Structural Drivers of Poland’s Renewable Energy Transition (2010–2023): Empirical Insights from Regression and Cluster Analysis

Abstract

This research investigates the structural drivers of Poland’s energy transition to decarbonization and wider sustainable development goals. With a focus on the period 2010–2023, we use longitudinal regression analysis and cluster-based segmentation to examine the dynamic interactions between investment expenditure, deployed renewable capacity, and innovation expenditure in driving renewable electricity production. Our findings suggest that although installed capacity continues to be the nearest cause of renewable energy output, innovation expenditure has an extraordinarily large marginal effect, acknowledging the system-transformational role of technology innovation in low-carbon systems. Regression specifications suggested that the establishment of Poland’s transformation process is not only guided by the growth in capital, but also by the systemic embedment of knowledge-driven innovation. Cluster analysis reveals three successive stages of sectoral development—initial growth (2010–2013), consistent expansion (2014–2019), and rapid transformation (2020–2023)—with blended policy actions and structural effects. Despite the long shadow of Poland’s coal-linked past and post-2015 stagnation in innovation, the results signal a major move towards a more low-emitting, resilient power system. The report offers empirical facts and prescriptive evidence to guide policy formulation supporting collective, innovation-driven approaches essential for driving energy change in coal-dominated economies.

1. Introduction

The transition to renewable energy has become one of the most urgent global imperatives of the 21st century. In the face of accelerating climate change, diminishing fossil fuel reserves, and growing demand for energy security, countries around the world are adopting renewable energy sources (RESs) as strategic pillars of their long-term energy systems [1]. Within this context, the European Union (EU) has recorded a steady increase in the share of renewables in its energy mix. Leading countries such as Denmark (88.8%), Portugal (87.4%), and Croatia (73.8%) exemplify the transformational potential of large-scale RES integration [2]. Poland, however, remains significantly behind these frontrunners (almost 30% RES in total energy production), largely due to its complex and historically entrenched dependence on coal. The country still relies on hard and lignite coal for nearly 70% of its electricity production [3], which presents a formidable political and economic challenge in the context of the EU goal of climate neutrality by 2050. The strategic orientation of Polish energy policy is defined in the Energy Policy of Poland 2024 (PEP 2024) [4], which rests on three pillars: just transition, zero-emission systems, and good air quality. The low-emission energy transformation is intended to support the modernization of the Polish economy, ensure energy security, and provide a fair distribution of costs while protecting the most vulnerable social groups.

Over the past three decades, Poland has made modernization efforts in its energy sector that have contributed to improving energy efficiency, expanding infrastructure, and reducing greenhouse gas emissions. Nevertheless, these efforts remain insufficient. Although renewable energy production has increased four-fold and installed capacity twelve-fold, innovation expenditure has stagnated since 2015, contradicting widely held assumptions about technology-driven progress [5,6,7]. Poland’s case offers valuable insight into the challenges of energy transition in coal-dependent economies. In response to such structural inertia, the EU has introduced new regulatory frameworks aimed at accelerating the deployment of renewables. The Renewable Energy Acceleration Act (REAA) 2025 [8], adopted as part of the revised European Green Deal, establishes binding national targets for RES electricity generation, streamlines permitting processes, and introduces the Innovation and Just Transition Fund to support coal-dependent regions. Member states are obliged to allocate at least 1.5% of their GDP annually to innovation and energy modernization initiatives, with particular emphasis on energy storage, smart grids, and digital infrastructure. By mandating innovation expenditure and enforcing coordinated policy planning, REAA 2025 directly addresses the structural asymmetries typified by the Polish case. This regulation marks a significant shift in EU energy governance—from a focus on emission targets toward a more interventionist approach that actively shapes the financial, technological, and institutional foundations of energy transition. This shift underscores the growing need for integrated, data-driven research that captures the complex interdependencies among investment, infrastructure, and innovation as key drivers of systemic energy transformation. The current EU climate targets (net zero by 2050) pose a substantial challenge for Central and Eastern European countries, whose economies remain heavily dependent on fossil fuels.

For 2010–2023, Poland’s energy transformation is a paradox: solar and wind electricity production increased four-fold and installed capacity twelve-fold but innovation expenditure normalized since 2015, contrary to expectations of technological advancement [5,6,7]. With intensive capital expenditure and infrastructure stock accumulation, transitioning to production efficiency is poorly explained. Existing literature is more likely to examine investment [9], innovation [10], and local systems [11] in isolation, as uninteractive drivers, rather than a dynamic interdependence. The central gap is still not addressed: no longitudinal, quantitative, integrated research yet has operationalized these drivers in aggregate terms to explain renewable energy outcomes. The role of innovation, as such, is always left out of macro-level modeling.

The gap is filled with regression models and cluster analysis to study simultaneously the roles of capital investment, installed capacity, and innovation in renewable electricity production in Poland from 2010 to 2023. It uncovers three evolutionary stages—early development (2010–2013), stable growth (2014–2019), and dynamic transformation (2020–2023)—and determines the changing role of each factor through time. The results are that innovation, while underestimated, statistically delivers more energy production than co-evolved capital, and hence paradigms grounded on investments are refuted. Poland therefore exhibits the EU-wide paradox: despite a fourteen-fold increase in RES since 2004 [12,13,14,15], Poland is still coal-dependent [3] with policy intransitivity and zero-to-low innovation following 2015 [5,6]. But recent developments—growing PV penetration, municipal cooperatives, and rising RES capacity [16]—indicate a tenuous but genuine trend toward integration.

We have formulated the following research questions:

• RQ1: How has the renewable energy sector evolved in Poland from 2010 to 2023?
• RQ2: What is the relationship between capital investment, installed capacity, and renewable energy output?
• RQ3: Does innovation expenditure significantly contribute to renewable energy production?
• RQ4: How do different development factors interact, and which have the greatest leverage?
• RQ5: What stages can be identified in the evolution of the renewable energy sector, and what characterizes them?

The paper’s main scholarly contribution is its empirically grounded, comprehensive model and explanation of the structural evolution of Poland’s renewable energy sector over a fourteen-year period with special focus on the dynamic relationship between capital investment, infrastructure, and innovation. By combining classical regression analysis with cluster analysis, the authors statistically quantify the comparative contribution of prime drivers of development to renewable electricity generation and identify unambiguous phases in sectoral transformation, thereby representing a sophisticated long-term vision of the process of energy transition. Adding innovation expenditure as an explanatory variable in the expanded regression model is a procedural advancement of great significance, bringing in the often-overlooked value-added technological innovation to enhance the level of energy production and efficiency. The integrated analytical strategy presents a model replicable to case studies in other countries and constitutes an astute contribution to the general debate about sustainable energy policy, system transition, and the several reasons behind decarbonization.

This paper is structured as follows. Section 2 contributes the research within sustainable development, renewable energy transitions, and energy system innovation literatures by establishing the theoretical context. Section 3 presents the data set and empirical analysis variables used, e.g., investment, innovation, and energy production indicators of Poland for the period 2010–2023. Section 4 details the methodology, i.e., building simple and multiple regression models with extended and simple structures and using cluster analysis in determining development stages in renewable energy. Section 5 provides empirical results, i.e., regression analysis, and clustering-based typology of sectoral transition. Section 6 explains the results within the framework of structural theories of energy transition, EU climate policy, and Multi-Level Perspective (MLP). Section 7 summarizes the main findings and makes conclusions in the form of policy suggestions towards an equitable innovation-driven transition to renewable energy in Poland.

