Natural Resource Rents and Income/Wealth Inequality in the European Union

Abstract

Starting with the debate on the “resource curse”, the main aim of this paper is to evaluate the impact of natural resource rents on income/wealth inequality in the European Union (EU) during the period from 1990 to 2023. Excepting the Gini index, natural resources rents reduced other measures of income and wealth inequality, and the results indicate that growth has a masking mediating effect on the Gini index, but no mediation role of GDP was observed in the case of the top 1% income/wealth share. The income inequality based on the top 1% share significantly increased in Denmark after the discovery of oil and gas relative to the control group composed of Finland and Sweden. Other control variables are considered, and some policy recommendations are proposed to reduce income/wealth inequality.

1. Introduction

Recent years have witnessed a surge in scholarly attention within economics, particularly concerning the dynamics of income inequality within nations, its subsequent impacts, and its underlying causes. Prominent contributions to this discourse include the work of Piketty and Saez (2013) and Alvaredo et al. (2017) [1,2]. A compilation of factors influencing income disparity, as identified by Li et al. (2016) encompasses advancements in technology favoring skilled labor, the effects of globalization in trade and finance, shifts in labor market regulations, and disparities in educational access [3].

Concurrently, a growing body of research has explored the relationship between natural resource endowment and income inequality, postulating that economies rich in natural resources tend to exhibit greater internal inequality compared to those with limited resources. Existing studies propose several mechanisms through which resources may exacerbate inequality. These include the concentrated ownership of resources [4], the promotion of institutional distortions through rent seeking and political control [5], and the creation of labor market imbalances such as the gravitation of labor towards less innovative sectors [6]. A recurring theme in this literature is the ‘resource curse’, which posits that natural resources can adversely affect institutional development, thereby undermining both political and economic structures. These institutional deficiencies, in turn, are thought to precipitate a range of suboptimal economic outcomes, including sluggish growth; diminished human capital; and, ultimately, heightened income inequality [7].

This study aims to contribute to the literature by directly evaluating the impact of natural resources rents on income/wealth inequality in the EU member states during the period of 1990 to 2023. Particular attention is paid to Denmark, where gas and oil deposits were discovered in 1966. To tackle endogeneity that might be also generated by reverse causality, this paper employs an econometric technique that solves the issue of endogeneity: Continuously Updated and Bias-Corrected (CUP-BC) estimators. Moreover, this study explores the mechanism through which natural resource rents influence income/wealth inequality via economic growth and assesses whether or not the oil and gas discovery in Denmark generated an increase in income inequality using data for the period of 1974–2023 and difference-in-differences (DID) estimators. The empirical findings are mixed. Natural resource rents increased the Gini index but reduced other measures of income inequality like the top 50% and top 1% income shares and measures of wealth inequality like the top 1% and top 10% wealth shares.

The late discovery of natural resources in already developed European nations could lead to a different set of outcomes compared to countries where resource wealth arrived much earlier in the development journey. A country with established institutions, a more diversified economy, and stronger social safety nets might absorb resource rents in a way that primarily benefits a broader segment of the population or allows them to be channeled through existing progressive tax systems. This could explain why this study found a reduction in some measures of inequality. On the other hand, in a less developed nation, the sudden influx of resource wealth might exacerbate existing inequalities. Weak institutions could be more susceptible to corruption and rent seeking, concentrating wealth in the hands of a few. The economy might become overly reliant on the resource sector, hindering diversification and potentially leading to a “resource curse” whereby other sectors stagnate and employment opportunities remain limited for many.

The remainder of this paper is structured as follows: The next section provides an overview of the existing literature on the natural resource–inequality nexus, including a detailed exploration of the theoretical channels through which resources can influence income distribution. Section 3 outlines the data and methodology, and Section 4 presents the empirical findings. In the last part of the paper, the results are discussed, and conclusions are formulated.

2. Literature Review

Studies examining the impact of natural resources have historically centered on the “resource curse”, investigating how resource wealth might impede national development [8]. Although the relationship between resource discoveries and income disparity has received less attention, a substantial body of research has explored the connection between resource abundance and inequality in wages and income [9]. This section provides a comprehensive review of relevant literature, detailing the mechanisms through which natural resources influence income inequality. The review is organized into two parts: the first summarizing studies that identify a positive association between natural resources and inequality and the second exploring those that find a negative or null relationship.

