprofit corporations: evidence from medical research charities, University of Arizona: Working Paper.

Eldenburg L., 1994, “The use of cost information in total cost management”, Accounting Review, 96-121.

Eldenburg L., Hermalin B.E., Weisbach M.S., Wosinska M., 2000, Hospital governance, performance objectives, and organizational form. Working Paper.

Fama E.F., Jensen M.C., 1983, “Separation of ownership and control”, Journal of Law and Economics, 301-325.

Gassler R.S., 1997, “The economics of the nonprofit motive: formulation of objectives and constraints for firms, nonprofit and workers’ enterprises”, Journal of Interdisciplinary Economics, 265-280.

Ham J., Losell D., Smielsiauskas W., 1985, “An empirical study of error characteristics in accounting populations”, Accounting Review, 387-406.

Handy F., 1995, “Reputation as collateral: an economic analysis of the role of trustees of nonprofits”, Nonprofit and Voluntary Sector Quarterly, 293-305.

Hansmann H., 1987, “Economic theories of nonprofit organizations”, in Powell W.W., 1987, ed., The nonprofit sector: a research handbook, New Haven: Yale University Press.

Herman R.D., Heimovics R.D., 1990, “The effective nonprofit executive: leader of the board”, Nonprofit Management and Leadership, 167-180.

Herman R.D., Heimovics R.D., 1991, Executive leadership in nonprofit organizations: new strategies for shaping executive-board dynamics, San Francisco: Jossey-Bass,.

Herzlinger R.E., Sherman H.D., 1980, “Advantages of fund accounting in ‘nonprofits’”, Harvard Business Review, 94-105.

Hewitt J.A., Brown D.K., 2000, “Agency costs in environmental not-for-profits”, Public Choice, 163-183.

Horngren C.T., Sundem G.L., Elliot J.A., 1993, Introduction to financial accounting, London: Prentice Hall.

Hylas R.E., Ashton R.H., 1982, “Audit detection of financial statement errors”, Accounting Review, 751-765.

Jegers M., Buijink W., 1987, “The realiability of financial accounting data bases: some Belgian evidence”, International Journal of Accounting, 1-21.

Jegers M., Houtman C., 1993, “Accounting theory and compliance to accounting regulations: the case of hospitals”, Financial Accountability and Management, 267-278.

Jensen M.C., Meckling W.H., 1976, “Theory of the firm: managerial behavior, agency costs and ownership structure”, Journal of Financial Economics, 305-360.

Jones R., Pendlebury M., 1996, Public sector accounting, London: Pitman Publishing.

Kinney W.R., McDaniel L.S., 1989, “Characteristics of firms correcting previsouly reported quarterly earnings”, Journal of Accounting and Economics, 71-93.

Krishnan J., Schauer P.C., 2000, “The differentiation of quality among auditors: evidence from the not-for-profitsector”, Auditing: a Journal of Practice and Theory, 9-25.

Lynk W.J., 1995, “Nonprofit hospital mergers and the exercise of market power”, Journal of Law and Economics, 437-461.

Maijoor S.J., 1991, The economics of accounting regulation: effects of Dutch accounting regulations for public accountants and firms, Maastricht: Datawyse.

Middleton M., 1987, “Nonprofit boards of directors: beyond the governance function”, in Powell W.W., 1987, ed., The nonprofit sector: a research handbook, New Haven: Yale University Press.

Nelson M.W., 1993, “The effects of error frequency and accounting knowledge in error diagnosis in analytical review”, Accounting Review, 804-824.

Newhouse J.P., 1970, “Toward a theory of nonprofit institutions: an economic model of a hospital”, American Economic Review, 64-74.

Ortmann A., Schlesinger M., 1997, “Trust, repute and the role of non-profit enterprise”, Voluntas, 97-119.

Razek J., Hosch G., 1985, Introduction to governmental and not-for-profit accounting, Englewood Cliffs: Prentice Hall.

Roomkin M.J., Weisbrod B.A., 1999, “Managerial compensation and incentives in for-profit and nonprofit hospitals”, Journal of Law, Economics and Organization, 750-781.

Rose-Ackerman S., 1987, “Ideals versus dollars: donor, charity managers, and government grants”, Journal of Political Economy, 810-823.

Salamon L.M., 1999, “The nonprofit sector at a crossroads: the case of America”, Voluntas, 5-23.

Salamon L.M., Anheier H.K., 1998, “Social origins of civil society: explaining the nonprofit sector cross-nationally”, Voluntas, 213-248.

Steinberg R., 1990, “Profits and incentive compensation in nonprofit firms”, Nonprofit Management & Leadership, 137-152.

Sunder S., 1997, Theory of accounting and control,Cincinatti, South-Western Publishing.

Trigg R., Nabangi F.K., 1995, “Representation of the financial position of nonprofit organizations: the Habitat for Humanity situation”, Financial Accountability and Management, 259-269.

Watts R.L., 1977, “Corporate financial statements, a production of the market and political process”, Australian Journal of Management, 53-75.

Watts R.L., Zimmerman J.L., 1986, Positive accounting theory, Englewood Cliffs: Prentice Hall,.

Williamson O.E., 1963, “Managerial discretion and business behaviour”, American Economic Review, 1032-1057.

Willingham J.J., Wright W.F., 1985, “Financial statement errors and internal control judgements”, Auditing: a Journal of Practice and Theory, 57-70.

Wolf C., 1988, Markets or governments: choosing between imperfect alternatives, Cambridge: MIT Press.

Young D.R., 1987, “Executive leadership in nonprofit organizations”, in Powell W.W., 1987, ed., The nonprofit sector: a research handbook, New Haven: Yale University Press.


Copyright © 2012 The International Center for Not-for-Profit Law (ICNL)
ISSN: 1556-5157