Factors for Development of Small Farms in Selected European Union Countries

Abstract

This research focused on the development of small farms, which in many countries form the basis of the agricultural sector. The specifics of this type of farm, as well as the way in which they operate, influence the possibilities for these farms to realise the model of sustainable agriculture. This study considers income and the rate of reproduction of fixed assets as the main measures of farm development, which are influenced by a number of endo- and exogenous factors. The research period covered 2017–2021, and the subjects of analysis were small individual farms located in Greece, Portugal, Lithuania, and Poland. The figures for the research were taken from the FADN system database. The purpose of this study was to assess the impact of endogenous agricultural factors on the development of small farms as measured by farm income and reproduction of fixed assets in four selected European Union (EU) countries, i.e., Greece, Portugal, Lithuania, and Poland. Spearman’s non-parametric rank correlation method was used to assess the impact of endogenous factors. Selected on the basis of correlation relationships, the farm development factors showed a significantly higher correlation with farm income than with the reproduction of the farm’s fixed assets. The analysis indicated that, irrespective of the location of the farm, factors significantly affecting income levels included the area of agricultural land and the number of full-time employees. Only in some countries was there a statistically significant correlation between farm income and the share of leased land, the number of full-time workers per 100 ha of UAA, the share of hired labour input, as well as the level of total farm subsidies received.

1. Introduction

The development of agriculture has historically been inextricably linked to the operation of small farms. In many countries, these entities formed the basis of food production for society, provided services to the rural population, cared for the environment, and cultivated culture and tradition [1]. Their role was significantly reduced in the second half of the 20th century, when, as a result of the market orientation of the economy, there was a strong industrialisation of agriculture, which led to structural changes in the countryside [2,3]. Today, small farms are at the centre of public policies in many countries. Their important role (environmental and social) in mitigating the negative effects of the marketisation of agriculture is noted, which is also in line with the concept of the sustainable development of rural areas [4,5].

The existence of small farms is one of the most important elements in the current development strategies of EU countries. To date, however, there is no complete consensus on supporting the existence of small farms within the agricultural policies of countries [6,7]. Proponents of their development emphasise their important role in the realisation of social, ecological, and food security functions [8]. It has also been pointed out that small-scale agriculture provides many other direct and indirect environmental, social, cultural, and economic benefits through improved crop diversification, job security, and self-sufficiency [9]. Some researchers also note that small-scale agriculture contributes to poverty alleviation and increased food security at the local as well as national level and biodiversity [10], highlighting the contribution of smallholders to protecting food security and alleviating poverty despite the constraints they face. They also emphasise their key role in global food production and point out that about 30–35% of total production comes from small farmers. In sub-Saharan Africa and Southeast Asia, small farmers are responsible for 70–90% of agricultural production; in China, they produce more than half of all domestic food commodities [11,12].

On the other hand, others have pointed out the inferior competitive position of small-scale farms in the market compared to large-scale farms, the dual profession of their users, that is not conducive to their basic functions, the high dispersion, and the limited connection with agricultural production. This leaves small farmers around the world facing serious challenges. In addition to the sector’s traditional inherent productive risks, including climate-related risks such as soil degradation, loss of biodiversity, water scarcity [13,14,15], prolonged droughts, floods, increased pests, and disease outbreaks [16,17,18,19], small farmers face socioeconomic challenges. The largest of these include poverty, as exemplified particularly in sub-Saharan African countries [20], household indebtedness, labour shortages, rising costs of agricultural inputs, lack of investment capacity, and low prices for sold agricultural products [21,22]. Socioeconomic challenges are common among small farmers around the world and are characterised by ageing populations, increasing socioeconomic pressures, and environmental problems. These adverse phenomena pose challenges to national policies in terms of increasing the adaptive capacity of smallholder farmers and improving food security in many countries. A study by Touch et al. [21] on small farmers in different regions of the world indicates the need for targeted interventions, given the strong diversity of this type of unit.