A significant body of research posits a direct correlation between natural resource abundance and heightened income inequality. Several explanatory mechanisms have been proposed, primarily centering on the impact of resources on labor markets, shifts in economic structure and export patterns, and the creation of institutional distortions.

Regarding labor markets, theoretical models, such as that developed by Leamer et al. (1999) [10], suggest that in countries with limited natural resources, labor-intensive production prevails, leading to a more equitable distribution of human capital and wages compared to resource-rich nations. Empirical evidence from Leamer et al. (1999) further supports this, showing that land-abundant Latin American countries exhibit a less skilled workforce and greater wage disparities than land-scarce Asian counterparts [10]. However, Spilimbergo et al. (1999) caution that wage inequality does not automatically translate to overall income inequality, as labor income constitutes only a portion of total income [11]. Expanding on Bourguignon and Morrison’s model [4], which implies that resource wealth exacerbates income inequality, Spilimbergo et al. (1999) [11] constructed a framework to analyze how the prices and ownership of production factors influence income distribution. Their empirical findings, based on a panel of nearly 100 countries from 1965 to 1992, indicate that countries rich in land and capital experience higher levels of income inequality.

Building upon labor market effects, researchers have also examined how natural resource-intensive production, particularly exports, contribute to inequality [12]. As noted by Spicker (2020), the increasing global demand for sophisticated products and services, relative to raw materials, compels resource-rich countries to amplify their exports of natural resource-intensive goods [13]. This trend concentrates income among resource owners. The seminal work of Bourguignon and Morrison (1990) corroborates this, demonstrating that resource-intensive exports increase inequality by concentrating wealth in the top 20% of income earners while diminishing the shares of the lower 40% and 60% [4]. Further research by Gylfason and Zoega [6] links natural resource abundance to greater wage inequality and reduced economic growth, attributing this to the unequal distribution of resource ownership. Buccellato and Alessandrini (2009) found that increased exports of ores and metals correlate with higher income inequality within exporting nations [14]. Auty (2007) argued that an over-reliance on resource-intensive exports hinders the absorption of surplus labor, leading to widening income disparities and an inflated public sector [15]. More recently, Farzanegan and Krieger (2019) showed a positive relationship between oil and gas revenues and income inequality in Iran, attributing this to increased imports, private sector credit growth, and higher real GDP per capita [16].

By examining panel data from the United States, the research of Berisha et al. (2021) revealed that the resource curse is, indeed, prevalent; the authors specifically investigated how oil resources impact income inequality, differentiating between the effects of oil abundance (production) and oil dependency (consumption) [17]. The findings indicate contrasting, non-linear effects. Increased oil production initially reduces inequality in states with low production but exacerbates it in states with high production. Conversely, greater oil dependency shows the opposite trend. The research suggests that increased rent seeking in oil-abundant states and vulnerability to commodity price shocks in oil-dependent states are potential mechanisms contributing to these observed inequalities.

Additionally, natural resource abundance has been observed to contribute to inequality by weakening institutional quality, thereby limiting individuals’ capacity to improve their economic prospects. Institutional deterioration primarily impacts human development, as evidenced by Carmignani (2013) [18], and leads to reduced educational investment, as shown by Cockx and Francken (2016) [19]. Caselli (2006) highlights that resource wealth triggers power struggles for control, diminishing incentives for long-term development investments [20].

Beyond human capital, the correlation between resources and weak governance is well documented. Bulte et al. (2005) demonstrated that natural resources impede the development of critical economic and political institutions across various economies [21], a phenomenon particularly pronounced in point-source-exporting countries [22]. Torvik (2002) modeled how resource abundance reduces income and welfare and increases inequality by decreasing the number of productive firms and fostering rent seeking [23]. In Russia, Buccellato and Alessandrini (2009) attributed rising income inequality to corruption and distorted economic institutions [14], while Ross (2001) provided a broader view of this impact on political institutions [24]. In China, Zhang et al. (2009) observed that state-owned enterprises capture resource gains [25], while household consumption declines due to rising prices, exacerbating income disparities. Caselli and Michaels (2013) found that increased governmental spending in Brazilian oil-rich municipalities does not translate to improved public goods or household incomes, suggesting rent seeking [26]. Irarrázaval (2023) showed how resource ownership fosters institutions that protect elites, perpetuating inequality [27]. Taking a historical perspective, Angeles (2007) argued that colonialism established political structures that consolidated resource wealth among colonizers, limiting indigenous access to capital and land [28]. Perez-Sebastian and Raveh (2016) demonstrate that fiscally decentralized developing countries are more vulnerable to the negative effects of resource booms due to rent seeking by local governments [29].

