Impact of Accountability on the Economic Performance of Hospital Entities: The Moderating Role of Gender

Abstract

This study examines on the association between the accountability and economic performance of Portuguese hospital entities after the entry into force of the new Public Accounting Standards, taking into account stakeholder theory, legitimacy theory, and institutional theory. Furthermore, we investigate whether accountability, underlying procurement, and other entity operations act differently in public and private hospitals. For this purpose, we collected data for 56 Portuguese hospital entities in the period from 2018 to 2022. We conducted a quantitative analysis, testing the hypotheses using econometric regressions estimated with panel data. The results show a reduction in the return on assets for both sectors. Our evidence suggests that being a public hospital entity harms economic performance. In this sector, size and gender positively affect economic performance. Still, the level of provisions negatively affects it. In contrast, in private hospital entities, the determinants of their economic performance are the number of directors (with a negative impact) and provisions (with a positive one). These results show that when hospital entities are committed to governance structures and social responsibility, or environmental, social, and governance (ESG) factors underlying sustainability strategies, they can enhance their performance. This article contributes by collecting evidence on the effect of accountability on the performance of public hospital entities, the features of which tend to be different from those of private hospital entities.

1. Introduction

Economic performance reflects whether entities manage their activities well, such as sustainable procurement, which can be measured with the financial statements reported. Given the evolution that has occurred, accounting standards in the public sector are similar to those in the business sector. The adoption of International Public Sector Accounting Standards (IPSAS) allows the comparability of financial statements to be increased and for financial funds with better conditions to be obtained [1]. Meanwhile, IPSAS is supposed to increase the quality of financial information, support external users’ decision-making, and improve transparency and accountability [2].

The Portuguese Accounting Standards of the Public Sector (SNC-AP) are based on the International Public Sector Accounting Standards (IPSAS) [3], the latter of which are based on the International Accounting Standards/International Financial Reporting Standards (IAS/IFRS). With the entry into force of the SNC-AP, the financial reporting of public sector institutions became closer to the reporting of the business sector [4]. According to [5], given that the SNC-AP is consistent with the accounting standards for the business sector, this allows greater comparability between public and private organizations and the IPSAS. Other benefits of SNC-AP accounting regulation include promoting greater accountability and more transparency in management activities and the allocation of public cash, according to [3,6].

Regarding the factors that can affect profitability in the public sector, the authors of Ref. [7] state that size reduction can be used as a strategy to improve organizational performance. In the same vein, the authors of Ref. [8] found evidence that medium-size hospitals are more efficient than their counterparts. In addition, the authors of Ref. [9] report that women continue to play an insubstantial role in the public sector in the United States (USA). Consistently, the authors of Ref. [10] state that gender inequalities still exist in both public and private entities concerning the pay gap’s persistence. The authors of Ref. [11] found that despite a significant increase in the number of women holding management positions, they still do not have top positions in the case of local public administration. Furthermore, the authors of Ref. [12] highlight the importance of social responsibility in the clinical sector. The authors of Ref. [13] performed a bibliometric analysis of transparency and accountability within the scope of the governance of public hospitals and identified that quantitative research is scarce.

In addition, hospitals have a significant impact on sustainability. Consequently, they are facing increasing pressure to improve their environmental, social, and governance performance and comply with the existing regulations regarding disclosing their ESG data. Compliance with ESG factors in healthcare involves rigorous risk management, implying that hospitals understand the relationship between financial and non-financial information and ensure compliance with emerging regulatory requirements [14]. In this sense, we focus this study on governance structure and social pillars. We provide evidence that while Portuguese hospital entities are becoming committed to these ESG risk factors, such as private/public owners, accountability concerning patients, and board diversity, they are in an earlier stage, taking into account their low significance for hospitals’ performance. In addition, these ESG factors produce different impacts on public versus private hospital entities’ performance.

Given the regulatory convergence of the public sector and the business sector, and based on stakeholder theory, legitimation theory, and institutional theory, this study aims to determine the role of accountability in hospital entities’ performance. In addition, to the best of the authors’ knowledge, this study is the first to investigate the potential moderating role of gender in the relationship between hospital entities’ governance structure and their performance.

