Intergovernmental Transfers as Determinants of Municipal Fiscal Sustainability: A Review of Theory and Empirical Evidence from Polish Municipalities

Abstract

Intergovernmental transfers play a crucial role in shaping the fiscal position of local governments, especially in countries where municipalities, such as those in Poland, exhibit a high dependence on central funding. Recent reforms and the increasing reliance on discretionary revenues transferred from the central budget have motivated a closer examination of how these instruments influence local fiscal sustainability. This article analyses how different types of transfers—general subsidies and targeted grants—affect the fiscal sustainability of Polish municipalities across several dimensions, including autonomy, solvency, efficiency and economic resilience. Using panel data, five sets of models test the crowding-out effect, developmental impact, pro-cyclicality, fiscal discipline, and fiscal replacement mechanisms. Results show that general subsidies crowd out local tax revenues, particularly in less developed municipalities, while targeted grants strengthen the tax base in rural areas. Transfers have mixed effects: targeted grants strongly stimulate investment and support local development but tend to increase debt; general subsidies weaken local tax capacity and reduce fiscal autonomy, although they improve short-term fiscal discipline. In municipalities with limited fiscal independence, transfers act as short-term compensatory tools, fostering dependence on state aid rather than self-reliance. A macroeconomic crowding-out effect also appears, as higher transfers reduce private sector resources. Regarding fiscal discipline, equalization and compensatory subsidies decrease debt levels, whereas targeted grants can raise debt in urban municipalities with co-financing obligations. General subsidies show fiscal replacement effects, substituting local revenue sources. The findings provide insights for designing transfer systems that balance financial support with incentives for local autonomy and sustainable development.

1. Introduction

Intergovernmental transfers play a key role in the financing of municipalities in most modern countries. The literature identifies a number of reasons for supporting municipalities from the national budget [1]. Fiscal equalization can provide an incentive to increase the supply (scope and quality) of public services in order to ensure their minimum standard for all inhabitants, to compensate for the costs of carrying out certain tasks delegated to local government units by the national administration, to stimulate local and regional development by financing investments, to internalize externalities and, above all, to equalize fiscal differences both vertically and horizontally. In addition, grants and subsidies can be an instrument of stabilization policy aimed at reducing the negative impact of external and economic shocks, as demonstrated by the extensive use of transfers during the 2007–2009 financial crisis [2] and the COVID-19 pandemic [3,4].

Poland is a notable example of a decentralized system that is highly dependent on intergovernmental transfers and has relatively low fiscal autonomy. Under the legal framework in force until the end of 2024, local governments predominantly relied on general subsidies (for education, equalization and compensation) and a wide range of targeted grants. These formed the backbone of local revenues. Persistent vertical and horizontal fiscal imbalances further reinforced this dependence. The reform of the local government finance system introduced in 2024 although officially intended to reduce fiscal dependence, brought only limited structural change and failed to address the fundamental problems of the equalization mechanisms. This institutional setting makes Poland an instructive case study for analyzing how transfer systems shape local fiscal behavior and the incentives faced by decentralized units.

Research on the role of corrective and compensatory mechanisms directed at local governments has recently received increasing attention in the academic community. A comprehensive review of the literature on this topic can be found in the works of: Boadway and Shah [5], Yilmaz and Zahir [5] and Lago et al. [6]. It should be noted that the results of previous studies on the fiscal equalization instruments are inconclusive and sometimes contradictory [7,8,9]. In contrast, relatively little space is devoted in the literature to the study of other effects associated with fiscal equalization. A review of the literature shows that most Polish and foreign studies on intergovernmental transfers focus on whether they achieve their redistributive objectives. What is often overlooked is the unintended consequences of these mechanisms, such as their impact on tax policy, the size of the local government sector, spending decisions, the impact on municipalities investment activity, or the cyclical nature of the market mechanism. This study seeks to fill this gap.

The aim of the paper is to assess the direction and strength of the relationship between the correctional-equalizing mechanisms applied in Poland, in the form of general subsidies and targeted grants (earmarked subsidies) from the state budget, and selected aspects of the financial management of municipalities. Intergovernmental transfers are an important component of local government budgets in Poland. Between 2007 and 2023, general subsidies accounted for 25.7% of local government unit (LGU) revenues, while targeted grants accounted for 23.6%.

The article contributes to research on the consequences of the use of intergovernmental transfers. First, the analysis focuses on the role of transfers as fiscal incentives that affect the tax and expenditure decisions of local governments, while ignoring the redistributive consequences of subsidies and grants. Second, empirical research on the above-mentioned effects of intergovernmental transfers has been conducted mainly in well-developed countries, including Australia, Belgium, Italy, Japan, Spain, Sweden and the US. This article examines these effects in the context of Polish municipalities and towns with county rights. Third, the study provides a multidimensional assessment of how different types of intergovernmental transfers shape the fiscal sustainability of municipalities. By jointly analyzing the crowding-out mechanism, developmental impacts, cyclical effects, fiscal discipline and fiscal replacement within one empirical framework, the research demonstrates how general subsidies and targeted grants influence various components of fiscal autonomy, solvency, stability and resilience. Fourth, the results of this study may have practical implications, as understanding the direct and indirect effects of grants and subsidies and the adjustment mechanisms implemented in the local government units is crucial for the proper design of intergovernmental transfers in the future.

The paper is structured as follows: Section 2 provides a literature review. Section 3 describes the institutional background of intergovernmental fiscal transfers in Poland, while Section 4 discusses the research method and data. Section 5 presents the results of the research and the discussion. The article concludes with a summary that highlights the practical implications of the research.

2. Research and Literature Review

The theoretical basis for determining the role and importance of intergovernmental transfers is provided by the concept of fiscal federalism, which focuses on fiscal relations between different levels of government. In this line of research, one can observe an evolution in the approach to corrective equalization mechanisms for LGUs, which is the consequence of the different assumptions made in the first and second generation theories of fiscal federalism (First Generation Federalism—FGF and Second Generation Federalism—SGF, respectively).

The FGF has the character of a rather idealistic concept in which public authority seeks to maximize the welfare of citizens and is not influenced by politically motivated factors [10,11]. Consequently, well-designed transfers ensure the effectiveness of decentralization by promoting the equitable satisfaction of local needs [12]. Research within the FGF stream focuses mainly on the role of transfers in reducing vertical and horizontal fiscal disparities and on the internalization of external effects.

The SGF, on the other hand, is more pragmatic and assumes imperfect information and self-interest on the part of politicians. In this view, transfers can act as fiscal incentives affecting the behaviour of local authorities and resource allocation. Thus, the impact of transfers on local governments’ fiscal strategies is analyzed, including tax policy [13], expenditure [14] or debt management [15]. In addition, the SGF draws attention to the potential distortionary effects of transfers, which can have a negative impact on economic development [5]. Research within the SGF stream focuses primarily on income and substitution effects of transfers to LGUs. Income effects are concerned with how changes in grants affect the overall availability of funds to LGUs. Substitution effects, on the other hand, analyze the impact of transfers on the prices of subsidized public goods (net of subsidies) and thus on LGU spending decisions.