2. Theoretical Background of Analysis

Renewable energy sources (RESs) play a key role in meeting the EU climate goals, increasing energy independence and developing low-carbon industries. By 2022, GHG emissions in the EU-27 will have fallen by 31% compared to 1990, with RES accounting for 23% of the energy mix. In 2023, RES accounted for 24.5% of final energy consumption, and in 2024 their share of net electricity generation reached 47.4% [2]. The main sources were wind power (39.1%), hydropower (29.9%), solar (22.4%), biofuels (8.1%) and geothermal (0.5%) [2].

Despite growing climate ambitions and the EU Renewable Energy Directive (RED) [17], coal remains a dominant energy source in several member states, notably Poland, the Czech Republic, and Germany. Poland is the EU leading producer of hard coal and, as of 2023, accounted for 34% of coal-based electricity generation in the Union [3]. While coal production in Poland fell by 25% between 2014 and 2020, consumption declined by only 15%, offset by increased imports. In 2020, Poland’s coal production was 108.5 million tons and consumption reached 111.6 million tons, ranking second after Germany (120.5 million tons produced and 151.6 million tons consumed) [3].

Total electricity production in Poland increased from 139.3 TWh in 2004 to 167 TWh in 2024 (Figure 1) with the share of RES increasing from 3.1 TWh in 2002 to 49.8 TWh in 2024 (Figure 1). Poland has made significant progress in integrating RES into its energy mix (Figure 2). The share of RES in electricity generation increased from 10% in 2015 to 30% in 2024, representing a sixteen-fold increase since 2004 [12,13,14,15]. Following the EU accession in 2004, the annual average electricity production from RES in Poland reached 19.9 TWh, with wind energy becoming the dominant source in recent years.

The structure of obtaining energy from renewable sources in Poland results primarily from the geographical conditions and resources that can be developed in this country. In Poland, renewable energy includes solar energy, water, wind, geothermal energy, energy generated from solid biofuels, biogas and liquid biofuels, as well as ambient heat obtained by heat pumps. In the long-term analysis, only four renewable energy sources in Poland are shown (Figure 2) because others were not reported by Statistics Poland (magnitude zero) [12,13,14,15].

After 2010, Polish statistics noted the share of sources such as solar energy, renewable municipal waste and bioliquids in the production of energy (Figure 3) with solar energy recording a significant increase. Photovoltaics is becoming an increasingly important component of the national electricity generation mix. The installed capacity of photovoltaic power plants in Poland currently exceeds 18 GW. During sunny days, this translates into an increasingly large share in the energy generation structure in Poland. According to data from the Energy-charts.info website [16], collected by the Fraunhofer ISE (Fraunhofer Institute for Solar Energy Systems) institute based on information from transmission system operators, in the whole of 2023 photovoltaic power plants in Poland generated 13.2 TWh of electricity—compared to 9.3 TWh in 2022, 4.6 TWh in 2021 and 1.8 TWh in 2020 [16]. The popularity of PV is due to the fact that, electricity from photovoltaic panels is becoming increasingly affordable: thanks to technological progress and scaling effects, the costs of PV modules have fallen by 90% since 2010. Currently, installation and system components of medium voltage offer the greatest potential for cost savings. By switching from low to medium voltage, the power of the subsystems can be significantly increased; at 1500 V, 10 to 12 MVA (Megavolt-Amperes) can be transmitted through the transformer, instead of the 3 to 5 MVA that are common today. Fewer transformers and fewer switching devices are needed for the same size of power plant; therefore, the specific construction and installation costs are reduced [16].

Renewable energy sources are used, in addition to electricity, to generate heat. In Poland, the average annual level of heat production from RES was 13629.505 GWh, of which solid biofuels accounted for 92% (Figure 4). Other sources are not among the main ones; therefore, the share of biogas is only 5%, and heat pumps only 0.03%. The latter source (heat pumps) is reported by Statistics Poland, only since 2014 (Figure 5).

According to PORT PC (Polish Organization for the Development of Heat Pump Technology) analysis, the share of heat pumps in the total number of single-family home heating appliances sold was 44%, while single-function gas boilers accounted for about 40% [18]. The Polish Organization for the Development of Heat Pump Technology PORT PC has recorded sales of heat pumps in Poland since 2012 (Figure 6). From that year to the end of 2023, the majority of heat pumps were sold in 2023: 203,300 units [18]. A significant share of these sales is accounted for by pumps for heating a dwelling (rooms, incubators) by more than 92% [18]. The decline in the overall sales of heat pumps, by about 40% in 2023, was caused by the increase in electricity prices, while the prices of other fuels used to heat buildings decreased. This price relationship has become unfavorable for heat pump technology, reaching a ratio of 4:1 between electricity and natural gas [18].

The capacity of power plants from RES in Poland increases from year to year. The share of RES in the production of both energy and heat translates into the efficiency (capacity) of producers, which is presented in Figure 7.

Despite a twelve-fold increase in installed RES capacity and a four-fold rise in renewable electricity generation between 2010 and 2023, coal remains dominant in Poland. In 2023, Poland accounted for over one-third of the EU coal-based electricity generation, significantly lowering its overall RES share and undermining its alignment with EU decarbonization goals. Poland must break its dependence on fossil fuels and embark on an innovation-based multi-dimensional policy incorporating investment, infrastructure, and regulatory convergence. Despite a sixteen-fold increase in RES electricity generation since 2004, stagnation in innovation investment since 2015 threatens this trajectory. This study responds to these concerns by presenting an empirical model linking investment, capacity, and innovation with energy outcomes, contributing to the discourse on sustainable energy system transformation.

3. Data Used for the Analysis

The data used in this research was drawn from the official Statistics Poland databases [12,13,14,15] and supplemented with annual reports concerning the energy and industrial sectors published between 2010 and 2023 [19,20,21,22,23,24,25,26,27,28,29,30,31,32,33]. The database includes main quantitative indicators such as the annual production of electricity from renewable sources (GWh), heat from renewable carriers (TJ), installed capacity of power plants from renewables (MW), total national primary energy production (PJ), renewable production (PJ), investment in the renewable industry (million PLN), and innovation expenses on the national scale (million PLN). All variables were cleaned and standardized to yield internal consistency and reliability over the fourteen-year period so that robust long-term statistical analysis was possible.

The decision to begin analysis in 2010 is motivated by the nature and availability of empirical data. Prior to that year, most statistical data had either hyphens or notations for the phenomenon as zero or existing in negligible quantities (“magnitude zero”) or not zero, but less than 0.5 of a unit. Such values in practice are considered non-representative for modeling purposes, particularly when applying regression or cluster analysis because they distort trends and undermine interpretive validity. Therefore, 2010 is the reference year from which renewable energy indicators became reliably measurable on a national scale and is an acceptable baseline year for quantitative analysis.

Table 1. Data used for the analysis.

