Guest editorial: From walking to running – What's the next step for earnings management research in public sector organizations?

Citation

Donatella, P., Bisogno, M., Cohen, S. and Stalebrink, O.J. (2024), "Guest editorial: From walking to running – What's the next step for earnings management research in public sector organizations?", Journal of Public Budgeting, Accounting & Financial Management, Vol. 36 No. 3, pp. 269-276. https://doi.org/10.1108/JPBAFM-06-2024-230

1. Earnings management research in different contexts and stages

The purpose of this special issue is to discuss and examine earnings management in public sector organizations. Earnings management can refer to either the discretion exercised by producers of financial reports within the accounting process in order to alter the financial performance presented to the organization’s stakeholders or the deliberate structuring of transactions with this specific aim (Healy and Wahlen, 1999; Ronen and Yaari, 2008). In the literature, the former is often referred to as “accounting” or sometimes “artificial earnings management,” while the latter is typically labeled as “real earnings management.”

Earnings management has generated a substantial body of literature over time. Its roots can be traced back to the 1980s and early 1990s, when the methodological foundation for this field was laid by scholars such as Healy (1985), Jones (1991) and others. Since then, earnings management has remained a prominent topic in accounting research. The significance and relevance of this phenomenon are underscored by the fact that, according to the Web of Science, six of the ten most cited articles in the history of accounting literature are related to earnings management or discretionary accruals (Runesson et al., 2022, p. 418), where the latter is a widely accepted proxy for artificial earnings management. In addition to the foundational articles mentioned earlier, the other earnings-management-related articles on the top ten list include works by Dechow et al. (1995), Dechow and Dichev (2002), Klein (2002) and Kothari et al. (2005). When the influence of time on citation counts is considered, this trend continues to be clear. In fact, an examination of time-adjusted citation counts reveals that the top five cited articles in the accounting literature pertain to earnings management or to discretionary accruals more broadly (Runesson et al., 2022, p. 419).

However, the extant body of literature has largely centered on earnings management in the context of private sector entities. Although earnings management in the public sector has been a subject of debate in the literature for some time (Anthony, 1985; Dafflon and Rossi, 1999; Stalebrink, 2007), it has only recently started to gain traction as an empirical research area. In a comprehensive literature review, Bisogno and Donatella (2022) mapped earnings management research in public sector organizations since the 1980s. They identified a total of 72 published articles, with a clear upward trend from 2010 onwards, signifying a growing body of research that builds upon and extends prior work. This emerging body of research has proceeded along three streams that are more or less separate: one focusing on national governments, another on subnational governments and organizations operating within these governments and a third on government-owned enterprises (Bisogno and Donatella, 2022).

Even though the most recent literature review on earnings management in the public sector covers the period up to 2020, it can be observed that articles on earnings management in public sector organizations continue to be published at a much higher rate than before 2010 and that recent research is primarily confined to two of the three identified streams. More specifically, recently published empirical research has primarily been conducted in the context of subnational governments (e.g. Ferreira, 2023; Goto and Yamamoto, 2023; Donatella et al., 2024), or organizations operating within these governments (e.g. Langella et al., 2021; Anessi-Pessina and Steccolini, 2024), and government-owned enterprises (e.g. Serra Coelho, 2022; Ruggiero et al., 2022). In addition, a handful of articles that fall within these two streams are published in this special issue.

In summary, we are witnessing a growing recognition of earnings management as a research area in the context of public sector organizations. Nevertheless, compared with research in the private sector, the current state of earnings management research in public sector contexts is still in its early (i.e. “walking”) stages, and there is potential for further progress and advancement (i.e. “running”). In the next section, we summarize and discuss what we have learned from the research conducted so far, with a particular focus on the articles included in this special issue. In the third and final section, we discuss the future of the literature, highlighting what we believe to be promising areas for future research on earnings management in public sector organizations.

2. Earnings management research in public sector organizations: the insights gained so far

The main take-away message from our examination of the earnings management literature is that earnings management is prevalent in public sector contexts. It is driven by a variety of political, regulatory and financial incentives and often revolves around utilizing accounting discretion to eliminate deficits and/or reduce excessive surpluses. The literature in this area is beginning to reach an initial level of maturity, with researchers building on and benefiting from each other’s work.