With regard to the consequences of intergovernmental transfers as fiscal and economic stimuli, the literature points, among others, to the following phenomena [6]:

• the crowding-out/fungibility effect, whereby financial support from the central budget can lead to a ‘crowding out’ of municipalities’ own tax revenues, discouraging local governments from improving tax efficiency because it is politically easier to rely on transfers;
• the flypaper effect, whereby an increase in transfers to municipalities leads to a greater increase in local government expenditure than if there is a corresponding increase in own revenues;
• the possibility of using fiscal transfers to stimulate local and regional development;
• the pro-cyclicality of fiscal transfers, as they increase when the economy is growing and decrease when it is contracting;
• the weakening of fiscal discipline, as transfers can lead to overspending, lower local tax efficiency, operating deficits and debt (Kornai’s ‘soft budget constraint’ hypothesis [16] and the ‘tragedy of the commons’);
• the fiscal replacement effect, which manifests itself in the asymmetric response of local governments to increases or decreases in transfers. According to this concept, local governments tend to compensate for possible cuts in transfers by increasing local taxes in order to maintain the existing level of local government expenditure. As a result, reductions in local government expenditure following a reduction in the level of transfers will be much smaller than increases in such expenditure when central budget support is increased [17].

Figure 1 presents a conceptual framework illustrating how income, substitution and incentive effects operate through intermediate mechanisms (fiscal space adjustment, relative price effects and behavioral and strategic responses) and translate into the main consequences of intergovernmental transfers discussed above.

Taken together, these mechanisms illustrate the various short- and medium-term consequences of intergovernmental transfers. However, their cumulative impact also determines the long-term fiscal sustainability of local governments. In line with recent literature, fiscal sustainability is defined as the enduring capacity of public finance to maintain solvency and fiscal balance over time. As Yuan and Yang [18] highlight, it is a multidimensional concept encompassing revenue, expenditure, fiscal risk and development objectives. Municipalities’ behavioral responses to intergovernmental transfers—such as changes in spending, debt or tax effort—can thus either enhance or undermine local fiscal sustainability. For example, the flypaper effect and the fiscal replacement effect may indicate a deterioration in fiscal sustainability by fostering dependence on external sources and reducing fiscal autonomy. Conversely, well-designed equalization and development grants can improve fiscal sustainability by stabilizing local revenues, improving fiscal discipline and supporting the continuity of essential public services. Therefore, analyzing intergovernmental transfers through the lens of local fiscal sustainability provides a more comprehensive understanding of the short- and long-term effects of different transfer instruments on the behaviour and financial resilience of municipalities.

However, it should also be noted that a number of other factors influence fiscal processes in local governments. In particular, the phenomenon of the political business cycle causes incumbent local authorities to manipulate capital expenditure, deficits and sometimes even tax levels in order to increase their chances of being re-elected. The political business cycle also influences additional fluctuations in local government debt [19]. Another behavioural political economy phenomenon is the occurrence of soft budget constraints. Notably, LGUs in Poland often support public utilities with off-budget instruments. The off-budget debt of LGUs is growing and constitutes a significant item in the balance sheets of consolidated local governments in Poland. Furthermore, changes to this debt are correlated with the electoral cycle [20]. Consequently, such phenomena disrupt the analysis of the impact of intergovernmental transfers presented in this article.

Table 1. Transfers as fiscal incentives: a literature review.

Effects of Transfers	Authors	Variables Studied	Transfer Type	Findings
crowding out/crowding in effect for own tax revenues	Mogues & Benin [13]	Own revenues (districts: Ghana)	Conditional grants	crowding out
	Masaki [25]	Own revenue less agricultural taxes (Boroughs: Tanzania)	General transfers, conditional grants	crowding out
	Banaszewska [26]	Fiscal effort in property tax (Polish municipalities)	Equalization transfers	crowding out (increased subsidies reduce the fiscal effort of municipalities)
flypaper effect	Koethenbuerger & Loumeau [27]	Local government expenditures, effective PIT rate (municipalities: Switzerland)	General transfers	confirmation of the flypaper effect
	Kluza &Wójtowicz [28]	Local government expenditures (Polish municipalities)	General subsidy and targeted grants	Confirmation of the flypaper effect in the case of targeted grants
local development stimulation	Aritenang [29]	Investment expenditures per capita (municipalities: Indonesia)	Targeted grants	Stimulation of development, especially in poorer regions
	Kańduła [30]	Total expenditures and investment expenditures (selected municipalities)	General subsidy and its components	Income equalization for municipalities is not a sufficient condition for their sustainable development
	Banaszewska [31]	Investment expenditures per capita (Polish municipalities)	General subsidies and targeted grants	Positive impact of transfers on investments
pro-cyclicality/counter-cyclicality of fiscal transfers	Blöchliger & Égert [8]	business cycle variable (OECD countries)	Conditional grants	Grants influence the procyclicality of expenditures
fiscal discipline	Pettersson-Lidbom [32]	public debt per capita (municipalities: Sweden)	Equalization transfers	Increased debt
fiscal replacement	Gamkhar & Shah [33]	Road expenditures per capita (states: USA)	Road grants	No asymmetry in response to changes in expenditures
	Shani et al. [34]	tax revenues and budget deficit (municipalities: Israel)	Unconditional grants	A decrease in grants causes taxes to rise twice as much as they decrease after a grant increase
	Deller &Maher [35]	Local government expenditures per capita (Wisconsin cities)	Intergovernmental transfers	Municipalities heavily dependent on shared revenues are most likely to cut spending when these revenues decrease

Source: own compilation based on literature cited.

shows that empirical studies on the role of transfers as a fiscal stimulus yield rather mixed results. A crowding out effect has been observed mainly in less developed countries where corrective equalization instruments have a relatively high share in LGU revenues [21]. Studies on tax competition generally show a weak relationship between correctional-equalization instruments and tax preferences granted by local governments [22,23,24].

The positive impact of transfers on local government spending has been successfully confirmed using the example of Central and Eastern European countries, among others [36]. A number of empirical papers have managed to positively verify the existence of a flypaper effect [37,38]. Some empirical studies have confirmed that budget transfers to LGUs contribute to the aggravation of the business cycle and that a pro-cyclical feature characterizes the fiscal equalization mechanisms in more than half of OECD countries [8]. Other studies have also shown that the heavy use of transfers as a source of financing for LGUs leads to significant problems of fiscal discipline [39,40]. Regarding the fiscal substitution effect, some studies have shown that the marginal propensity of LGUs to spend public funds when transfers increase is higher than the propensity to cut spending when transfers decrease [34,41,42].

The observed ambiguity of the previous findings in the literature encourages to verify the described consequences of transfers using the example of Polish municipalities.

3. Institutional Background of Intergovernmental Fiscal Transfers in Poland

Intergovernmental fiscal transfers are a fundamental component of Poland’s system of public finance, ensuring the redistribution of resources among different levels of government. The decentralization reforms introduced in the late 1990s established a three-tier system of local governance—boroughs, counties and provinces, each with distinct responsibilities and expenditure needs. In Poland, counties (powiaty) form the intermediate tier of local government between municipalities (gminy/boroughs) and regions (voivodeships/provinces) and are governed by an elected county council and an executive board headed by the starosta.

