Exploring the Determinants of Energy Vulnerability in Micro-Enterprises: Insights from the Croatian Case Study

Abstract

Micro-enterprises are vital to the European economy, including in Croatia, where they make over 88% of the total number of businesses. Despite their significance, they face substantial energy vulnerability due to factors like small size, limited financial resources, and high energy costs. This paper investigates the determinants of energy vulnerability among Croatian micro-enterprises, employing a survey of 470 micro-enterprises. The study covers firms across all Croatian NUTS2 regions and ensures geographic and sectoral representativeness. Key findings reveal that enterprises with higher energy expenditures relative to revenue are most susceptible to energy vulnerability, which is aligned with our assumption. On the other hand, businesses that own their premises, have more employees, and have been operational longer are more likely to invest in energy efficiency measures, thereby reducing vulnerability. Notably, a significant proportion of micro-enterprises report that energy costs adversely affect their household finances, highlighting the nature of business and personal economic stability. The paper underscores the need for targeted policies and support mechanisms to enhance the energy-related resilience of micro-enterprises, considering their unique structural and financial constraints.

1. Introduction

Micro-enterprises are a crucial part of economies worldwide, especially in Europe, where they play a significant role in fostering economic growth and creating employment opportunities [1]. Eurostat’s Structural Business Statistic shows that micro-enterprises make a substantial contribution to Europe’s economy, generating nearly 17% of the EU’s total net turnover. Within the broader SME sector, they are responsible for 29.8% of total employment [2]. When examining energy consumption, defining an average for micro-enterprises is virtually impossible due to the significant variation based on sector and business activity. Even at the broader SME level, most countries lack dedicated datasets, and European energy statistics do not differentiate by company size. As a result, research findings vary widely depending on the methodologies used and the EU countries included in the sample. Estimates range from 9% to 18% of total gross inland consumption (GIC), with some outliers reporting figures as high as nearly 30% [3]. Obviously, among other factors, their stability is strongly influenced by energy-related aspects, particularly the impact of energy vulnerability [4].

Energy vulnerability of micro-enterprises refers to the challenges and risks these small businesses face in terms of their access to affordable, reliable, and sustainable energy. It also refers to their limited capacity to maintain stable operations and financial viability in the face of rising or volatile energy costs, constrained access to efficient energy infrastructure, and reduced ability to invest in energy-saving technologies. It encompasses both exposure to energy-related shocks and the ability (or lack thereof) to adapt. This vulnerability can result from a variety of factors, including a mix of external influences and internal determinants. We have been exploring factors looking into available statistical datasets in our research, like external factors related to energy and climate; external factors related to finances, market and development; business determinants affecting vulnerability and those affecting adaptation, forming a composite index.

In this research, we aimed to identify specific determinants that contribute to the energy vulnerability of micro-enterprises. We analysed independent factors such as location, business position, sector size, company age, and energy characteristics, informed by the existing literature and expert prioritisation of vulnerability indicators. Previous findings show that energy costs as a percentage of revenue and higher energy expenditures compared to sector peers might be the most critical determinants of energy vulnerability in micro-enterprises. With further insights, we developed dependent determinants—identifiers of energy vulnerability in micro-enterprises. We utilised these insights to design a survey for micro-enterprises and received 503 (470 eligible) responses from various businesses across Croatia, collected through an EU-survey tool, distributed with the help of the national stakeholders.

In comparative terms, SMEs across the EU are subject to a variety of energy efficiency policies, yet these are rarely designed to accommodate micro-enterprises specifically. A review of 64 national and regional policies shows that while most are well intentioned, many overlook administrative simplicity and firm-specific constraints. Their conclusions underscore the importance of context-sensitive approaches, reinforcing the need for targeted research on micro-enterprises, such as the one presented in this study [3].

The first section presents a conceptual framework with the literature review on the common characteristics of micro-enterprises that contribute to their energy vulnerability. This is followed by an introduction to the research concept and a description of the methodology used. The Results section provides the information on the statistically significant determinants of vulnerability. The paper concludes with a discussion that interprets the results, provides policy recommendations, and outlines potential ideas for future research.

Although the concept of energy poverty has been widely studied in households, its relevance in the business context, especially for micro-enterprises, remains underexplored. This study bridges this gap by evaluating the determinants that capture firm-level and owner-level vulnerability. It thereby links operational fragility with personal wellbeing, contributing novel insights to energy vulnerability theory.

2. Conceptual Framework

A micro-enterprise is defined as a business employing fewer than 10 people and having an annual turnover or balance sheet not exceeding EUR 2 million, in accordance with Articles 3 to 6 of Annex I of Commission Regulation (EU) No 651/2014 [5]. A key challenge in assessing the impact of policies on vulnerable micro-enterprises is that their energy vulnerability has only recently been acknowledged and is not yet clearly defined. Research on energy poverty and the broader vulnerability of the small business sector to energy, is virtually absent from the literature, no matter this recognition in the EU legislation [6]. Some empirical analyses focused on examining the level of involvement of micro- and small enterprises in the energy transition and the impact this has on their market and financial performance [7]. They found a clear link between the characteristics of micro- and small enterprises, and the energy efficiency measures they adopt. Therefore, we build our research on our previous findings, expert inputs, and literature on energy poverty of households and energy-related challenges in SMEs. To identify indicators describing vulnerable micro-enterprises, we examined various potential factors such as location, sector size, company age, and energy characteristics, drawing insights from the existing literature. The analysis also relies on the experts’ prioritisation of vulnerability indicators, which highlighted that energy costs as a percentage of revenue (indicating a high share of energy expenditure relative to total revenue) is the most critical determinant of energy vulnerable micro-enterprises, followed by higher energy expenditures compared to sector peers (Figure 1).