As depicted in Table 1, the five articles included in this special issue contribute to advancing the state of knowledge on this topic in several ways. The article from Haugdal et al. (2024) investigates whether and to what extent hard or soft regulatory control can affect earnings management behavior. More specifically, it investigates the previously untested hypothesis that decentralization of control increases opportunistic financial reporting. The study is motivated by the introduction of a decentralized control regime in the Norwegian context. Conducted at the local government level, the researchers compare the degree of earnings management under different regulatory control regimes. Using both the Jones model (Jones, 1991) and its modified version (Dechow et al., 1995), they find that the introduction of a soft control regime does not affect financial reporting discipline in municipalities. The study’s findings are notable as they draw attention to the potential role that different control regimes may have on earnings management at a local government level. These findings also contradict the private control benefit line of reasoning, according to which hard control regimes are expected to limit the ability of local governments to achieve benefits in managing their accounts. Hence, the article suggests that motivations – rather than the control regime – can explain earnings management behavior.

The study by Ferreira (2024) investigates whether mayors tend to manage earnings motivated by their re-election. As such, the article seeks to shed additional light on a largely unexplored – albeit important – topic in earnings management: the role of earnings management in election politics. Previous research on this topic at the local government level is confined to a study by Cohen et al. (2019), which showed that there is an inclination in Greek municipalities to engage in earnings management the year before the election year and that, when the mayor of a municipality is re-elected, earnings management is higher, ceteris paribus, in comparison with that in municipalities where the mayor is elected for the first time. The work by Ferreira (2024) is based on a sizable sample of Portuguese municipalities for the period 2005–2016, covering three election cycles. As her point of departure, Ferreira utilizes the Jones model (Jones, 1991) to estimate discretionary accruals and the distribution approach developed by Burgstahler and Dichev (1997) to identify discontinuities in reported net income. Her findings suggest that Portuguese municipalities generally engage in earnings management, with this behavior becoming more pronounced in pre-election years. However, no qualitative differences in earnings management were identified when comparing municipalities run by first-time-elected mayors with those run by re-elected mayors.

In a similar vein, the article by Malkogianni (2024) adds to the body of work centering on the role of political factors in earnings management. Conducted in the context of Greek municipalities, the article examines the role of political factors coupled with specific budget execution items, considered as proxies of vulnerability and sustainability, in earnings management. The study covers the period 2011–2019 and examines 1831 financial and budget execution statements, documenting that earnings management is motivated by the municipalities’ dependence on subsidies, especially during a pre-election year. This article offers a valid contribution to the earnings management literature as it is the first study investigating budget execution items. It is also noteworthy in that it provides evidence of the role of political factors, similar to the work of Ferreira (2024), along with additional variables linked to the financial sustainability of municipalities.

The article by Capalbo et al. (2024) investigates municipally owned entities (MOEs), with a focus on the Italian context. Motivated by the introduction of strict budget regulations to control local government spending behavior, this study contributes to the growing body of research on earnings management in MOEs by focusing on the effect that reducing the spending capabilities of controlling municipalities can have on the manipulation of financial statements by controlled entities. The researchers hypothesize that such regulation can be positively associated with earnings management practices. Applying Benford’s law (Benford, 1938) to reported net income figures, in addition to Burgstahler and Dichev’s (1997) distribution approach, the study indicates that the extent of data manipulation grows as the municipality’s ownership stake increases.

The contribution by Müller and Sidki (2024) is a conceptual article focusing on earnings management in the context of MOEs. The article adds to the theoretical foundation that can help guide future empirical research in these types of hybrid organizations, for which theoretical frameworks applicable to sub-government entities do not apply. With political economic theory as a starting point, the authors discuss and launch propositions concerning the role of election cycles, interest fragmentation and partisan ideology in relation to earnings management in MOEs.