However, as in many decentralized systems, subnational governments in Poland face horizontal and vertical fiscal imbalance, where their revenue-generating capacity remains insufficient to fully cover the increasing demands for public services. To address these challenges, a structured system of intergovernmental fiscal transfers was implemented through the Act on Revenues of Local Government Units from 2003 [43], which was in force from 2004 to 2024. Under the Act, Poland’s intergovernmental fiscal transfer system consisted primarily of general subsidies/general-purpose grants and targeted grants. General subsidies (GS) were designed to provide local governments with unrestricted financial support, enabling them to allocate resources based on local needs while ensuring a degree of fiscal equalization. The general subsidy system comprised three key components: the educational subsidy (SED), the equalization subsidy (SEQ), and the compensatory subsidy/regional subsidy (SCO).

The education subsidy was the largest component and helped municipalities, counties and provinces to finance public education. It was distributed according to a formula that took into account factors such as the number of pupils, specific educational needs and regional cost differences [30]. However, concerns about the adequacy of funding remained, as many local authorities argued that the allocation did not fully cover rising education costs, leading to financial shortfalls at the municipal level [44].

The equalization subsidy aimed to reduce disparities between wealthier and poorer LGUs by redistributing resources on the basis of fiscal capacity indicators. Boroughs, counties and provinces with lower-than-average per capita tax revenues received additional funding, allowing them to maintain essential public services despite a weaker local tax base.

The compensatory subsidy was targeted at municipalities and counties with specific structural challenges, such as declining population, economic downturns or historically weaker economic development. Unlike the equalization subsidy, which was based on general fiscal capacity, the compensatory subsidy took into account additional social and demographic factors affecting local government finances. In addition, the grant functioned as a Robin Hood tax, as its source of funding came from mandatory contributions from above-average wealthy areas. These wealthier jurisdictions were required to transfer some of their revenues to support financially weaker authorities, thus ensuring greater equity in local government financing.

During the period covered by the 2003 Act on Revenues of Local Government Units, Poland’s system of targeted grants (TG) provided earmarked funding for specific projects and public services. These grants covered a wide range of policy areas, including infrastructure development, health care and social assistance. Many targeted grants were co-financed by the European Union’s Structural and Cohesion Funds, which play a crucial role in financing regional development projects.

However, the deteriorating financial situation of municipalities, exacerbated by external and largely unpredictable factors—such as the COVID-19 pandemic and the war in Ukraine—led to the enactment of a new Act on Revenues of Local Government Units from 2024 [45]. The changes are manifested in an increase in the local government’s share of central income taxes and a new mechanism for allocating the general subsidy, which is no longer linked to the redistribution of national income. Unlike the previous system, its amount is no more based on parameters reflecting the fiscal situation of a given local jurisdiction. Instead, it depends on the growth rate of income tax-sharing revenues in relation to the financial needs calculated for each jurisdiction [46].

Under the new system, the calculation of local financial needs includes five separate categories: equalization, education, development, environment and additional needs. The new system has not changed the allocation formulas of targeted grants, which still remains an important source of intergovernmental transfers. The 2024 reform marks a fundamental change in Poland’s transfer system, moving away from a formula-based, multi-component grant structure to a mechanism based on revenue growth. While the new approach aims to bring greater stability and predictability to subnational finances, its long-term impact on fiscal autonomy and equity remains controversial. Figure 2 shows how the significance of intergovernmental revenues has evolved in relation to local government revenues in different LGU categories: rural boroughs (RB), municipal-rural boroughs (MRB), municipal boroughs (MB) and towns with county rights (TCR).

4. Methods

The research on the role of compensatory transfers in Polish boroughs and towns with county rights used panel data models covering 2470 units. Data for the years 2007–2023 were included, giving a total of about 39,500 observations. This period was chosen in order to cover a sufficiently long time series with complete data at the municipal level from the Ministry of Finance and the GUS Local Database.

The explanatory variables in the models were both GS and TG as well as the components of the general subsidy: SEQ, SCO and SED (subsidy decomposition models). The existence of five of the previously discussed theoretical effects (the flypaper effect was already confirmed in a study by Kluza and Wójtowicz [28] was statistically verified by separate models. A list of these, together with a description of the dependent variables, is shown in

Table 2. List of model numbering, effects analyzed and description of dependent variable.

Model Group No.	Effects of Intergovernmental Transfers to Be Verified	Dependent Variables (Y) in Each Model	Rationale for Variable Selection
1	Crowding-out/crowding-in effect	Local tax revenues (excl. PIT and CIT)	Measure of local fiscal effort; PIT and CIT excluded due to lack of any tax autonomy granted to LGUs.
2	Local development stimulation	LGU investment	A quantifiable indicator of local development; local investment directly reflects the impact of LGUs on growth, as it results from their policy decisions and financial priorities
3	Pro-cyclicality/counter-cyclicality of fiscal transfers	Number of business entities within given LGU	Proxy of economic growth fluctuations (the Polish national statistical office—Statistics Poland—does not provide GDP growth data for boroughs, only for the provinces)
4a & 4b	Improving/weakening fiscal discipline	a. LGU outstanding debt b. Operating balance (surplus/deficit) in LGU	a. Reflects changes in the ability to service long-term obligations b. Reflects compliance with the golden budget rule, which requires current revenues to cover current expenditures
5	Fiscal replacement effect	LGU total expenditures	A measure of overall fiscal policy response; reflects LGUs spending adjustments

Source: authors’ elaboration.

.

The analytical procedure included the following steps a.–h.:

• Compilation of a panel of data, including elimination of 7 units where territorial conversions and two outliers occurred. For comparability of LGUs, all financial categories have been expressed on a per capita basis.
• Preliminary assessment of descriptive statistics, correlations and stationarity. Due to the non-stationarity of the series, all variables were transformed to their first differences.
• Conducting a collinearity analysis using variance inflation factors (VIF) and the Belsley–Kuh–Welsch method [47]. The variables (their first differences) showed no collinearity.
• Model construction for panel data. For each group of models 1–4, models were constructed according to two specifications (see equations I and II below). Each was then recalculated for all LGUs and separately for each LGU category, resulting in the estimated 10 models in each group (50 models in total). In addition, for group 5, models were also constructed for all LGUs and for their subcategories (5 models in total) according to the specifications in point III below.

Due to the very likely time-lag effect of transfers on the phenomena studied, models were estimated with lags for independent variables of up to two periods.

Model specifications:

I.

Equation formula—Models 1–4 with aggregate subsidies:

$$Y_{it}=\alpha +{{\beta }_{0}GS}_{it}+{{\beta }_{1}GS}_{it-1}+{{\beta }_{2}GS}_{it-2}+{{\gamma }_{0}TG}_{it}+{{{\gamma }_{1}TG}_{it-1}+{{\gamma }_{2}TG}_{it-2}+{\delta }_{t}YEAR}_t+{ϵ}_i$$

(1)

II.