Year	Generation (Production) of Electricity from Renewables [GWh]	Production of Heat from Renewable Energy Carriers [TJ	Capacity of Power Plants from Renewables [GW]	Production of Total Primary Energy [PJ]	Production Energy from Renewable Sources [PJ]	Investment Outlays [mln PLN]	Production of Electricity Total [GWh]	Expenditures on Innovation [mln PLN]
2010	10,888.8	12,231	2.178	2824.0	287.0	16,464.9	102,101.3	3272.3
2011	13,137.0	13,452	3.018	2882.4	312.8	18,030.7	108,690.3	3272.4
2012	16,878.9	19,052	4.093	3038.9	356.1	19,295.6	118,240.9	3579.3
2013	17,066.5	15,949	5.116	3005.5	357.0	20,150.4	116,395.4	2791.3
2014	19,841.8	14,275.3	5.637	2803.6	339.8	22,952.4	114,153.5	4181.8
2015	22,684.1	12,885.2	6.927	2831.3	371.0	30,573.3	118,873.0	7264.2
2016	22,807.4	13,979.2	7.902	2783.2	379.6	23,653.4	115,502.6	6472.2
2017	24,122.1	13,047.9	8.011	2682.6	383.2	21,550.4	112,071.9	4474.4
2018	21,617.2	14,809.1	8.343	2570.6	371.6	21,701.1	117,093.3	4474.5
2019	25,458.8	17,644.6	9.406	2645.6	513.6	23,300.9	124,795.6	4474.6
2020	28,226.6	21,204.8	12.325	2475.8	524.1	23,300.9	115,595.8	4474.7
2021	30,568.5	23,511.4	16.502	2537.8	534.9	24,537.8	147,715.9	4474.8
2022	37,688.6	23,353.2	22.567	2519.3	563.1	26,181.8	247,889.0	4474.9
2023	45,853.4	24,751.4	27.980	2365.0	580.4	33,216.3	314,196.1	4474.10

Source: Own elaboration based on data from Statistics Poland [12,13,14,15,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33].

reflects the overall outlook for the process of development of the Polish renewable energy market over 2010–2023. Six indicators in total are indicated: electricity from renewables (GWh), heat from renewable carriers (TJ), capacity of installed power plants using renewable sources (GW), total primary production of energy (PJ), renewable energy output (PJ), and investment expenses on RES (million PLN). Collectively, they have offered a multi-faceted image of the development of renewable energy in the country, both as a system of production and as an infrastructure source.

The data indicate strongly rising trends everywhere. Renewables generation increased over four times from 10,888.8 GWh in 2010 to 45,853.4 GWh in 2023. Installed capacity increased even more steeply—from 2178 MW in 2010 to 27,980 MW in 2023—indicating good proof of investment in renewable plant. Heat generation and renewable generation also indicated a widely increasing trend, although the odd year-on-year downward spikes are very likely the result of market place, weather, or policy conditions.

Table 1 also completes the above data with two most important characteristics of Poland’s climate of innovation and energy for 2010–2023: the total production of electricity (in GWh) and domestic innovation investment (in million PLN). The statistics show that Polish total electricity generation had a general upward trend, from 102,101.3 GWh in 2010 to the record high of 314,196.1 GWh in 2023. Overall, electricity generation increased from approximately 102 TWh in 2010 to over 314 TWh in 2023, almost three times in less than ten years.

From 2010 to 2023, the installed capacity of renewable power plants in Poland increased from 2.178 GW to 27.980 GW—more than a twelve-fold increase—and renewable electricity output grew from 10,888.8 GWh to 45,853.4 GWh, a four-fold increase. Nevertheless, year-over-year percentage growth shows the most precipitous growth in capacity from 2020 to 2023, a period of growing EU climate policy and higher domestic deployment of photovoltaics. Interestingly, the productivity of capacity (in the form of GWh per GW installed) also varied widely over the period, touching peaks in some years, e.g., 2012–2014, and then plateauing. This reflects that capacity addition does not automatically translate into proportional production, and this underlines the significance of gaining efficiency and the integration of the system.

Concurrently, innovation spending and investment spending trends diverge. Spending on renewable energy in 2023 was at historical records of PLN 33,216.3 million—twice that of 2015—but innovation spending leveled off from 2016 to around PLN 4474 million per year. When measured in terms of the output generated per million PLN of investment, the marginal rate of investment decreases without an analogous rise in innovation, visible in particular after 2016. Decoupling of such a nature would mean a decreasing return on investment unless augmented by technological upgrading. Ratio analysis indicates that investment in innovation has a higher marginal return than capital investment, as demonstrated by the regression equation (510 MWh vs. 250–280 MWh per million PLN, respectively). Thus, a relational and temporal analysis of indicators not only reveals hidden structural bottlenecks—similar to the post-2015 slowdown in innovation—but also highlights the imperative of breakthroughs at the same time in infrastructure, investment, and innovation to guarantee long-term development in Poland’s renewable energy industry.

4. Methodology

The empirical basis for this analysis is a comprehensive 2010–2023 data set focused on the key indicators reflecting the development of the renewable energy sector in Poland. The data were assembled on the basis of national statistics and authors’ own first elaborations in order to be coherent, complete, and useful for long-term analysis. The following quantitative indicators were used: (1) electricity production from renewable energy sources per year [GWh], (2) heat production from renewable energy carriers [TJ], (3) installed power plant capacity based on renewable energy sources [GW], (4) overall primary energy production [PJ], (5) renewable energy production [PJ], (6) investment in the renewable energy industry [million PLN], (7) overall national electricity production [GWh], and (8) national innovative expenditure [million PLN].

These were chosen to give a multivariate account of industry growth—ranging from manufacturing, infrastructure, economic investment, to technology. All the data were normalized before analysis to make it comparable across years and variables.

The choice of the six unique quantitative metrics monitoring the development of Poland’s renewable energy sector was inspired by their composite ability to reflect the advanced dynamics of energy transition under a national context. The indicators of renewable carrier electricity generation, renewable carrier heat production, renewable energy plant capacity installed, total primary energy output, renewable energy output (PJ), and investment spending were selected to capture infrastructural, systemic, and economic dimensions of sectoral development in 2010–2023. The authors used a multivariate analytical approach deliberately in order to evaluate output indicators and transformation drivers of structure. These indicators constitute a combined empirical foundation for cluster analysis and regression and rely on official national statistics, which from 2010 became sound and comparable. These were chosen with a conceptual aim to triangulate production performance (GWh and TJ), resource mobilization (innovation and investments), and infrastructural capacity (installed capacity) in order to allow for the analysis of growth, efficiency, and strategic orientation of Poland’s renewable energy market combined.

The choice adheres to internationally accepted templates utilized by the European Commission, Eurostat, International Renewable Energy Agency (IRENA), and International Energy Agency (IEA) which periodically observe and contrast energy transitions through similar benchmarks. For instance, the EU Regulation on Energy Union Governance [17] defines installed RES capacity, gross electricity produced from renewables, and renewable proportion in gross final energy consumption as fundamental indicators for Member States. Additionally, IRENA and IEA use aggregate production of total energy, capacity growth, and investment streams in sectoral diagnosis and projections. The measures hence ensure cross-country comparability as well as policy purposefulness, keeping the study in line with modern European analytical traditions.

The selected indicators capture some features of Poland’s energy mix and institutional structure. With Poland’s traditionally coal-dominated system and concurrent requirement for decarbonization in the electricity and heat sectors, coverage of renewable carriers’ heat (TJ) is especially warranted. Contrary to a few EU countries with lower heating requirements or district heating dominance, Poland’s energy system calls for end-to-end monitoring of renewable thermal energy, particularly from biomass, biogas, and heat pumps. Similarly, the inclusion of investment spending captures the overall significance of capital mobilization to overcome infrastructural inertia, consistent with Poland’s efforts to mobilize EU structural funds and domestic institutions of investment.