2.1. Independent Variables of Energy Vulnerability in Micro-Enterprises

• Location of the micro-enterprise

The location of the micro-enterprise affects multiple aspects of energy vulnerability:

• Heating, cooling degree days

Important social aspects of energy are accessibility, affordability, and efficiency [8]. Affordability depends on energy needs and price. There is a linear relationship between heating energy needs and heating degree days [9]. A building’s energy performance is influenced by the climate conditions, and to assess a building’s energy performance and correlate energy demand with the climate context, it is essential to have access to weather data and the accurate HDD [10].

b.

Energy dependence, transport network, and infrastructure

Energy dependence can increase a region’s vulnerability across multiple domains, the more dependent a region is on external or non-renewable energy sources, the more it faces risks from supply disruptions and price volatility [11]. However, this aspect is hard to use for specific purpose, due to the fact that the prices and imports are regulated on the national level. On the other hand, regions that rely on outdated or inefficient energy infrastructure (e.g., older coal-fired power plants, aging grids) may be more prone to energy disruptions and inefficiencies, influencing stable supply needed for business operations [12]. Transportation is more straightforward on a smaller scale, as it not only makes individuals vulnerable but also impacts micro-enterprises by limiting their access to essential services needed for business operations [13].

c.

Economic development

The economic development of a region can both increase and decrease its energy vulnerability. While development may increase energy demand and reliance on external sources, it can also provide the resources and technology needed to diversify energy sources, improve energy infrastructure, and invest in energy efficiency. It also ensures higher financial capacity of the individuals and micro-companies as adaptation factors for risks of energy-related disruptions. However, in the case of significantly underdeveloped regions, there is undeniably higher risk for energy-related vulnerabilities [14,15,16].

In conclusion, the location of a micro-enterprise plays a critical role in determining its energy vulnerability through various interconnected factors.

• Company sector

• Consumption per sector

Energy consumption intensity differs significantly across economic sectors, influencing the vulnerability of micro-enterprises. Enterprises in manufacturing, construction, and transport consume substantially more energy due to process-driven operations and heavy reliance on fossil fuel-based logistics or machinery. For example, textile micro-enterprises often report energy expenditures exceeding 10% of their operating costs and food-processing industries display distinct energy consumption patterns, with individual sectors exhibiting significant variations during production processes. Conversely, sectors like professional services and information technology tend to have lower energy consumption profiles. This variation underscores the importance of sector-specific strategies when addressing energy efficiency and vulnerability in micro-enterprises [17].

b.

Sectoral position in the supply chain

The position of a micro-enterprise within the supply chain can influence its exposure to energy-related risks. Enterprises operating upstream, such as raw material suppliers, may face different energy challenges compared to those downstream, like retailers or service providers. For instance, upstream enterprises might be more affected by fluctuations in energy prices due to their reliance on energy-intensive production processes, while downstream enterprises could be more sensitive to disruptions in energy supply that affect their ability to meet customer demands. Understanding these dynamics is crucial for developing targeted interventions to enhance energy resilience across the supply chain. Studies emphasise the need for multi-level supply chain management strategies to support micro- and small enterprises in achieving sustainability [18].

• Company size (number of employees)

The age and size of a micro-enterprise are critical factors influencing its energy vulnerability. Younger enterprises may lack the financial resources and experience to invest in energy-efficient technologies, making them more susceptible to energy price volatility. Similarly, smaller enterprises, often characterised by limited staff and capital, may find it challenging to implement energy-saving measures or negotiate favourable energy contracts. These constraints can exacerbate their exposure to energy-related risks, highlighting the need for supportive policies that facilitate access to energy efficiency resources for newer and smaller micro-enterprises. Research indicates that enhancing energy resilience in manufacturing enterprises is vital for operational stability, especially for smaller firms [19].

Understanding the average number of employees in micro-enterprises across European countries provides essential insights into structural business dynamics, human resource limitations, and firm-level resilience. Micro-enterprises, defined by the European Commission as those employing fewer than ten individuals, constitute the vast majority of enterprises in Europe (particularly in Southern economies such as Spain and Croatia), and their internal human capital configuration plays a pivotal role in shaping their economic viability, energy management capacity, and capacity to absorb exogenous shocks. The average number of employees in micro-enterprises varies across countries, reflecting differences in economic structures and business practices. For example, in the European Union, micro- and small enterprises employed approximately 77.5 million people in 2022, accounting for nearly half (48%) of the total employment in enterprises [20].

In Spain, micro-enterprises constitute 94.4% of all firms, with the vast majority employing between one and two workers. Sánchez-Infante Hernández et al. highlight that the average firm size for Spanish MSMEs leans heavily toward the micro category, often operating with minimal staff, which limits their ability to implement strategic measures such as energy management systems, sustainability frameworks, or Corporate Social Responsibility (CSR) programmes [21].

A similar structural profile exists in Croatia, where Financial Agency (FINA) data show that the average number of employees per micro-enterprise is 1.98. These firms, often sole proprietorships or family-run businesses, operate with severely constrained administrative and technical capacity, thereby facing inherent barriers to participation in formal sustainability schemes or accessing energy transition financing instruments.

The scholarly literature further confirms that the limited size of micro-enterprises is not merely a descriptive statistic but a critical determinant of firm behaviour and strategic performance. Firm size significantly moderates the relationship between CSR implementation and economic performance: enterprises with more employees demonstrate higher adoption rates of structured sustainability practices and perform better across economic indicators. Conversely, smaller micro-enterprises exhibit “silent social responsibility” practices that, while morally aligned with CSR principles, remain undocumented and non-strategic due to resource and reporting constraints [21].

