Building European Cities, Shaping Economies: The Roles of Infrastructure and Demographics in Urban Economic Performance (2017–2022)

Abstract

This study examines the interplay between urban policy interventions, infrastructure investments, and economic and labor market outcomes in large metropolitan areas from 2017 to 2022. We have employed empirical research to unravel the phenomena among these domains by employing a comprehensive dataset comprising gross value added (GVA), compensation of employees, gross fixed capital formation, environmental taxes, trade, employment, and time series data for demographic cohort transitions. Through a linear regression analysis, we uncovered strong positive associations between gross fixed capital formation and GVA and employment per capita, highlighting the critical role of infrastructure in driving urban economic growth. We also scrutinize how population dynamics—natural population changes and net migration—affect economic performance, offering valuable insights for evidence-driven urban policy. This research informs sustainable urban economic development policies and contributes significantly to the urban economics discourse.

1. Introduction

Issues of infrastructure development and upgrading, as well as the impact of investment financing on infrastructure, poses challenges to all national economies worldwide. In the countries of the South, the deficiencies are mainly centered to the lack of basic infrastructure, whether economic or social, as well as the rudimentary development—or even the non-existence—of transport and communication networks. In developed countries, the global competition in terms of improving economic efficiency and promoting the digital economy is imperative for global metropolises and the European urban agglomerations, without, however, underestimating the urgent need for upgrading the already outdated infrastructure.

The aim of the present study is precisely to reveal the essential role of infrastructure upgrading in the economic performance of European urban agglomerations, as commonly reported in other relevant empirical studies. The critical role of public infrastructure investment in enhancing economic productivity has been extensively established in the literature, with Aschauer [1] demonstrating its significant contribution to output growth. The contribution of this study to the relevant literature becomes even more relevant, especially if taking into account the challenges and limitations set by the urban tissue itself on infrastructure deployment in European cities. The latter may be due to historical reasons or to the need to preserve the cultural heritage of traditional European urban centers, instead of a “tabula rasa” infrastructure development strategy in modern Asian metropolises [2]. These potential linkages, whether revealing a correlational or causational relationship, remain to be examined in the present analysis.

Demographic challenges are an equally important factor affecting urban economies. Shifts in population compound demands on the city infrastructure and public services. Considering those demographic challenges is critical when they impact population sheds and hence the demands for labor, housing, education facilities, and health services. An aging population and slowly declining birth rate in many of the metropolises within Europe raises more pressing needs for customized human resource management policies. The rise in migration significantly amplifies the demand for social inclusion and access to essential services.

More studies are needed regarding the correlations between infrastructure investment, demographic dynamics, and economic performance in large metropolitan areas, especially in the European context. Although several studies have examined the effect of infrastructure on economic growth [3,4] or demographic changes individually [5,6], the combined analysis of these two factors and their links to economic performance and employee earnings still need to be researched.

The literature documents infrastructure as a critical factor in economic growth, as it facilitates the flow of goods, services, and people, thereby helping to boost productivity. However, limited research precisely quantifies the impact of a specific infrastructure, such as ports and roads, on the economic performance of major cities in Europe. This creates a knowledge gap about how infrastructure affects different urban environments and how it can be used as a policy tool for sustainable urban development. At the same time, while the impact of demographic dynamics, such as natural population growth and immigration, on the economy has been recognized, the analysis often remains fragmented and isolated from other critical factors, such as infrastructure. More empirical studies must examine how demographic changes interact with existing and new infrastructure to affect economic growth and employee earnings. This interaction is critical since demographic changes, such as population growth or decline, directly affect the demand for infrastructure, services, and jobs.

Moreover, studies examining the complex interaction between infrastructure and demographic changes in the same analysis are quite limited. Although both factors are essential for economic development, their interaction and impact on the urban economy have yet to be sufficiently explored. This gap limits our understanding of how policies focused on both infrastructure and population management can combine to create sustainable and resilient urban economies.

This article fills this gap with an interdisciplinary and synthetic approach, combining data on infrastructure, demographics, and economic performance. The proposed analysis seeks to provide new insights and evidence-based proposals for development policy and is focused on a more comprehensive understanding of how infrastructure and demographic changes interact to shape economic performance and employees’ well-being in major European cities.

2. Literature Review

It is commonly accepted in the existing literature that urban development and the formation of urban agglomerations are considered to fuel economic growth across national economies [7]. The role of infrastructure is deemed crucial for coordinating private investments, along with the need for adequate financing tools that enable urban growth [8]. The assessment of infrastructure’s effects on economic performance should focus within the metropolitan areas, as mentioned by Immergluck [9], calling for economic development planning at all public administration levels. Maket et al. [10] propose a measure of urban agglomeration by employing Herfindahl–Hirschman Index estimates in order to assess the impact of urban agglomeration changes on economic performance connectedness. They support that the linkages are rather country-specific, thus distinguishing a differentiated effect when it comes to developing economies compared to developed ones.

Investments in infrastructure by definition entail high initial costs, or even comparatively higher costs in relation to their respective operating costs, which nevertheless aim to boost economic activities in local economies, thus exerting a direct or even indirect impact on economic performance. The World Bank Guidance paper [11] provides a typology of the socioeconomic impacts on infrastructure investments, namely, by distinguishing the direct and indirect impacts, and the induced and catalytic impacts. The direct impacts refer to the effects of the deployment, operation, and maintenance of the infrastructure itself, while the indirect ones could be related, for example, to spillover effects on local employment opportunities. The induced impacts go a step further and incorporate the effects of increasing employment opportunities on household spending, while catalytic impacts may relate to long-term spillovers in other economic sectors (e.g., the tourism industry).