Indicators were chosen because of their analytical character in developing regression and cluster models on tracing structural change over time. Each of the indicators is used to characterize the sector in different aspects: installed capacity (GW) as a proxy for physical capital; energy production (GWh, TJ, PJ) as a proxy for productivity; and investment (PLN) as a proxy for financial input. They facilitate meaningful sensitivity analysis and temporal segmentation using cluster modeling. These figures are normally supplied by Statistics Poland and follow-up post-2010 national energy reports and are therefore apt to statistically sound, year-on-year comparison.

Each of the six indicators points to a distinctive yet connected element in Poland’s transition towards renewables, all in-aggregate an energy sector evolution system model by production dynamics, infrastructure, and investment.

• Renewable electricity production (GWh)—It is a measure of the production of electricity from renewable energy sources such as wind, sun, hydro, and bioenergy on an annual basis. It is the most important empirical estimation dependent variable, capturing the effectiveness of structural and financial inputs in producing decarbonized energy.
• Renewable heat production (TJ)—It gives the amount of thermal energy per annum produced from renewables (i.e., heat pumps, biogas, solid biofuels). It extends the energy analysis to cover heat, in addition to electricity, by giving the total proportion of renewables in the domestic energy balance and in the heat sector, which under Polish climatic conditions is very significant.
• Installed capacity of renewable energy power stations (GW)—Installed capacity is the aggregate potential output of all renewable energy plants that are operating in any given year. Installed capacity is an infrastructural driver for renewable heat and electricity generation. In principle, more capacity should automatically generate more energy, provided that maximum utilization and no curtailment are achieved.
• Total primary energy production (PJ)—This indicator provides a baseline by which one can contrast the entire domestic energy supply from every source (renewable and fossil). It allows one to determine the relative and absolute share of renewables in the broader context of national energy production, which is useful in decarbonization and diversification analysis.
• Renewable energy production (PJ)—As a part of the subset of the total primary energy indicator, the value indicates the amount of energy produced by renewables in all types (electricity, heat, etc.). The indicator acts as a key parameter to follow the transition towards energy and align national trends with EU and worldwide sustainability norms.
• Investment expenditures in the renewable energy sector (mln PLN)—This indicator reflects monetary expenditures focused on the development of infrastructure for renewable energy, such as the construction of plants, grid interconnection, and importing technologies. Investment is an enabling factor that supports the expansion of capacity and indirectly affects the output of production.

The research strategy is the combination of exploratory data analysis and traditional statistical analysis in an attempt to reveal deeply embedded trends and causality behind renewable energy installation development in Poland. The two general categories of methods employed by the study are (1) regression modeling and (2) cluster analysis.

To examine the determinants of renewable electricity production, a model of multiple linear regression was formulated. The model investigates the influence of capital expenditure, installed renewable capacity, and innovation spending on the volume of electricity produced from renewable power.

Two models were fitted: a parsimonious model with investment and capacity alone as explanatory variables, and an expanded model with innovation spending. The two models were contrasted on coefficient of determination (R²), analysis of residuals, and parameter significance based on normal OLS standards.

To identify distinct phases in development of the renewable energy market, cluster analysis was conducted with the k-means algorithm—a non-hierarchical method that partitions observations into k groups of similar observations. For purposes of analysis, the number of clusters a priori was specified as k = 3 based on visual inspection of data distribution and interpretive ease. It was grouped according to normalized values of key indicators such as renewable energy capacity, installed capacity, cost of innovation, and capital cost.

K-means algorithm clusters years into homogeneous growth periods through minimizing within-cluster variation and maximizing between-cluster variation. The resulting cluster was then qualitatively described to allow for the determination of three substantive periods: the early growth period (2010–2013), the stable growth period (2014–2019), and the acceleration period of expansion (2020–2023). The segmentation was then used to situate regression estimates and policy decisions.

In the numerical analysis, Statistica 13.3 and Python 3.9 were used.

5. Results

5.1. A Multiple Linear Regression Model

A linear regression model constructed on the basis of data from the years 2010–2023 aims to explain the impact of two key factors—capital investment (in million PLN) and installed capacity of renewable energy power plants (in GW)—on the annual production of electricity from renewable sources (in GWh). The selection of these two independent variables is justified both logically and empirically. Capital investments form the foundation for infrastructure development and technological modernization, while installed capacity defines the physical potential of the energy production system.

Model specification:

I—Investment;

IC—Installed Capacity.

Y = β0 + β1⋅I + β2⋅IC + ε

(1)

where:

• Y—annual renewable energy production [GWh];
• Investment—capital expenditures [million PLN];
• Installed Capacity—capacity of renewable energy sources in a given year [GW];
• ε—residual error (random component).

This means that each increase of PLN 1 million in investment leads to an average increase in production of 0.280 GWh, while each additional MW of installed capacity increases annual energy production by approximately 1010 MWh.

Final form of the estimated model:

REP—Renewable Energy Production.

REP (GWh) = −4321.57 + 0.28 × I (PLN million) + 1.01⋅IC (MW)

(2)

• β1 = 0.28: Every additional million PLN of investment increases production by approximately 280 MWh.
• β2 = 1.01: Each additional MW of installed capacity boosts production by approximately 1010 MWh annually.
• β0 = −4321.57: The intercept (theoretical production at zero investment and capacity) has no practical interpretative meaning.

The model actually fits historical data. The years 2010–2020 all have a very low relative error, which in most instances is no greater than 5%. That is, the two variables utilized explain a lot of the variability in the production of renewable energy for the period under consideration. Particularly between 2014 and 2020, the model has a good fit for actual values in model validation for predicting past interdependence between investments and the energy performance of the renewable sector.

In sustainable development, the model demonstrates that in practice it works: it provides a clear instrument of estimation of the influence of the investment policy and infrastructure development on environmental performance in the form of more active production of green energy. It establishes that until 2020 the Polish renewable energy market evolved in a structured manner and that was predictable, as the development was stimulated by investment, along with the development of the capacities. Implications of this model would be able to assist strategic planning, e.g., projecting the implications of hypothetical future new investment programs or designing public investment in energy plants at the climate policy level.

Sensitivity analysis (

Table 2. Sensitivity analysis relationship between capital investments and the installed capacity.

Scenario	Investment (mln PLN)	Capacity (MW)	Forecasted Production (GWh)	Change vs. Baseline
Baseline	33,216	27,980	52,771	—
−10% investment	29,894	27,980	51,155	−3.06%
+10% investment	36,538	27,980	54,386	+3.07%
−20% investment	26,573	27,980	49,540	−6.12%
+20% investment	39,860	27,980	56,002	+6.13%
−10% capacity	33,216	25,182	49,470	−6.26%
+10% capacity	33,216	30,778	56,071	+6.26%
−20% capacity	33,216	22,384	46,169	−12.49%
+20% capacity	33,216	33,576	59,373	+12.51%
+10% investment and +10% capacity	36,538	30,778	57,687	+9.30%
+20% investment and +20% capacity	39,860	33,576	62,276	+17.98%
−20% investment and −20% capacity	26,573	22,384	43,681	−17.21%

) of the linear regression model provides some idea regarding the effect of change in two crucial variables—investment in renewable energy and installed capacity—on estimated annual production of electricity from renewable sources. The base case, calculated based on 2023 data, includes investment at PLN 33,216 million, installed capacity at 27,980 MW, and estimated energy production at approximately 52,771 GWh. The base case was used to model different alternative conditions such as single and combined 10% and 20% increases and decreases in investment and capacity.