This paper is structured as follows. The next section reviews the relevant literature on the links between different determinants of economic performance for public and private hospital entities. Section 3 describes the research methodology adopted, including hypothesis development, methods, and data-collection procedures. Section 4 presents the empirical results of the research and their discussion. Section 5 ends this article with conclusions summarizing this study’s main contributions and practical relevance.

2. Literature Review

2.1. Performance Determinants

The public sector does not focus on economic benefits like private entities but on providing social benefits for the well-being of citizens, as the authors of [7] mention. Consistently, the authors of Ref. [15] argue that the public and private sectors show different performances because private operators tend to offer only services with higher returns rather than those that citizens most need. However, the structures and processes of public administration have been changing to become more flexible and citizen-oriented and move away from hierarchies. The authors of Ref. [16] say that there is silence regarding management performance in public services, justifying it with the complexity of the context compared to the private and non-profit sectors. However, the same authors indicate that in the literature, the principle of the “effective public manager” exists and that this can produce results. The authors of Ref. [1] argue that the adoption of IPSAS aims to gain or maintain legitimacy. Furthermore, these authors mention that following IPSAS allows for higher performance to be achieved or to receive better debt contract conditions and institutionalist perspectives, specifically normative and mimetic isomorphic pressures to gain and maintain organizational legitimacy.

Regarding the performance of public hospitals, Ref. [12] argues that factors such as clinical effectiveness, productivity, social responsibility, and satisfaction should be considered. According to the same authors, given that the public organization is oriented toward public well-being, the performance management of public hospitals should identify areas that require performance improvement, establish goals, and continue monitoring [15] to find empirical evidence that private hospitals are more efficient than public hospital using a Data Envelopment Analysis (DEA) methodology in the Italian Health System. These authors argue that private hospitals may contract the best medical employees because they can access more significant funds and manage their human resources without wages or normative national labor constraints.

Meanwhile, Ref. [17] classifies the factors determining performance assessment as internal and external. Internal ones are linked to efficiency and effectiveness, with efficiency indicators associated with financial factors. The external ones are related to the economy, legal, social, environmental, political, technological, and megatrends in healthcare. In the Portuguese case, Ref. [18] argues that despite the efforts to mitigate bureaucratic culture in public hospital sectors, namely, via the legal transformation of hospitals into Public Enterprise Entity Hospitals, decentralization at an intermediate level was not achieved, and a lack of transparency remains.

Ref. [19] emphasized that procurement practice has been identified as particularly important for creating efficiencies in health service delivery. In addition, monitoring the management performance of public hospitals is associated with the institutional logic underlying these entities and the fact that there is professional administration [20]. In the same vein, Ref. [21] argues that in National Health Systems, sustainability depends on the ability to optimize the use of resources, which requires better health professionals in a manager function. Ref. [22] classified the different types of administration in hospitals, connecting them to the kind of governance in each country. Portugal was categorized within the same batch of countries as France, Spain, Italy, and Greece, forming the European–Napoleonic group. In this group, the traditional administrative structure is characterized by the rule of law, with significant separation between society and unitary government, centralization of power, and little autonomy for local government [23].

Ref. [22] mentions that hospital management must see its performance analyzed in six areas: size, clinical effectiveness, patient care, safety efficiency, staff guidance, and responsible governance. Size has also been identified as a determinant of performance, according to [24]. Regarding size, Ref. [7] states that downsizing in the public sector tends to be an opportunity to reorganize services, reduce costs, and make services more efficient. Ref. [13] emphasizes the positive association between transparency and stakeholder participation as an element of good governance in corporations and a mechanism to enhance quantitative performance. According to these authors, transparency is one of the dimensions of accountability and is based on the idea that an organization must assume and declare its actions without omitting its mistakes to avoid public scrutiny. Another factor that impacts profitability is gender. Ref. [9] states that women perform management roles in entities with reduced performance. In fact, Ref. [10] mentions that several member states of the European Union began to consider the application of legislative measures that would require employers to balance gender representation on the board of directors. Ref. [10] argues that gender stereotypes are one of the causes of the gender pay gap in women. This author mentions that while in most European countries, public organizations are considered more favorable to women, in Portugal, the opposite seems to be the case. In addition, gender inequalities are still present in public and private entities. Ref. [11] finds that despite the significant increase in the number of women holding management positions, there are substantial differences with regard to the hierarchical level of the Portuguese local public administration. It was found that women mostly hold management positions at lower hierarchical levels—that is, at the base of the hierarchical pyramid, unlike men, who are concentrated in positions at higher hierarchical levels—that is, at the top of the hierarchical pyramid. For the UK, Ref. [25] collected data that reveal vertical and horizontal occupational gender segregation.