Urban agglomerations now host and induce technological innovation, facilitating and providing incentives for the deployment of industries that take advantage of the dense urban transport and communication networks for their economic activities. Proximity to the abundant urban labor force and a network of partner businesses and suppliers, as well as the visibility to a broader consumer base, are some of the advantages of urban agglomeration economies. Interconnectedness at the regional, national, and international level also matters for economic performance. The degree of interconnectivity between partner industries in the case of Germany, according to Shutters et al. [12], is associated with improving economic performance, being able to cause larger shocks in the cases of more coherent industrial structures. The authors refer to the challenges posed by the dilemma between higher exposure to shocks due to a higher degree of interconnectedness, on the one hand, and higher economic resilience as a result of looser ties between different industries.

The empirical studies extend beyond the upgrading of the existing infrastructure in urban areas across the European cities due to obsolescence, mentioning also the need to update them in order to provide a prompt response to the upcoming digital transformation of production processes. In this context, the German Standardisation Strategy [13] proposes in Objective 3 the introduction of new platforms related to the promotion of “smart cities” and other issues linked to the competitiveness of cities in the near future.

When it comes to country-specific case studies and trying to delve into the ways the European urban agglomerations evolve and become more efficient, in economic terms, Dijkstra et al. [14] highlight the beneficial role of improvements in the access to broadband services, especially in the periphery of large cities. On the contrary, they point out that pollution and congestion costs exert a clear repulsive effect toward smaller urban centers and rural areas. Wang et al. [15] examine the linkages between the presence of the Chinese National Economic and Technological Development Zones (NETDZs) and urban economic performance, highlighting the role of foreign investment and investments in fixed assets on the Chinese industrialization process. Finally, Wu [16] tries to evaluate the way infrastructure investment financing is related to urban economic performance by also introducing lagged variables in order to identify any short-term and medium-term effects.

3. Data and Methodology

3.1. Study Area and Data

The main purpose of this research is to investigate the impacts of infrastructure investment and employees’ compensation on the economic growth of the EU-27 metropolitan areas from 2017 to 2022.

The present study employed 244 major cities (Appendix A) across the 27 European Union (EU) countries based on criteria that enhance scientific value and research targeting. The selected cities are the main economic activity centers in the EU, gathering a significant proportion of the population and economic output. Because of their size and complexity, megacities face unique challenges and opportunities, particularly in infrastructure and employment, making them ideal laboratories for analyzing the impacts of urban development policies and investments. The selected cities differ in demographic composition, geographic location, and level of development. This enables the empirical investigation of various parameters affecting the urban economy, such as infrastructure investment, labor market policies, and demographic changes. Analyzing such a diverse sample provides significant comparative advantages for drawing generalizable conclusions that can contribute to the design of effective policies.

The European Union, a single political and economic entity, has several common objectives for urban development, such as promoting sustainability, economic resilience, and social cohesion. Policies implemented at the EU level directly or indirectly affect city development, making the member states an ideal context for evaluating the effectiveness of these common strategies. Finally, the study of EU cities offers a valuable opportunity to compare the effects of infrastructure investment and demographic change on different economic and social conditions, enhancing the understanding of the mechanisms that drive economic growth in modern cities.

3.2. Data Sources

The data employed in the analysis were derived from the Ardeco Database of the European Commission, as well as the Eurostat and the International Monetary Foundation (IMF) databases, providing city-level and NUTS2-level estimates for a range of factors affecting urban economic performance. The datasets refer to the EU members’ urban agglomerations from 2010 to 2022, aiming to assess the linkages between urban infrastructure investments, demographic changes, and economic outcomes. Initially, we assembled a panel dataset for 241 major cities in EU-27 countries from 2010 to 2022. Due to missing data and to ensure a balanced panel for the regression analysis, we reduced the year coverage of the sample to 2017–2022.

3.3. Determinants of Urban Economic Performance

Gross value added (GVA) (Figure 1) is a crucial indicator of economic performance at a spatial level, such as cities and regions. GVA measures the total contribution of various economic sectors to the gross domestic product (GDP), giving a clearer picture of economic activity, as it includes the net income from producing goods and services. In the present analysis, data on GVA at constant prices are used to eliminate the effect of inflation, allowing comparisons between different periods and cities based on real economic growth. This variable examines how infrastructure investment, demographic development, and other economic factors affect economic growth in significant EU cities.

Employees’ compensation (COMPENS) (Figure 2) is a critical labor market metric that indicates economic health and employee welfare. COMPENS at constant prices avoids distortions arising from inflation in the observed changes to employee compensation and thus provides a precise evaluation of whether employees are attaining enhancements or encountering declining earnings. This variable assesses the relationship between infrastructure investment productivity and employee compensation in EU cities.

This is pertinent for our strategic approach. The two dependent variables—GVA and COMPENS (

Table 1. Variables.

Variables	Description	Spatial Level	Source
$GVA$	GVA at constant prices	city	Ardeco Database, European Commission
$COMPENS$	Compensation of employees at constant prices	city	Ardeco Database, European Commission
$GDP$	Gross domestic product (GDP) at current market prices	city	Eurostat
$FIXEDCAP$	Consumption of fixed capital at constant prices	city	Ardeco Database, European Commission
$ROIGT$	Gross fixed capital formation at constant prices	city	Ardeco Database, European Commission
$SNPNN$	Natural balance change in the population	city	Ardeco Database, European Commission
$SNPCN$	Total population change	city	Ardeco Database, European Commission
$MIGR$	Net migration	city	Ardeco Database, European Commission
$CEDUC$	Students in higher education	city	Ardeco Database, European Commission
$CENV$	Municipal waste generated	city	Ardeco Database, European Commission
$CLIVCON$	Number of conventional dwellings	city	Ardeco Database, European Commission
$UNEMP$	Unemployment rate	city	Ardeco Database, European Commission
$PDENS$	Population density, inhabitants per km2	city	Ardeco Database, European Commission
$IAAC$	Percentage of households with access to the internet at home	NUTS-2	Eurostat
$HRST$	Persons with tertiary education (ISCED) and/or employed in science and technology	NUTS-2	Eurostat
$TRANS$	Motorways, kilometers per thousand square kilometers	NUTS-2	Eurostat
$RNLHTP$	Hours worked per capita	city	Ardeco Database, European Commission
$PORT$	Port	city	Ardeco Database, European Commission
$ENVTAX$	Environmental taxes	country	Eurostat
$GOVDEBT$	Government gross debt	country	IMF
$EXIMPORTS$	Extra-EU27 imports	country	Eurostat
$FISCALB$	Government fiscal balance	country	Eurostat