• Survival rate of companies (age)

The average lifespan of micro-enterprises offers additional perspective on their stability and resilience. Enterprises with longer operational histories may have established customer bases and more robust financial structures, potentially enabling them to better absorb energy cost increases. In contrast, newer enterprises might lack such buffers, making them more susceptible to energy-related disruptions. Therefore, strategies aimed at enhancing energy resilience should consider the age distribution of micro-enterprises to ensure that support mechanisms are appropriately targeted.

• Turnover and profit of a company

• Total/compared to peers

Turnover and profit are fundamental indicators of a micro-enterprise’s financial health, directly influencing its capacity to manage operational costs, absorb external shocks, and invest in strategic areas such as energy efficiency. These metrics reflect not only the enterprise’s current financial performance but also its resilience and adaptive capacity relative to its sectoral peers.

A commonly used indicator in this context is net turnover per company, calculated by dividing the total net turnover of a group of micro-enterprises by the number of firms in that group. This average value provides a useful benchmark for comparing financial performance across regions and sectors. A lower net turnover per micro-enterprise typically reflects limited income-generation capacity and may suggest higher exposure to energy-related costs due to constrained investment capabilities.

Recent evidence from both institutional reports and academic literature underscores growing pressures on SMEs, including micro-enterprises, in the post-pandemic and energy crisis context. According to the European Central Bank’s Survey on the Access to Finance of Enterprises, euro area firms have reported deteriorating profitability and muted turnover growth [22]. While only a minority of firms experienced an increase in turnover, a significantly larger share signalled declining profits amid mounting cost pressures, including those linked to energy, wages, and raw materials.

The SMEunited Barometer for the first half of 2025 corroborates these trends. It reveals that SMEs across Europe are operating in a challenging macroeconomic environment, marked by inflationary pressures and weakened demand. As a result, many firms, particularly micro-enterprises with limited financial buffers, reported reduced profitability and increasing difficulty in maintaining stable turnover levels. This situation has further constrained their capacity to reinvest in productivity-enhancing areas, including decarbonisation and energy efficiency upgrades [23].

The academic literature reinforces these insights by showing a clear relationship between energy prices and firm-level profitability. A recent study examined the impact of rising electricity prices on business margins across the European Union and found that sustained increases in energy costs significantly erode profit margins, particularly for smaller firms with less capacity to hedge against price volatility or invest in alternative energy solutions [24]. The authors also observed that sectors with high energy intensity, such as manufacturing, construction, and transport—face more acute profitability risks.

Similarly, the OECD’s 2023 report on SME policy responses to the energy crisis identifies profitability deterioration as one of the most immediate consequences of the energy price spike triggered by geopolitical instability. The report warns that without targeted policy support and affordable financing instruments, many SMEs may struggle to maintain financial stability, undermining national recovery and climate transition goals. The OECD calls for proactive investment in energy efficiency as both a cost-containment measure and a resilience strategy [25].

In summary, turnover and profit provide important insight into the financial situation of micro-enterprises and are closely linked to their energy-related challenges. Businesses with lower profits are often less able to handle increases in energy costs or to invest in measures that would reduce those costs over time, such as improving energy efficiency. This can lead to ongoing financial strain. Monitoring these indicators, especially in relation to sectoral norms, is thus critical for designing effective support policies for micro-enterprises navigating the energy transition.

b.

Energy in turnover

The proportion of energy costs within a company’s turnover is a crucial indicator of its financial resilience, especially for micro- and small enterprises. When energy expenses constitute a significant share of turnover, even minor fluctuations in energy prices can substantially impact profitability.

This relationship becomes even more relevant when considered alongside net turnover per company. Micro-enterprises with relatively low net turnover are more likely to experience disproportionate financial stress when energy prices rise, as energy costs represent a larger slice of their revenue base. In such cases, businesses may struggle to maintain operational efficiency or to make the necessary investments in energy-saving technologies.

A study analysing micro- and small enterprises in Mexico found that a 1% increase in fuel prices could lead to a profit reduction exceeding 1% for firms with low profit margins and high energy cost shares [26]. This effect was particularly pronounced in the transport sector, where fuel costs represent a substantial portion of operating expenses.

In the European context, the energy crisis has similarly affected small businesses. A survey by Business at OECD (BIAC) reported that SMEs experienced an average energy cost increase of 159%, with some facing hikes exceeding 600% [27]. These surging costs have compelled many SMEs to raise product prices, delay investments, or reduce operations to manage expenses. Such financial pressures underscore the importance of energy efficiency measures. By investing in energy-saving technologies and practices, SMEs can reduce the proportion of energy costs in their turnover, enhancing profitability and resilience against future energy price volatility.

• Urban development and energy vulnerability

The distribution of employment opportunities between rural and urban areas is an important structural factor influencing the energy vulnerability of micro-enterprises. In regions where employment is concentrated in urban centres, rural businesses often encounter operational challenges, including longer travel distances, reduced access to qualified labour, and limited proximity to support services. These factors can contribute to higher transport and logistics costs, increasing overall energy expenditure.

Conversely, a more even distribution of employment across the rural–urban spectrum, more aligned lower vulnerability, can reduce these burdens. Enterprises located in areas with better employment alignment tend to benefit from shorter commuting distances, stronger local labour markets, and improved access to infrastructure. Research has shown that spatial employment balance contributes to more stable economic conditions and lowers dependency on energy-intensive mobility [28]. The European Commission, with the EU rural vision, similarly recognises the importance of employment alignment in its rural development strategy, noting its relevance to local economic resilience and energy efficiency [29].