Awoa et al. (2024) investigated how economic complexity influences the relationship between natural resources and income inequality across 111 developed and developing countries between 1995 and 2016 [30]. Utilizing system GMM analysis, the study found that greater economic complexity negates the positive impact of natural resource dependence on income inequality. This finding holds true when differentiating between dependence on concentrated resources like fossil fuels and minerals and dispersed resources such as raw agricultural materials, as well as when considering overall resource abundance. Furthermore, the study identified significant variations in these effects based on a country’s level of ethnic fragmentation and democratic institutions.

While much research has focused on the impact of resource abundance on inequality, the role of resource discoveries and exploitation booms remains less explored. Ross (2001) found that while resource abundance may not affect income inequality, resource booms can increase vertical inequality by shifting labor toward resource extraction and related services [24]. Imperfect labor mobility may then lead to unemployment and increased inequality. Additionally, if exploitation begins in affluent regions, regional disparities can be exacerbated, contributing to horizontal inequality.

Iacono (2016) demonstrated that resource windfalls drive income inequality in Norway [31]. Similarly, Loayza and Rigolini (2016) showed that mining booms in Peru increase inequality between and within mining regions by attracting educated labor to mining-related sectors [32]. Marchand (2015) observed similar effects in Western Canada, albeit in a U-shaped pattern [33], where energy booms raise wages in both lower- and upper-income deciles. Smith and Wills (2018) found that exogenous oil price shocks and major discoveries widen the gap between urban and rural economic activity and fail to reduce rural poverty [34], indirectly contributing to inequality. This aligns with findings that resource discoveries, while boosting GDP per capita [35], also increase unemployment and child labor, negatively impact school attainment [36], reduce education quality [37], and induce brain drain [38]. Harding et al. (2020) showed that oil discoveries lead to exchange-rate appreciation, the reallocation of labor from traded to non-traded sectors, and productivity growth disparities, potentially contributing to income inequality [39].

Contrary to the “resource curse” narrative, some research indicates that natural resources can mitigate income inequality. Hartwell et al. (2019) highlighted the importance of robust institutions and government accountability in harnessing the benefits of resource wealth [40]. Parcero and Papyrakis (2016) further suggested that oil abundance can correlate with lower inequality in nations with equitable rent distribution, strong institutions, and low ethnic diversity [41]. While total natural resource rents had a significant negative impact on income inequality in the short run in Nigeria, suggesting that aggregate rents help reduce income disparity, rents from oil and natural gas individually exacerbated income inequality in the short term, as Kakain and Ewubare (2022) suggested [42].

Labor markets provide another potential avenue for reduced inequality. Ross (2001) posited that mineral resource revenues can generate public sector employment, potentially compressing income disparities in the short term [24]. Additionally, if resource extraction commences in economically disadvantaged regions, it can reduce horizontal inequality. However, Lay et al. (2008) found no significant impact of gas discoveries on income inequality in Bolivia, as countervailing forces neutralized each other [43]. Similarly, Allcott and Keniston (2018) observed that oil and gas booms in the US boosted local wages and welfare in extraction areas without affecting manufacturing productivity [44].

Education and household well-being can also improve with resource abundance. Kim and Lin (2018) demonstrated that oil abundance correlates with lower income inequality, attributing this to enhanced educational attainment and health outcomes, although rent seeking can diminish these benefits [45]. Ampofo (2021) found that oil extraction in Ghana improved household wealth and electricity access and reduced income inequality in extraction zones [46].

Country-specific studies have yielded mixed results. Howie and Atakhanova (2014) reported reduced income inequality following oil discoveries in Kazakhstan, a relatively homogenous society [47]. Farzanegan and Habibpour (2017) found similar effects on poverty and inequality in Iran [48]. Zabsonre et al. (2018) observed that gold mining in Burkina Faso reduced poverty but not income inequality [49]. Tano et al. (2016) noted that Sweden’s mining boom increased labor income across various sectors [50]. Goderis and Malone (2011) found that oil and mineral resource booms decreased income inequality in developing countries in the short run but had no long-term effect [51]. Employing the system GMM estimator for low-income, lower–middle-income, and resource-rich countries during the period of 2009–2019, Gemicioğlu et al. (2024) observed an inverted U-shaped relationship between natural resource rents and income inequality [52]. This suggests that resource rents initially worsen inequality up to a certain point, after which they begin to reduce it.