Moreover, institutional theory explores the impact of social structures, norms, and regulations on organizational behavior and performance. In the context of public administration performance, institutional theory suggests that public institutions’ performance is influenced by the social institutions and norms within which they operate [26,27,28]. Furthermore, institutional theory also emphasizes the role of institutional isomorphism, the process through which organizations within an institutional field become structurally and behaviorally similar. This concept can be applied to the study of gender in public administration, highlighting how gender norms and roles are institutionalized within organizations and how they can shape organizational practices and performance [29].

2.2. Evolution of Public Accounting Standards

In Portugal, the first step toward accounting standards is attributed to the entry into force of the Official Public Accounting Plan (POCP) and its sectoral plans [30]. This new accounting regulation aligns with public management reforms that have occurred in several countries, namely, by applying the IPSAS [2]. The SNC-AP appears with the publication of [31] based on IPSAS. Ref. [32] highlights that the implementation of changes produces effects that depend on the institutional environment of each location, namely, the cultural predisposition to accept change. In this sense, an approximation to international standards improves the acceptance of change. With the SNC-AP, it is expected to eliminate the fragmentation and inconsistencies in the different areas of activity in the public sector. Under this new accounting regulation, besides budgetary accounting, public entities disclose Financial Accounting information, including not only treasury but also integrating the accrual regime in the Financial Statements, as well as Management Accounting, contributing to the preparation of useful financial information for decision making. This follows the accrual accounting systems in IPSAS, which allows the creation of comparable, reliable data to facilitate auditing and fight corruption, according to [2]. Regarding the application of the SNC-AP, Ref. [33] highlights the additional information that is transferred to the Financial Statements, which includes the requirements not only for the initial measurement but also for the subsequent measurement, namely, the recognition of impairments, depreciation, and the possibility of using the revaluation model for tangible fixed assets. Adopting the SNC-AP will make it possible to overcome conceptual gaps in the prior regulations [34].

Thus, the SNC-AP presents a conceptual structure consisting of a budget reporting standard, the Public Accounting Standard (NCP) 26, twenty-five NCPs in the financial area, and one NCP in Management Accounting (NCP 27). Additionally, it presents a conceptual structure with the structuring concepts expressed in the NCP, Multidimensional Accounts Plan (PCM), replacing the various sectoral plans. The Framework Notes and demonstration models are also part of the SNC-AP.

3. Research Methodology

3.1. Sample Selection Procedures

The sample was obtained through the Iberian Balance Sheet Analysis System (SABI) INFORMA database. We selected the entities that belong to the industry activity code (CAE) of 86 hospitals over the period from 2018 to 2022. We considered 2018, given that the Accounting Standards for Public Sector changed and entered into force for hospital entities in 2018. The year 2022 is considered because it is the last year in which the financial statements are available, resulting in 35,433 entities.

As an additional criterion to include entities, we required that entities have the status of being active, which led to a reduction in the number of hospital entities to 24,074.

Furthermore, entities were included if they were located in the continental territory, reducing the number to 23,207 entities.

Moreover, we require that these entities’ size be at least EUR 50,000 in the year 2022 in terms of total assets. This requirement led to a significant reduction in the sample’s size, maintaining a total of 56 hospital entities.

Therefore, the final sample considered was a total of 280 entity-year observations.