• Building (dwelling) and energy

The energy efficiency of buildings occupied by micro-enterprises significantly influences their operational costs and overall energy vulnerability. Many micro-enterprises operate in older structures that often lack modern insulation, efficient heating and cooling systems, and up-to-date energy management technologies. These deficiencies can lead to increased energy consumption, higher utility bills, and reduced comfort for occupants.

A comprehensive analysis examined challenges and opportunities for improving energy efficiency in small and medium enterprises (SMEs) across Europe [7]. The findings highlighted that barriers such as limited access to information, financial constraints, and lack of technical expertise hinder SMEs from adopting energy-efficient practices. Conversely, drivers like staff training, facilitation of energy audits, development of corporate policy measures, and collaboration among SMEs within the same supply chain were identified as key mechanisms to improve energy efficiency uptake.

• Car and other vehicles and fuels

Transport-related energy consumption is a significant contributor to the operational costs of many micro-enterprises, particularly those active in services, logistics, delivery, repair, and construction. The reliance on vehicles, whether for transporting goods, accessing customers, or carrying out mobile services, makes these businesses highly sensitive to fuel costs and energy market fluctuations.

The type of vehicle, age of the fleet, and fuel used are key variables in understanding this dimension of energy vulnerability. Micro-enterprises that own older or poorly maintained vehicles often face higher fuel consumption rates, increased maintenance expenses, and elevated exposure to emissions-related regulations. Petrol and diesel vehicles, still dominant in many rural and semi-urban areas, are especially prone to cost volatility due to global oil price fluctuations. Firms located in remote areas, where alternatives such as public transport or electric charging infrastructure are limited, tend to depend even more heavily on such vehicles. The research emphasises that electric vehicles (EVs) can offer significant reductions in greenhouse gas emissions compared to conventional internal combustion engine vehicles, especially when the electricity used is generated from low-carbon sources. This underscores the potential benefits of adopting EVs for micro-enterprises aiming to enhance energy efficiency and sustainability.

However, the transition to energy-efficient or alternative-fuel vehicles often requires upfront investment that many micro-enterprises cannot afford without public support. While national and EU-level funding schemes (e.g., green mobility grants, tax incentives, scrappage programmes) aim to address these gaps, awareness and access remain limited, especially among the smallest businesses.

Furthermore, the lack of suitable infrastructure (such as electric vehicle charging stations) in non-urban locations adds to the hesitancy. This is particularly relevant in rural or economically lagging regions, where vehicle dependency is high but fleet renewal is slow due to capital constraints. Support mechanisms should therefore not only promote cleaner fleets but also target local accessibility to supporting infrastructure and financial instruments tailored to micro-enterprise capacities.

2.2. Energy Vulnerability of Micro-Enterprise

We also sought to answer the question of how we can determine if these factors result in energy vulnerability, by first understanding what exactly constitutes this vulnerability. We grounded our dependent variables on research related to household energy poverty and validated them with experts to identify key indicators of a micro-enterprise’s vulnerability. These indicators include challenges with bill payments (direct effect on the micro-company). We use this indicator in two ways, asking about current and prospective challenges, showing companies in absolute and future risk. Another one is risks to households associated with micro-enterprises (indirect welfare effect), which the literature finds inevitable for micro-enterprises and family companies [30,31,32]. We also ask about the energy efficiency paradox (inability or missing resources to invest in energy efficiency by those who need it the most) [33,34]. These insights contribute to a comprehensive understanding of the variables that define the energy vulnerability of micro-enterprises and help shape the relevant assessment.

3. Case Study—Croatia

In 2023, there were 156,145 entrepreneurs in Croatia, of which the majority were micro-entrepreneurs, a total of 137,950 (accounting for 88.3%). With 273,474 employees (26.6%), they generated total revenues and expenses of about 11.5% of all companies [35]. These shares are a bit lower than the European average, where micro-enterprises account for 93.6% of companies’ share. However, EUROSTAT methodology is limited to size (not including the aspect of turnover); therefore, if both are using the same methodology, results would be more in line with the EU [2].

Moody’s, Bureau van Dijk ORBIS database contains information on 84,663 Croatian micro-enterprises, which means the information about 61% of micro-companies and could be considered representative sample to show the overall financial status of micro-enterprises. After filtering out outliers using the 1.5 IQR rule, the average net profit/loss account (net income) is approximately 6745.99 EUR and the median is around 2954.91 EUR. The spread, as shown by the summary, suggests that most companies have P/L values between approximately 110 EUR (25th percentile) and 12,640 EUR (75th percentile). This shows a relatively low level of profit [36].

In 2021, the average energy expenditure of micro-enterprises in Croatia was 21,166 HRK (equivalent to EUR 2806.61). According to the Croatian Bureau of Statistics, the energy price index rose from 105.3% in 2021 to 123.5% in 2024 (with 2015 as the base year at 100%), reflecting a 17% increase [37]. This suggests that, under current conditions, a median-profit level micro-enterprise may struggle to afford energy costs. This makes a Croatian case study highly relevant for the topic.

4. Methodology

4.1. Data Collection

To identify the determinants of vulnerability among micro-enterprises, we developed a survey that evaluates both independent and dependent variables. We have investigated what could affect the vulnerability of micro-enterprises, considering a range of internal and external determinants. The independent variables represent the structural characteristics of businesses, such as sector, size, location, energy use, and infrastructure, that may influence their exposure or sensitivity to external shocks, as described in the conceptual framework. The dependent variables assess actual or perceived vulnerability, focusing on areas such as past investments in energy efficiency, resilience to energy price fluctuations, or inability to cover energy costs. The survey was developed both to support the design of Croatia’s Social Climate Plan (SCP) and for research. The SCP requires evidence-based identification of vulnerable groups. As such, the sampling strategy was purposive and stratified, aiming to capture a diverse and policy-relevant picture of micro-enterprises across Croatia, and included assistance from the national stakeholder (relevant ministry). As many researchers emphasise, accessing micro-enterprises for data collection is notoriously difficult, as they often have minimal administrative capacity, low participation in formal research, and limited awareness of energy-related programs. For this reason, institutional distribution channels and policy purpose was crucial [38,39].