These diverse findings underscore the context-dependent nature of the relationship between natural resources and inequality. A thorough understanding necessitates a focused examination of individual countries’ economic structures and historical trajectories.

3. Methods and Data

The data used in this study cover the years of 1990 to 2023 and include the EU-27 countries, contingent on data availability. Income inequality is quantified using the Gini index, sourced from the World Bank. Wealth inequality is assessed through indicators from the World Inequality Database (WID) based on net personal wealth: the top 10% wealth share (p90p100 wealth), indicating the percentage of total wealth held by the richest decile; the bottom 50% wealth share (p0p50 wealth), representing the proportion of total wealth owned by the least wealthy half of the population; and the top 1% wealth share (p99p100 wealth), illustrating the percentage of total wealth concentrated among the wealthiest percentile.

More control variables were added to the models. Inflation is determined using the consumer price index (CPI), with 2010 as the base year (CPI 2010 = 100). This index captures changes in the price level of a representative set of consumer goods and services. The CPI data were obtained from the World Development Indicators database. The rest of the variables are presented in

Table 1. Data description—explanatory variables.

Definition of Variable	Notation	Data Source
Total natural resource rents (% of GDP)	rents	World Bank database
GDP per capita (constant 2015 USD)	GDP
Consumer price index (2010 = 100)	inflation
Secondary school enrollment (% gross)	edu
Urban population (% of total population)	urban
Net foreign direct investment inflows (% of GDP)	FDI
Trade (% of GDP)	trade
Population growth (annual %)	pop
General government final consumption expenditure (% of GDP)	gov
Domestic credit to private sector (% of GDP)	credit
Control of corruption: estimate	corruption	Worldwide Governance Indicators website

Source: own description.

. With the exception of the control of corruption index, all data were transformed using the natural logarithm, and variable names reflect this transformation.

Let us consider the basic model to explain the income/wealth inequality–natural resource rent nexus:

$${inequality}_{it}={\alpha }_i+\beta ·{rents}_{it}+\sum _{j=1}^m{\gamma }_j·X_{it}+{\epsilon }_{it}$$

(1)

inequality—measure of income/wealth inequality;

m: number of control variables;

i: index for state;

j: index for control variable;

t: index for year;

${\alpha }_i$, $\beta$, and ${\gamma }_j$: parameters;

${\epsilon }_{it}$: error;

X: control variable.

To analyze the long-term connections between income/wealth inequality and natural resource rents, this research employs continuously updated and bias-corrected (CUP-BC) estimators. This method offers the advantage of providing unbiased estimates for cointegrated variables without needing to specify exogeneity or instrumental variables. Furthermore, it remains reliable even if certain variables are not included in the cointegrating relationship. The subsequent analysis is based on the regression models outlined in (1) and on the following:

$${\epsilon }_{it}={\lambda }_i^T·f_t+v_{it}$$

(2)

$${\epsilon }_{it}^{\prime }={\lambda }_i^{\prime T}·f_t^{\prime }+v_{it}^{\prime }$$

(3)

$v_{it}$ and $v_{it}^{\prime }$—idiosyncratic errors;

$f_t$ and $f_t^{\prime }$—vectors of unobserved common factors that impact the dependent variable and natural resource rents;

${\lambda }_i$ and ${\lambda }_i^{\prime }$—loading factors that change across countries.

Following the approach of Bai et al. (2009) [53], CUP-BC estimators that are robust to the integration order of the underlying factors are used, accommodating both stationary (I(0)) and non-stationary (I(1)) processes.

The descriptive statistics and matrix of correlation are reported in Appendix A. GDP is strongly correlated with credit, corruption, and inflation. Inflation is correlated with edu. Credit and corruption are strongly correlated.

