Programme de la conférence

1. Purpose of the paper

The municipal bond market is a key area in public sector financial management, as municipal bonds serve as a major funding tool for local governments. The volume and cost of borrowing directly affect financial condition. Prior studies have shown that accounting disclosures by local governments influence interest rates in municipal bond markets. Other research highlights the importance of fiscal transparency, financial sustainability, and institutional design in shaping borrowing costs and investor perceptions.

Municipal bonds are issued in two main markets: public offering bonds (public market) and private placement bonds (private market). Public markets involve a wide range of investors, with pricing determined by market dynamics, while private markets involve a limited number of investors and rely on bilateral negotiations. Local governments must decide which market to access, and understanding how accounting information contributes to pricing in each context is critical.

In corporate bond research, it is well-established that firms with lower accounting quality or less rigorous disclosure policies tend to prefer the private placement market, where disclosure requirements are less demanding. However, few studies—whether in corporate or municipal contexts—have directly compared the role of accounting information in pricing between public and private bond markets. This study addresses that gap.

2. Research methodology

We identify two methodological challenges in comparing the impact of accounting information across markets. First, ordinary least squares (OLS) regression assumes linearity and may fail to detect complex interactions. Moreover, statistical significance in both markets may obscure differences in the relative importance of accounting variables. Second, empirical analysis requires large-scale, standardized data, which are generally scarce for private markets due to their negotiated and decentralized nature.

To overcome these challenges, we apply machine learning to capture nonlinear relationships and use Shapley Additive Explanations (SHAP) to measure the contribution of accounting variables to interest rate determination. We focus on Japan, which has the second-largest municipal bond market globally. Nearly all private placements are managed by a single securities firm and recorded in a centralized database (JSPRICE), providing rare access to comprehensive private market data. Japanese local governments also follow standardized accounting rules, ensuring comparability across entities.

3. Main findings and implications

Our analysis constructs a dataset covering both public and private municipal bond markets in Japan and applies machine learning to assess the contribution of accounting information in each. The results show that while accounting information is relevant in both markets, its relative contribution is smaller in the private market. This indicates that private market investors rely more on non-accounting or soft information when pricing risk.

The implications are twofold. From a research perspective, this finding suggests that private bond investors possess advanced capabilities in processing private information, resulting in structurally distinct pricing mechanisms. From a public financial management standpoint, the results imply that financially distressed local governments may be better served by accessing the private market—where accounting information plays a smaller role—and by strategically disclosing relevant private information during negotiations.

Purposes and objectives: Prior studies call for more empirical studies that investigate the use of cost and MA information for different purposes to attend to the needs of information of wide stakeholders. In the context of Portugal, the legal and normative frameworks are pressuring local governments to develop a full cost accounting system since 2000 and MA systems since 2020. But, prior literature evidence problems and delays in the implementation process of the formal and normative requirements (Carvalho, et al., 2012; Jorge et al., 2020), especially after the adoption of the new IPSAS-based accounting system (SNC-AP), which become mandatory for local governments since 2020. Complementing the institutional pressures for public sector accounting changes with the specific contingencies in the Portuguese context, our study aims two research objectives: (i) to understand how the new accrual-based MA system has been developed in the particular context of Portuguese municipalities, and (ii) for what purposes are MA information being developed.

Design/methodology: After describing the context of research and how legal and normative frameworks have evolved in the last 20 years for Portuguese municipalities, our findings are based on the perceptions of the producers of information (CFO - Chief Financial Officers) collected by a survey addressed to all the 308 municipalities in 2024. This is the first dataset that aggregates the CFO perceptions about the development of MA systems after the adoption of the SNC-AP in 2020 and its use for different purposes. Our sample is representative of the whole population (49% of responses). The original survey was built on a literature review basis and was pilot-tested by academics and practitioners.

Main Findings and Implications: The main findings highlight the different rhythms to which municipalities are developing MA. The maturity in the implementation of the new accounting changes and the adaption of the structures and staff to operationalize cost and MA information evidence a positive and significant influence, which reinforces the role of the contingency theory. In addition, CFO perceived different uses of management accounting information for different purposes. Although conceptually, MA is more related to internal decision-making processes, our findings highlight the role in promoting accountability for the wide stakeholders. This empirical research enhances the theoretical arguments on the intertwined between management accounting, accountability and governance (Napier & Carnegie, 2023; Liguori & Kelly, 2023; Major et al., 2023). In addition, the paper provides an innovative study on the use of MA information looking at the perception of local CFO.

In the last decades, the study of the use of financial information by politicians has raised interest among academics (e.g. Van Helden, 2016; Nogueira et al., 2023).