Geographically, respondents included are from all Croatian NUTS2 regions, reflecting the regional distribution of micro-enterprises based on available national statistics. This ensured the inclusion of both urban and rural businesses, as well as those operating in areas of special state concern (e.g., mountainous or island regions). Sectorally, the survey includes key economic activities relevant for both the SCP and energy transition efforts, such as services, manufacturing, construction, and transport.

The EU Survey tool was chosen for its GDPR compliance and institutional reliability and it was distributed with the assistance from the Ministry of Environment and Green Transition, as part of efforts to determine energy vulnerability in specific social and economic contexts of micro-enterprises. All responses were fully anonymous and validated through consistency checks prior to analysis.

4.2. Sample

The sample consists of 503 responses, out of which 470 are valid and correspond to micro-enterprises. The sample offers strong geographic coverage across Croatia. This regional distribution reflects a well-balanced representation of micro-enterprises nationwide, as shown in Figure 2. The majority of companies (78.8%) do not fall into the country’s special care category areas (regions of particular state concern, supported regions, islands, mountain regions).

The sample reflects a diverse sectoral distribution across various industries. The services sector, including consulting, represents the largest portion at 36.5% of respondents. Other sectors—such as culture and creative industries, the digital sector, construction, and tourism—each account for approximately 7% of the sample. Transport-related businesses, encompassing both passenger and freight services, collectively make up around 10%. They also have highest energy (transport fuel included) expenditure in total expenditure as shown in Figure 3. Additionally, manufacturing and specialised industries, including textiles, electronics, and energy, are present in smaller yet notable proportions.

Most companies have 1 employee, followed by a decreasing number as the employee count increases. The average number of employees is 1.93 (including the owner), and the median number is 1. Based on the dataset from Croatian Financial Agency (FINA), the average number of employees in micro- enterprises is 1.98 in 2023, while the median is also 1; therefore, we can consider the dataset representative from the perspective of size.

The distribution of annual turnover indicates a predominance of companies with small annual revenue. The majority of companies, totalling 331 or 71.3%, report a turnover ranging from 0 to 100 thousand EUR. A significant proportion of these companies (over 70%) report profits below EUR 20,000. We can say this aligns perfectly with the description of the case study, stating low average profits of Croatian micro-enterprises.

Most companies operate from premises designated for residential use or situated in predominantly residential buildings. A total of 124 companies (26.4%) possess an energy certificate, with the majority classified in energy classes B or C. Given the limited dataset, we also inquired about the age of the company premises. Most of these buildings are relatively old and likely require energy renovation (if they have not implemented it already).

Companies utilising natural gas for heating and domestic hot water represent 34.3% of the total, around 33% use electricity, 22% wood, and around 6.2% district heating. Other energy carriers are minor. Most companies incur energy expenses ranging from 500 to 1000 EUR or exceeding 1000 EUR annually. Most vehicles operate on diesel fuel, a lesser number use petrol, while a minority use alternative fuels. As anticipated, the findings indicate that companies within the transport (freight and passenger) and service sectors incur higher fuel cost shares (exceeding 20% of total expenses).

4.3. Analysis of Variables

The dataset is structured to examine the determinants of energy vulnerability among micro-enterprises. It includes a comprehensive set of independent variables, as detailed in

Table 1. Independent variables for statistical analysis of vulnerability, authors.

No	Variable	Type
1	Region of Croatia (NUTS2)	Categorical
2	Special care area (determined by law)	Categorical
3	Primary sector of the company (NACE)	Categorical
4	Company age	Continuous
5	Number of employees (including owner)	Continuous
6	Annual revenue (last year)	Continuous
7	Annual profit (last year)	Continuous
8	Type of business premises	Categorical
9	Ownership status of premises	Categorical
10	Energy certificate class of building	Ordinal
11	Year of building construction (if no certificate)	Continuous
12	Main energy source for heating/hot water	Categorical
13	Total cost of heating and hot water	Continuous
14	Number of vehicles total (including all types)	Continuous
	Number of vans/minibuses, buses	Continuous
	Number of light duty vehicles (up to 3.5 t)	Continuous
	Number of heavy-duty vehicles	Continuous
	Number of work machinery	Continuous
15	Vehicle fuel types	Categorical
16	Cost of all fuels (including gas and oil) in total expenses	Continuous
17	Average vehicle age	Continuous
18	Average km per vehicle per year	Continuous

, that describe various business characteristics, operational conditions, and contextual factors.

The dependent variables in this dataset represent key indicators of energy vulnerability among micro-enterprises and are shown in

Table 2. Dependant variables detecting vulnerability, authors.

	Variable	Type
1	Investment in energy efficiency (buildings/vehicles)	Binary
2	Problems paying energy bills last year	Binary
3	Energy price changes as business risk	Binary/Ordinal
4	Energy price changes affecting household	Binary

. They are used to assess how various business characteristics influence a company’s exposure to energy-related challenges, financial strain, and capacity to adapt. We have explained in the conceptual framework why these indicators are used.