It should be noted that of these 56 hospital entities, 38 are from the public sector, and 18 are from the private sector. Additionally, in 2022, 19 of these entities had a woman as the primary person responsible for management, which represents 34%.

3.2. Research Hypotheses

Given the literature review carried out in this study, the research model is defined, as shown in Figure 1.

With a larger size, it is possible to achieve economies of scale and present more economic resources that allow entities to increase their activity and be more profitable. In the same vein, Ref. [35] argues that when hospital entities increase their assets, it helps entities restructure their operations, efficiency, and profitability over time.

In contrast, Ref. [7] argues that size reduction can be a strategy to improve organizational performance in the public sector. According to these authors, reducing size can be an opportunity to reduce costs and make services more efficient. In a similar vein, Ref. [8] found results indicating that medium-sized hospitals are more efficient than their counterparts. Smaller hospitals are more efficient than medium and large facilities, with the efficiency inversely related to size [36]. Following this last trend, we formulate the first hypothesis:

Hypothesis 1.

Size has a negative impact on performance of hospital entities.

Another factor analyzed was the governance of the entities, focusing on the dimensions of gender and the number of directors.

Regarding gender, the United Nations defends equality and diversity at all levels and sectors of public administration. There is also the argument that diversity makes governance more reactive and responsible toward public interests, with benefits for the quality of services provided, which leads to an increase in trust in public organizations [37].

According to [9], women continue to be underrepresented in the US public sector. Additionally, these authors report that women are more likely to occupy leadership positions in organizations with declining performance or circumstances that create greater professional risk and are normally hired when it is necessary to make a cultural break from previous practices. Ref. [38] finds that just 23.6% of Directors in public hospitals were women. On the other hand, there are differences between men and women in motivation to work in the public sector [39]. The differences between the two genders can improve organizational aspects and help with their understanding as well as understanding the characteristics and social factors of each sector. Despite some mitigation of the inequalities of gender in public administration, there are still differences both in top management positions and payment [10,11,25]. Therefore, we posit the second hypothesis:

Hypothesis 2.

Top management by women has a negative impact on the performance of hospital entities.

According to [14], a sustainability strategy involves replacing traditional management practices with new and active practices, assuming social responsibility and environmental impacts. Meanwhile, Ref. [38] found that public hospitals needed to establish a corporate governance structure such as a board of directors and management committee. These authors argue that corporate governance structure in public hospitals is useful to separate the power and responsibilities between the directors and the government.

As for the number of directors, in the study by [40], it is clear that controversies have a negative impact on the financial performance of entities. In the same sense, Ref. [41] shows that controversies on the boards of directors of entities tend to have a negative impact on the reporting of the entities’ social performance. Therefore, we define the following hypothesis:

Hypothesis 3.

The greater number of directors has a negative impact on the performance of hospital entities.

To test the moderating effect of gender in the relation between the number of directors and hospital entities’ performance, we rely on the argument of [42] that masculinity is linked with a society with a preference for achievement, heroism, assertiveness, and material success, while femininity is associated with a society with a preference for relationships, modesty, caring for the weak, and quality of life. Therefore, we expect that women would decrease hospital entities’ performance, and we posit the following hypothesis:

Hypothesis 4.

Female gender has a negative moderating effect on the performance of hospital entities.

In addition, Ref. [12] argues that one of the factors that affects the performance of hospital entities is social responsibility. This way, we clarify whether greater accountability leads to lower profitability, as it results in higher expenditure. Conversely, Ref. [13] mentions that the transparency of governance can positively influence corporate performance through stakeholders’ participation. Given these opposite arguments, we formulate the fourth hypothesis:

Hypothesis 5.

Greater accountability has a negative impact on the performance of hospital entities.