The first set of experiments examined the effect of investment changes alone, with constant installed capacity. An increase in investment of 10% leads to the estimated rise in production of approximately 3.07%, and a decrease of 10% leads to a decrease of 3.06%. Increasing the changes to 20% leads to output changes of ±6.13%. These proportionate and symmetrical changes are characteristic of the model’s linearity. But they also demonstrate that while money inputs are vast, their proportion of output is relatively modest—that investment alone is not the dominant force behind energy production.

The investment fixed variations in capacity are the ones with the larger effect. An increase in capacity by 10% results in an increase in energy produced by 6.26%, and a decrease by 10% produces an equivalent decrease. Both these effects add up to ±12.5% at the 20% level. This confirms the fact that installed capacity is the more significant and direct force behind power generation. In contrast to investment, which has an indirect effect such as the build-up of infrastructure, installed capacity has a direct and quantifiable effect on the extent to which energy can be produced. Strategic planning efforts should therefore aim at optimizing and building production infrastructure.

The combined scenarios, where both investment and capacity were altered simultaneously, indicate the combined impact of the two drivers. A rise of 10% in both variables is computed by the model to raise energy production by 9.3%. A rise of 20% in both raises it by 18%, and a fall of 20% in both lowers it by 17.2%. These findings highlight the cumulative nature of the model and that output gains are largest where investment and infrastructure building are simultaneous and each feeds back on the influence of the other.

Sensitivity analysis puts a premium on the relative significance of the two inputs to the model. Even though investment is the facilitating key role, installed capacity becomes the short-run instrument for output attainment increases. This has the following implication for energy policymakers and planners: that simply making arrangements for massive fiscal outlays will prove insufficient. The challenge is in converting investment into tangible capacity with the prospect of reaping material benefits in energy yield. As it happens, the linear regression model—simple as it is—is a convenient tool for modeling a strategic situation and informing renewable energy development policy in an empirical, pragmatic manner.

5.2. Extended Linear Regression Model of Renewable Energy Production with Consideration of Innovation

The extended linear regression model is a statistical tool used to quantitatively estimate the impact of three key factors on electricity production from renewable energy sources. In contrast to simpler two-dimensional models, this model considers both investment outlays, installed capacity, and—representing its novelty—expenditures on innovation. Each of these variables has its independent influence, but only their combined consideration allows for capturing the more complex mechanisms of the RES sector’s development.

The results for the extended multiple linear regression model are as follows:

• Intercept: 5091.31;
• Coefficient for Investment Outlays: 0.25;
• Coefficient for Installed Capacity: 1.09;
• Coefficient for Innovation Expenditures: 0.51;
• R-squared (R²): 0.98.

This extended model, which now includes expenditures on innovation, provides some interesting shifts compared to the previous model:

• Baseline Electricity Generation: The intercept has increased to 5091.31.
• Impact of Investment: The coefficient for investment outlays has decreased from 0.41 to 0.25. This suggests that when innovation expenditure is considered, the direct impact of investment outlays on electricity generation is somewhat reduced.
• Impact of Installed Capacity: The coefficient for installed capacity has slightly increased from 1.03 to 1.09, indicating that its influence remains strong and has become even more pronounced in the extended model.
• Impact of Innovation: The coefficient for innovation expenditures is 0.51, suggesting that innovation also plays a significant role in increasing electricity generation from renewable sources.
• Model Fit: The R-squared remains very high at 0.98, indicating that the extended model still explains a large variance in electricity generation.

The long regression model has a coefficient of determination that is very high and R² = 0.98; therefore, the three independent variables of innovation expenditure, capital expenditure, and installed capacity explain 98% of variability in renewable energy production per year from 2010 to 2023. The very high fit level bears witness to the strength of the model and its utility as a diagnostic and prediction tool for energy policy planning. Not surprisingly, the inclusion of innovation expenditure did not reduce the performance of the model—instead, it contributed to its explanatory capacity, revealing the multifaceted dynamics of physical investment, financial inputs, and technological advancement.

The coefficient of renewable energy investment is 0.25, or one PLN million invested into renewable energy provides approximately 250 MWh more to annual output. This is lower in the truncated two-variable model (where the coefficient was between 0.28 and 0.41), in order to control the investment in innovation; some of the rise in production hitherto explained by investment in capital is actually caused by the effect of improvement in efficiency brought about by innovation. The model thereby corrects an overestimation and unravels the relative contributions of money and technology drivers.

Installed capacity remains the dominant force behind renewable energy production, with a coefficient of 1.09. That is, for each added 1 MW of installed renewable energy capacity, the average level of electricity produced per year is 1090 MWh. This coefficient is slightly greater in earlier models but reflects the reality that technological improvement and system efficiencies—typically stimulated by innovation—allow each unit of capacity to produce more. This outcome makes physical infrastructure the simplest and prevailing generation level generator and has the potential to gain from the increasing technological progress.

The price of innovation is also uncovered to be a major part of the extended model with a coefficient of 0.51. It finds that with every million PLN increase in innovation, the outcome is the respective average increment of 510 MWh in renewable electricity production. Most notably, this contribution is more than double the contribution of capital investment, highlighting the fact that innovation has a primary and possibly underrated contribution to make in the transition to energy. The contribution of innovation is presumably through enhancing generation technologies, system efficiency, digitalization, energy storage, and grid integration. Its timing highlights that a shift towards knowledge- and technology-driven energy policy is not just welcome but long overdue if long-term growth is to be achieved.

The findings of the extended model require a harmonized and balanced growth strategy in which the expansion of infrastructure is coordinated with investment finance and continuing innovation. While installed capacity brings short-run immediate production gains, its potential is best exploited by investment in capital and—above all—technological improvements. This determines that policies targeting only infrastructure expansion can be very limited in scope unless supported by a firm commitment to research and innovation. Strategically, the model provides a strong analytical tool to organize public and private resource mobilization to attain environmental and economic payoffs in the renewable energy economy.

From a sustainable development point of view, the extended regression model provides an insight into how different types of investment propel a cleaner and more sustainable energy system. The model’s explanatory power conclusively positions strategic investment—namely, in infrastructure, innovation, and capacity building—as a measurable driver of renewable electricity production. It is in harmony with some implied notion of sustainability: long-term environmental objectives are reconciled with collective economic effort. The model demonstrates not only physical progress (installed capacity), but also knowledge-led progress (innovation investment) as crucial to achieve measurable progress in curbing carbon. Here, innovation is not a secondary or ancillary driver but one of the primary supports for change in the environment.

The model also breaks away from conventional investment models of capital investment in infrastructure as the chief method of expanding energy. In measuring the role of innovation, it makes it unmistakable that systemic sustainability also relies to the same extent on research environments, institutional learning, and technology development. Innovation cleans and enhances energy production efficiency, reduces raw material dependence, and allows for more straightforward deployment of renewables onto complex energy grids. Thus, sustainable energy development is not merely an issue of expanding hardware, but also expanding knowledge base, compressing innovation cycles, and bridging policy, investment, and technology feedback loops. The extended model argues forcefully for an integrated, multi-dimensional approach to energy transition.