)—allow us to assess economic performance levels and employee well-being simultaneously, providing insights into the impacts of urban policies or infrastructure investments at the city level across the European Union.

The present study employs a set of macroeconomic indicators, such as the government gross debt (national level) and the gross domestic product (NUTS2 level), while also including city-level indicators, namely, the unemployment rate, the gross fixed capital formation, and, alternatively, the consumption of fixed capital, at constant prices. The methodological choice to incorporate proxies representing estimates at the NUTS2 level was deemed necessary in order to capture the links between urban economic performance and the applied national and regional policies in the economic field.

Governments’ high debt exposure hampers the investment potential of both central and regional authorities in order to finance large new infrastructure or upgrading projects through lending. This macroeconomic indicator at the national level also has implications for financing and upgrading urban infrastructure. National or regional-level economic performance should imply greater potential in the implementation of appropriate economic policies, thus exerting a positive impact on the attractiveness of urban economies. For that reason, we introduce a proxy for the gross domestic product (GDP) at current market prices (NUTS2 level) and the gross fixed capital formation and consumption of fixed capital at the city level.

Networks are fundamental to economic performance [17,18,19]. Improving the terms of interconnectivity implies more efficient channels for transferring products and human resources, but also facilitates the flow of information (“virtual accessibility”), the latter is also from an economic point of view. The literature makes a distinction between physical and virtual accessibility, where the first type refers to the tangible networks for facilitating the flow of goods and human resources, while the second type refers to the intangible networks and the role of ICT infrastructure and the internet (for example, [20,21]). For that reason, we represent the frequency of access to the internet by introducing a proxy corresponding to the share of households with access to the internet at home (NUTS2 level).

As regards the tangible networks, we further make use of a proxy representing the density of the motorway network at the NUTS2 level, measured in kilometers per thousand sq. km. The expansion of road networks minimizes the daily commuting time to work, at the same time extending the range of work opportunities for urban dwellers. New highways are seen to attract investment interest from businesses to settle along them, following the ribbon development pattern [11]. The existence of port infrastructure, represented here by the corresponding dummy, is considered key to economic performance, as it provides additional trading routes for both European countries and local industries [22,23].

Cities located in the same country with international trade hubs are expected to benefit from this indirect access to international markets. This centrality to the global trade scene is reflected here by a proxy representing the EU members’ import intensity from non-EU countries, based on the Eurostat data. In this context, we support a positive link between the cities’ localization in major non-EU product importing countries—or in other words, proximity to major trading ports—and urban economic performance.

Higher population densities are associated with a larger urban market size, with all that this entails for the size of the urban labor market, for the economies of scale that firms seek to take advantage of, for labor availability, and for the city-level consumer base. Controlling the effect of population density aims precisely to assess the impacts of the aforementioned factors on economic efficiency, and therefore we expect a positive link. The urbanization degree is linked to an increase in per capita growth indicators, such as the GDP per capita, and this finding is also confirmed in the case of African cities [24].

Urban economic growth is also fueled by population growth, as long as rural-to-urban population movements are the driving forces of the increasing labor supply and expanding domestic consumer base, thus improving the terms of attracting urban economic activities. But even though the causal relationship between urbanization and economic growth can be well documented by the relevant literature, Nguyen and Nguyen [25] identify an urbanization threshold in the case of ASEAN countries, above which urbanization may impede economic growth. In this light, we introduce two proxies for changes in the city-level population size and natural balance, thus seeking to re-examine this relationship in the context of the European cities. Finally, the latter variable also aims to capture the effects of urban attractiveness, in terms of migratory movements towards the cities, on economic outcomes, as also supported by the relevant literature [26,27,28].

The share of the university-educated population—including persons engaged in science and technology—in the total labor force is employed to capture the effect of skills development on economic efficiency, and the data are available at the NUTS-2 level. Inversely, the unemployment rate at the city level is introduced in order to assess the effect of imbalances between the labor supply and demand on economic outcomes. In addition, greater working time per worker, ceteris paribus, does not necessarily imply higher economic performance. This can be easily confirmed at the national level among the EU members to the extent that southeastern countries commonly record a greater number of working hours per employed person, according to recent data [29].

3.4. Stationarity Tests and Transformations

Econometric models analyzing time-series data need to undergo a thorough examination to ensure their accuracy and reliability. This study tested stationarity using the Augmented Dickey–Fuller (ADF) test for all time-series variables. Variables that were not stationary at their levels were transformed using first differences to achieve stationarity. Specifically, the variables consumption of fixed capital at constant prices and fiscal balance were non-stationary in their original form, and the first differencing was applied. Subsequent ADF tests confirmed that the transformed variables were stationary, thus ensuring the robustness and reliability of the model. The stationarity test results are summarized in Appendix B.