3. Reflection on the next step for earnings management research in public sector organizations

Although the three streams of earnings management literature in the public sector context have been relatively separate thus far, they share some important similarities. Aside from their shared interest in the same empirical phenomenon, it is evident that research in this area is predominantly quantitative and tends to favor economic theories (Bisogno and Donatella, 2022). In other words, this literature has primarily been influenced by the positivistic research tradition. The evolution of this literature typically follows a distinct research cycle, in which initial ideas are introduced and subsequently tested with increasingly rigorous methods and data. With this in mind, it can be argued that the literature would benefit if we – as authors, reviewers and editors – continue to allow the testing of fundamental earnings management ideas in order to ensure the validity and reliability of results produced by earlier studies. Future research could further investigate the determinants of earnings management practices, scrutinizing incentives due to earnings-based targets, analyzing political and external factors and examining the mediating role of supreme audit institutions and auditors. Examinations of the implementation of whistleblowing frameworks and legislation on earnings management practices could be another promising research area. Furthermore, future research could examine the consequences of earnings management practices on the cost of debt capital as this has been documented in the private sector context (Dechow et al., 2010) but has not yet been tested in the public sector context. Closely related, earnings management has nevertheless been examined in the context of municipal bond issuances (Beck, 2018).

Considering that public sector organizations do not aim to make a profit, citizens and other stakeholders are not only interested in earnings; therefore, other (non-financial) information – including that provided on a voluntary basis – can also play a role. Future studies could then investigate whether an association exists between earnings quality and decisions to voluntarily disclose additional information regarding sustainability or environmental issues. This may serve to enrich the theoretical foundation of public sector earnings management research by referring to systems-oriented theories such as the stakeholder, legitimacy and institutional theories. Studies could go even further to analyze whether there is an association between the quality of financial information and that of nonfinancial information.

That said, it is equally important to acknowledge that incremental progress is an essential aspect of the research cycle. As mentioned earlier, the literature is beginning to display promising signs of initial maturity, with researchers building upon and benefiting from each other’s work. However, the detection of earnings management through discretionary accruals is still based on models that were developed for the private sector, such as the Jones and modified Jones models. Cases in which earnings management is proxied with less conventional methods are usually context specific and cannot be easily applied outside of the context in which they have been developed. The theoretical justification and development of earnings management detection models that take account of the financial particularities of public sector entities might be an interesting way forward. Understanding which models better capture earning management behaviors and under what conditions might also enhance the quality and comparability of studies.

There are, of course, not only similarities but also differences among the three streams. While the research stream focusing on national governments has its roots in economics, the stream focusing on subnational governments and organizations operating within these governments and the stream examining government-owned enterprises share a common foundation in the accounting literature (Bisogno and Donatella, 2022). As previously mentioned, it is the latter two streams that have been most active in recent years. In terms of research design, the standard approach employed in these two streams involves single-country studies, with only a few exceptions (e.g. Cohen et al., 2019), whereas earnings management research in the national government context tends to be comparative in nature. There is certainly potential to advance the literature through more single-country studies, especially by taking into account different periods with exceptional characteristics related to cutback management, austerity policies or extensive funding (e.g. the health COVID-19 crisis), which may alter earnings management motives and attitudes. However, we believe that it would be especially valuable to conduct future country-comparative earnings management research at the sub-national level and for government-owned enterprises. Still, the performance of quantitative analyses at all levels of government presupposes access to a complete set of financial data time series. Unlike the private sector, financial statement information about public sector entities is not always easily available, and it is not uncommon for researchers to develop hand-collected datasets. As a considerable evolution in the availability of financial information for public sector entities is currently underway, the execution of more sophisticated, comparative and multidimensional earnings management studies will become possible.

Thoughtfully designed comparative studies can, for example, enhance our understanding of the role of administrative traditions in earnings management. Other highly relevant points for comparison could include the prevalence of earnings management under different accounting bases (e.g. modified accrual vs. full accrual), standards (e.g. International Public Sector Accounting Standards vs. national public sector accounting standards) and auditing and oversight systems. Examining how vulnerable a variety of systems and standards – or combinations thereof – are to earnings management would carry significant practical relevance. Undoubtedly, the adoption of International Public Sector Accounting Standards or European Public Sector Accounting Standards (in the EU member states) by public sector organizations is expected to facilitate comparative studies.