Natural resource rents are likely to influence income/wealth inequality by enhancing economic growth. Therefore, it is necessary to propose a causal model that highlights the role of economic growth as a mediating factor:

$${GDP}_{it}=a_0+b_1·{rents}_{it}+b_2·X_{it}+c_i+d_t+v_{it}$$

(4)

$$\begin{array}{l}{inequality}_{it}=e_0+e_1·{rents}_{it}+e_2·{GDP}_{it}+e_3·X_{it}+c_i+d_t\\ \hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}\hspace{1em}+{\pi }_{it}\end{array}$$

(5)

$v_{it}$ and ${\pi }_{it}$—errors;

$c_i$—country fixed effect;

$d_t$—year fixed effect.

To investigate whether natural resource rents affect income or wealth inequality via economic growth, this paper utilizes the mechanism effect test method established by Baron and Kenny (1986) [54]. The initial step involves verifying the significance of the $f_1$ parameter in the following equation:

$${inequality}_{it}=f_0+f_1·{rents}_{it}+f_2·X_{it}+c_i+d_t+e_{it}$$

(6)

If $f_1$ in Equation (6) is significant, we should check if $b_1$ and $e_2$ are significant. If both coefficients are significant, a mechanistic effect is valid. Then, if $e_1$ is non-significant, the mechanistic effect is total; otherwise, it is partial.

The main hypotheses are outlined as follows:

H1: 

Natural resource rents have a negative impact on income/wealth inequality in the European Union.

H2: 

Economic growth mediates the relationship between natural resource rents and income/wealth inequality in the European Union.

Particular attention is assigned to Denmark. In the case of Denmark, oil and gas reserves were discovered in 1966, with production starting in 1972, while the event year is 1982 [35]. Therefore, this paper employs difference-in-differences (DID) estimation to assess the impact of natural resource rents on income/wealth inequality in Denmark compared with other Nordic countries in the EU (Sweden and Finland), where oil and gas were not discovered. The event year is 1982, exactly as in paper by Smith (2015) [35]. While the initial discoveries were made earlier, high-quality, consistent data on indicators became reliably available starting around 1990, aligning with the panel data analysis period. Denmark likely possesses a long and detailed history of such data collection compared to some other EU nations with resource wealth. Even if the initial discovery was earlier, the period from 1990 to 2023 might represent a significant and mature phase of resource rent generation for Denmark, making its impact on inequality more pronounced or observable during this time. This paper argues that the sustained flow of rents within the analyzed period is more relevant than the initial discovery.

4. Results

Cross-sectional dependence is supported for all series, excepting inflation, at a 5% significance level. The series for Gini, credit, urban, gov corruption, p90p100 income, and p99p100 income are stationary in the first difference, while the rest of the data series are stationary in terms of level at the 1% significance level, as

Table 2. The results of preliminary tests.

Variable	CD-Test Stat.	Pesaran’s CADF Test
		Data in Level	Data in the First Difference
Gini	4.34 ***	0.237	−6.456 ***
GDP	−3.64 ***	−11.191 ***	−18.716 ***
credit	−6.45 ***	0.346	−8.269 ***
trade	−2.87 ***	−9.766 ***	−16.566 ***
urban	25.18 ***	1.866	−1.698 **
gov	83.73 ***	−0.321	−10.301 ***
corruption	5.65 ***	−0.249	−9.367 ***
edu	−2.09 **	−5.258 ***	−9.930 ***
inflation	−1.36	−7.781 ***	−17.473 ***
p90p100 income	24.64 ***	0.413	−14.310 ***
p0p50 income	21.81 ***	−3.383 ***	−14.266 ***
p99p100 income	29.03 ***	0.048	−13.600 ***
p90p100 wealth	4.57 ***	−6.417 ***	−8.090 ***
p0p50 wealth	7.78 ***	−4.100 ***	−7.807 ***
p99p100 wealth	21.67 ***	−7.568 ***	−9.568 ***
rents	28.14 ***	−2.009 **	−12.960 ***
FDI	27.84 ***	−5.656 ***	−16.184 ***
pop	2.02 **	−4.759 ***	−13.633 ***

Source: own calculations in Stata 15. Note: ***, and ** indicate significance at 1%, and 5%, respectively.

indicates.

4.1. Baseline Estimations

The results of estimations indicate that natural resource rents increased the Gini index but reduced the top 1% and 50% income shares. Inflation increased the Gini index and the top 1% share. Education, credit, and trade reduced income inequality, while FDI, population growth, and government expenditure increased it. Urban population reduced the top 50% share, while control of corruption reduced the Gini index and the top 10% share (see

Table 3. CUP-BC estimators to explain income inequality.