A sensitivity analysis was grounded on a very rich linear regression model with a very high explanatory power (R² = 0.98). One of the objectives was to evaluate whether and how each one of the three explanatory variables—innovation expenditure, installed capacity, and capital investment—would vary because of each one of ±10% and ±20% improvements, and how this would influence estimated renewable electricity generation. The normal level of production stood at 52,842.4 GWh and served as the reference point against which all the deviations were measured. The scenarios involved single parameter deviations and composite deviations in all three directions (

Table 3. Sensitivity analysis renewable energy production with consideration of innovation.

Scenario	Investment Change	Capacity Change	Innovation Change	Forecasted Production (GWh)	Change vs. Baseline (%)
Baseline	0%	0%	0%	52,842.4	0.00%
−10% Investment	−10%	0%	0%	52,510.8	−0.63%
−20% Investment	−20%	0%	0%	52,179.3	−1.25%
+10% Investment	+10%	0%	0%	53,174.0	+0.63%
+20% Investment	+20%	0%	0%	53,505.5	+1.26%
−10% Installed Capacity	0%	−10%	0%	49,866.6	−5.63%
−20% Installed Capacity	0%	−20%	0%	46,890.9	−11.27%
+10% Installed Capacity	0%	+10%	0%	55,818.2	+5.62%
+20% Installed Capacity	0%	+20%	0%	58,794.0	+11.27%
−10% Innovation Expenditures	0%	0%	−10%	50,643.4	−4.16%
−20% Innovation Expenditures	0%	0%	−20%	48,444.4	−8.32%
+10% Innovation Expenditures	0%	0%	+10%	55,041.4	+4.16%
+20% Innovation Expenditures	0%	0%	+20%	57,240.4	+8.33%
−10% All Variables Simultaneously	−10%	−10%	−10%	47,150.8	−10.77%
+10% All Variables Simultaneously	+10%	+10%	+10%	58,534.1	+10.77%
−20% All Variables Simultaneously	−20%	−20%	−20%	41,459.1	−21.54%
+20% All Variables Simultaneously	+20%	+20%	+20%	64,225.8	+21.55%

).

The research established that capital investment change had quite limited effects on energy output. A 10% reduction in investment reduced anticipated output by merely 0.63%, and a 20% rise increased output by only 1.26%. The sensitivity is low, which implies that capital spending on its own, without being supported by accompanying rises in capacity or innovation, has minimal effects on energy outcomes. Capital investment is more of an enabler that facilitates the application of technology and infrastructure but not necessarily creating equal production benefits by itself.

Installed capacity was the optimum predictor variable of the model. A ±10% capacity variation produced effects of ±5.6% in the production, and ±20% variations produced over ±11% effects. The results confirm the fact that physical infrastructure is the ultimate spur for production—additional megawatt of installed capacity is directly and proportionally linked to additional output. This reinforces the necessity of mass deployment of RES systems and calls for consistent growth and capacity expansion in generation in order to meet long-term sustainability objectives.

Incorporation of the innovation spending into the model was quite significant. The 10% rise in the innovation spending pushed the level of energy production up by 4.16%, while the 20% rise pushed the level of energy production up by 8.33%. These signify that organizational, digital, and technology innovation contributes significantly towards raising the performance of the renewable energy sector. Innovation provides more optimal utilization of existing assets, promotes integration into the grid better, develops new storage technologies, and optimizes management of production assets. Innovation in such a manner can be seen as a multiplier with higher returns on investment and capacity addition.

The largest alterations in estimated output are when all three variables were changed at once. A 20% rise in investment, capacity, and innovation resulted in a 21.55% increase in renewable energy production, and a 20% reduction in all three resulted in a 21.54% drop. These proportional and symmetrical impacts are a result of the model’s linearity and illustrate a key observation: that returns to sustainability are maximized when finance, infrastructure, and innovation are developed out simultaneously. To address one in the round in isolation yields minimal returns; success with the energy transition requires concerted, coordinated action on all of the main pillars of development.

5.3. Model Comparison

Comparison between simple and complex models of regression gives a keen picture of shifting trends of renewable energy production in Poland. The simple model contains only two independent variables: capital expenditure costs and installed capacity of the renewable energy sources. It is an adequate fit to historical data because both variables contain statistically significant as well as logically anticipated coefficients. The most important of these is installed capacity, where every extra megawatt translates into an additional approximate 1010 MWh of renewable electricity per year. Capital expenditure, being positively related, is lower, adding up to a mean of 280 MWh per million PLN invested. The model is very explanatory, especially for 2014–2020, exhibiting the explanatory capacity of infrastructure variables in the process of systematic expansion of the sector.

The expanded model includes a third explanatory variable—national spending on innovation—and properly increases the level of analysis. The model has a high coefficient of determination (R² = 0.98) similar to the base model but allocates the weight of explanation among the independent variables differently. Most importantly, the capital input coefficient declines from 0.28 to 0.25, which means that some of the gain previously attributed to financial inputs is actually caused by enhanced efficiency due to innovation. The installed capacity coefficient increases marginally to 1.09, further establishing itself as the most proximate determinant of output. The introduction of innovation expenditure introduces an additional variable: for every additional million PLN invested in innovation, there is an additional 510 MWh of electricity produced. That is, innovation is not a contained, but a standalone and significant variable for sectoral performance.

Cross-comparison of both models illustrates a shift in the growth logics of the renewable energy sector. Physical infrastructure remains pivotal but the extended model indicates the way investment success is tremendously enhanced through exposure to innovation. To the existing view, capital and capacity fall short in ensuring long-term growth unless there is simultaneous progress in technology, system efficiency, and digitalization. The extended model therefore reflects a more advanced level of sectoral development, where marginal rates of return to investment are determined increasingly by knowledge-based and innovation-led processes. For policymakers and stakeholders, moving to the extended model represents a strategic necessity: low-carbon energy growth will no longer simply rely on assets growth, but upon embracing systemic innovation as central to the energy policy and planning agenda.

5.4. Cluster Analysis

Cluster analysis was employed as an exploratory data analysis method to identify natural clusters of years according to comparable development trends in the renewable energy sector in Poland. The k-means algorithm with a fixed number of clusters (K = 3) was applied so that the 2010–2023 period could be divided into three stable time periods with various structural characteristics. Classification criteria were major quantitative factors such as the volume of renewable energy output, installed capacity of renewable plants, levels of capital investment, and R&D expenditure. This approach permitted the demarcation of three differentiated stages of development—encompassing from the first stage of the energy transition, via one of established expansion, to that of active technological enlargement—enabling therefore the carrying out of a more differentiated evaluation of innovation speed and quality in the green energy sector.

Table 4. Cluster analysis.