Equation formula—Models 1–4 with decomposed subsidies:

$$\begin{gathered} Y_{it}={\alpha }^{\mathrm{’}}+{{\beta \mathrm{’}}_{0}SEQ}_{it}+{{\beta \mathrm{’}}_{1}SEQ}_{it-1}+{{\beta \mathrm{’}}_{2}SEQ}_{it-2}+{{\beta \mathrm{’}\mathrm{’}}_{0}SCO}_{it}+{{\beta \mathrm{’}\mathrm{’}}_{1}SCO}_{it-1}+{{\beta \mathrm{’}\mathrm{’}}_{2}SCO}_{it-2}+{{\beta \mathrm{’}\mathrm{’}\mathrm{’}}_{0}SED}_{it} \\ +{{\beta \mathrm{’}\mathrm{’}\mathrm{’}}_{1}SED}_{it-1}+{{\beta \mathrm{’}\mathrm{’}\mathrm{’}}_{2}SED}_{it-2}+{{\gamma \mathrm{’}}_{0}TG}_{it}+{{{\gamma \mathrm{’}}_{1}TG}_{it-1}+{{\gamma \mathrm{’}}_{2}TG}_{it-2}+{\delta \mathrm{’}}_{t}YEAR}_t+{ϵ}_i \end{gathered}$$

(2)

III.

Equation formula—Model 5 (Fiscal Replacement Effect):

$$Y_{it}=\alpha {+{\beta }_{0}GS}_{it}+{{\beta }_{1}GS}_{it-1}+{{\beta }_{2}GS}_{it-2}+{{\gamma }_{0}TG}_{it}+{{{\gamma }_{1}TG}_{it-1}+{{\gamma }_{2}TG}_{it-2}+{\lambda }_n\mathrm{C}\mathrm{H}\mathrm{G}{\mathrm{S}}_n+k_n\mathrm{C}\mathrm{H}{\mathrm{T}\mathrm{G}}_n{+\delta }_{t}YEAR}_t+{ϵ}_i$$

(3)

where:

Y_it—dependent variable in the respective models as described in Table 2:

independent variables for all models:

GS_it—general subsidy,

TG_it—targeted grants,

independent variables specific for models 1–4:

SEQ_it—equalization subsidy,

SCO_it—compensatory subsidy,

SED_it—educational subsidy,

0–1 variables specific for the model 5:

CHGS_n—set of four 0–1 variables (n = 1, 2, 3, 4) depicting annual changes of GS per capita in each LGU:

ch_GS_min10—GS decreased by 10% or more,

ch_GS_min5—GS decreased by 5% or more, up to 10%,

ch_GS_plus5—GS increased by 5% or more, up to 10%,

ch_GS_plus10—GS increased by 10% or more,

CHTG_n—set of four 0–1 variables (n = 1, 2, 3, 4) depicting annual changes of TG per capita in each LGU:

ch_TG_min10—TG decreased by 10% or more,

ch_TG_min5—TG decreased by 5% or more, up to 10%,

ch_TG_plus5—TG increased by 5% or more, up to 10%,

ch_TG_plus10—TG increased by 10% or more.

β, β′, β″, β‴—regression coefficients for GS, SEQ, SCO, SED variables, including those lagged by 1 and 2 periods (sub-index in the formula)

γ, γ’—the regression coefficients for TG, including those for this variable lagged 1 and 2 periods

YEAR—variables denoting individual years t

δ_t, δ’_t—regression coefficients for individual variables YEAR_t

λ_n, k_n—regression coefficients for individual 0–1 variables from the set of CHGS_n and CHTG_n

i—number of entities in a given panel

t—time dimension of the panel

α, α′—constants

${ϵ}_i$—random component—in a form suitable for WLS (this specification was ultimately chosen).

e.

Choice of model type between Fixed Effects, Random Effects and Pooled OLS based on tests for the combined significance of inequalities in group means (F-test for fixed effects in panels) and Hausman tests. The tests indicated the choice of Pooled OLS.

f.

Verification of the validity of the models obtained—both White and Wald tests showed a heteroskedasticity problem.

g.

Test of the nature of the variance of the OLS residuals in the above models.

This procedure comprised:

i.

extracting the residuals $\widehat{ϵ}$ from the OLS models,

ii.

running the regressions between independent variables (GS, TG, and other respective independent variables) and a logarithm of the squared residuals $\mathrm{l}\mathrm{n}\left({\widehat{ϵ}}^2\right)$ as dependent variable,

iii.

verifying statistical significance of regression coefficients.

These procedures showed that the independent variables were the source of heteroskedasticity of variance in the models analyzed, allowing the weighted least squares method (WLS) to be used—see Supplementary Materials Part S1.

Applying WLS eliminates the occurrence of heteroskedasticity and produces more correct model estimates than using robust standard errors [48].

h.

Final estimation of models from Equations (1)–(3) using WLS. When applying WLS, the models’ forms remain the same, but the estimation of coefficients take into account the weights. The weights are applied to minimize the weighted sum of squared residuals. We calculated the weights based on per-unit error variances rather than for the entire set, in order to capture the characteristics of individual entities more precisely. The general formula for estimating coefficients is then:

$$\widehat{\rho }=(X^{T}WX{)}^{-1}{X}^{T}Wy$$

(4)

where:

$\widehat{\rho }$—the vector of all estimated coefficients.

X—the matrix of all explanatory variables (including a column of ones if an intercept is included).

y—the vector of observed dependent variable values.

W—the diagonal weight matrix, where each diagonal element w_i is the weight for the i-th observation.

The weights w_i are typically chosen to be inversely proportional to the variance of the error term for each observation, i.e., the higher the variance, the lower weight it gets.

All models presented in this article are based on normalized data with exception of model group no. 3. This allows for easier comparison of modelling results in the models where both independent and dependent variables are pecuniary indicators. The modelling procedures were also carried out with non-normalized data, and similar results were obtained. They are not presented in this article purely due to publication size limits.

5. Results and Discussion

The WLS estimation results corroborate the previously discussed effects. For the sake of clarity, the estimates of the constant, the fixed effects for years and, only for the subsidy decomposition models, the coefficient estimates for TG are omitted. The full forms of these models are presented in Supplementary Materials Part S2.

The results of the estimations for model group 1 generally confirm the existence of a ‘crowding out’ effect of own tax revenues in the case of the general subsidy (GS), which shows a statistically significant negative relationship with the level of local taxes. It can therefore be assumed that additional support from the state budget weakens the incentives for local governments to increase the efficiency of their own tax collection. Moreover, this effect is quite permanent, as higher subsidies received in the past are associated with lower tax revenues in the current period. Considering the cumulative effect, this phenomenon is the strongest in municipal boroughs (MB) and the weakest in rural boroughs (RB). This can be explained to some extent by the design of the GS algorithm, the amount of which (especially in the case of the equalization subsidy) depends on the evolution of the per capita tax revenue ratio. The results recalculated into monetary effects of GS changes are illustrated in

Table 3. Impact of PLN 100mln change in General Subsidies on local taxes—calculation for 2023.

	All LGUs	MB	MRB	RB	TCR
General subsidy (PLN mln)	71,510	8181	16,168	22,705	24,457
Local taxes (PLN mln)	37,204	5566	9506	9921	12,212
The 3-year cumulative impact of a PLN 100 million increase in general subsidies on local taxes:
Change of local taxes in PLN mln	−10.8	−23.5	−13.6	−7.2	−9.1
Effect magnitude relative to local taxes	−21%	−35%	−23%	−17%	−18%

Source: authors’ elaboration.

. They show relative importance of that effect on LGUs revenues. The decomposition of the subsidies confirms these observations, with the education subsidy (SED) having the weakest ‘crowding out’ effect, which shows a slight ‘supporting’ effect for all boroughs.