Contemporary public sector accounting research encompasses a wide array of methodological approaches. Considering this, we see untapped potential for the earnings management literature to evolve by involving scholars who favor more qualitatively oriented methodologies. Such research can certainly complement the current streams of literature and contribute to advancing the research agenda. For instance, there have been calls for research in the public sector context that explores the ethical or moral aspects of earnings management (Hodges, 2023; van Helden et al., 2023). According to these scholars, such aspects can only be meaningfully explored through qualitative methods.

Another opportunity for advancing the understanding of earnings management in public sector contexts is to draw from findings generated from experiments in the field of behavioral economics. Behavioral economics is a subfield of economics that combines “… elements of economics and psychology to understand how and why people behave the way they do in the real world” (Witynski, n.d.). Experiments show that individuals systematically make decisions that are at odds with rational decision-making, particularly when faced with uncertainty. Under such conditions, they tend to revert to subjectively chosen heuristics (Tversky and Kahneman, 1974), fail to fully recognize the limitations of their own decision-making (Tversky and Kahneman, 1974) and exhibit “loss aversion” behavior – that is, they are often willing to forego long-term gains to avoid risk (Kahneman and Tversky, 1979). Related to these classical findings, which challenge theories that rely on assumptions of rational decision-making, recent behavioral public administration research has identified a wide range of biases when politicians, managers, employees and the general public process information in the government context (Grimmelikhuijsen et al., 2017; Battaglio et al., 2019). Given that studies of earnings management tend to be grounded almost exclusively in the rational decision-making paradigm, the application of such insights might offer a rich foundation for expanding the existing body of research, from both the producer and user perspectives.

Despite a steady increase in earnings management studies in the public sector during the last 20 years, this area is dwarfed by its counterpart in the private sector. We believe that, in the years to come, evolution in earnings management detection models, the availability of adequate data and inquiry through innovative research questions will provide enough impetus to transition this research area from “walking” to “running”.

Table 1

Overview of articles in special issue

Author(s)	Title	Research context	Data	Contribution to the earnings management literature in public-sector organizations
Ane Haugdal, Frode Kjærland, Levi Gårseth-Nesbakk and Are Oust	Earnings management in local governments under a soft control regime	Local governments (Norway)	Panel data for the years 2003–2018	The introduction of a soft control regime seems to have little impact on municipalities’ financial reporting discipline. These empirical findings underscore that the motivations, not the control regime, may serve as decisive factors in explaining the prevalence of earnings management
Augusta Ferreira	Mayor re-election and earnings management: evidence from Portuguese municipalities	Local governments (Portugal)	Panel data for the years 2005–2016	The article provides empirical evidence regarding the role of earnings management in election politics. The findings indicate that earnings management is more pronounced in pre-election years. Still, no qualitative differences in earnings management were observed between municipalities led by first-time-elected mayors and those led by re-elected mayors
Ioanna Malkogianni	Earnings management detection through budget execution. Insights from Greek municipalities	Local governments (Greece)	Panel data for the years 2011–2019	The article utilizes budget execution items to comprehend the drivers behind earnings management. The empirical evidence highlights the consequential role of political factors in understanding earnings management within the local government context
Francesco Capalbo, Luca Galati, Claudio Lupi and Margherita Smarra	Proportional appropriation systems and financial statement quality in municipally owned entities: empirical evidence from Italy	Municipally owned entities (Italy)	Panel data for the years 2010–2019	The empirical findings suggest that rules and regulations that reduce the spending capabilities of municipalities can result into controlled entities manipulating financial statements. The article offers insights into municipalities’ incentives to exert pressure on controlled entities
Harry Müller and Marcus Sidki	The political economy of earnings management in municipally owned enterprises	Municipally owned entities	Not applicable*	The article elaborates on theoretical foundations to guide future empirical research on earnings management in municipally owned entities

Note(s): *This is a conceptual article not using any data