Variable	Gini		p90p100 Income		p0p50 Income		p99p100 Income
rents	0.013 *	0.007 *	0.001	0.001	−0.007 **	−0.004 *	−0.003 *	−0.005 *
inflation	0.032 *	-	0.001	-	0.007	-	0.002 **	-
credit	−0.004 *	-	−0.003 *	-	−0.001 *	-	−0.011 *	-
trade	−0.100 ***	−0.106 ***	−0.029 *	−0.014 *	−0.002 *	−0.004 *	−0.005 *	−0.002 *
urban	−0.284	−0.488	−0.107	−0.169	−0.344 *	−0.316 *	−0.083	−0.069
gov	0.034 *	0.056 *	0.072 *	0.122 **	0.057 **	0.074 ***	0.030 *	0.069 *
corruption	-	−0.068 ***	-	−0.066 *	-	−0.003	-	−0.004
edu	-	−0.053 **	-	−0.066 *	-	−0.022 *	-	−0.014 *
FDI	0.009 *	0.012 *	0.002	0.003	0.002 *	0.002 *	0.002 *	0.002 *
pop	0.033 **	0.029 ***	0.039 *	0.032 *	0.017 *	0.037 *	0.039 *	0.017 *
constant	4.212	7.748	−0.914	−1.904	2.998 **	3.337 ***	−0.273	−0.826

Source: own calculations in Stata 15. Note: ***, **, and * indicate significance at 1%, 5%, and 10%, respectively.

).

The results presented in

Table 4. CUP-BC estimators to explain wealth inequality.

Variable	p90p100 Wealth		p0p50 Wealth		p99p100 Wealth
rents	−0.008 **	−0.006 **	0.003	0.002	−0.008 **	−0.007 **
inflation	0.005	-	−0.003	-	0.002 *	-
credit	0.002 *	-	0.003	-	0.002 *	-
trade	0.002 *	0.003 *	0.004 *	0.002 *	0.002 *	0.005 *
urban	−0.256	−0.484	0.135	0.111	−0.151	−0.523
gov	0.042 *	0.034 *	0.014 *	0.008 *	0.041 *	0.080 *
corruption	-	−0.006	-	0.001	-	−0.001
edu	-	−0.005	-	0.003	-	−0.001
FDI	0.004 *	0.003 *	0.008 *	0.006 *	0.004 *	0.002 *
pop	0.006 *	0.003 *	0.002 *	0.003 *	0.003 *	0.006 *
constant	0.493	1.769	−0.085	−0.573	−0.215	0.534

Source: own calculations in Stata 15. Note: **, and * indicate significance at 5%, and 10%, respectively.

suggest that natural resource rents reduced the top 1% and top 10% wealth shares. Inflation and credit increased the top 1% wealth share. FDI, population growth, trade, and general government final consumption expenditure increased wealth inequality.

For a robustness check, non-linear connections were considered, but none of the non-linear models based on square resources was valid.

4.2. Mechanisms Mediating the Role of Economic Growth

To explore the mechanism through which natural resource rents influence income/wealth inequality via economic growth, this paper employs the system of recursive equations proposed by Wen and Ye (2014) [55]. The results presented in

Table 5. The mediating role of economic growth.

Variable	Gini	GDP	Gini	p99p100 Income	GDP	p99p100 Income	p99p100 Wealth	GDP	p99p100 Wealth
rents	0.01 *	−0.049 *	0.08 *	−0.002 *	−0.049 *	−0.002 *	−0.006 *	−0.049 *	−0.006 *
trade	−0.105 ***	0.214 ***	−0.101 ***	0.009 *	0.214 ***	0.017 **	0.006 *	0.214 ***	0.001 *
urban	0.088	−7.027	−0.303	0.222	−7.027	0.205	−0.396	−7.027	−0.395
gov	0.185 *	0.288 *	0.017 *	0.001 *	0.288 *	0.006 *	0.061 *	0.288 *	0.069 *
edu	−0.079 ***	0.204 *	−0.094 ***	−0.004 *	0.204 *	−0.017 *	−0.002	0.204 *	−0.004
FDI	0.010 *	0.059 **	0.011 *	0.002 *	0.059 *	0.002 *	0.002 *	0.059 *	0.005 *
pop	0.088 *	−0.155 **	0.026 **	0.001 *	−0.155 **	0.002 *	0.009 *	−0.155 *	0.009 *
GDP	-	-	−0.103 ***	-	-	−0.005	-	-	0.002
constant	−0.654	41.389	7.046	−0.858	41.389	−0.833	0.468	41.389	0.232

Source: own calculations in Stata 15. Note: ***, **, and * indicate significance at 1%, 5%, and 10%, respectively.

show that economic growth did not mediate the effect of natural resource rents on the top 1% share based on income and wealth inequality, but growth had a partially mediating effect on the Gini index, although in the form of a masking effect. This result suggests that there are other factors that mediate the natural resource rent–Gini index nexus.