Table 4. Group 1–3 Models testing the existence of a ‘crowding out’ effect, impact of transfers on economic development and pro-cyclicality of transfers.

	Models Based on Equation (1)						Models Based on Equation (2)—(Only Estimates for Subsidies Are Presented)
predictors ⟶	GS	GS-1	GS-2	TG	TG-1	TG-2	SEQ	SEQ-1	SEQ-2	SCO	SCO-1	SCO-2	SED	SED-1	SED-2
LGU category ↓
	Group 1: ‘crowding out’/’crowding in’ effect; (Y: change in LGU own tax revenues, excluding PIT and CIT shares, per capita)
all LGUs	−0.019 ***	−0.035 ***	−0.059 ***	0.016 ***	0.000	0.001	−0.005 ***	−0.019 ***	−0.016 ***	−0.002 *	0.004 ***	−0.003 **	0.004 ***	−0.005 ***	0.000
MB	−0.021	−0.129 ***	−0.097 ***	0.034 ***	−0.010	0.008	−0.018 **	−0.030 ***	−0.019 **	−0.000	0.002	−0.014 *	0.010	−0.003	−0.018 **
MRB	−0.022 ***	−0.039 ***	−0.082 ***	0.018 ***	−0.000	−0.016 ***	−0.011 ***	−0.015 ***	−0.016 ***	−0.007 ***	0.000	−0.006 ***	0.006 *	−0.007 **	−0.001
RB	−0.013 ***	−0.021 ***	−0.042 ***	0.007 ***	0.009 ***	0.004 **	−0.001	−0.012 ***	−0.007 ***	−0.002	0.002	−0.002 *	0.001	−0.010 ***	−0.002
TCR	−0.026 *	−0.032	−0.038 *	0.003	−0.013	0.018	−0.001	−0.069 ***	−0.025	0.003	0.011 **	−0.005	0.012	−0.019 **	0.001
	Group 2: impact of transfers on economic development; (Y: change in per capita investment expenditure)
all LGUs	0.020 ***	−0.019 ***	0.082 ***	0.620 ***	0.044 ***	−0.004	0.023 ***	0.007 **	−0.001	−0.003	0.007 **	0.005 *	−0.005	0.010 **	0.004
MB	0.069 **	−0.019	−0.004	0.680 ***	0.045 ***	−0.006	0.005	0.023	−0.002	−0.029 **	0.030 **	−0.010	0.020	−0.006	−0.014
MRB	0.041 ***	−0.006	0.055 ***	0.634 ***	0.041 ***	−0.027 **	0.026 ***	0.006	−0.004	−0.003	−0.006	0.019 ***	0.004	0.010	0.006
RB	0.019 ***	0.012	0.042 ***	0.609 ***	0.023 ***	−0.010	0.019 ***	0.010 **	−0.002	−0.010 **	0.009 **	−0.002	−0.003	0.012 **	−0.005
TCR	−0.076 **	0.081*	0.041	0.668 ***	0.109 ***	−0.121 ***	0.004	0.038	−0.070	0.011	0.012	0.003	−0.037 *	0.025	0.006
	Group 3: counter-cyclicality/pro-cyclicality; (Y: annual change in the number of enterprises)
all LGUs	0.0001	−0.0003 ***	−0.0004 ***	0.0000	−0.0001 ***	0.0000	0.00012	−0.0008 ***	−0.0002	−0.0017	−0.0029 *	−0.0064 ***	0.0037 ***	−0.0008 ***	−0.0015 ***
MB	0.0012 **	0.0002	−0.0001	0.0001	0.0001	0.0002	−0.0024 **	−0.0021 *	−0.0005	0.0416 ***	−0.0111	0.0030	0.0075 ***	−0.0018	−0.0021
MRB	0.0004 **	−0.0001	0.0000	0.0000	−0.0001 **	0.0000	0.0004	0.0000	0.0003	0.0027	−0.0056 *	−0.0116 ***	0.0038 ***	−0.0011 **	−0.0010 *
RB	0.0000	−0.0006 ***	−0.0009 ***	−0.0001 *	−0.0002 ***	0.0000	0.0001	−0.0013 ***	−0.0009 ***	−0.0053 ***	−0.0017	−0.0065 ***	0.0036 ***	−0.0005 *	−0.0016 ***
TCR	0.0001	−0.0022 ***	−0.0016 **	−0.0001	0.0000	0.0001	−0.0029	−0.0092 ***	−0.0053 **	−0.0061	−0.0103	−0.0062	0.0050 ***	−0.0015	−0.0035 **

Common notations for Table 4, Table 5 and Table 6: MB—municipal boroughs, MRB—municipal-rural boroughs, RB—rural boroughs, TCR—towns with county rights, GS—general subsidy (comprising all subsidies), SEQ—equalization subsidy, SCO—compensatory subsidy, SED—educational subsidy, TG—targeted grants, The subscripts ‘-1’ and ‘-2’ for GS, TG, SEQ, SCO and SED denote the parameter estimates for lags t−1 and t−2, respectively. Significance levels: *** p < 0.01, ** p < 0.05, * p < 0.10. Notes: Group 1: ‘Crowding out’ effect is highlighted in light grey, ‘Crowding in’ in dark grey. Group 2: The effect of promoting development (investment) has been highlighted in dark grey and the opposite in light grey. Group 3: The effect of countercyclicality has been highlighted in dark grey and the effect of pro-cyclicality in light grey. Source: authors’ calculations.

presents a summary of the model estimates, while detailed results are provided in Supplementary Materials Part S2.

The strongest crowding-out effect observed for RBs reflects differences in fiscal autonomy and economic strength across LGU categories: rural boroughs have the lowest fiscal autonomy and rely most heavily on transfers in their overall fiscal capacity, which amplifies the sensitivity of their own-source revenues to changes in general subsidies.

Targeted grants (TG), on the other hand, are a transfer that has the opposite effect on LGU tax revenues compared to subsidies. The results of the estimates indicate that there is a positive relationship between grants and local taxes, which is most pronounced in the case of RBs. Subsidies, which often finance investment projects or provide social support to the local community, indirectly build the tax base, suggesting their pro-development character, especially in less prosperous LGUs with an agricultural profile. Characteristically, this effect is almost imperceptible in TCRs. In summary, the effect of ‘crowding out’ of own tax revenues by corrective equalization mechanisms from the national budget is strongest in less developed LGUs where transfers account for a significant proportion of budget revenues.

The Group 2 models examined the relationship between transfers and the economic development of the LGUs studied (Table 4), measured by the change in per capita investment in the LGU. The analysis shows that the largest positive impact of transfers on investment activity concerns targeted grants (in this case, both current and one-year lagged effects can be observed). Also from the perspective of general subsidies, a positive relationship (including effects from previous years) can be observed with LGUs investment. This is not the case for TCRs. Looking at the different types of grants, SEQs play a particularly developmental role (although the relationship is much weaker than for grants). On the other hand, other parts of the general subsidy such as SED and SCO show a neutral relationship with the investments made in a given financial year, especially in MBs and RBs, although in the long term the relationship is mostly positive. For TCRs, on the other hand, a slightly negative relationship can be observed.

