Innovative SMEs in Italy. Explaining profitability patterns in inner areas

Abstract

Purpose

This paper investigates whether and to what extent operating in inner areas affects the profitability of innovative Italian small and medium-sized enterprises (SMEs) over 2012–2018.

Design/methodology/approach

Guided by the National Strategy for Inner Areas and the “Investment Compact,” this study distinguishes between inner and core innovative SMEs. It employs various econometric models to estimate a regression for the return on assets of SMEs, differentiating between firms operating in inner and non-inner areas of northwest, northeast, centre and south Italy.

Findings

Findings reveal that innovative SMEs in inner areas generally exhibit lower profitability compared to those in non-inner municipalities. However, huge heterogeneity in results is observed across the country. Specifically, innovative SMEs in the inner areas of the south register lower profitability than those operating in non-inner zones. Conversely, innovative SMEs located in the inner municipalities of northwest and northeast Italy show higher profitability than their peers in non-inner areas. The results imply that targeted policies for inner areas are crucial. However, due to the diversity of local impacts, a differentiated approach, depending on the geographic context, is necessary.

Originality/value

The study aims to explore the relationship between inner areas and the performance of innovative SMEs in Italy. More precisely, it examines the effect of operating in a municipality located within an inner area on the profitability of innovative SMEs. This issue has been overlooked in existing literature. Importantly, we aim to determine whether there is a heterogeneous impact based on geographical localisation, specifically in the Northwest, the Northeast, the Centre and the South of the country. Therefore, this paper contributes to the literature by investigating the factors influencing the performance of innovative SMEs and suggesting new policy recommendations for developing inner areas in Italy.

Citation

Aiello, F., Errico, L. and Rondinella, S. (2024), "Innovative SMEs in Italy. Explaining profitability patterns in inner areas", Journal of Economic Studies, Vol. 51 No. 9, pp. 306-322. https://doi.org/10.1108/JES-02-2024-0094

1. Introduction

For decades, studies focusing on the Italian growth model have investigated the gap between northern and southern regions. In more recent years, as a result of the pronounced spatial inequalities in economic prosperity and welfare spreading through several economies (Martin et al., 2021), researchers have also devoted attention to the core-peripheral divide across the country (Bonanno et al., 2022; Viesti, 2020; Vendemmia et al., 2021; Mariotti et al., 2023; Thomä, 2023). One advantage of disaggregating the analysis is the ability to verify whether the localisation in urban versus non-urbanized areas is a matter for consideration. This new approach moves beyond the North-South dualism, emphasizing the importance of business location decisions, as the diverse challenges faced by firms have varying impacts on firm performance (Lever, 2002; Farja et al., 2017). This study therefore aims to investigate whether the profitability of innovative Italian firms is influenced by their operation within the contrasting contexts of inner and non-inner areas, while also examining how this localisation effect varies across different regions of Italy.

The notion of “inner areas” was formally introduced in 2012 with the launch of the Italian Government’s National Strategy for Inner Areas (NSIA), aimed at addressing issues such as depopulation and inadequate access to services in these territories (Barca et al., 2014). Serving as a platform for integrated local development and institutional advancements (Basile and Cavallo, 2020; Zolin et al., 2020), the NSIA responds to challenges unique to non-urban zones, which constitute approximately 60% of Italy’s total land area. This phenomenon also affects the European Union (EU) more broadly, where urban areas encompass roughly 25% of the total inhabited territory (Perpiña Castillo et al., 2018) and are inhabited by 28% of Europeans (Augère-Granier, 2018) [1]. As such, sustainable strategies are needed to address the specific traits of these areas, which exhibit environmental, economic, cultural, and demographic differences (Zolin et al., 2020). Similar to those of the EU, development strategies for peripheral regions in Italy prioritize social, political, and cultural needs, aiming to counteract depopulation and marginalization by enhancing essential services and stimulating local growth initiatives (Basile and Cavallo, 2020).

Under the Italian initiative NSIA, inner areas are defined as parts of the Italian territory significantly distant from essential services such as education, health, and rail services, despite being endowed with environmental and cultural resources (Barca et al., 2014). The primary criterion for identifying these territories as inner areas is the distance, measured in terms of travel time, from a given municipality to the nearest agglomeration offering services of general interest (Copus et al., 2017). According to NSIA criteria, there is some overlap between inner and rural areas [2]. Consequently, discussions about these areas often focus on their vulnerabilities and opportunities, aiming to highlight the potential of inner areas. Unfortunately, this potential often remains untapped, leading to an underdeveloped socio-economic fabric across the country (De Renzis et al., 2022).

The literature extensively investigates inner areas, analysing their strengths and weaknesses. Compagnucci and Gabriele (2021) propose contextualizing territorial capital to combat abandonment in inner areas. Ottomano Palmisano et al. (2022) assess policy actions to enhance recovery in Puglia’s inner areas. De Renzis et al. (2022) examine demographic growth in remote municipalities, while Sonzogno et al. (2022) analyse young people’s migration tendencies, shedding light on motivations and challenges in peripheral areas. Bonanno et al. (2022) explore income inequality in Italian municipalities, noting disparities in inner and ultra-peripheral regions.