3.5. Models’ Specification

This research uses two econometric models to analyze the impacts of infrastructure investments and employees’ compensation on economic growth and employment in major European cities from 2017 to 2022. The Ordinary Least Squares (OLS) method was used to estimate the model’s parameters. Introducing proxies referring to different spatial levels, such as in the present study (country, NUTS2, and city-level data), is certainly not novel, as long as it is a methodological choice also adopted by the European Commission in order to evaluate regional competitiveness across the European Union. Specifically, the recently proposed Regional Competitive Index (RCI) is an indicator composed of the major factors of competitiveness at the NUTS-2 level [30].

As already mentioned, the model was applied to two distinct dependent variables, namely, the gross value added and the compensation of employees, which were logarithmically transformed to assess the percentage changes in economic performance. The covariates were also logarithmically transformed, while Inverse Hyperbolic Sine (IHS) transformations were incorporated for the demographic variables to address the issue of zero and negative values and differences and lag to capture the temporal relationship and the effects of previous periods and the current ones on economic performance. The variable lag(RNLHTP) (hours worked per capita) was included in the model to capture the delayed effects of labor inputs on economic output (GVA) and employee compensation. This approach aligns with the theoretical expectations that labor-related adjustments influence economic outcomes over time rather than instantaneously [31]. Alternative lag lengths (e.g., one-year and two-year lags) were tested, with the one-year lag yielding the best model performance based on the statistical criteria, including AIC and adjusted R². Robustness checks confirmed that models including lag(RNLHTP) provided superior explanatory power compared to those without lagged variables. We also ensured data stationarity by transforming economic indices into the first difference.

The two econometric models are formulated as follows:

$$\begin{array}{ll}\mathit{log}\left(GVA\right)=a_0& +{\beta }_0\mathit{log}\left(\mathit{GDP}\right)+{\beta }_1\mathit{lag}\left(\mathit{RNLHT}\right)+{\beta }_2\mathit{log}\left(\mathit{GOVDEBT}\right)+{\beta }_3\mathit{log}\left(ROIGT\right)+{\beta }_{4}log\left(UNEMP\right)\\ & +{\beta }_5\mathit{log}\left(ENVTAX\right)+{\beta }_6\mathit{log}\left(EXIMPORTS\right)+{\beta }_{7}PORT+{\beta }_{8}log\left(TRANS\right)\\ & +{\beta }_9\mathit{IHS}\left(SNPNN\right)+{\beta }_{10}\mathit{IHS}\left(SNPCN\right)+{\beta }_{11}\mathit{IHS}\left(MIGR\right)+{\beta }_{12}log\left(HRST\right)\\ & +{\beta }_{13}\mathit{log}\left(IAAC\right)+{\beta }_{14}log\left(CENV\right)+{\beta }_{15}\mathit{log}\left(CLIVCON\right)+{\beta }_{16}\mathit{log}\left(PDENS\right)+{\epsilon }_i\end{array}$$

(1)

$$\begin{array}{ll}\mathit{log}\left(COMPENS\right)& ={{a_0+\beta }_0\mathit{log}\left(\mathit{GDP}\right)+\beta }_1\mathit{lag}\left(RNLHT\right)+{\beta }_2\mathit{log}\left(GOVDEBT\right)+{\beta }_{3}DIFF\left(FIXEDCAP\right)\\ & +{\beta }_{4}DIFF\left(FISCALB\right)+{\beta }_{5}log\left(UNEMP\right)+{\beta }_{6}log\left(ENVTAX\right)+{\beta }_{7}PORT+{\beta }_{8}log\left(TRANS\right)\\ & +{\beta }_9\mathrm{IHS}\left(SNPCN\right)+{\beta }_{10}log\left(HRST\right)+{\epsilon }_i\end{array}$$

(2)

We further tested indicators of robustness and validity. The Durbin–Watson test was used to assess the residuals for autocorrelation. We evaluated Variance Inflation Factor (VIF) (Appendix C) and tolerance values to test for multicollinearity among the independent variables. To test for heteroscedasticity in the regression models, we used White’s test. A fixed-effects model was conducted to account for unobserved time-invariant heterogeneity across cities (Appendix D). The results from the fixed-effects model are consistent with those obtained using OLS.

The analysis centers on a comprehensive examination of the statistical correlations among various key variables. It aims to uncover the relationships and interactions between these variables by employing rigorous statistical methods. By analyzing the data, the study seeks to identify patterns and trends that can provide deeper insights into how these variables influence one another, ultimately contributing to a more nuanced understanding of the underlying dynamics, such as infrastructure investment and GVA. While correlations offer useful information, they do not imply causation. Moving ahead, causal inference methods (e.g., IV methods using exogenous shocks or DiD approaches) could be employed to substantiate causal assertions.

4. Empirical Results

This research provides new knowledge about how infrastructure investments interact with demographics and economic performance in metropolitan areas. The analysis examines gross value added and employee compensation in order to produce a comprehensive understanding of the forces that drive urban economic outcomes and labor dynamics.

The detailed descriptive statistics (

Table 2. Descriptive statistics.