Another group of models (Group 3 models in Table 4) was used to check whether subsidies and grants were pro-cyclical. Due to the impossibility of obtaining data on GDP dynamics at the LGU level, the yearly change in the number of registered business entities was used as an explanatory variable in the model. The data in Table 4 shows that an increase in transfers over time is associated with a decrease in the number of enterprises operating in the municipality. This may indicate the existence of a macroeconomic crowding-out effect, whereby a growing public sector reduces the activity of the private sector (as a result of competition for the same resources, especially specialized human resources). Of the budget transfers analyzed for LGUs, only the SED does not have such an effect, which may be due to the fact that it is targeted at specific expenditures, which limits the competition for resources with the private sector in this case.

Regarding the possible relationship between transfers and fiscal discipline (

Table 5. Group 4 models verifying improvement or weakening of fiscal discipline due to transfers.

	Models Based on Equation (1)						Models Based on Equation (2)—(Only Estimates for Subsidies Are Presented)
predictors ⟶	GS	GS-1	GS-2	TG	TG-1	TG-2	SEQ	SEQ-1	SEQ-2	SCO	SCO-1	SCO-2	SED	SED-1	SED-2
LGU category ↓
	Group 4.a: Impact of debt; (Y: Change of LGU outstanding debt per capita)
all LGUs	−0.021 ***	−0.025 ***	−0.081 ***	0.051 ***	0.043 ***	0.039 ***	−0.032 ***	−0.022 ***	−0.026 ***	−0.010 ***	0.004	0.004	0.023 ***	0.017 ***	0.023 ***
MB	−0.035	−0.003	0.026	0.087 ***	0.068 ***	0.029	−0.041 **	−0.009	0.002	−0.061 ***	−0.004	−0.013	0.027	0.001	0.034 *
MRB	0.003	−0.026 *	−0.058 ***	0.111 ***	0.084 ***	0.081 ***	−0.023 ***	−0.014 *	−0.020 ***	−0.001	0.004	0.004	0.019 **	0.011	0.012
RB	−0.036 ***	0.029 ***	−0.072 ***	0.036 ***	0.034 ***	0.031 ***	−0.026 ***	−0.020 ***	−0.020 ***	−0.012 ***	0.002	0.006	0.008	0.007	0.011 **
TCR	−0.144 **	−0.205 ***	−0.053	0.241 ***	0.293 ***	0.132 ***	−0.054	−0.126 **	−0.183 ***	−0.010	0.034 *	0.025	−0.011	−0.044	0.059
	Group 4.b: Operating balance impact; (Y: Change of operating balance per capita)
all LGUs	0.008 *	0.109 ***	−0.010 *	0.034 ***	−0.026 ***	−0.029 ***	0.115 ***	−0.013 ***	0.003	0.003	0.000	−0.006 **	0.068 ***	−0.022 ***	−0.008 **
MB	0.127 ***	−0.083 ***	−0.024	0.024 **	−0.018	−0.011	0.085 ***	−0.008	−0.011	0.041 ***	−0.027 **	0.004	0.059 ***	−0.013 **	−0.027
MRB	0.065 ***	0.081 ***	−0.020 *	0.045 ***	−0.028 ***	−0.047 ***	0.102 ***	−0.014 ***	−0.001	−0.001	0.004	0.001	0.066 ***	−0.011 **	0.003
RB	0.067 ***	0.055 ***	0.002	0.016 ***	−0.023 ***	−0.023 ***	0.129 ***	−0.016 ***	−0.002	0.006 **	−0.004	−0.001	0.077 ***	−0.032 ***	0.000
TCR	0.174 ***	0.136 ***	−0.106 ***	0.026	−0.011	0.055 **	0.099 ***	0.039	−0.020	0.019 *	0.013	−0.012	0.087 ***	−0.022	−0.039 **

Note: The effect of improving fiscal discipline is highlighted in dark grey and the effect of worsening fiscal discipline is highlighted in light grey. Source: authors’ calculations. *** p < 0.01, ** p < 0.05, * p < 0.10.

), models with an explanatory variable of per capita debt and operating balance were tested separately. The impact of subsidies and grants on the debt of the LGUs studied varies. Equalization subsidies and compensatory subsidies promote fiscal discipline by reducing per capita debt, while TGs weaken this discipline. The strongest positive impact of SEQs and SCOs on debt reduction is observed in RBs. A particularly negative effect of TGs on fiscal discipline is observed in TCRs. The positive correlation between the level of TG and debt per capita may in this case be due to the need to borrow in order to contribute to grant co-financed investment. A difference in the impact of TGs is also observed when assessing fiscal discipline through the operating balance per capita, as TGs improve the balance in the current fiscal year but may lead to operating deficits in the longer term. This may be related to the lagged effects of grant co-financed investments, which generate recurrent expenditure in subsequent periods for maintenance of newly released infrastructure facilities and equipment.

The Group 5 models (

Table 6. Group 5 Models testing the fiscal replacement effect (Y: change in total expenditure per capita).

	Models Based on Equation (3)						Coefficients for 0–1 Variables Describing the Strength of Change in Transfers in Equation (3)
predictors ⟶	GS	GS-1	GS-2	TG	TG-1	TG-2	ch_GS_min10	ch_GS_min5	ch_GS_plus5	ch_GS_plus10	ch_TG_min10	ch_TG_min5	ch_TG_plus5	ch_TG_plus10
LGU category ↓
all LGUs	0.0142 ***	0.0667 ***	0.0762 ***	0.6674 ***	0.0341 ***	−0.0005	−0.0622 ***	−0.0465 ***	0.0555 ***	0.1055 ***	−0.0077	−0.0017	−0.0114	−0.0158 **
MB	0.1078 ***	0.0455	−0.008	0.725 ***	0.0453 ***	0.0435 ***	−0.0681	−0.1059 ***	0.0161	0.0271	−0.02	−0.0013	−0.0003	−0.002
MRB	0.0362 ***	0.0737 ***	0.0668 ***	0.6695 ***	0.0446 ***	−0.0101	0.0045	0.002	0.0264 **	0.0845 ***	−0.0287 *	−0.0015	−0.0053	−0.01
RB	0.0078	0.0556 ***	0.0587 ***	0.6571 ***	0.0245 ***	−0.0048	−0.0899 ***	−0.0488 ***	0.0771 ***	0.1078 ***	0.0164	0.0093	−0.0104	−0.0217 **
TCR	−0.0843 **	0.0872 **	0.1214 ***	0.6479 ***	0.1083 ***	−0.101 **	−0.0466	−0.1096	0.0489	0.2082 ***	−0.0178	−0.0297	−0.0988 *	−0.0191

Note: The fiscal replacement effect is highlighted in dark grey. The fiscal replacement effect is verified by estimating the regression coefficients on the right-hand side of the table. Source: authors’ calculations. *** p < 0.01, ** p < 0.05, * p < 0.10.

) confirm the existence of fiscal substitution effects for general subsidies. First, there is an asymmetric response of LGUs to changes in the level of transfers. The propensity to spend public funds when transfers increase is higher than the propensity to cut spending when transfers decrease. Second, higher absolute rates of change are also associated with higher regression coefficients, suggesting that the magnitude of fiscal adjustment is related not only to the direction, but also to the magnitude of the change in transfers. The fiscal substitution effect is most visible in RBs. In these municipalities, the ability to reduce expenditure without compromising basic administrative and social functions is quite limited under conditions of declining transfers. At the same time, the analysis provides some evidence, that the fiscal replacement effect may not occur in the case of targeted grants.