4.3. The Case of Denmark: Difference-in-Differences (DID) Estimators

This paper employs a DID estimator with Denmark as the treatment unit and other Nordic countries in the EU as the control group. The event year is considered the year of intervention, and it corresponds to the first year when oil and gas output grew by 0.5 barrels per capita [40]. The control variables in the models with data provided by the World Bank for the period of 1974–2023 are GDP per capita (constant 2015 USD), trade (% of GDP), population growth (annual %), and domestic credit to the private sector (% of GDP). The data for inequality measures are provided by the World Wealth & Income Database, while total natural resource rents (% of GDP) are provided by the World Bank. The results presented in

Table 6. Difference-in-differences estimation results for income inequality.

Outcome Variable	Top 10% Share	Top 1% Share
Before
Control	0.279	0.072
Treated	0.269	0.066
Diff (T-C)	−0.01 (0.363)	−0.006 (0.485)
After
Control	0.303	0.098
Treated	0.302	0.107
Diff (T-C)	−0.001 (0.894)	0.010 (0.014)
Diff-in-diff	0.009 (0.435)	0.016 (0.093)

Source: authors’ calculations in Stata 15. Note: p-value in brackets.

indicate that income inequality based on the top 1% share significantly increased in Denmark compared to the control group (Finland and Sweden) after the discovery of oil and gas, as reported by Smith (2015) [35]. The result is contrary to that reported by Hartwell et al. (2019) [40], who checked the same hypothesis using synthetic control measures and other, larger control groups. However, no significant impact was observed in the case of the top 10% share. The results are, indeed, conditioned by the method of analysis, period, and control group.

5. Discussion

Greater revenue from natural resources (e.g., oil and minerals) widens the overall income gap in a country based on the Gini index. This could be due to the concentration of wealth in the hands of a few, corruption, or a lack of effective redistribution policies. However, while overall inequality is increased, the very top earners and the upper–middle class have a slightly smaller slice of the pie. It could be that resource rents create a wider gap between the very poor and the middle class, thereby raising the Gini index but not significantly changing the very top percentages. Sawadogo and Ouoba (2024) showed that natural resource rents reduce income inequality mostly in countries with political stability [56]. Our results are also in line with those of Hartwell et al. (2019) [40], who showed that natural resource discoveries reduced income inequality in Denmark, the Netherlands, and Norway. However, our comparison with Sweden and Finland revealed an increase in income inequality due to oil and gas discovery.

Resource rents might disproportionately benefit a segment of the population that is not in the top 1% or even the top 10% but is significantly wealthier than the lower-income deciles. While top management and ownership might remain concentrated, the resource sector and its supporting industries (e.g., specialized services, logistics, and some manufacturing) could create relatively well-paying jobs for a significant portion of the middle class. This would pull the middle class further away from the lower income groups, increasing overall dispersion and, thus, the Gini index. While not enough to shrink the overall gap, resource revenues might fund social programs or tax policies that disproportionately benefit the lower and middle classes, increasing their income relative to the poorest but not significantly impacting the very top. It is conceivable that corruption and rent-seeking activities associated with natural resources might primarily benefit individuals and groups below the very apex of the wealth distribution—for instance, well-connected business owners or politically influential individuals who are not necessarily the absolute wealthiest. This could inflate the income and wealth of the upper–middle class more than those of the top 1%.

High inflation disproportionately affects lower-income individuals, as their purchasing power diminishes. This can exacerbate overall inequality and concentrate wealth at the very top. The results are in line with those of Law and Soon (2020) [57], who showed that inflation increased income inequality in 65 states during the period of 1987–2014. Education, credit, and trade, when well-managed, tend to promote economic mobility and broader participation in the economy, leading to a more equitable distribution of income.