While previous research on inner areas has provided valuable insights, a distinct contribution to the debate arises from studies on firm location, which shed light on the expectations regarding the link between firm performance and localisation in inner areas. These studies have extensively examined the structural factors impacting a firm’s localisation decisions, including logistics, traditional location determinants, environmental considerations, social factors, and institutional contexts (Noonan et al., 2020; Arauzo-Carod, 2021). Notably, the innovative potential and entrepreneurial vibrancy of an area act as attractors for new R&D firms (Autant-Bernard, 2006; de Felice, 2014). By integrating insights from inner area studies and firm location literature, a comprehensive picture emerges, highlighting the multifaceted determinants that shape the dynamics between inner areas and economic activities. This synthesis contributes to a nuanced perspective, emphasising the potential for strategic interventions to leverage the unique attributes of inner areas and foster sustainable economic development.

These considerations provide a suitable foundation for investigating the impact on performance of a firm’s location within an inner area. We do so by drawing insights from studies focusing on firms operating in rural zones, given the strong correlation between inner and rural areas (Dezio et al., 2021). Much like urban vs rural firms, enterprises active in non-inner vs inner areas face different challenges. On the one hand, firms in core centres grappling with diseconomies of scale and agglomeration cope with a lack of space for conducting activities or expanding business, higher rents and wages, and intensified local competition. On the other, firms in inner areas confront challenges related to geographic remoteness from markets, difficulties in labour recruitment, and limited access to professional and technical information networks (Curran and Storey, 1993). As a result, a firm’s performance profile may be directly influenced by competition, space limitations, and proximity to markets. Urban firms face higher costs due to intense competition for factors such as land and labour, potentially making non-urbanized firms more profitable. However, proximity to markets benefits urban firms, allowing them to adapt to market changes and improve performance (Glancey, 1998). It is worth mentioning that remoteness and rurality do not always represent an obstacle to performance, as evidenced by various studies (Álvarez-Coque et al., 2012; Zouaghi et al., 2017).

In addition to examining the relationship between peripheral areas and firm features and performance, many scholars also focus on small and medium-sized enterprises (SMEs). The latter represent approximately 99% of total enterprises in Italy, contributing around 65% of overall value-added and employing roughly 70% of the national workforce (ISTAT, 2023). Due to their deep integration into local communities (Agostino et al., 2022), SMEs are particularly affected by contextual factors, including their location in inner areas. This renders the aforementioned arguments on challenges facing remote firms particularly critical (e.g. Margarian, 2021; Thomä, 2023). Despite the increase in empirical literature on Italian SMEs in recent decades, broader understanding of innovative SME performance remains limited, especially when considering the spatial localisation in peripheral territorial contexts [3].

Based on these considerations, this study aims to explore the relationship between localisation in an inner area and the performance of innovative SMEs in Italy. More precisely, we examine how operating in a municipality within an inner area affects the profitability of innovative SMEs [4]. To the best of our knowledge, this aspect has been overlooked in the existing literature. Moreover, our study aims to identify any heterogeneous impacts contingent upon geographical localisation. Consequently, this paper adds value to the literature by scrutinizing the determinants of innovative SME performance and proposing novel policy recommendations to revitalize inner areas in Italy. Our analysis connects studies on inner areas with research on firm location, providing a more comprehensive view of the relationship between inner areas, firm operations, and business performance. The focus on innovative SMEs can help policymakers identify new opportunities and interventions that leverage geographical location to promote innovation, which is widely recognized as vital for national economic growth (Bae and Yoo, 2015; Santacreu, 2015).

We perform our analysis by using data from the sample of firms registered as innovative SMEs in 2018 by the Italian Chamber of Commerce, spanning the period from 2012 to 2018. To identify innovative SMEs, the study leverages the so-called “Investment Compact” of 2015. Expanding the interventions introduced for innovative start-ups with the Italian Startup Act of 2012, the 2015 law extends benefits to a broader spectrum of SMEs with a significant innovation component, explicitly defining a distinct entity (e.g. innovative SME) eligible for specific advantages. However, unlike start-ups, innovative SMEs benefit from support without time limits, provided they annually update their information and certify adherence to the law’s requirements. The implication is that, despite the potential for increased competitiveness, profitability and export (Nadotti, 2014), innovative SMEs may face challenges similar to those encountered by firms trying to finance innovation for the first time (Aiello et al., 2020). In other words, sustaining high-standard operational activities remains crucial for innovative SMEs to enhance their chance of securing credit.

The main findings suggest that the impact on profitability of SMEs active in inner areas varies across the country, with distinct patterns emerging in the northern and southern regions. Specifically, innovative SMEs operating in the inner areas of the Mezzogiorno (southern Italy) experience a downturn in their financial performance compared to those in non-inner zones. In contrast, innovative SMEs located in the inner zones of northwest and northeast Italy exhibit higher profitability compared to counterparts in non-inner municipalities. Therefore, this evidence highlights a significant and contrasting correlation between inner areas and the performance of innovative SMEs in northern vs southern Italy. Thus, beyond Italy’s well-known North-South divide, a distinct gap emerges between inner and non-inner areas in terms of the profitability of innovative SMEs, yielding more significant challenges for those located in more marginalised regions of the country.