The results confirm that intergovernmental transfers can discourage local governments from increasing their fiscal effort, which is consistent with previous findings in the literature [7,10,49,50,51]. This effect is strongest in lower-income LGUs (MRBs and RBs) and weaker in wealthier ones, which maintain high fiscal efficiency regardless of transfers received. Aragón and Gayoso [52] also emphasize that poorer LGUs tend to substitute their own revenues with transfers due to the higher marginal benefits from public spending. Moreover, in less developed LGUs, transfers are often perceived as a politically easier and less costly alternative to local taxation, partly due to the influence of local “lobbies” [25,53]. The results of our study also reveal the relationship between the crowding out effect and the type of transfer. GSs constrain local tax effort more than TGs, suggesting that greater flexibility in resource allocation does not necessarily favour greater mobilization of own revenues. This finding differs from the findings of some researchers [54], but is in line with studies by Rougier [55] and Mookherjee and Ray [56]. Our findings also indicate that the transfer allocation formula in Poland reinforces the crowding out effect. This is because the equalization subsidy is conditional on the per capita tax income ratio, such that an increase in tax effort leads to a decrease in the support received. Similar mechanisms have previously been identified in the literature as an important contributor to the crowding out effect [25,53]. It seems that this could be limited to some extent by converting equalisation transfers into lump-sum block grants that are not linked to tax effort. Under the current system, LGUs that are fiscally prudent receive lower payments and are, in effect, penalised for maintaining a robust revenue base. Therefore, switching to unconditional block grants could help reduce this distortion.

Our research results generally confirm previous findings in the literature on the impact of intergovernmental transfers on local investment, emphasizing their pro-development potential. Similar mechanisms have been observed across countries. Tang et al. [57] showed that an increase in transfers in the Philippines contributed to an increase in infrastructure investment. Lago-Peñas [58] showed that capital grants to Spanish regions led to a doubling of investment spending. In Ghana, on the other hand, increased transfers improved the quality of public services, which led to an increase in local governments’ own revenues and allowed funds to be reinvested in infrastructure development [59]. Similar long-term effects were found in Brazil by Filho and Litschig [60]. Our findings show that targeted grants (TG) had the strongest positive effect on local government investment, while general subsidies (GS)—especially equalization subsidies (SEQ)—had a weaker pro-development impact. This greater effectiveness of TG is consistent with Dellmuth and Stoffel [61], who argue that targeted transfers, being tied to specific projects or sectors, better align LGU investment with funding priorities. Such targeting can support strategic development in key areas such as infrastructure or social services, as seen in the case of the European Union’s Structural Funds, among others.

In the case of examining the pro-cyclical/counter-cyclical effect of intergovernmental transfers, we adopted a slightly different research perspective in our study compared to previous studies [8]. We combined fiscal variables with variables characterizing the real economy. The dependent variable was a proxy for changes in the business cycle (i.e., the change in the number of firms in LGUs), while intergovernmental transfers played the role of an independent variable. Despite the different approach, our results confirm the negative relationship between fiscal transfers and economic activity in the municipalities studied, although this effect occurs with a lag. These findings suggest that in the longer run, both GS and TG may contribute to the amplification of business cycle fluctuations. However, while these relationships are statistically significant, they are not particularly strong.

Table 7. Standardised beta coefficients for models from Groups 1–3. The results for all LGUs, without breaking down the General Subsidy into subcategories, are presented.

	Group 1—‘Crowding Out’/’Crowding In’ Effect				Group 2—Impact on Economic Development				Group 3—Counter-Cyclicality/Pro-Cyclicality
	Beta	std. err.	t	p > t	Beta	std. err.	t	p > t	Beta	std. err.	t	p > t
GS	−0.0194	0.0020	−9.61	0.000	0.0203	0.0060	3.40	0.001	0.0058	0.0040	1.46	0.146
GS−1	−0.0297	0.0019	−15.58	0.000	−0.0159	0.0058	−2.74	0.006	−0.0137	0.0040	−3.43	0.001
GS−2	−0.0392	0.0020	−19.94	0.000	0.0549	0.0047	11.66	0.000	−0.0155	0.0038	−4.06	0.000
TG	0.0165	0.0021	7.77	0.000	0.6203	0.0055	112.42	0.000	−0.0001	0.0024	−0.03	0.974
TG-1	−0.0004	0.0016	−0.26	0.796	0.0367	0.0049	7.45	0.000	−0.0082	0.0031	−2.64	0.008
TG-2	0.0009	0.0015	0.63	0.528	−0.0034	0.0045	−0.75	0.452	0.0015	0.0027	0.56	0.574

Source: authors’ elaboration.

provides an additional summary of the strength of the relationships in models from Groups 1–3, presenting standardised beta coefficients and their respective levels of significance.

Our research provided empirical evidence on the differential impact of intergovernmental transfers on LGU fiscal discipline, as GS generally improved it, while TG led to its relaxation. The results obtained differ from most previous literature findings, which argue that GS—despite their flexibility—weaken fiscal discipline by discouraging local governments from tax effort and increasing LGUs’ dependence on external funds [62,63]. According to these findings, the lack of strict guidelines on the use of GS allows local policymakers to use them, among other things, to finance electoral promises [64], which may lead to excessive public spending. Moreover, some studies argue that equalization grants, which depend on per capita tax revenues, may have a negative impact on fiscal discipline [10]. However, our research showed the opposite relationship, as in Poland SEQ and SCO were correlated with lower debt levels. The strongest impact of this category of transfers on debt reduction was observed mainly in the least developed and financially independent rural boroughs. This suggests that in these cases GS played mainly a stabilizing role, allowing more predictable budget management, which helped to strengthen fiscal discipline.

Moreover, the literature has shown that TGs can have a positive impact on reducing local government debt, especially when they are targeted at specific economic sectors, such as social care, health care or infrastructure [65,66]. The differences in our research findings may be due to the specific nature of targeted grants for LGUs in Poland. They require a significant own contribution, which means that LGUs have to cover up to 80% of the cost of implementing the task in order to receive funding, most often by issuing debt. The contribution of our research is that TGs have different effects on the operating balance of LGUs, depending on the time horizon chosen. In the short run, they promote fiscal discipline, but their delayed effect is negative. These results are consistent with previous findings that high reliance on transfers leads to excessive current expenditure growth in the long run [58,63,67]. Our study is also related to that of Iacuzzi et al. [68], who examined not transfers per se but rather the role of fiscal autonomy in shaping the broader fiscal health of local governments. Their evidence suggests that high fiscal autonomy can present difficulties in the short term with regard to operating balances, and they emphasise the importance of temporal dynamics in the relationship between fiscal autonomy (or fiscal dependency) and the financial condition and fiscal discipline of LGUs.