Cluster	Years	Interpretation	Key Characteristics
K = 3	2010–2013	Early Development	Characterized by lower levels of renewable energy production, capacity, and investment compared to later years.
	2014–2019	Steady Growth	Exhibits consistent rise in all numbers, suggesting a period of continuous expansion for the renewable energy sector.
	2020–2023	Accelerated Expansion	Characterized by the sudden surge in renewable energy production, capacity, investment, and spending on innovation, which represents a rate of rapid growth and policy focus.

includes results of cluster analysis where 2010–2023 data were distinguished into three well-defined periods in Poland’s growth of the renewable energy market: the initial development phase (2010–2013), the growth period (2014–2019), and the stage of intensive development (2020–2023). The first cluster is characterized by low levels of renewable energy output, small installed capacity, and relatively modest investment expenditures reflecting the infancy stage of sector development. The second cluster reflects systemic and balanced development in all its indicators, reflecting efforts consolidation and integration of renewables into the mainstream of national energy policy. The third cluster outlines a period of dynamic growth, marked by swift leaps in production, investment, and innovation, hinting at serious institutional and political commitment towards energy transition. The study confirms that Poland’s renewable sector has been incrementing stepwise through phases, with the current phase being not only marked by scale but also sophistication and maturity in its development.

Cluster analysis, conducted on the 2010–2023 data, made possible the recognition of three stages in Poland’s renewable energy sector development. The initial cluster, 2010–2013, has been determined to be the time of “early development.” The cluster is represented by fairly modest renewable energy output levels, low installed capacity, and moderate investment costs. This phase can be interpreted as the installation or establishment phase of the renewable energy system, institutionally and infrastructurally. Policy responses at this time were experimental in nature, with pilot experiments first being conducted and experimentation with financing mechanisms, prior to and while public and political awareness of renewable energy was still being formed.

The second segment, 2014–2019, is a phase of “steady growth.” Here, all the key indicators—the production of renewable energy, installed capacity, and investment—show a steady increase. It is a phase of industry maturity, when the shift occurs from isolated pilot activity to more concerted and strategic expansion. Public policy begins to welcome more systematic practices, and the renewable energy market becomes increasingly attractive to private capital. The distinguishing feature of this stage is stabilization of the pattern of growth—the industry no longer becomes a marginal industry and begins to become a part of the national energy complex as an equal component.

The final cluster, dated 2020–2023, is distinguished as a stage of “accelerated expansion.” It features a dramatic increase in all the dimensions considered here: energy output, installed capacity, volume of investment, and—most important—spending on innovation. This is the phase of proactive development, promoted by fresh climate policies, EU funding mechanisms, and growing pressure worldwide for energy transition. As statistics show, these years are the milestone of Polish development pace and its path to energy-intensive modernization. The fact that innovations were among the most distinguishing drivers in this growth period states that besides new capacities of construction and infrastructure expansion, this process meant systemic changes—digitalization, automation, interconnection into the smart grids, and installing new models for the energy management.

The cluster analysis confirms that the evolution of constructing the renewable energy sector in Poland, although initially gradual and incomplete at first, with time has gradually evolved into one of systemic maturity. This reflects a sign of a shift from quantity to quality—i.e., from a mere expansion in capacity and investment to adopting an approach of innovation, efficiency, and technological integration. According to this, the results of the analysis indicate that the development of the renewable energy sector in Poland is increasingly consistent with the very nature of sustainable development—economically, socially, and environmentally.

6. Discussion

The crucial implication from the regression models carried out and cluster analysis is the primary contribution of installed capacity as the primary driver of renewable electricity production in Poland during the time frame of 2010–2023. The simple linear model reveals that although capital investment has a positive contribution towards energy production, it is installed capacity that captures the most explicit and robust influence. This is in line with previous work that emphasized infrastructure development as the driver of renewable performance, though often without complete inclusion of innovation as a co-determinant [9,10,11]. The results here confirm that infrastructural expansion, measured in megawatt increase, remains the most powerful predictor of renewable output annually and that Poland’s energy revolution, to this point, has been driven by rising capacity rather than structural technological shift.

The extended regression model with innovation expenditure provides a core counter to infrastructure-based theories of energy transition. The findings suggest that innovation spending—albeit flat since 2015 in Poland—exerts a considerable independent influence on renewable electricity output, even surpassing the marginal contribution of the funds invested. This resonates in broader literature that signifies the central place of innovation ecosystems in the process of energy transition, particularly through enhancing energy efficiency and enabling the integration of power grids through clever technologies [34,35,36,37]. The fact that for every extra million PLN innovation delivers approximately 510 MWh, whereas a simple investment guarantees 250–280 MWh, offers some empirical support for the status of innovation not as a servicing function, but as a guiding axis of renewable energy policy. Authors credited arguments in favor of AI-based and technologically emphasized transformation paths in industries [34,35,38,39]. In Poland, the huge problem is energy intensity of industries, e.g., steel [40,41] and lack of nuclear sources of energy. Moreover, coal mining realizes the strong restructuring process [42] according to the strategy of moving away from coal as a key energy source by 2050 and decommissioning mines.

The cluster analysis reveals the staged and non-linear dynamics of Poland’s renewable energy development. By dividing the period 2010–2023 into three stages—early development (2010–2013), stable growth (2014–2019), and accelerating growth (2020–2023)—the analysis refutes the notion of linear development in the energy transition. Instead, the analysis suggests that the intensity of policies, investments volumes, and innovation involvement fluctuate over time and are contingent on political, economic, and institutional factors. These stages parallel shifting EU climatic and energy independence policies emphasized by the European Environment Agency and Eurostat [1,2,43] and reflect how global and domestic institutional framework shapes the country’s capacity for energy transmutation.

While the results are promising, they also reflect structural weaknesses in Poland’s long-term capacity to sustain innovation-driven growth. While renewable production has doubled and installed capacity has grown twelve times over, the breakdown in innovation spending since 2015 is worrisome. Without a resumption of national policy for innovation, there is a risk that increments to physical capacity will return diminishing dividends. Poland’s energy policy needs to transcend capacity development and guarantee an integrated strategy which brings together infrastructure, capital streams, and sustained R&D. Only a coordinated model can gain congruence with the greater goals of sustainable development and carbon neutrality as set out in the IEA and IRENA (International Renewable Energy Agency) forecasts [44,45,46].

An examination of the results of this research through the framework of the Multi-Level Perspective (MLP) illustrates a complicated interplay between technological niches, socio-technical regimes, and overarching landscape pressures [47,48]. The early growth period of Poland’s renewable energy sector (2010–2013) corresponds to niche-level experimentation, where emerging technologies such as solar and wind energy began experimenting, albeit still within a policy and market environment controlled by fossil fuels. At this stage, as the regression model indicates, capital investments and innovation impact on real renewable production were low-key, testifying to the resilience of niche innovations in a stable regime that is coal-driven. The early low policy coherence and low public awareness created structural inertia that preserved dominant configurations, which is in accordance with the MLP insight that transitions begin with decoupled innovation and not with systemic rebellion [49,50]. The middle period (2014–2019), acknowledged by cluster analysis as one of steady growth, is the consolidation of niche innovations and the emergence of cracks in the existing socio-technical regime. During this period, renewable technologies began to gain economic credibility and policy support—which was further accelerated by increasing EU climate expectations and improved funding mechanisms—enabling higher levels of alignment of actors, institutions, and infrastructure. In MLP parlance, this describes the gradual remaking of the regime level, where existing players such as utilities and state energy organizations begin adapting to, rather than resisting, the inclusion of renewables [51,52,53]. The capacity role of installations increases more pronouncedly in this stage, as the regression analysis testifies, but only if complemented by an accompanying rise in innovation expenditure, which is indicative of lock-in on current infrastructural solutions rather than a movement toward systemic flexibility or learning. This phase does show the regime opening up to innovation; however, this paves the way for a broad transition.