The remainder of this work is organised as follows. Section 2 describes the empirical setting. Section 3 presents the data used in the analysis. Section 4 focuses on the results. Section 5 concludes the study.

2. Empirical setting

As in other studies (e.g. Alarussi and Alhaderi, 2018; Fasano et al., 2022; Domma and Errico, 2023), here we adopt an accounting measure of profitability to evaluate firm performance. The estimated equation is specified as follows:

(1)Profitabilityit=β0+β1Inner Aream+β2Northwesti+β3Northeasti+β4Centrei+β5Inner Aream*Northwesti+β6Inner Aream*Northeasti+β7Inner aream*Centrei+ϕXit+φZmt+∑tφtYEARt+∑pγpPROVp+∑sγsINDs+εit

where the dependent in this analysis is the Profitability of the i-th innovative SME at time t. It is expressed as the Return on Assets (ROA), computed using profit/loss before taxes. [5] Focusing on the right side of equation (1), the key explanatory variable is the Inner Area, which takes the value of one if firms operate in a municipality (m) located within an inner area. Northwest, Northeast, and Centre are binary variables equal to one if the i-th firm is located, respectively, in north-western, north-eastern, and central regions (with Mezzogiorno being the base group) [6]. Finally, equation (1) is augmented by three product terms resulting from the interaction of the Inner Area dummy variable with Northwest, Northeast, and Centre. Therefore, the full model specification aims to examine whether firm profitability varies not only between geographical regions but also within them, specifically assessing whether firms in inner areas perform differently than core firms. The vector X comprises a set of micro-level variables typically used to account for internal determinants, being factors influenced by management decisions. The vector Z includes controlling variables to account for demographic and economic conditions at the municipal level [7]. Finally, the regressions also include dummy variables to control for time, space, and sectorial fixed effects (YEAR, PROV, and IND). εit is the error term.

To exploit the panel data structure of the dataset, we first estimate Equation (1) using a Random Effect estimator, which controls for unobserved time-invariant heterogeneity at the firm level. Additionally, since unobservable features such as cultural and historical determinants could affect the context and performance of a given firm, the application of Lewbel (2012) IV approach will address concerns about omitted variables [8].

3. Data and descriptive statistics

We drew our sample of firms from the Italian Chamber of Commerce, which offers a complete list of innovative SMEs registered in a designated section of the Business Register and active in Italy in 2018 [9]. Utilising this sample, we then retrieved balance sheet data for the 2012–2018 timeframe from the ORBIS database maintained by Bureau van Dick. This dataset provides detailed information for measuring profitability and various other firm characteristics [10]. Information on the Italian National Strategy for Inner Areas classification comes from the National Agency for Territorial Cohesion website [11]. Finally, data on municipal demographic characteristics employed as control variables are sourced from the Italian National Institute of Statistics (ISTAT) [12]. The fiscal declarations, crucial for computing the Gini Index, are drawn from the Ministry of Economy and Finance [13]. Lastly, the Civil Protection Department provides information on the municipal seismic risk class [14].

Table 1 provides a detailed description and main summary statistics of the variables used in the econometric analysis. Tables D1 and D2 in Online Appendix D present a comparative analysis of socioeconomic characteristics' mean values across different Italian macro-regions, distinguishing inner and non-inner areas, and show all pairwise correlation coefficients of variables used in the benchmark model, respectively.

Figure 1 provides an overview of the spatial distribution of inner areas (Panel A) and Innovative SMEs (Panel B) in 2018. Inner areas are mostly present in the Centre and Mezzogiorno, while in northern Italy, municipalities are mainly classified as non-inner areas. It also emerges that the non-inner areas of northern Italy exhibit a higher concentration of Innovative SMEs than other regions. In particular, Milan hosts more than 100 Innovative SMEs, followed by Rome and Turin, each with a range of 50–100 firms. Instead, most municipalities within the Centre and the Mezzogiorno host a total number of innovative SMEs falling within the 0–5 range (depicted in yellow), except for Naples, Bari, and Catania (depicted in green and orange).

Table 2 outlines the distribution of Innovative SMEs by NSIA classification and geographical regions. In 2018, Innovative SMEs were predominantly situated in non-inner areas across all macro-regions, indicating a concentration of innovative activity outside urban cores. Northwest stands out with the highest number and percentage of these enterprises, emphasizing its role as an innovation hub in Italy. Mezzogiorno, though hosting fewer Innovative SMEs overall, shows a higher percentage in inner areas, suggesting efforts to stimulate innovation in less developed regions. Specifically, 71 Innovative SMEs operate in inner areas, representing about 8% of the total, while 847 firms (92%) were located in non-inner zones. In Northwest, non-inner areas significantly outnumber inner areas (97.83 vs 2.17%). Similarly, the Northeast and Centre regions exhibit higher concentrations of Innovative SMEs in non-inner areas compared to inner areas. Mezzogiorno displays a notable proportion (14.97%) of Innovative SMEs in inner areas, though non-inner areas still comprise the majority (85.03%). This analysis highlights the divergence in innovative activity distribution between inner and non-inner areas within each region [15].