Variables	Mean (St. Dev.)	Min	Max	CV
GDP	31.337 (14.058)	2.787	189.379	0.49
Hours worked per capita	754 (131)	455	1.405	0.17
Government gross debt	71 (29)	4	213	0.41
Gross fixed capital formation	6.868 (11,690)	214	178.692	1.70
Consumption of fixed capital at constant prices	5.452 (9.241)	166	141.599	1.69
Fiscal balance	−34.081 (57.164)	−208.236	65.623	−1.68
Unemployment rate	9 (5)	1	37	0.60
Environmental taxes	6 (2)	3	15	0.26
Imports from non-EU countries	11 (8)	0	23	0.72
Motorways	36 (32)	1	205	0.91
Natural change in the population	684 (7994)	−59.471	113.686	11.69
Total population change	4.245 (12.568)	−106.408	160.491	2.96
Net migration	3.435 (11.086)	−92.105	165.862	3.23
Share of students in higher education	111 (70)	2	465	0.63
Households with access to the internet at home	79 (17)	21	100	0.21
Municipal waste generated	429 (4.683)	20	206.339	10.92
Number of conventional dwellings	549,735 (9,455,161)	11.521	548,808,316	17.20
Persons with tertiary education (ISCED) and/or employed in science and technology	40 (10)	11	71	0.24
Population density	490 (498)	36	8.097	1.02
	Frequency
Port	Yes 26.7%

) offer an overview of the economic, social, and demographic variables included in this analysis of metropolitan cities across EU member states. This aims to identify significant trends that enhance our understanding of the diversity in these major European cities.

Both models were highly explanatory (Table 3). R² values ranged from 0.802 to 0.884 for both GVA and employee compensation. Durbin–Watson statistics reached acceptable levels, implying no major issues of autocorrelation, whereas the F-statistics confirmed the significance of the econometric models. The empirical results offer a solid and empirically grounded explanation for how infrastructure investments, labor dynamics, and demographic shifts may operate together to drive macroeconomic performance within large metropolitan areas.

4.1. Gross Value Added (GVA)

In Model 1, gross fixed capital formation significantly impacted the rise in GVA (β = 0.430; p < 0.001). Growth is driven by infrastructure investments, making this result important. Urban areas with higher gross fixed capital formation will have greater economic output, as transport, utilities, and other fixed capital investments will increase productivity and development. The natural population change also positively and significantly impacted GVA (β = 0.007, p < 0.001), indicating that cities with increasing populations through natural demographic processes, including births and exceeding deaths, tended to have higher economic performances. Nevertheless, population density had a detrimental effect on GVA financially (β = −0.103, p < 0.05), and was possibly linked to significant overcrowding and resource pressure in large cities with many migrants. In metropolitan areas, municipal waste production—a measure of material use in economies—is positively and significantly associated with economic activity/consumption. The number of conventional houses also significantly contributes to GVA (β = 0.174, p < 0.05), emphasizing the roles of urbanization and a higher housing demand in economic growth. More housing brings construction jobs and real estate work that spur economic activity. Finally, persons with tertiary education and/or employed in science and technology highly impact GVA (β = 1.490, p < 0.001). This shows how much a college education and job in a science- or tech-related field affect economic progress. High levels of skilled cities can have high investment simultaneously in innovation and increased productivity and contribute much more to increase the gross value added.

The results from Model 2 confirm the relevance of infrastructural investment toward the gross value added (GVA) determination since the gross fixed capital formation measure is found to be a significant determinant of economic growth at the metropolitan level (β = 0.412; p < 0.001). The coefficient of infrastructure investment, with a positive and statistically significant sign, indicates that when a city’s fixed assets (e.g., transport networks or public services) are developed, they can increase its production capacity. The significance of expanding or updating urban infrastructure becomes even more evident throughout the EU urban agglomerations, given the stringent regulatory framework often imposed by the urban planning authorities but also the dense urban tissue itself, as a legacy of the centuries-old history of several European metropolises.

The most intriguing finding was the detrimental impact of environmental taxes (β = −0.121; p-value < 0.001), which undoubtedly slow economic growth by increasing business costs in operations. While these taxes play a role in environmental policies that aim to achieve sustainability, they seemingly hurt cities’ productivity and economic performance in the short run. This exactly reflects a direct economic cost that urban economic actors have to bear in order to achieve long-term economic efficiency and sustainability. Moreover, imports from non-EU countries had a positive and statistically significant effect on GVA (β = 0.111, p < 0.05), indicating that external trade connections allowing for markets of various goods to interconnect are beneficial in terms of the megacities’ development. Apart from the linkages between the urban market size and the import intensity, local industries in the greater urban agglomerations are also in need of importing raw or intermediate products in order to expand production activities, a reasoning that justifies the empirical results.

Both the port and motorway variables contain significant knowledge, enabling an understanding of factors influencing economic performance. The presence of a port infrastructure impacts GVA (β = 0.189, p < 0.05), which means that the port infrastructure can support urban economic activities in a city area. Cities invest in their port facilities to enhance international trade, allowing them to accumulate wealth more rapidly. Motorways also show a positive and significant effect (β = 0.235, p < 0.05), indicating that a well-developed motorway network enhances the economic performance of cities by improving connectivity with other urban areas in a short period. Total population change is negatively and significantly related to gross value added (GVA) (β = −0.017, p < 0.05). This relationship indicates that rapid and unplanned population growth can hinder the economic potential if the local infrastructure and public services are insufficient to meet the increased demand from incoming migrants.

4.2. Compensation of Employees

In the analysis of compensation of employees, the first model’s results are consistent with a significant positive relationship between lagged per capita hours worked and employee compensation (β = 0.027; p < 0.001). The labor inputs of previous periods are essential in explaining current wage levels since past productivity is driving wage growth. Conversely, unemployment reduces compensation, possibly because a high unemployment rate will tend to depress wages since there is an oversupply of workers (β = −1.039, p < 0.01). Interestingly, shifts in fiscal balance are associated with modest yet positive changes in wages (β = 0.001, p < 0.05). This suggests that even small adjustments to a city’s financial situation can lead to favorable outcomes in salaries. These improvements gain more significance when it comes to recovering from global crises, such as the conjuncture of the pandemic crisis.