Finally, we analyze whether the factor of being a public hospital entity harms its profitability. According to [7], the public sector aims to provide social benefits in order to increase the well-being of citizens and is not aimed at profit. These authors add that the economic rationality of privatization lies in improving efficiency and operational and financial performance. Differences between private and public hospitals are also addressed by [21], who mention that the public health system relies on a heavily bureaucratic approach, which harms changes in operational organizations, even when it is clearly obsolete, which is associated with a lack of managers’ responsibilities. In the same vein, Ref. [15] finds evidence that private hospitals perform better than public hospitals both in terms of productivity and cost saving. In the same vein, Ref. [13] argues that there is a growing disappointment with the performance of public hospitals because they are still very bureaucratic and traditional, while private sector and market mechanisms tend to be more efficient.

However, Ref. [43] argues that the majority of evidence suggests that public hospitals are as efficient or more efficient than private hospitals and that the latter tend to react more significantly to incentives, particularly financial ones, than public hospitals. Additionally, as healthcare is considered more relevant than economic performance, this leads to lower profitability in public hospital entities. Similarly, Ref. [8] finds evidence that public hospitals are significantly more efficient than private hospitals. Given the existence of contradictory arguments, we present the following hypothesis:

Hypothesis 6.

Public hospital entities hurt profitability.

3.3. Econometric Model

To test the hypotheses, we estimate the following econometric regression using panel data with fixed effects over the years. Random effects cannot be used since the number of periods is equal to the number of independent variables.

$${\mathrm{R}\mathrm{O}\mathrm{A}}_{\mathrm{i},\mathrm{t}}={\mathsf{\beta }}_1+{\mathsf{\beta }}_2{\mathrm{S}\mathrm{i}\mathrm{z}\mathrm{e}}_{\mathrm{i},\mathrm{t}}+{\mathrm{G}\mathrm{e}\mathrm{n}}_{\mathrm{i},\mathrm{t}}+\mathrm{N}{\mathrm{D}\mathrm{i}\mathrm{r}}_{\mathrm{i},\mathrm{t}}+{\mathrm{G}\mathrm{e}\mathrm{n}}_{\mathrm{i},\mathrm{t}}\times \mathrm{N}{\mathrm{D}\mathrm{i}\mathrm{r}}_{\mathrm{i},\mathrm{t}}+{\mathrm{P}\mathrm{r}\mathrm{o}\mathrm{v}}_{\mathrm{i},\mathrm{t}}+{\mathrm{P}\mathrm{u}\mathrm{b}\mathrm{H}\mathrm{E}}_{\mathrm{i},\mathrm{t}}+{\mathsf{\epsilon }}_{\mathrm{i},\mathrm{t}}$$

(1)

where

i,t, represent hospital entity i in period t.

ROA is the operational result relativized by the total assets of the hospital entity.

Size is given by the logarithm of total assets, with the expected sign being negative; that is, when the size of the hospital entity reduces, it becomes more profitable.

Gen is the gender that corresponds to a dummy variable—that is, one being female and zero in other cases. The expected sign is negative, given that the literature review carried out indicates that women tend to take over driving in more difficult situations.

NDir corresponds to the number of directors, with the expected sign being negative given that the greater the number of directors, the greater the burden on corporate bodies and the level of controversies.

Gen × Ndir is the term for the moderating effect of gender on the relationship between the number of directors and hospital entities’ performance.

Prov represents the ratio of accumulated provisions to total assets for the hospital entity, representing the entities’ level of responsibility. Thus, the expected sign is negative, as the greater the commitments present, the greater the losses recognized, and consequently, the lower the profitability will tend to be.

PubHE is a dummy variable with a value equal to one if it is a public hospital entity and zero if it is private. The expected sign is negative, as public entities essentially aim to offer services that increase the well-being of citizens.

ε is the estimation residual.

We estimate Equation (1) using the entire sample, which we denominated Model (1). For a more detailed analysis, we estimate Equation (1). For the subsample related to public hospital entities, we use Model (2), and for the subsample related to private hospital entities, we use Model (3).

As a robustness test for the results, we estimated a second equation, changing the dependent variable to net income for the period, as there may be differences in financing and tax costs between public and private hospital entities.

4. Empirical Results and Discussion

4.1. Descriptive Statistics

Figure 2 shows the evolution of the average profitability of public and private hospital entities given by EBIT relativized by total assets (ROA), allowing them to be compared.