4. Econometric results

Table 3 shows the estimates for the period 2012–2018 using ROA (based on profit/loss before tax), as the dependent variable. Columns 1–2 refer to the results obtained from the Random Effect model, while Columns 3–4 display those from Lewbel IV. The odd columns report the results without interactions. These interactions are then added to the models displayed in the even columns. Finally, to corroborate the results, we also consider the sub-sample of innovative SMEs involved in high-tech sectors (Columns 5–6) [16] and perform a robustness check by including additional explanatory variables (Columns 7–8).

Some model diagnostics are presented. The assumption of heteroscedasticity in the error term is tested to justify the use of the Lewbel approach. The heteroskedasticity test statistics reject the null hypothesis of homoscedasticity (Columns 1 and 2 at the bottom of Table 3) [17]. Additionally, the Hansen J statistic for the instrument’s validity is satisfied (Columns 3 and 4 at the bottom of Table 3). It is important to note that the endogeneity test on the key variable Inner Area is not statistically significant (Columns 3 and 4 at the bottom of Table 3), indicating that the null hypothesis of exogeneity of the instrumented variable cannot be rejected. This suggests that the Random Effect model is reliable.

Before delving into the results of inner areas, it is useful to discuss the outcomes concerning other explanatory variables included in the model. From Table 3 (Column 2) it emerges that past profitability is positively correlated with current profitability, suggesting improved access to resources like liquidity, customer relations, and market share. (e.g. Goddard and Wilson, 1999). The profitability of Innovative SMEs tends to increase with age. The profitability of these firms rises with maturity, as older firms tend to attract investors and allocate resources to R&D, which boosts their competitiveness and profitability (Samosir, 2018). Similarly, working capital and firm growth have a positive influence on the profitability of Innovative SMEs. Firms with additional liquidity are better positioned to invest in R&D and embrace new technologies, thereby enhancing their financial performance. In addition, maintaining high liquidity levels enables SMEs to mitigate economic fluctuations, reduce default risk on short-term debts, and capitalize on profitable investment opportunities (Yazdanfar, 2013). Conversely, the level of leverage harms profitability, consistent with prior studies demonstrating an inverse correlation between these micro-level variables and firm profitability (e.g. Jensen and Murphy, 1990). SMEs with high levels of leverage encounter difficulties in accessing funds, especially during economic downturns, which can impede their ability to make profitable investments (Serrasqueiro and Maçãs Nunes, 2008).

Moreover, the estimates associated with municipal socio-economic controls are worth exploring. A rise in population density, often seen as a potential source of efficiency gains and knowledge exchange, here is linked to a decrease in firm profitability. This suggests that densely populated areas involve increased competition and higher operational costs, leading to lower profitability (Mariotti et al., 2023). Intriguingly, the seismic risk variable shows a positive association with profitability. This suggests that despite challenges, natural disasters can present entrepreneurial opportunities, such as infrastructure reconstruction projects and new ventures aimed at alleviating suffering (Williams and Shepherd, 2018). Moreover, SMEs can develop innovative earthquake preparedness solutions, leading to increased profitability through the introduction of new products and services (Salvato et al., 2020). The above results on control variables are confirmed in all specifications.

Let us now shift our attention to the main objective of our study. We will proceed by analysing the outcomes related to the impact of inner areas, with a particular emphasis on their role in shaping the profitability of firms, both between and within geographical macro-areas. The estimates reveal several patterns concerning the influence of inner areas on the performance of innovative SMEs. At first glance, location within an inner area seems to have no significant effect on Italian innovative firms. In fact, in Columns 1 and 3 of Table 3, the coefficient associated with Inner Area is negative but not statistically significant. However, the scenario changes notably when the analysis is disaggregated through the model with interactions. When examining Columns 2 and 4, the correlation between inner areas and firm profitability shows variation based on geographical location. Specifically, estimates indicate that in Mezzogiorno, innovative SMEs located in inner areas exhibit lower profitability compared to firms in non-inner areas. This is evident from the negative coefficients associated with the Inner Area, which are −0.0355 in Model 2 and −0.0334 in Model 4. Both coefficients are statistically significant at the 5% level. Conversely, the interaction terms between the Inner Area and Northwest and Northeast present a contrasting picture in the northern regions. In fact, the positive and statistically significant coefficients of the relevant product terms indicate that SMEs in inner areas in the north generally achieve higher profitability compared to those in non-inner areas [18].

Overall, these findings highlight a significant and distinct association between inner areas and the profitability of innovative SMEs in northern Italy as compared to the Mezzogiorno. The results indicate that innovative SMEs operating in inner regions of more marginalized parts of the country experience a decline in financial performance compared to non-inner areas. Essentially, in the peripheral areas of the Mezzogiorno which comprise a higher number of municipalities classified as inner areas where the socio-economic fabric is not fully developed, innovative SMEs face reduced profitability in comparison to their urban counterparts. These firms may be negatively affected by traditional location factors such as distance from the final market, limited infrastructure accessibility, higher transportation costs, labour costs, and lack of skilled labour. In other words, the challenges in this part of the country, including tangible and intangible networks, lead to increased mobility costs for firms, a reduced ability to capitalize on opportunities arising from digitalization, and consequently lower profitability.