The economic output, represented here by the GDP proxy, exerts a positive and significant effect (β = 14.329, p < 0.01), which means that the increase in the gross domestic product is associated with an evolution of workers’ wages. As the GDP grows, it results in greater productivity and demand by businesses, which usually raises wages across the board, with most workers earning more money. In other words, the higher the levels of available capital per worker—which can be translated into the renewal of production equipment—the more effective the technology integration into production processes, which leads to more sophisticated jobs and, lately, to higher wages.

Conversely, environmental taxes also hurt employee compensation (β = −0.092, p < 0.01), most likely capturing the higher costs to businesses following their imposition that should delay any potential wage hikes. This is consistent with the main result from GVA models that environmental taxes depress productivity, at least during the early beginning of the energy transition. A business may face an environmental tax for a failure to meet a government-set standard, which has the effect of passing off some or all pollution costs to its employees. As expected, the motorway variable was significant and positive (β = 0.493, p < 0.01), indicating a relationship between an efficient road network linked across distances and worker compensation. High-quality transit systems facilitate smooth travel for employees. By enhancing accessibility in various areas, they can foster economic development and boost business competitiveness. This factor, in turn, results in a better working environment for businesses, enabling them to pay higher wages.

Model 2 for the compensation of employees shows a combination of economic variables that influence wages in large cities, such as unemployment and infrastructure, among other demographic movements. The second model found a similar significance of lagged hours worked per capita (β = 0.022, p < 0.001), further confirming that labor productivity is a crucial determinant of compensation levels. In contrast, the model also indicated a substantial negative correlation between government gross debt and employee compensation (β = −0.001, p < 0.01). This analysis could reveal how rising public debt affects governments’ fiscal space for wage growth, potentially impacting both government employment and private sector wages. The unemployment rate significantly affected employee compensation (β = −0.705; 95% CI: −1.039 to −0.371), indicating that higher rates are associated with lower household income levels. This necessarily reduces wages elsewhere or spoils their growth prospects; in economic terms, this means wage depression. Unemployment is likely a proxy for more significant regional economic uncertainty, which implies that individuals earn less.

Port infrastructure has a statistically significant, adverse effect on the compensation of employees (β = −0.242, p < 0.001). While the existence of ports finally leads to higher GVA levels, which is a measure of economic growth, it also seems that their presence results in workers receiving lower pay, possibly because more low-paid jobs are concentrated in some port-related sectors. This could stem from a higher concentration of low human capital jobs typically paying lower wages. In contrast, the motorway variable had a positive and highly significant influence on the compensation of employees (β = 0.194; p < 0.001). This suggests that road infrastructure investment eventually leads to higher wages for employees. Better road links reduce transport costs, generating a significant demand for better connectivity and making businesses more productive. This, in turn, helps generate higher income due to high corporate profits and a lower commuting time by the workforce. The negative coefficient for total population change (β = −0.175, p < 0.05) suggests that change in the total population size of a county has an inverse relationship with employee earnings.

Over time, the combined impacts of changes in population through natural growth and migration can put pressure on infrastructure and services, affecting what remains for wages. Cities with significant population increases will likely incur greater costs in adjusting infrastructure and social services, potentially leaving less economic resources available for wage hikes. The share of students in higher education also significantly negatively impacts the compensation of employees (β = −0.139, p < 0.001). While higher education is critical for long-term knowledge and innovation, this result implies that cities with a more significant share of students may also have more young people who have not yet wholly entered the labor market, and so future skill absorption drives down average wages in those locations. Additionally, students typically hold lower-paying part-time jobs that can impact overall earnings.

Table 3. OLS models.

Dependent Variables	Gross Value Added		Compensation of Employees
Covariates\Models	Model 1	Model 2	Model 1	Model 2
(Constant)	−1.153 (−0.956)	1.152 (0.408)	7.159 (1.773) *	4.342 (19.161)
GDP	0.098 (0.178)		14.329 (2.580) **	12.250 (2.994) ***
Hours worked per capita (LAG)			0.027 (4.858) ***
Government gross debt		0.000 (−0.221)	−0.000 (−0.440)	−0.001 (−2.327) ***
Gross fixed capital formation	0.430 (6.057) ***	0.412 (4.067) ***
Consumption of fixed capital at constant prices (DIFF)			−2.492 (−0.845)	−1.221 (−0.576)
Fiscal balance (DIFF)			0.001 (0.057) *	0.000 (0.116)
Unemployment rate	0.111 (1.291)	−0.114 (0.664)	−1.039 (−2.815) **	−0.705 (−4.478) ***
Environmental taxes		−0.121(5.223) ***	−0.092 (−2.206) **
Imports from non-EU countries		0.111 (2.297) **
Port		0.189 (2.087) **	−0.075 (−0.671)	−0.242 (−3.859) ***
Motorways	0.008 (0.235)	0.235 (1.988) *	0.493 (2.510) **	0.194 (2.410) ***
Natural change in the population	0.007 (3.332) ***	0.015 (2.191) **	−0.137 (−1.279)	−0.175 (−2.294) ***
Total population change		−0.017 (−1.976) *
Net migration	−0.001 (−0.072)	0.019 (2.230) **	0.028 (4.858) ***	0.022 (5.136) ***
Share of students in higher education	−0.006 (−0.608)	0.067 (2.363) **		−0.139 (−4.977) ***
Households with access to the internet at home	0.446 (0.791)	0.942 (0.645)
Municipal waste generated	0.176 (4.719) ***
Number of conventional dwellings	0.174 (2.086) **
Persons with tertiary education (ISCED) and/or employed in science and technology	1.490 (6.467) ***
Population density	−0.103 (−2.536) **
R2	0.865	0.868	0.802	0.884
Baltagi and Wu LBI Statistic	1.75	1.60	1.78	1.85
Durbin–Watson	1.782	1.515	1.818	2.058
F	84.293	22.416	12.398	25.317

* for p < 0.10, ** for p < 0.05, and *** for p < 0.01; t-statistics are shown in parentheses.