The profitability of public hospital entities was always negative. In contrast, that of private entities was always positive from 2018 to 2022, being in both cases lower in 2022 compared to 2018. This may be due to the bureaucratic features and the promotion of citizen welfare in the case of public hospital entities. Furthermore, it is possible to verify that the evolution of the profitability of public hospital entities evolves in the opposite direction to that of private entities; it should be noted that in public hospital entities, it decreased in 2020, while in private hospital entities, it rose that year, which may be a result of the COVID-19 pandemic. In 2022, there was a convergence between the public and private sectors.

Table 1. Descriptive statistics.

	Mean	Median	Standard Deviation	Máximo	Mínimo
Panel A: Entire sample
ROA	−0.0899	−0.1113	0.1773	0.5918	−0.4912
RLOA	−0.1007	−0.1104	0.1648	0.5910	−0.4911
Size	139,094	103,573	95,580	516,929	20,818
NDir	25	26	9	40	5
Prov	0.0195	0.0133	0.0240	0.1456	0
Painel B: Public hospital entities
ROA	−0.1872	−0.1721	0.1214	0.4088	−0.4912
RLOA	−0.1868	−0.1715	0.1207	0.4096	−0.4911
Size	141,407	107,693	97,079	516,929	39,269
NDir	25	26	9	40	5
Prov	0.025	0.0200	0.0247	0.1456	0
Painel C: Private hospital entities
ROA	0.0898	0.0824	0.1117	0.59182,	−0.1311
RLOA	0.0584	0.0399	0.1066	0.5910	−0.1063
Size	134,823	93,093	93,165	420,416	20,818
NDir	24	26	9	39	5
Prov	0.0096	0.0030	0.0191	0.1121	0

presents the descriptive statistics of the total sample in panel A—that is, for public and private hospital entities. Statistics are presented for public hospital entities in panel B and for private ones in panel C.

Hospital entities present, on average, a negative return on assets, which results from negative profitability of around 18% of assets in private hospital entities, while in private hospital entities, the profitability is positive, close to 9%. The average size and provisions scaled by total assets are higher in public hospital entities, with the average number of directors being the same in both sectors. The dispersion of values is higher in public hospital entities in all variables except in the number of directors and the proportion of provisions, which are identical in both sectors.

Table 2. Correlation matrix.

	ROA	Size	Gen	NDir	Prov
ROA	1
Size	−0.2230	1
Gen	−0.0474	0.2685	1
NDir	−0.3321	0.1390	0.1614	1
Prov	−0.0912	−0.0869	0.0854	0.4142	1

contains the correlation coefficients between two variables when the dependent variable is EBIT.

The highest correlations are found between the dependent variable and size and between the dependent variable and provisions. As for the correlations between the independent variables, they are all reduced. Therefore, multicollinearity problems are not expected when estimating the regressions. It should be noted that by changing the dependent variable to net results for the period relativized by total assets, the correlation matrix remains identical.

4.2. Results and Analysis

Table 3. Factors determining the ROA of hospital entities.

	Expected Sign	Model (1)	Model (2)	Model (3)
Interception		0.0734	−0.5870 ***	0.4549 **
		(0.1391)	(0.1901)	(0.1968)
Size	(-)	0.0065	0.0378 **	−0.0209
		(0.0124)	(0.0167)	(0.0179)
Gen	(-)	0.0327	0.1206 *	−0.0434
		(0.0574)	(0.0693)	(0.1087)
NDir	(-)	−0.0021 *	−0.0006	−0.0056 ***
		(0.0012)	(0.0016)	(0.0017)
Gen × NDir	(-)	−0.0013	−0.0045 *	0.0020
		(0.0021)	(0.0025)	(0.0039)
Prov	(-)	−0.5193	−1.0701 ***	1.3900 *
		(0.3384)	(0.3948)	(0.6807)
PubHE	(-)	−0.2657 ***
		(0.0167)
No. Obs.		242	157	85
R2		0.5814	0.1095	0.2081
F-statistic		32.0881 ***	2.0081 **	2.1897 **

The numbers in ( ) represent standard errors, giving the precision of coefficient estimate statistics; ***, **, *, statistical significance at 1%, 5%, and 10%, respectively. Model (1) refers to entire sample, Model (2) refers to public hospital entities, and Model (3) refers to private hospital entities.

documents the results for estimating the regression with the ROA as the dependent variable. As a robustness test, we use a different dependent variable. Specifically, we change EBIT by net income and scale by total assets. For estimating regressions, we use panel data with cross-fixed effects.