4.4. Methods for Statistical Analysis

Since our dataset includes 18 independent variables and 4 binary dependent variables, our first step was to assess whether there is a significant correlation among the dependent variables. This would help determine the most appropriate statistical analysis approach. To begin, we cleaned and prepared the dependent variables. For the variables related to investment, we recoded responses to reflect their impact on vulnerability: all positive answers (indicating investment) were coded as 0, as they suggest lower vulnerability, while negative answers were coded as 1, indicating higher vulnerability. For the other three dependent variables, such as those related to payment difficulties, we applied the opposite logic, coding positive responses (e.g., experiencing payment issues) as 1, to reflect higher vulnerability. We applied the chi-square test on this dataset, using appropriate formulas.

When this test is performed for the first variable (investment in energy efficiency), there is no expected correlation with other variables, so it can be seen separately. However, there is a statistically significant relationship between having payment problems and perceiving energy price changes as a business risk. Companies with payment problems are much more likely to see energy price changes as a risk, as visible from

Table 3. Relation between dependent variables, authors with Python 3.9.

DV1	DV2	Chi-Square	p-Value
Payment problems	Business risk	36.861	2.186 × 10−7

.

In “effect on household”, we have three types of answers, with many companies answering with “I don’t know”, as shown in

Table 4. Answers related to dependent variables, authors from survey.

DV1	DV4 Answers About Risks on Households
Payment Problems	I Don’t Know (2)	No (0)	Yes (1)
0	64	37	242
1	12	2	111

.

Most companies experiencing payment problems responded with “yes” or “don’t know”, while only two answered “no”. Among companies without payment issues, the majority also responded “yes”, but there was a greater variety of responses, including “don’t know” and “no effect on the household”. Although a correlation exists between the two variables, its statistical significance is notably lower compared to the relationship between payment issues and business risks. Based on this, we decided to proceed with separate analyses: for DV1 and DV4, with an intention to apply individual logistic regression models using the independent variables; for DV2 and DV3, which showed stronger interdependence, we used a multivariate model.

5. Results

5.1. Which Companies Invest in Energy Efficiency?

The analysis shows that companies are more likely to invest in energy efficiency or vehicles if they own their premises, have been around longer, or employ more people. On the other hand, having older vehicles or lacking an energy certificate is linked to lower investment in these areas. While building age negatively correlates with energy efficiency investments, this relationship should be interpreted cautiously. It remains unclear whether older buildings directly discourage energy upgrades, or whether other hidden factors, such as liquidity constraints or landlord–tenant dynamics, mediate this outcome. As visible from

Table 5. Independent variables with statistical significance to investment in energy efficiency, authors, R+.

Independent Variable		r	b	R2	p-Value	Interpretation of Results
4	Company age	0.123	0.044	0.015	0.0079	Older companies are more likely to have invested in energy efficiency or vehicles, possibly due to greater resources, financial stability, experience, or established business practices.
5	Number of employees	0.111	0.033	0.015	0.016	Companies with more employees are more likely to have invested in energy efficiency or vehicles, reflecting greater capacity and need.
9	Ownership status	0.168	0.138	0.027	0.00025	This is the most significant IV. Businesses that own their premises are more likely to have invested in EE measures or vehicles (the dependent variable). Ownership provides more control and incentive for long-term investments.
10	Energy certificate class of building	−0.124	−0.023	0.016	0.0075	These relations respond to simple logic, companies with newer vehicles, and better energy certificates are more likely to have invested in energy efficiency or vehicles. It is a consequence more than a determinant.
17	Average vehicle age	−0.143	−0.011	0.020	0.0043

, all significant relationships have low R-squared values, indicating that while the relationships are statistically significant, each variable alone explains only a small portion of the variance in the dependent variable.

• r—corelation coefficient
• b—regression estimate
• R²—coefficient of determination
• p-value—statistical significance

The analysis shows that companies with more employees, longer operational history, and ownership of their premises are significantly more likely to invest in energy efficiency or vehicle upgrades. These characteristics reflect greater financial capacity, stability, and autonomy, which reduce energy vulnerability.

This is consistent with findings by Caporale et al., who demonstrate that financial constraints, particularly limited access to external capital, are among the most significant barriers for SMEs undertaking energy-saving investments, even when awareness or intent is present [40]. The findings on the lack of ownership as a barrier are aligned with similar findings about barriers to energy efficiency being more pronounced in firms that do not own their premises, for example in Slovenia [41].

5.2. Companies with Problems Paying the Bills and Expected Risks of Future Price Changes

As visible from the distribution in

Table 6. Payment issues and perceived risk based on location, authors from survey.

NUTS 2 Region	Payment Issues (Proportion of Businesses Reporting Payment Problems)	Risk Perception (Proportion of Businesses Perceiving Energy Price Risks)
City of Zagreb	33.0%	86.8%
Pannonian Croatia	29.3%	80.6%
Adriatic Croatia	23.6%	79.3%
Northern Croatia	23.4%	78.9%

, there is obviously no significant difference in payment issues or risks based on location of the company, therefore no need to evaluate statistical significance. Regarding special areas, the highest rates of payment issues and risk perception are observed in mountainous and special state concern areas. However, none of the special care area categories have p-values below 0.05, meaning there is no statistically significant difference in the odds of payment issues between the reference group and any special care area group.

As for the sector differences, both dependent variables p-values are well above the significance threshold of 0.05 (payment issues: p-value = 0.3962, risk perception: p-value = 0.6878), indicating no statistically significant association between sector and either payment issues or risk perception. The highest rates of risk perception are observed in transport of both freight and passengers, but these differences are not statistically significant.

For payment issues, none of the company age groups show a statistically significant difference compared to the reference group, although newly opened companies had most payment issues. For risk, companies older than 10 years have higher odds of perceiving risk related to energy prices (p-value = 0.016). This is likely due to perception of risk, not so much the reality of the issues.