Table 3. OLS models.

Dependent Variables	Gross Value Added		Compensation of Employees
Covariates\Models	Model 1	Model 2	Model 1	Model 2
(Constant)	−1.153 (−0.956)	1.152 (0.408)	7.159 (1.773) *	4.342 (19.161)
GDP	0.098 (0.178)		14.329 (2.580) **	12.250 (2.994) ***
Hours worked per capita (LAG)			0.027 (4.858) ***
Government gross debt		0.000 (−0.221)	−0.000 (−0.440)	−0.001 (−2.327) ***
Gross fixed capital formation	0.430 (6.057) ***	0.412 (4.067) ***
Consumption of fixed capital at constant prices (DIFF)			−2.492 (−0.845)	−1.221 (−0.576)
Fiscal balance (DIFF)			0.001 (0.057) *	0.000 (0.116)
Unemployment rate	0.111 (1.291)	−0.114 (0.664)	−1.039 (−2.815) **	−0.705 (−4.478) ***
Environmental taxes		−0.121(5.223) ***	−0.092 (−2.206) **
Imports from non-EU countries		0.111 (2.297) **
Port		0.189 (2.087) **	−0.075 (−0.671)	−0.242 (−3.859) ***
Motorways	0.008 (0.235)	0.235 (1.988) *	0.493 (2.510) **	0.194 (2.410) ***
Natural change in the population	0.007 (3.332) ***	0.015 (2.191) **	−0.137 (−1.279)	−0.175 (−2.294) ***
Total population change		−0.017 (−1.976) *
Net migration	−0.001 (−0.072)	0.019 (2.230) **	0.028 (4.858) ***	0.022 (5.136) ***
Share of students in higher education	−0.006 (−0.608)	0.067 (2.363) **		−0.139 (−4.977) ***
Households with access to the internet at home	0.446 (0.791)	0.942 (0.645)
Municipal waste generated	0.176 (4.719) ***
Number of conventional dwellings	0.174 (2.086) **
Persons with tertiary education (ISCED) and/or employed in science and technology	1.490 (6.467) ***
Population density	−0.103 (−2.536) **
R2	0.865	0.868	0.802	0.884
Baltagi and Wu LBI Statistic	1.75	1.60	1.78	1.85
Durbin–Watson	1.782	1.515	1.818	2.058
F	84.293	22.416	12.398	25.317

* for p < 0.10, ** for p < 0.05, and *** for p < 0.01; t-statistics are shown in parentheses.

The analysis was based on four distinct econometric models, namely, two models for capturing the determinants of gross value added (GVA) and the other two for assessing the drivers of employees’ compensation levels. These models provide valuable insights into the factors influencing economic performance and employee earnings in major European metropolises. Each econometric model incorporates different variables, highlighting the complex and multidimensional nature of factors affecting economic growth and the labor market.

In the two models for gross value added (GVA), gross fixed capital formation emerges as one of the most critical drivers of economic performance, demonstrating the importance of infrastructure in improving cities’ productivity and economic performance. Additionally, demographic variables such as natural population growth and population changes highlight the effect of population trends on economic growth, with metropolitan areas experiencing natural population growth performing better in economic terms. The negative effect of population density is quite intriguing, as it can be related to overcrowding problems and increased pressure on the urban infrastructure. At the same time, trade, in the form of imports from non-EU countries, also proves to be a positive factor for boosting GVA, demonstrating the role of external trade flows in improving the competitiveness and productivity of cities.

The two models for compensation of employees, as a dependent variable, show a difference in the independent variables employed, providing a more specialized picture of the factors influencing remuneration. In the first model, GDP and hours worked per capita from previous periods are essential in explaining wages. At the same time, variables such as unemployment and fiscal balances affect workers’ pay negatively or positively, respectively. In the second model, environmental taxation exerts a negative effect, demonstrating that additional environmental costs burden firms and limit the potential for wage increases. At the same time, transportation infrastructure, such as highways, positively affects wages and improves accessibility and productivity. In contrast, the port variable affects wages negatively, possibly due to the concentration of low-skilled jobs.

The models highlight the fundamental importance of infrastructure investment in improving economic performance and employee compensation. At the same time, factors such as demographic changes, unemployment, environmental taxation, and trade flows play critical roles in shaping the urban economy, demonstrating that economic growth and social well-being in large cities depend on complex policies and interventions at multiple levels.

5. Discussion and Implications

The results of this analysis emphasize the importance of infrastructure investment, especially gross fixed capital formation, in driving urban gross value added and employee compensation. Gross fixed capital formation also consistently appears as a significant positive factor in all models, suggesting that general investments, such as transportation and utility infrastructure, have broad impacts on urban economic performance. The reports of these findings should be comprehended as authenticating relationships, not causations.

We find complicated interaction effects from the GVA model on infrastructure investments, demographic characteristics, and economic growth in metropolitan areas. The strongly positive relationship between gross fixed capital formation and gross value added supports the view that urban productivity thrives with infrastructure investment. Infrastructure lays the groundwork for economic activities by supporting natural networks [32]; more efficient transportation and other utilities promote production and distribution, leading to greater urban output. Evidence from Western Europe has demonstrated the robustness of the relationship between construction flows and economic growth [33]. The infrastructure in European cities is becoming outdated, undermining productivity and competitiveness [34]. These results underscore the critical role of public infrastructure investment in enhancing regional economic performance and reducing spatial inequalities, as has been demonstrated in prior research on the spatial and intertemporal effects of public expenditure [35].