In contrast, firms operating in inner areas of the northern regions may potentially benefit from specific features of the context that contribute to increased profitability. Because the northern regions are more developed in both infrastructure and mobility, the specific obstacles arising in an inner context may not pose a problem for firms located in the northern peripheries. Instead, firms in inner areas can take advantage of a more developed context to enhance their performance. In the north, evidence indicates that operating in an inner area does not hinder the profitability of innovative SMEs. This implies that the higher costs of the urban context, such as higher production factor prices (e.g. land and labour) and limited space availability, outweigh the challenges related to a peripheral context, resulting in higher profitability levels for firms in inner areas [19].

To summarize the results, Table 4 reports the fitted values of profitability derived from Model 2, categorized by the location of SMEs (non-inner area vs inner area) and region (Northwest, Northeast, Centre, and Mezzogiorno). Results show that firms in inner areas register lower fitted profitability compared to their counterparts in non-inner areas. This result is driven by the Centre and Mezzogiorno. However, the trend is reversed in the Northwest and Northeast, where inner area SMEs exhibit higher fitted profitability than non-inner area SMEs.

Moreover, SMEs in the Northwest and Northeast generally demonstrate higher profitability than those in the Mezzogiorno. Specifically, in northwest Italy, both non-inner and inner area SMEs display positive fitted profitability values, albeit slightly lower for non-inner area (0.0046 vs 0.0260). That is, inner-area SMEs have a higher profitability level of 0.0214. Similarly, in the northeast, SMEs in inner areas outperform those in non-inner areas: the differential of profitability is 0.0177. However, in the central region, while non-inner area SMEs show positive profitability values (0.0050), inner area SMEs exhibit negative values (−0.0186), indicating lower profitability levels of 0.0236. Lastly, both non-inner and inner area SMEs in the Mezzogiorno show on average negative fitted profitability values, with inner area SMEs having a significantly lower value than urban SMEs (−0.0686 vs −0.0331), showing a lower expected profitability of 0.0355. Furthermore, when comparing the predicted profitability of firms located in inner areas of both northwest and northeast Italy with that of firms operating in non-inner areas of the Mezzogiorno, results indicate that the former group maintains higher levels of profitability. In sum, the analysis highlights varying patterns of profitability across regions and areas of operation for SMEs in Italy. While inner-area SMEs often exhibit lower profitability, exceptions exist in the northwest and northeast, further emphasizing the importance of context-specific strategies to enhance SME performance.

Specifically, in the Northwest, SMEs in inner areas show a fitted profitability value of 0.0260, while non-inner areas have a value of 0.0046. This indicates that inner-area SMEs in this part of the country have a higher profitability level of 0.0214. In the Northeast, inner-area SMEs have a negative profitability value of −0.0092, while non-inner-area SMEs have a value of −0.0268. Although profitability is negative in the Northeast, there is a differential of 0.0176 in favour of inner areas. Conversely, in the Centre, non-inner area SMEs display a positive profitability value of 0.0050, whereas inner-area SMEs exhibit a negative value of −0.0186, indicating a lower profitability level of 0.0236 for inner areas. In the Mezzogiorno, both non-inner and inner-area SMEs show negative fitted profitability values, with inner-area SMEs having a significantly lower value of −0.0686 compared to −0.0331 for non-inner areas, showing a lower expected profitability of 0.0355.

When comparing geographical locations while keeping the location in core/inner areas constant, SMEs in the Northwest and Northeast generally demonstrate higher profitability than those in the Centre and Mezzogiorno. For instance, in non-inner areas, SMEs in the Northwest (0.0046) and Centre (0.0050) show positive profitability values, whereas those in the Northeast (−0.0268) exhibit negative values. The lowest profitability is registered by SMEs in the Mezzogiorno (−0.0331). Similarly, in inner areas, SMEs in the Northwest (0.0260) and Northeast (−0.0092) perform better than those in the Centre (−0.0186), and significantly better than those in the Mezzogiorno (−0.0686). Notably, the Northwest stands out with the highest profitability among inner-area SMEs (0.0260), and it also shows one of the highest profitability levels among non-inner-area SMEs (0.0046), following the Centre (0.0050).

These comparisons indicate that the geographical effect (North/South) tends to be greater than the “inner/core” effect. For example, when comparing the differences in profitability between the Northwest and the Mezzogiorno, we observe a difference of 0.0377 in core areas and 0.0946 in inner areas. This suggests that the regional effect is stronger than the core/inner effect. Specifically, in the Mezzogiorno, the latter effect is 0.0355, while in the inner areas of the Northwest, it is only 0.0214. Although more attention should be devoted to this issue, our findings highlight the complexity of regional dynamics and the necessity of implementing strategies tailored to specific contexts in order to enhance SME performance. In general, our analysis demonstrates the significance of the North/South geographical divide as well as the crucial role played by the “inner/core” dimension. This is especially evident in the Mezzogiorno, where the effect in inner areas is most prominent, resulting in the largest absolute difference in profitability between inner and non-inner areas (0.0355). Consequently, it is essential to consider both geographical location and area type when striving to effectively support SMEs throughout Italy.