The analysis shown in

Table 7. Relation between significant independent variables (IVs) with payment issues and risk perception (DV), authors, Python.

IV	DV	Chi2	p-Value
6. Annual revenue (last year)	Payment issues	13.1830	0.10569913
6. Annual revenue (last year)	Risk perception	8.9150	0.71015888
7. Annual profit (last year)	Payment issues	15.7500	0.00441240
7. Annual profit (last year)	Risk perception	18.3283	0.00546198
13. Total cost of heating and hot water	Payment issues	15.7869	0.04553277
13. Total cost of heating and hot water	Risk perception	46.2662	0.00000625
16. Cost of all fuel in total expenses	Payment issues	11.9028	0.15559380
16. Cost of all fuel in total expenses	Risk perception	40.3935	0.00006186

clarifies that profit is a sensitive indicator of vulnerability. Regarding the energy and fuel costs, there is a strong link between expenses and vulnerabilities.

The results clearly demonstrate that both energy and fuel costs are strongly associated with how businesses perceive energy-related risks. The relationship between energy costs and risk perception is the strongest in the dataset, with an exceptionally high chi-square value and a p-value near zero, indicating an almost certain link. Fuel costs are also significantly related to risk perception. Additionally, energy costs are significantly associated with reported payment difficulties, although this relationship is not as strong as those related to risk perception. In contrast, fuel costs do not show a statistically significant link to payment issues.

These findings support the argument that energy expenditure plays a critical role in shaping both financial strain and future risk anticipation among micro-enterprises. However, we acknowledge that, as with any cross-sectional analysis, the observed relationships may be influenced by unmeasured confounding variables. For example, prior exposure to subsidies, owner-level financial literacy, or sector-specific energy norms may simultaneously affect both cost structures and perceived vulnerability. While such variables were not directly captured in this study, the use of chi-square analysis and correlational methods remains appropriate for identifying indicative patterns in our research.

Of all the other variables, which we assume have lower significance in the context of energy-related vulnerabilities, companies with more heavy-duty vehicles are more likely to experience payment problems, possibly due to higher operational costs. Businesses with older vehicle fleets show greater concern about energy price risks, likely due to lower fuel efficiency. Other variables (building type, energy certificates, etc.) do not show significant relationships with payment problems or energy price risks. However, the number of companies included with ownership over heavy duty vehicles is small and therefore results might be biased.

5.3. Energy Vulnerability of Micro-Enterprise Affecting Household

For the effect on household, we wanted to only consider relevant variables, including companies’ location, age, size, financial situation, total costs and ownership of dwellings and vehicles. A total of 353 respondents (75.1%) answered “Yes” to the dependent variable, indicating strong agreement. A total of 77 respondents (16.4%) selected “I don’t know”, while 40 respondents (8.5%) answered “No”. This means that most participants confirmed the statement, while a notable minority, roughly one in six, expressed uncertainty, and a smaller portion disagreed. A clear majority of respondents confirmed that rising energy costs in their business have a direct impact on their household, highlighting the financial vulnerability of micro-enterprise owners and the blurred lines between business and personal budgets. Meanwhile, an obvious 16.4% expressed uncertainty, indicating a lack of clarity or awareness about how business expenses influence their home life, a gap that may point to the need for better support. Only a small share reported no link between business energy costs and household finances, suggesting that full separation of business and household economics is rare in this group. This determinant cannot be used for evaluation of statistical relevance due to minimum number of negative answers. However, it can help us with conclusions in explaining the importance of energy vulnerability of micro-enterprises.

6. Discussion

The findings of the analysis reveal several key determinants that contribute to the energy vulnerability of micro-enterprises, as shown it Figure 4. Companies that own their premises are significantly more likely to invest in energy efficiency or vehicle upgrades, likely due to their ability to make long-term infrastructure decisions. Company size (number of employees) and company age show a positive relation with energy-related investments, suggesting that more established and larger businesses possess greater financial stability or operational capacity (or both) to manage energy investments. These characteristics collectively reflect a lower level of vulnerability. Referring to the literature, for example, special issue compilation on papers on firm age and performance gives us a bit more explanation on how to interpret our findings. The relationship between firm age and performance is complex and context dependent, highlighting the importance of considering multiple factors when assessing firm performance over time. However, older firms in general have higher financial stability and address long-term investments (like energy efficiency) more [42].

In contrast, companies that lack energy certificates, have older vehicle fleets, or do not own their properties are less likely to invest, pointing to greater exposure to energy-related financial stress. The most relevant finding on the link between split ownership and investment provides us insights for a new policy area in split incentives between premises and companies’ owners, as already recognised in the case of rented vulnerable households [43,44].

This might be an area where education could prove beneficial and provide assistance.

Profit levels were found to be significantly related to both payment difficulties and perceived risk, making it a more sensitive indicator of financial vulnerability than turnover/revenue alone. Additionally, total energy costs, especially for heating and hot water, emerged as the strongest predictor of both perceived risk and payment issues, while fuel costs were strongly linked to risk perception but not to actual payment difficulties. Sector, region, and company location did not show statistically significant effects, though certain groups (e.g., businesses in special concern areas or the transport sector) had higher observed levels of vulnerability. Some operational characteristics, such as the number of heavy vehicles, showed a possible link to higher payment issues, but the sample size was too small for reliable inference.

Finally, responses to whether energy price changes in micro-enterprises affect households underscore the deep personal impact of energy vulnerability of companies. With 75% of respondents affirming that energy costs influence their household finances, it is clear that energy vulnerability in micro-enterprises often transcends the business sphere and directly affects wellbeing, so this aspect should also be considered.