The results of our research confirmed the existence of asymmetries in the budgetary responses of LGUs to changes in transfers, with the nature of this disproportionality depending on the type of transfer and the type of LGU. Our results confirmed the existence of the fiscal replacement effect in the case of general subsidies, which is consistent with previous findings in the literature [34,35,41,42]. We also show that RBs have a stronger propensity to engage in fiscal substitution, which may be due to their low fiscal autonomy and difficulties in reducing spending without compromising key public functions. This is in line with the earlier findings of, among others, Rios et al. [41], who showed that budgetary flexibility in terms of adjustments to changes in the level of transfers depends on the degree of fiscal autonomy of LGUs. At the same time, our analysis provides some evidence that TGs may be less vulnerable to asymmetric budget responses, confirming earlier findings by Deller & Maher [35], Gamkhar & Shah [33] and Goodspeed [69], who emphasized that asymmetries in the impact of changes in the level of transfers vary depending on the type of transfer. The summary of the verification results of individual theories is presented in

Table 8. Verification results of individual theories.

Model Group No.	Theory	Results
1	Crowding-out/crowding-in effect	Crowding out effect confirmed for general subsidies; the most visible for equalization subsidy and compensatory subsidy. Crowding in effect confirmed for targeted grants.
2	Local development stimulation	Stimulation effect confirmed for targeted grants (strong impact) and general subsidies (modest impact), mostly visible for equalization subsidy.
3	Pro-cyclicality/counter-cyclicality of fiscal transfers	Counter cyclicality effect of intergovernmental transfers with some mitigating effect of educational subsidy in current year.
4a & 4b	Improving/weakening fiscal discipline	Effect of improved fiscal discipline for all kinds of general subsidies. Targeted grants causing higher debt and lower operating balance.
5	Fiscal replacement effect	Fiscal replacement effect confirmed for general subsidies.

Source: authors’ elaboration.

.

Based on these findings, the overall impact of intergovernmental transfers on local fiscal sustainability is ambivalent. While counter-cyclical transfers, equalization mechanisms, and general subsidies contribute to long-term stability by enhancing fiscal discipline, crowding-out effects, fiscal replacement, and debt growth associated with targeted grants may weaken it by limiting local autonomy and increasing financial vulnerability.

The empirical patterns observed in Poland are consistent with fiscal developments reported across EU Member States in the post-COVID period. The crowding-out effects confirmed for general subsidies—especially the equalization and compensatory subsidy—reflect a growing dependence of LGUs on central transfers, which is also noted in European studies [4]. Likewise, the pro-cyclicality of intergovernmental transfers in Poland and the weakening of fiscal discipline associated with targeted grants correspond to a broader post-pandemic trend of increasing central influence over local public finances. Across the EU, the expansion of central support instruments during the COVID-19 crisis, combined with the uneven recovery of local revenues, has led to a recentralization of fiscal power. In this context, our results show that Poland follows a similar trajectory, with LGUs becoming more reliant on central transfers and showing behavioral responses consistent with reduced fiscal autonomy.

In light of these findings, it is possible to link the identified effects to specific dimensions of local fiscal sustainability. The crowding-out effects of general subsidies primarily weaken fiscal autonomy. Targeted grants, through their positive impact on investment, also tend to increase debt, resulting in an ambivalent effect. On one hand, they may support long-term resilience and development capacity; on the other hand, they can weaken solvency due to higher borrowing needs. The pro-cyclicality of intergovernmental transfers may undermine stability, especially in economically vulnerable LGUs. By contrast, the stabilizing impact of general subsidies on debt and operating balances—particularly in rural LGUs—strengthens solvency and contributes to greater stability over time. Taken together, these mechanisms show that intergovernmental transfers affect different components of fiscal sustainability in divergent ways, simultaneously supporting resilience and solvency in some areas while constraining autonomy and stability in others.

6. Conclusions

Our study confirmed that the effects described in the theory also occur in Polish municipalities, although the intensity and direction of the effects of intergovernmental transfers sometimes differ from the results obtained in other countries. It is important to note that the negative effects of intergovernmental transfers are most pronounced in the smallest and poorest municipalities, which are characterized by limited financial independence. It is precisely in relation to these municipalities that the design of the current transfer system needs to be reconsidered. The system should promote financial self-sufficiency and reducing long-term dependence on external support. However, taking into account the differences in the revenue volumes as well as marginal effects, the local tax rates should be increased on average by 71% to compensate a loss of general subsidies in local budgets (ranging from +112% increase in rural boroughs to +12% in municipal boroughs). Such a tax hike, although rational, would be extremely difficult to implement from political perspective. From this point of view, it can also be seen that the incentive to apply transfer instruments may stem from tax design and collection mechanisms. Transfers are primarily funded by indirect taxes levied by the central government. Consequently, these revenue sources are less economically distortive, particularly in the context of local tax competition, and they do not impose collection burdens on local authorities.

Our results suggest that transfers (mainly general subsidies) not only crowd out own tax revenues, but also do not support the development of a local tax base in the future. As shown, they can have a negative impact on the economic situation by contributing to a decrease in the number of enterprises in municipalities. In addition, the expanding public sector restricts the activity of the private sector, inter alia, by competing for the same resources. Furthermore, the study of the fiscal replacement effect confirmed that municipalities tend to make their fiscal policy dependent on transfers, which in the long run further reduces their fiscal autonomy.

Contrary to previous findings, the study shows that general subsidies contribute to improving fiscal discipline, while targeted grants contribute to increasing debt. This is due to the fact that GSs in Poland mainly play a stabilizing role, enabling less affluent municipalities to finance key current tasks despite limited revenues. However, their impact on investment is relatively weak compared to targeted grants.

Our research also shows that targeted grants can have a positive impact on both the financial and economic development of municipalities. They help support the local tax base and have a strong and lasting impact on local government investment.

The results point to the need to move away from a one-size-fits-all transfer system for all LGUs, regardless of their type, level of financial independence or functions. A one-size-fits-all approach that does not take into account the specificities of each municipality may lead to inefficient use of funds. Local government capital expenditure should certainly be financed mainly through earmarked grants. The support provided should also be more targeted, with the mechanisms for allocating funds specifying the direction of spending to a greater extent. Perhaps they should be based on criteria related to expenditure needs and not only tax potential, which would facilitate the mobilization of own revenues and improve the financial independence of local governments.

From a broader perspective, the findings also contribute to the discussion on local fiscal sustainability. The study shows that intergovernmental transfers influence the short- and long-term fiscal behavior of municipalities. Although equalization and counter-cyclical mechanisms can support fiscal sustainability, the confirmed crowding-out and fiscal replacement effects emphasize the risks associated with long-term reliance on external funding. Strengthening fiscal sustainability therefore requires a transfer system that promotes local revenue mobilization, encourages efficient expenditure management and establishes a stable foundation for sustainable municipal finances.

Although our research has provided important conclusions, there are some limitations that may affect their interpretation. First, we focus on the specific Polish institutional and economic context, so the observed effects cannot be fully generalized to local government systems with different governance structures and different levels of fiscal decentralization. It should be noted, however, that a model similar to the Polish one operates in many other EU countries, such as Spain, France, the Netherlands, and the Czech Republic.

Secondly, the lack of some economic statistics at the municipal level is a certain limitation. The limited availability of data has forced the use of proxy measures of economic activity, which may affect the precision of the estimates and the interpretation of the results.

Despite the limitations mentioned, the results of the study conducted can have important implications for practice. Understanding the direct and indirect effects of the provision of subsidies and targeted grants, as well as the adjustment mechanisms adopted by municipalities, is crucial for the proper design of the future structure of intergovernmental transfers.