Model 1 presents an R² of 58.14%, suggesting that the included variables explain more than half of the profitability of hospital entities. This result is robust, as it does not depend on the explained variable used for profitability when compared to the value documented in

Table 4. Factors determining the RLOA of hospital entities.

	Expected Sign	Model (1)	Model (2)	Model (3)
Interception		0.0233	−0.5931 ***	0.3853 **
		(0.1363)	(0.1889)	(0.1909)
Size	(-)	0.0082	0.0384 **	−0.0181
		(0.1221)	(0.0166)	(0.0173)
Gen	(-)	0.0334	0.1244 *	−0.0691
		(0.0563)	(0.0689)	(0.1054)
NDir	(-)	−0.0021 **	−0.0006	−0.0050 ***
		(0.0011)	(0.0016)	(0.0016)
Gen × NDir	(-)	−0.0014	−0.0046 **	0.0026
		(0.0020)	(0.0025)	(0.0038)
Prov	(-)	−0.5846 *	−1.0405 ***	0.9191 *
		(0.3316)	(0.3924)	(0.6603)
PubHE	(-)	−0.2330 ***
		(0.0164)
No. Obs.		242	157	85
R2		0.5346	0.1100	0.1820
F-statistic		26.5324 ***	2.0189 **	1.8538 **

The numbers in ( ) represent standard errors, giving the precision of coefficient estimate statistics; ***, **, *, statistical significance at 1%, 5% and 10%, respectively. Model (1) refers to entire samples, Model (2) refers to public hospital entities, and Model (3) refers to private hospital entities.

.

Regarding the size variable, it is not statistically significant. Thus, it does not allow for the support of Hypothesis 1. In the case of government structure, the direction being assumed by a woman does not present statistical significance. However, the variable number of directors is statistically significant at a 10% level, and its estimated sign is negative, consistent with the expected consideration [40,41]. Therefore, when the number of directors increases, performance tends to decrease. Therefore, these results do not allow us to validate Hypothesis 2, but they support Hypothesis 3. Thus, gender does not tend to affect performance, but a higher number of board members tends to reduce the entities’ performance. In Model 1, the moderating effect of gender is not statistically significant. Thus, Hypothesis 4 is not supported. Provisions are not statistically significant; thus, this evidence does not support Hypothesis 5. Regarding the variable that distinguishes public and private hospital entities, it is statistically significant at the 1% level. Additionally, the sign is negative, as expected. Thus, the empirical evidence shows that public hospital entities tend to perform worse than private hospital entities. This result is consistent with [7,15], which allows us to support Hypothesis 6.

Model 2 focuses on the subsample of public hospital entities and presents an R² of 10.95%. In this model, all variables are statistically significant, except for NDir. However, both size and Gen are significant at a 5% and 10% level, both exhibiting opposite signs from that expected, which do not support Hypotheses 1, 2, and 3. Therefore, regarding size, our result shows that larger public hospital entities may achieve economies of scale and present more economic resources that allow them to increase their activity and be more profitable. In public hospital entities, the moderating effect of gender is significant at a 10% level, and it has a negative sign as expected, allowing us to support Hypothesis 4. This evidence suggests that when women are in top management positions, the negative impact of the number of directors on performance is increased. Provisions are statistically significant at a level of 1%, presenting a negative sign as expected; that is, an increase in provisions tends to reduce the performance of hospital entities. This means that although increased responsibility may be beneficial by increasing the level of trust among users of hospital services, in the short term, this effect is outweighed by the increase in associated costs. This result is in line with [12], who emphasize responsibility in the clinical sector, allowing Hypothesis 5 to be confirmed.