5. Discussion and concluding remarks

This paper contributes to the ongoing debate on the regional divide in Italy by examining how location in inner areas affected the profitability of innovative SMEs from 2012 to 2018. With its specific focus on Italy, the study addresses a significant gap in the existing literature by evaluating the role of peripheral areas within the country’s economic landscape. This investigation is highly relevant to both national and regional political debates as it highlights the crucial implications of Italy’s polycentric urban structure. This structure encompasses diverse areas with different levels of spatial periphery which influence business activities and performance.

The findings of the study shed light on the specific role which inner areas, classified according to the NSIA framework, play for innovative SMEs. In particular, when comparing inner areas to non-inner areas, the profitability of SMEs decreases in the Mezzogiorno. This result suggests that the underdeveloped infrastructure in the South hampers innovative SMEs from capitalizing on their proximity to markets, thus hindering their performance. Despite significant progress, infrastructure improvements are still required in the Mezzogiorno, including the expansion of rail networks and road connections between inner areas and economic centres, the improvement of local public transport, the maintenance and enhancement of local road mobility infrastructure, the implementation of high-speed connections, and reduction of the digital divide. Addressing these infrastructure challenges is imperative to creating an environment that promotes the growth and success of innovative SMEs in the inner areas of southern Italy.

On the other hand, innovative SMEs in northern inner areas outperform their counterparts in non-inner areas. This could be attributed to a more developed socioeconomic environment and lower operating costs in the inner areas as compared to metropolitan poles.

These interpretations are dependent on the level of socio-economic development that characterizes the country as a whole and contributes to the ongoing North-South divide. In the Mezzogiorno, firms located in inner areas may face higher overhead costs due to limited infrastructure accessibility, increased transportation expenses, and difficulties in recruiting skilled personnel. As a result, their profitability may be reduced. In other words, the findings related to the Mezzogiorno reveal limitations in mobility networks, which include both tangible infrastructures and intangible assets (SRM, 2022; Mazzoni et al., 2023). On the other hand, businesses operating in inner areas of the northern regions may be able to take advantage of the unique characteristics of their surroundings to enhance their profitability. Since the northern regions have higher levels of development, including better infrastructure, the challenges faced by firms in inner areas may be less significant. They can instead leverage the region’s higher levels of development to improve their performance. Therefore, it is evident that peripheral settings do not hinder the profitability of innovative SMEs in northern Italy. This suggests that the costs associated with urban operations outweigh those associated with peripheral environments, leading to higher levels of profitability. Consequently, in addition to the well-known North-South divide in Italy, a distinct gap emerges between inner and non-inner areas in terms of the profitability of innovative SMEs. This gap poses greater challenges for those located in more marginalised regions of the country.

This study acknowledges its limitations and suggests avenues for future research. First, the investigation only considers inner areas as a whole. It would be interesting to examine each subcategory separately, as inner areas consist of three groups: intermediate, peripheral, and ultra-peripheral municipalities. Second, this study focuses on a specific type of firm which significantly contributes to Italy’s economy, namely innovative SMEs. Future research should assess whether these findings apply to other types of enterprises. Additionally, it would be worthwhile to evaluate whether these dynamics change when considering the performance profiles of other types of firms. Lastly, the timeframe of the study does not allow us to assess the impact of the COVID-19 pandemic on the operation of innovative SMEs in inner areas, as the pandemic may have disrupted production mechanisms and encouraged entrepreneurship in less central locations. Further investigation of these factors could provide additional and valuable contributions to policy assessment and application at the national level.

Figures

Figure 1

Inner areas based on NSIA classification (Panel A) and the total number of innovative SMEs in 2018 (Panel B)

Table 1

Summary statistics

Variable	Description	Mean/Relative frequency	StdD	Min	Max	Obs
Roa	Profit or loss before taxes over Total Assets (in percentage)	−0.95	21.04	−81.38	59.61	5,664
Roa2	Net income over Total Assets (in percentage)	−2.32	18.01	−77.12	40.6	5,670
Inner area	Dummy = 1 if a firm operates in a municipal having the characteristic of an inner area	0.08	0.27	0	1	6,426
Northwest	Dummy = 1 if a firm operates in North-western regions	0.4	0.49	0	1	6,426
Northeast	Dummy = 1 if a firm operates in North-eastern regions	0.21	0.41	0	1	6,426
Centre	Dummy = 1 if a firm operates in Central regions	0.19	0.39	0	1	6,426
Age	Current year minus firm’s year of establishment (in unit)	9.4	9.91	0	89	6,138
Leverage	(Current plus non-current liabilities) to total assets (in percentage)	65.13	25.21	3.52	98.21	5,759
Working capital	(Currents assets minus current liabilities) to total assets (in percentage)	14.1	24.95	−58.94	77.14	5,713
Firm growth	(Current period sales minus previous period sales) divided previous period sales (in percentage)	0.16	0.42	−3.2	3.89	5,387
Population density	Population over municipal surface (in square km) (in unit)	2884.96	2859.84	18.58	8,107.28	6,426
Branch density	Number of banks branches over population (municipal level)	0.06	0.02	0	0.23	6,426
Gini index	The Gini concentration index (municipal level)	0.44	0.05	0.32	0.54	6,426
Senior index	Inhabitants ≥ 65 over inhabitants ≤ 14 (municipal level)	1.78	0.41	0.49	6.5	6,426
Seismic risk	Dummy = 1 if the municipality in which a firm is located has a high/medium-high earthquake risk	0.23	0.42	0	1	6,426