7. Conclusions and Policy Implications

In conclusion, energy vulnerability among micro-enterprises is primarily driven by structural and financial factors: ownership, size, profit, and energy expenses, with energy costs playing a central role. These findings highlight the need for targeted support policies that recognise the specific vulnerabilities of smaller, younger, and financially constrained enterprises. Results also show that energy price increases are not only a challenge for business operations but also pose a real threat to household wellbeing, especially for the most financially exposed small business owners. This brings important conclusions, outlined below.

7.1. Need for Data Collection for Informing Policies

The sample we evaluated is representative and unique, but small, and further data collection is essential. Effective data collection is important for understanding and mitigating energy vulnerability among micro-enterprises. For now, Eurostat’s energy statistics provide insights into energy consumption across various sectors but do not offer detailed breakdowns for micro-enterprises. Similarly, the Structural Business Statistics (SBS) database includes data on SMEs but often lacks granularity at the micro-enterprise level.

By gathering data on energy consumption patterns, financial health, and operational practices, countries can identify specific vulnerabilities and tailor interventions accordingly. For instance, data can reveal which enterprises lack energy-efficient equipment or are burdened by high energy costs relative to their revenue.

Further insights can be implemented through the micro-financing of simplified energy audits. Research indicates a cost–benefit advantage to conducting audits even when energy expenditures are less than 10% of total costs (or 2% for smaller companies) [45]. Also, firms that have implemented an energy audit generally have internal drivers for energy efficiency more intensely than firms that have not [41].

Moreover, data collection facilitates the monitoring of intervention outcomes, ensuring that support measures are effective, and resources are allocated efficiently. It also contributes to broader policy development by highlighting systemic issues affecting micro-enterprises across regions or sectors. For example, the Household Budget Survey (HBS) provides a much more detailed and structured model for collecting data on energy-related expenditures and vulnerability at the household level. This approach offers valuable lessons for improving data collection on micro-enterprises. In particular, the HBS demonstrates how to systematically capture disaggregated information on expenditure, including energy and household characteristics. This has brought huge progress in energy poverty research and policy development. A similar methodology adapted for the business context could enhance the quality relevance of data on micro-enterprises, helping to better assess their energy needs and target support measures more effectively. Such insights enable the development of targeted support programs, such as subsidies for energy-efficient upgrades or training on energy management (and we see the link between companies that, for example, do not have energy certificate and do have higher vulnerabilities).

7.2. Split Incentive in Micro-Enterprise

We have noted the importance of ownership of the infrastructure of the company for the investment in energy efficiency. A key structural barrier to improving energy efficiency in micro-enterprises, especially those operating in rented premises, is obviously the presence of a split incentive.

Research on split incentives in households has shown that this issue significantly reduces the uptake of energy efficiency investments. These findings are equally relevant to the commercial sector, particularly for micro-enterprises that often lease space and have limited negotiating power or capital for improvements. Learning from the household sector gives us information on further steps.

Tailored surveys targeting both micro-companies and landlords or renting facilities can help identify existing barriers, awareness, and attitudes toward cost-sharing in energy efficiency upgrades. In addition, modelling tools can be used to quantify how the costs and benefits of energy-saving measures are distributed between property owners and business tenants, as they were developed for households.

Lease agreements also offer an opportunity to encourage joint responsibility for energy performance improvements. Recognising the limited administrative capacity of many micro-enterprises, public authorities and business support organisations should provide advisory services, model contracts, and mediation support to help implement these solutions. Given that this segment often faces the dual constraint of limited capital and limited control over their premises, addressing this challenge is essential for enabling their participation in the energy transition.

7.3. Energy Policies and Enabling Tools

Beyond financial and structural constraints, the study highlights a notable awareness gap among many micro-enterprise owners, particularly in understanding the direct and indirect effects of energy costs on business and household finances. This lack of awareness limits their ability to assess risks, make informed energy-related decisions, and access available support schemes. Several studies show that employee awareness programmes are effective in lowering energy efficiency barriers and energy consultations are improving energy efficiency [41]. Developing accessible educational resources, offering basic energy management training, and integrating advisory services into support structures (e.g., chambers of commerce, SME helpdesks) can significantly enhance decision-making capacity.

Creating tailored guides and energy performance benchmarks for sectors like retail, hospitality, and small manufacturing could be beneficial as companies highly refer to benchmarking. Micro-enterprises are often overwhelmed by general advice and sector-specific materials make the recommendations more relevant and easier to implement.

An important aspect is the focus on multiple benefits of energy efficiency, as suggested in the Energy Efficiency Directive Art 11. This might be one of the reasons older companies recognise the strategic aspect of long-term investment in energy efficiency. For micro-enterprises, recognising and quantifying these benefits can shift the perception of energy upgrades from a cost to a strategic investment. Integrating these insights into policy and communication efforts can also increase uptake, as proposed by the EED [46].

7.4. Further Research

Building upon this understanding of energy vulnerability in micro-enterprises, several avenues for further research could be interesting. Sector-specific analyses could uncover unique vulnerabilities in different industries, enabling the development of tailored strategies to address sectoral needs. Investigating behavioural aspects, such as decision-making processes and risk perceptions related to energy use, would offer valuable insights into the adoption of energy-efficient practices.

Developing energy vulnerability composite indicators that include indicators like energy cost threshold and financial health could facilitate the identification of at-risk enterprises and the allocation of resources. Evaluating the effectiveness of existing policies and programs aimed at mitigating energy vulnerability would help in refining strategies and ensuring that support reaches those most in need.

Exploring the intersection between business operations and household energy use is crucial, especially for micro-enterprises where personal and professional often overlap. This could reveal compounded vulnerabilities and inform integrated support mechanisms (joint energy poverty/energy efficiency measures).