Model 3 focuses on the subsample of private hospital entities, presenting an R² of 20.81%. In this model, the variables with statistical significance are NDir and Prov, at 1% and 10% levels, respectively. However, Prov has an opposite sign from what is expected. In turn, more provisions tend to enhance the private hospital entities, which may result from stakeholder recognition of the benefits of higher accountability. The negative sign of NDir is consistent with [40,41] because more members of the Board of Directors tend to increase the number of controversies and thus decrease the performance of the private hospital entity. Therefore, these results allow us to support Hypothesis 3. This lack of statistical significance may result from the low number of observations of private–public entities.

The results of the robustness tests are documented in Table 4.

Provisions are statistically significant at a level of 10%, presenting a negative sign as expected; that is, an increase in provisions about its assets tends to reduce the performance of hospital entities. This means that although increased responsibility may be beneficial by increasing the confidence level among patients, in the short term, this effect is outweighed by the increase in associated costs. This result is in line with [12], who emphasize responsibility in the clinical sector, allowing Hypothesis 5 to be supported. The results for the other coefficients of the independent variables are consistent with the first dependent variable, but with a slight increase in statistical significance.

5. Conclusions

The evolution of accounting standards in the national public sector, converging with IPSAS, allows comparisons across public sector entities. It also provides a comparison with entities in the business sector, given that accounting regulation of the public sector is similar to business accounting regulation. Despite the fact that it is fundamental to consider that the public sector aims to increase the well-being of citizens, economic performance should not be neglected. Moreover, in average terms, we find empirical evidence that the profitability of public hospital entities was consistently negative in the period considered, from 2018 to 2022, while in private hospital entities, it was always positive. This evidence emphasizes the need to improve procurement and other entities’ operations. On the other hand, returns have evolved in different directions, namely, diverging in 2021 with the COVID-19 pandemic and converging in 2022.

Regarding the determinants of economic performance, the results show that while size and gender tend not to have a significant impact on private hospital entities’ profitability, they are significant for the economic performance of public hospital entities. Furthermore, we find that more prominent public hospital entities tend to exhibit better economic performance, consistent with the benefits of scale economies. In the case of gender, it is significant just for public hospital entities and has a positive effect on economic performance. In this sector, gender has a moderate effect on the impact of the number of directors on economic performance, supporting its negative influence. Meanwhile, the government variable relating to the number of board members is significant for the economic performance of private hospital entities, and the empirical results suggest that the greater the number of directors, the lower the economic performance, which may be due to the increase in controversies. Social responsibility tends to reduce the economic performance of public hospital entities, reflecting that despite the benefits of increasing the confidence of users of hospital services, in the short term, this effect is exacerbated by the increase of associated costs. In private hospital entities, it positively impacts economic performance, suggesting that their users appreciate the commitment to social responsibility. Additionally, the results show that being a public hospital entity has a negative impact on economic performance, which is consistent with the primary objective of these institutions to provide services that promote social well-being and bureaucratic features.

Overall, the results show that public hospital entities have lower economic performance compared to private hospital entities, and the ESG factors that affect their performance are different: in public hospital entities, size and gender produce a mediating effect as well as a moderating effect, but in private hospital entities, the determinant ESG factors are accountability related to patients and the number of directors. Thus, this study contributes to the existing literature by providing empirical evidence on some ESG factors underlying a sustainability strategy that affect hospital profitability, giving useful information to several stakeholders, such as public hospital managers, accounting regulators, citizens in general, and the users of hospital services in particular.

The main limitation of this research is related to information collection concerning earlier years, namely, the public accounting standards change in 2018 and, on the other hand, the scarce disclosure of governance structure.

For future research perspectives, it would be important to increase the number of observations, both in the number of hospital entities and the number of years. It would also be important to include other determinants, such as specific measures for procurement, patient profile, variety of provided services, and other sustainability indicators, such as CO₂ emissions and compliance with ESG reporting regulations.