Overall, those studies have essentially addressed the types of information politicians use, their purposes, and the circumstances influencing their preferences. Authors have mainly focused on accessibility and usability of financial and accounting information (type, format, media, …) (e.g. Buylen & Christiaens, 2015; Caruana & Farrugia, 2018; Giacomini, 2020; Jorge et al., 2016, 2019), including a users’ needs standpoint, and how those and other issues affect information use (e.g. Guarini, 2016; Liguori & Steccolini, 2018; Sinervo & Haapala, 2019; van Helden et al., 2023). Most studies focused on local and central governments, while comparative-international studies have been limited. Also, qualitative approaches have prevailed (Nogueira et al., 2023).

Existing literature has not focused much on measuring the level of use of financial information by politicians. Some proposals came from Haustein et al. (2021), Faber & Budding (2022) and Pajković et al. (2023). Recently Jorge et al. (2024) proposed the FIU Index, to assess the degree of use of financial information by politicians in Portuguese small to medium size municipalities. The FIU index considered the number of times a certain accounting term (both cash and accrual-based) is mentioned by politicians during the meetings approving the annual budget and the annual accounts, both in local executive and council bodies.

This paper aims to further validate the FIU Index by exploring its application to a small group of large municipalities (200,000 to 600,000 inhabitants) from Portugal (Cascais, Lisboa, Porto and Vila Nova de Gaia) and Italy (Genova, Verona, Bari and Messina) in the period from 2020 to 2023. The research applies a multiple case study methodology (Yin, 2014), following a descriptive qualitative approach, relying on documental analysis of the minutes of the local councils (first stage), expected to be complemented with some online interviews (second stage) with the politicians in the local councils and some aldermen (executive).

The study explores the potential of the index to both qualitative and quantitative analysis. In particular, we intend to calculate the FIU Index for the municipalities considered, to assess them according to the level of use of financial information measured through an FIU Index, adapted to incorporate semantic differences that exist across countries and jurisdictions.

Thus, our research questions are:

1. What is the level of use of financial information by politicians in local councils, measured through the FIU Index, considering different countries?

2. Is financial information used more in the budgeting or reporting phases?

3. Which organizational and institutional circumstances/drivers seem to affect the level of use of financial information by politicians in municipalities?

We expect to contribute to the literature by supporting the validation of the measurement tool FIU Index. Moreover, the study brings qualitative implications for practice providing insights about issues that may hinder or enhance the use of financial information by local politicians, included the municipalities’ organizational and institutional context.

In recent years, there has been a growing body of research on earnings management (EM) in the public sector. Much of this literature relies on established EM models and employs quantitative methods to examine accrual-based EM, highlighting evidence of such practices (Bisogno & Donatella, 2022).

This study examines real EM practices by local government employees in the allocation of COVID-19 grants from central to local governments. In doing so, it sheds light on the understudied phenomenon of real EM in the public sector, illustrating how financial decision-making during times of crisis can affect public sector accountability. Based on 11 interviews with employees from the financial department of a Norwegian municipality, complemented by archival research, the paper identifies discretionary actions involving cash flow adjustments and post-year-end entries.

Understanding how local government employees exercise discretion in financial reporting is essential for ensuring transparency in financial management and for effectively monitoring performance (Beck, 2018). By focusing on real EM in local governments, this study seeks to complement the literature on accrual-based EM in the public sector, thereby addressing a gap in our knowledge of financial practices within these organizations. The paper investigates potential EM activities, the scope of discretionary behavior, and the underlying incentives driving such actions. The exceptional context of the COVID-19 pandemic, alongside extraordinary reimbursements from central governments, provides a suitable setting for this exploration.

Public choice theory serves as the theoretical framework for understanding potential real EM in this study. It offers a lens through which to examine the motivations and incentive structures that may drive financial decision-makers to alter reporting practices in pursuit of specific outcomes. By focusing on individual and organizational self-interest within the public sector, the theory illuminate the strategic use of discretion in financial reporting. While alternative frameworks, such as institutional theory, could provide valuable insights into how norms, routines, and organizational culture influence financial behavior, this study argue for, and adopts, the EM perspective to explore how municipal managers may deliberately adjust financial reports to meet performance benchmarks.

The main finding of this study is that employees exercise discretion in their financial decision-making with the intent of achieving specific goals - most notably, in ways that result in actual cash flow alterations. This discretionary behavior is primarily motivated by a desire to meet financial targets and avoid the negative consequences of underperformance. While research on real EM in local governments remains limited, the findings of this study are consistent with prior research in the not-for-profit sector (Eldenburg et al., 2011; Wen et al., 2019; Habib et al., 2022).

This study underscores the importance of transparent financial management and highlights the need for nuanced and context-sensitive regulatory approaches. Its alignment with findings from private-sector research; and more notably, research on not-for-profit organizations and municipally owned entities, suggests a potential universality in the incentives that drive earnings management. These parallels offer valuable insights into the financial behavior of local governments and reinforce the relevance of applying EM frameworks in public-sector contexts.