The phase of accelerated growth (2020–2023) aligns with a tipping point in MLP [54,55] terms: increasing landscape pressures—e.g., the EU Green Deal, fluctuating energy prices, and geopolitical tensions—couple with the maturity of niche innovations to bring destabilizing pressure to bear on the incumbent regime. The results indicate that in this stage, innovation expenditure begins to generate large productivity effects, confirming Poland’s energy transition had reached a level at which technology and institutional design were able to support more rapid reorientation. Yet the slowdown of innovation investment since 2015 suggests that Poland’s regime is still partially stabilized in an infrastructural rather than knowledge-based development model. From an MLP perspective, this also emphasizes the risk of “shallow transitions”—more extensive deployment of green technology without more fundamental changes in institutional logic, user practices, or cultural values [56,57]. To propel a full regime change, Poland now needs to consolidate its gains through continuous innovation, policy linkages, and long-term vision that incorporate not just technological substitution but systems change.

The greatest political and regulatory challenges facing Poland’s energy transition are Poland’s structural reliance on coal, which still affects the energy mix composition and policymaking agenda of the country. Even with the significant increase in renewable capacity and output since 2010, Poland was still responsible for 34% of the EU coal-fired electricity production in 2023, a testament to the continued dominance of fossil fuels in domestic energy planning [3]. Such a legacy has built deep institutional inertia, and it becomes hard to bring in combined and forward-looking regulation systems. Therefore, attempts to decarbonize the power sector are increasingly troubled by fractured governance, regulatory unpredictability, and lack of proper national coordination, particularly in new frontier regions like offshore wind and geothermal development [5,6,7].

The next problem arises from conflicts between EU-level renewable energy directives and the requirement for a less harmonized, sovereignty-based model of energy regulation in Poland. Whereas the European Green Deal and Renewable Energy Directive (RED) incentivize action that is harmonized at the level of the member-state, Poland has frequently pushed back against top-down governance models for the sake of discretion at the level of the nation in designing and implementing support mechanisms [58]. This duality creates an incoherent subsidy system, behind-schedule infrastructure deployment, and lower appeal for long-term private financing. As highlighted in EU policy discourse, this incoherence undermines Poland’s ability to be able to achieve both its decarbonization targets as well as wider climate resilience goals [1].

Also the suspension of innovation spending since 2015—having its tested track record for increasing the production of renewable energy—also indicates evidence of an existence mismatch between politic rhetoric and policy setting of priorities. When Poland raised the investment in innovation up to 2015, levels have stabilized since then to around PLN 4.5 billion per year, indicating an inability of sustained political will to boost technical competence [5,6,7]. This is especially concerning considering the context that innovation expenditure, as determined by regression analysis, records higher marginal rates of returns on power generation than capital investment (Table 4). Without strategic shifts towards building a robust institution for R&D, digitalization, and smart grid integration, Poland is likely to be left even further behind EU leaders like Denmark, Portugal, and Croatia in the energy transition [59,60,61,62,63,64]. Political will and regulatory consistency, then, are needed so that investment in the infrastructure is bolstered by equally strong support for innovation and structural transformation.

7. Conclusions

7.1. Summary of Main Results

Between 2010 and 2023, Poland’s renewable energy sector transitioned in a multi-stage process from the pilot-scale and periphery phase to emerging as a strategically important part of the country’s energy mix. The industry has traversed distinctive phases of early experimentation, incremental and step-by-step expansion, and high-growth expansion, each characterized by rising levels of investment, deployment of infrastructure, and policy action. Most importantly, the post-2020 era saw a qualitative change with installed capacity and generation of renewable electricity improved, reflecting increased system integration and paradigm shift through innovation.

The most important contribution of this work is its complete empirical model predicting relative share of capital expenditure, installed capacity, and cost of generation of renewable electricity. While still the largest short-run production motivator, the long-run regression model illustrates that innovation makes the most marginal unit of contribution per unit spent. The evidence goes against prevailing infrastructure-based perspectives through the establishment of the central position of technological change, efficiency of systems, and learning in bridging institutions in driving energy transition. Cluster analysis also provides perspective on similar dynamics by decomposing the transition into analytically distinct phases of development.

Placing its analysis in the Multi-Level Perspective (MLP) framework, the paper offers theory-driven and policy-oriented explanation of Poland’s transition to renewables. The paper demonstrates how landscape drivers like EU climate policy and changing markets emerge with regime inertia and niche innovations. The novelty of the study lies in the simultaneous application of longitudinal statistical modeling and development typologies to structural and temporal dynamics of energy transition. The findings conclude that Poland as a country needs to invest overall in infrastructures, finances, and innovations in order to adhere strictly to the targets of sustainable development and catch up with developed EU countries in the renewable energy sector.

Table 5. Summary of main findings.

Research Area/Question (RQ)	Key Findings
RQ1—Evolution of the RES sector in Poland (2010–2023)	The industry passed through three phases: formative development (2010–2013), stable growth (2014–2019), and vibrant growth (2020–2023). Electricity generation increased more than four-fold, and installed capacity more than twelve-fold.
RQ2—Relationship between capital investment, installed capacity, and energy output	There are high statistical interdependencies. An increase in investment by PLN 1 million raises output by approximately 280 MWh, and an increase in installed power by 1 MW adds approximately 1010 MWh per year.
RQ3—Contribution of innovation expenditure to RES electricity production	Innovation has a very strong positive impact: increasing innovation by PLN 1 million increases production by approximately 510 MWh—over double the marginal impact of the capital input.
RQ4—Interactions and relative influence of development factors	Installed capacity has the most effect. The largest contributions are where capacity, investment, and innovation all increase simultaneously—each 20%; therefore, a total of 21.55% of output increased.
RQ5—Developmental stages of the renewable energy sector	There are three stages based on the cluster analysis: (1) initial growth (2010–2013) with low investment and production; (2) steady growth (2014–2019) with balanced development; (3) rapid growth (2020–2023) with innovative dynamism and investment.

summarizes the core empirical findings of the study to indicate the diversified and dynamic characteristics of Poland’s market of renewable energies between the period 2010–2023.

7.2. Future Research Directions

One of the prospective study fields in the future could be the integration of digitalization and artificial intelligence (AI) with Polish renewable energy installations, i.e., the improvement of innovation efficiency and grid management. With this report’s findings—especially with the vastly disproportionate share of innovation spending heading to energy generation—there is a pressing need to investigate the way in which digital technologies like AI-driven predictive maintenance, self-driving smart grids, and energy demand forecasting can help further boost the productivity of installed infrastructure. In addition, research can investigate the extent to which decentral energy infrastructures, i.e., local storages, prosumer communities, energy cooperatives, can be algorithmically controlled to facilitate energy resilience in transition economies. Finally, comparative longitudinal studies siting Poland’s transition in the wider EU policy context—i.e., front-running Denmark or Portugal—would provide acutely required institution, regulation, and technology gap analyses differentiated for different decarbonization pathways.

7.3. Limitations of the Paper

Although the data set is large, the study is undermined by its reliance on aggregate yearly data, which may hide intra-year volatility and inter-regional heterogeneity. Second, innovation spending was employed as a single national figure without sectoral disaggregation, which may potentially under- or over-estimate their relevance to energy-specialized innovations. Despite such limitations, the methodology provides a robust and unambiguous foundation for assessing long-term structural change in the renewable energy industry.