Source(s): Authors own work

Table 2

Number and percentage of innovative SMEs in 2018 by NSIA classification

	Italy		Northwest		Northeast		Centre		Mezzogiorno
NSIA classification	Number	%	Number	%	Number	%	Number	%	Number	%
Non-inner areas	847	92.27	361	97.83	172	90.53	155	90.12	159	85.03
Inner areas	71	7.73	8	2.17	18	9.47	17	9.88	28	14.97
Total	918		369		190		172		187

Source(s): Authors own work

Table 3

Results for innovative SMEs profitability

					High-tech		Robustness check
	Random effect		Lewbel IV		Random effect		Random effect
	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
Inner area	−0.0106	−0.0355**	−0.0123	−0.0334**	−0.0146	−0.0300*	−0.0363**	−0.0372**
	0.009	0.015	0.009	0.015	0.011	0.016	0.015	0.015
North West	0.0447**	0.0377**	0.0441**	0.0379**	0.0363*	0.0483*	0.0378**	0.0329*
	0.018	0.019	0.02	0.019	0.02	0.029	0.019	0.019
North East	0.0143	0.0063	0.0139	0.003	0.0105	0.0067	0.0067	0.0062
	0.019	0.02	0.023	0.023	0.023	0.021	0.020	0.020
Centre	0.0441***	0.0381**	0.0437**	0.0379**	0.0427**	0.0322*	0.0378**	0.0393**
	0.016	0.017	0.018	0.018	0.018	0.019	0.017	0.017
Inner area*Northwest		0.0570*		0.0519**		0.0404	0.0578*	0.0578*
		0.034		0.025		0.043	0.034	0.034
Inner area*Northeast		0.0532**		0.0560**		0.0464*	0.0521**	0.0511**
		0.023		0.023		0.028	0.023	0.023
Inner area*Centre		0.0119		0.0094		−0.0025	0.0117	0.0105
		0.024		0.017		0.030	0.024	0.024
Roa_1	0.5475***	0.5390***	0.5475***	0.5459***	0.5205***	0.5148***	0.5389***	0.5391***
	0.014	0.014	0.022	0.023	0.016	0.016	0.014	0.014
Roa_2	0.1323***	0.1281***	0.1323***	0.1321***	0.1258***	0.1223***	0.1281***	0.1280***
	0.014	0.014	0.020	0.020	0.015	0.015	0.014	0.014
Age	0.0009***	0.0010***	0.0009***	0.0009***	0.0010***	0.0008**	0.0010***	0.0010***
	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000
Leverage	−0.0239**	−0.0208**	−0.0239**	−0.0241**	−0.0275**	−0.0246**	−0.0208**	−0.0208**
	0.010	0.010	0.012	0.012	0.011	0.011	0.010	0.010
Working capital	0.0693***	0.0749***	0.0693***	0.0707***	0.0717***	0.0765***	0.0751***	0.0749***
	0.010	0.010	0.010	0.010	0.012	0.012	0.010	0.010
Firm growth	0.0945***	0.0953***	0.0945***	0.0944***	0.1086***	0.1083***	0.0953***	0.0954***
	0.006	0.006	0.009	0.009	0.007	0.007	0.006	0.006
Population density	−0.0037**	−0.0039**	−0.0037**	−0.0039**	−0.0027	−0.0028	−0.0039**	−0.0036*
	0.002	0.002	0.002	0.002	0.002	0.002	0.002	0.002
Branch density	0.0514	−0.0277	0.0527	−0.0311	0.0735	0.0495	−0.0253	−0.0124
	0.16	0.162	0.122	0.126	0.193	0.203	0.163	0.163
Gini index	−0.1296	−0.101	−0.1314	−0.124	−0.1148	−0.0978	−0.0955	−0.1031
	0.086	0.087	0.082	0.083	0.098	0.103	0.088	0.089
Senior index	0.0051	0.006	0.0052	0.0054	0.0041	0.0034	0.0054	0.0059
	0.008	0.008	0.007	0.007	0.009	0.010	0.008	0.008
Seismic risk	0.0228**	0.0260***	0.0231**	0.0243***	0.0278**	0.0215*	0.0256**	0.0233**
	0.010	0.010	0.009	0.009	0.012	0.011	0.010	0.010
Rural areas							0.0048	0.0031
							0.010	0.011
Altitude								1.08E−05
								0.000
Coastal municipality								−0.0143
								0.012
Intercept	−0.0281	−0.0272	−0.0269	−0.0211	−0.0131	−0.0089	−0.0287	−0.023
	0.040	0.040	0.037	0.037	0.046	0.048	0.041	0.041
Observations	4266	4266	4266	4,266	3,446	3,446	4,266	4,266
Model test	3,964.353	4024.145	116.3	97.95	2847.34	2849.52	4023.596	4025.197
Model test p-value	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000
R-squared	0.486	0.490	0.455	0.456	0.459	0.461	0.490	0.490
Heteroskedasticity test	615.35	598.34
Heterosked. test p-value	0.000	0.000
Number of groups/clusters	863	863	863	863	694	694	863	863
Endogeneity test			1.219	1.377
Endogeneity test p-value			0.269	0.241
Hansen J test			65.98	74.14
Hansen J test p-value			0.646	0.474