Foreign direct investment (FDI) can sometimes create a dual economy, with benefits concentrated in specific sectors or regions, leaving others behind. Rapid population growth can strain resources and create competition for jobs, potentially widening income disparities. Government expenditure, if not targeted effectively, can also increase inequality, for example, if the expenditure is used to benefit already wealthy segments of the population.

A larger urban population, while potentially driving economic growth, may not necessarily benefit the upper–middle class disproportionately. It could mean that the growth of urban environments creates more lower- or middle-class jobs, thereby lowering the percentage of the top 50% income share.

Effective control of corruption promotes transparency and fairness, reducing the ability of elites to accumulate wealth through illicit means, thereby reducing overall inequality and the income share of the top 10%.

Important results were observed when the impacts of various factors on wealth inequality were analyzed. The finding that natural resource rents reduce the top 10% and top 1% wealth shares implies that resource wealth is being distributed in a way that slightly moderates extreme wealth concentration, perhaps through government programs or taxation. However, the magnitude and sustainability of this effect are very low. This result is contrary to that reported by Tadadjeu et al. (2023) [58], who observed that natural resource rent enhanced wealth inequality in 45 developed and emerging economies in the period of 2000–2014. Government programs and taxation are potential channels moderating extreme wealth concentration in the EU. The effectiveness and progressivity of these policies likely differ significantly between the EU and the broader sample considered by Tadadjeu et al. (2023) [58]. Trade, credit, FDI, population growth, and government expenditure have all been shown to increase wealth inequality. This aligns with many economic theories. Trade and FDI can benefit those with existing capital, widening the gap. Tadadjeu et al. (2023) also showed that trade increased wealth inequality, but FDI reduced it [58]. Increased credit access may favor those who are already wealthy, leading to further asset accumulation. Population growth can strain resources and potentially exacerbate existing inequalities. Government expenditure, if not carefully targeted, could also increase wealth inequality.

The finding that inflation increases the top 1% wealth share is notable. This could be due to various causes, such as the wealthy owning assets that appreciate with inflation (e.g., real estate and stocks), their ability to hedge against inflation more effectively than the less wealthy, and the fact that the wealthy have more power to increase their income to match or exceed the inflation rate. Altunbaş and Thornton (2022) also highlighted the role of inflation in increasing inequality in the case of 121 states in the period of 1971–2015 [59].

More attention should be assigned to Denmark. The oil and gas industry often generates substantial profits. These profits can become concentrated in the hands of a relatively small number of individuals and corporations. This concentration can lead to a significant increase in the wealth and income of the top 1%. Even with strong social welfare systems, the sheer volume of wealth generated by resource extraction can outpace redistributive efforts. Smith (2015) also found that oil and gas discovery enhanced income inequality [35].

While Denmark has a strong social welfare system, the “resource curse” is a phenomenon whereby reliance on natural resources can lead to economic distortions. These distortions can include increased focus on the resource sector, potentially neglecting other sectors of the economy; fluctuations in wealth due to volatile commodity prices; and potential for corruption and rent-seeking behavior, which can exacerbate inequality. Profits from oil and gas may be invested in financial markets, leading to capital gains that disproportionately benefit the wealthy [60]. Those who already possess capital are better positioned to take advantage of new investment opportunities created by the resource boom. While the oil and gas sector create jobs, these jobs may be highly specialized and concentrated in certain regions. This can lead to wage disparities between those employed in the sector and those in other industries.

6. Conclusions

The results highlight the importance of policies that promote education, financial inclusion, and good governance while mitigating the negative effects of natural resource dependence and uncontrolled inflation. In essence, this study also suggests that while natural resources may have a slightly equalizing effect, factors related to globalization, financial markets, and macroeconomic conditions tend to exacerbate wealth concentration. While Denmark’s social welfare system aims to mitigate inequality, the concentrated nature of resource wealth can still lead to a widening gap between the very rich and the rest of the population.

This paper also presents limitations that could be tackled in future studies. For example, we considered only the case study of Denmark, but another country that could be considered is the Netherlands, with an initial oil and gas discovery in 1959. The first year of production in the Netherlands was 1963, and the event year is 1966. For a smaller set of data without missing values, one may apply the synthetic control method (SCM) considering a control group comprising the other 25 EU member states, with Denmark and the Netherlands considered separately as treatment units.