The positive effect of natural population changes on GVA indicates that demographic growth nurtures the demand for goods and services, but also the labor market, where companies find the pace to grow their economic activities. This result aligns with findings that emphasized that natural demographic growth enhances urban output by increasing consumption and growing the labor supply [27]. In contrast, total population change (including migration) negatively affects GVA. This negative impact is probably due to the stress of large and unplanned migration on urban resources and facilities. Uncontrolled migration can strain urban systems excessively [36]. The findings suggest that cities aiming to promote natural growth instead of relying solely on migration should develop policies that ensure an adequate infrastructure to accommodate varying population pressures.

The model also reveals a negative relationship between the population density and GVA, suggesting that dense urban environments may suffer from overcrowding and the super-consumption of finite resources. Except for economies of scale in population density, many urban areas that become too dense will need more saturation points of infrastructure and experience declining economic efficiency. This result is consistent with reports that population density negatively impacts economic growth if it is over some critical value, as the scarcity of infrastructure leads to more severe congestion and more significant depreciation [37]. These findings highlight a crucial consideration for planners aiming to balance urban density with the capacity of surrounding systems, thus promoting efficiency and avoiding excess failure-prone nodes that hinder agglomeration and economic output in cities.

There are also strong negative correlations between environmental taxes and both GVA and employee compensation, indicating that while these taxes are vital for sustainable development, they may inflict economic costs in the short run by raising firms’ operating expenses. This is consistent with Siedentop and Fina [38], who argued that environmental regulations could reduce short-term profit-enhancing efforts by increasing costs in urban economies [39]. Policies aligned with halting or even slowing down the ongoing climate crisis translate into direct or indirect economic costs, which national and local authorities, as well as businesses, are called upon to bear. Findings on the negative relationship between environmental tax levels and economic performance are expected to diverge as economic actors and governments adopt common strategies for the green transition. In the meantime, this means that the green transition may imply eventual policy trade-offs between short-term economic growth and sustainability in the long run.

Imports from outside the EU’s impositions have an effective influence, pointing to the significance of international trade as an element in urban economic success. Urban areas, being centers for trade and commerce, benefit from the availability of goods from abroad, which provide supply chains that offer businesses and consumers a wider variety of available goods to complement local economic endeavors. Consistent with this, Garcia [40] argue that cities with solid trade linkages mark increased economic dynamism.

City performance is closely tied to the port and motorway infrastructure. We find a positive relationship between the likelihood of having port facilities and GVA, implying that maritime connectivity plays a vital role in urban economic activities through trade and logistics. Motorways, similarly, have positive effects on GVA and employee compensation, which are consistent with the literature, demonstrating that solid infrastructure networks facilitate regional economic and productivity integration [41].

If human capital drives economic performance, tertiary education and highly skilled labor will significantly and positively contribute to GVA. Human capital, or the attribute of well-trained workers that stimulates innovation and productivity, forms the basis of a knowledge-driven economy, which is critical for urban success. “Well-educated” cities are more likely to draw high-value industries and benefit from long-term economic development [42].

The coincidence of GVA with municipal waste generation and its positive association with rich feedback may reflect higher consumption and production activity occurring more broadly, involving the health of the urban area. The study correlates with the trends reported by the World Bank [11], which state that material consumption increases in a similar ratio in urban areas along with economic development. More conventional residences positively impacted GVA, indicating that both urbanization and housing spurred development activity with an emphasis on construction and real estate. This highlights how housing policies that accommodate (rather than bet against) population growth create the conditions for more economic activity associated with urban intensification.

One of the main limitations of this study is the focus on a snapshot of data (covering the timeline of 2017–2022). As a result, the findings do not capture long-term structural and cyclical economic trends. This fact provides further scope for limitations, acknowledging that significant economic disruptions occurred during the COVID-19 years. Such disruptions have a transitory impact on economic performance, possibly obscuring the longer-term effects of infrastructure investment and demographic trends. Despite the findings on the interactions between infrastructure, demographics, and economic outcomes in European metropolitan areas, the results should be taken with caution, bearing in mind the specificities of the pandemic period. An analysis extending beyond 2022 would be helpful for this type of conclusion to capture broader economic trends and reduce the influence of shocks, capturing more of these dynamics.

Although the analysis was performed at the city level, we comprehend the possibility of spatial dependence and spillover among neighboring cities. Future research on spatial econometric models shall effectively capture the spatial dynamics of the data.

6. Conclusions

This study offers valuable insights into the interconnections between infrastructure investments, metropolitan demographics, and the economy in Europe. It empirically demonstrates infrastructure’s significant roles in urban economic growth, employee compensation, and gross fixed capital formation. The consistently strong positive correlation between infrastructure investment and gross value added (GVA) indicates that cities with high levels of investment in transportation, utilities, and other forms of fixed capital tend to experience greater productivity and economic output.

Natural population growth and migration flows are mixed, as the analysis uncovers apparent trade-offs between gains from higher shares of older or younger populations. Natural population change has a significant positive impact on gross value added (GVA). Conversely, unplanned or rapid population growth, primarily due to migration, can lead to the overstrain of resources, potentially hindering urban growth. To this extent, urban planning must be conducted carefully, considering natural demographic dynamics and economic objectives.

Environmental taxes are increasingly viewed as essential for sustainability; however, our findings indicate they may negatively impact both economic output and wages in the short term. The undesirable impacts of unemployment and environmental taxes are vital indicators showing why a balance between development or environment-friendly policies is essential to achieving long-term urban economic success.

Overall, the results provide insights into how infrastructural growth and regional restructuring, demographic trends, and labor market dynamics influence economic development at diverse scales in European agglomerations. The results should provide useful lessons for urban policymakers, showing that every city needs to be strong economically but also socially inclusive. This spatial, place-based evidence is essential for policy design, helping metropolitan areas ensure their future prosperity in light of demographic and economic changes.