Note(s): ***p < 0.01, **p < 0.05, *p < 0.1. The standard errors are reported in italics. The dependent variable is ROA calculated using P/L before tax. Sector and year dummies are always included but not reported

Source(s): Authors own work

Table 4

Fitted values of profitability derived from model 2

	Non-inner area	Inner area
Northwest	0.0046	0.0260
Northeast	−0.0268	−0.0092
Centre	0.0050	−0.0186
Mezzogiorno	−0.0331	−0.0686

Source(s): Authors own work

Notes

1. As documented in Martin et al. (2021), Italy’s inner areas fall within the broader phenomenon of forgotten territories which, despite specific national and subnational features, affects many different countries including those of the European Union, the UK, and the United States. Examples of this phenomenon are: “La France périphérique” (peripheral France), “abgehängte Regionen” (suspended regions) in Germany, “Krimpgebieden” (shrinking areas) in the Netherlands, “la España vaciada” (hollowed-out Spain), and “legacy cities” and the “rustbelt” in the United States.

2. In Italy, there is significant overlap between rural and inner areas, although it is not absolute. Using data on population size and density provided by ISTAT/EUROSTAT, Italian municipalities can be classified into cities, towns, suburbs, and rural areas. According to this classification, 5,044 Italian municipalities (representing 63% of the total) are classified as rural. Among these rural municipalities, 3,379 are also classified as inner areas, meaning that 1,665 rural municipalities are considered core in the SNAI classification. These data show that while there is some overlap between rural municipalities and inner areas, it is not complete. The correlation between being a rural municipality and an inner area municipality is 0.41 and varies by region, ranging from 0.34 in the Northwest to 0.48 in the Mezzogiorno. Figure A1 in the Online Appendix A displays a map of rural and inner areas.

3. For further details on the institutional framework and theoretical background, refer to the Online Appendix B.

4. One of the most important measures of firms' financial performance is profitability. Studies on the determinants of firm profitability can be categorized into two groups: internal factors influenced by management decisions, and external factors that reflect the market, business, and economic environment in which the firms operate (Pattitoni et al., 2014).

5. As a robustness check, the econometric analysis is replicated by employing the dependent variable ROA obtained using net income. Details are in Table 1.

6. A map of Italian macro-areas is provided in the Online Appendix C.

7. Variables included in the vectors X and Z are described in the Online Appendix C.

8. The Lewbel (2012) method is particularly suited when finding valid external instruments is challenging or impossible. This approach enables the identification of structural parameters in regression models with endogenous variables by exploiting the heteroscedasticity of the error term. The greater the degree of heteroscedasticity in the structural equation error process, the higher the correlation between the generated instruments and the endogenous variables.

9. https://www.registroimprese.it/

10. https://www.bvdinfo.com/it-it/le-nostre-soluzioni/dati/internazionali/orbis

11. https://www.agenziacoesione.gov.it/strategia-nazionale-aree-interne/

12. https://www.istat.it/

13. https://www1.finanze.gov.it/finanze/pagina_dichiarazioni/public/dichiarazioni.php

14. https://rischi.protezionecivile.gov.it/en/

15. Table D3 in the Online Appendix D reports an overview of the total number of Innovative SMEs in Italy for the year 2018, categorised by NSIA classification and sector.

16. Eurostat employs a methodology that aggregates the manufacturing industry by technological intensity, based on NACE Rev. 2 at the 2-digit level. This allows for the compilation of aggregates related to high-technology, medium-high-technology, medium-low-technology, and low-technology sectors. Following a comparable approach, Eurostat defines sectors as either knowledge-intensive services (KIS) or less knowledge-intensive services (LKIS), each further subdivided into sub-sectors. For further details, see:https://ec.europa.eu/eurostat/cache/metadata/Annexes/htec_esms_an_3.pdf

17. The heteroskedasticity test is the Breusch-Pagan/Cook-Weisberg test performed using the “estat hettest” command in Stata software, which is only available for OLS regression. The results from a pooled OLS estimator are consistent with those discussed in this section, but are not reported for conciseness.

18. We provide comments on the results for the sub-sample of firms operating in high-tech industries and other robustness checks in Online Appendix E

19. The robustness of the findings is further demonstrated through the repetition of the econometric analysis using net income-based ROA as the dependent variable (see Table E1 in the Online Appendix E).

Competing interests: The authors declare that they have no conflict of interest.

Data availability: The authors declare their use of data from ORBIS (Bureau van Dick) which are subject to licensing restrictions.

Appendix

The supplementary material for this article can be found online.