However, considering the fact that not only market goods but also public goods are important for the development of agriculture, the functioning of small farms is justified because, as Żmija notes [23], the specificity of small farms as well as the way they function affect the possibilities of these farms to implement the model of sustainable agriculture. Hence, the study of the factors of the development of this type of entities is justified and can be helpful in developing appropriate development plans for agriculture and the economy of a given country. Such research was conducted by Touch et al. [21], who pointed to the need to construct appropriate support systems and livelihoods for small farmers, in particular promoting sustainable intensification, diversified income strategies, climate-resilient agricultural practices, investing in water management infrastructure, increasing access to timely and accurate climate information, implementing social protection measures, and constructing effective social support systems.

Farm Development Factors

In terms of the economic analysis of agricultural development issues, including farms, the research area primarily covers the efficiency of the management of production factors in agricultural production processes, the economic (market) and non-economic conditions of agricultural production, and the efficiency of agricultural policy.

Agricultural development should be understood as quantitative and qualitative changes taking place in this section of the national economy [24]. The analysis of development paths of agriculture in the world indicates that the basic driving factor of its development was and still is the maximisation of efficiency (productivity) of the management of production factors [25]. According to Sroka and Dacko [26], the possibilities of farm development lie in the farm itself, and this is determined primarily by its economic potential in the form of an appropriate state and structure of resources, as well as the ability to use them. Determinants of farm development therefore include changes in the area of agricultural land, as well as changes in the value of commodity production, the condition and structure of fixed assets, or the level of obtained economic effects [26]. Various categories of financial results, including, above all, agricultural income, are considered to be the basic measures of efficiency [26,27,28]. This is also confirmed by the research of Piet and Hérault [29], which emphasises that income is both an indicator of the efficiency of farms and their ability to produce value. The higher its level, the greater the willingness of farmers to invest in fixed assets or to innovate. Such activities contribute to the efficient and effective functioning of farms, the development of which is not possible without the ability to reproduce assets that contribute to improving the efficiency of the total factors of production [26]. This is also indicated by the results of a study by Urban and Kowalska [30], which shows that investing results in an increase in production, which makes the development of an agricultural holding possible. The main reason why farms invest is the expectation of higher income in the future than the investment costs [31]. Although income is a category with many weaknesses [32,33], it is considered the best and most widely used measure of business development [34].

The driving force behind agricultural development is farms, which are the basic and oldest unit of production in agriculture. However, there is no single concrete definition of an agricultural holding. The Court of Justice of the European Union (CJEU) has stated that ‘(…) that definition (…) varies according to the purpose of specific acts’ [35]. The simplest definition of an agricultural holding is provided by the Regulation of the European Parliament and of the Council [36], which states that an agricultural holding is all units used for agricultural activities and managed by a farmer, located on the territory of the same Member State. In turn, according to the Economic and Agricultural Encyclopaedia, “an agricultural holding (…) is an organised set of productive forces necessary for the production of agricultural products” [37], whereas in Polish law, according to the definition used for agricultural tax purposes in the Journal of Laws [38], “an agricultural holding is considered to be an area of land (…) with a total area exceeding 1 ha or 1 ha of calculation, owned or held by a natural person, a legal person or an organisational unit, including a company, without legal personality”. Small farms, which, as indicated above, play an important role in the development of agriculture if only by providing an impetus for sustainable development, are also referred to differently [8]. It is relatively often stated that they are farms with a small area of agricultural land and/or not much economic power [39,40]. In this study, these farms are defined according to the second of these criteria. According to the typology of the Farm Accountancy Data Network (FADN) (from 2025, the FADN system (Farm Accountancy Data Network) is to be replaced by the FSDN system, i.e., Farm Sustainability Data Network [41]), small farms include very small and economically small units, i.e., farms with an economic size of between EUR 4000 standard output (SO) and less than EUR 25,000 SO (the economic size of an agricultural holding is the total standard output of the holding expressed in EUR. This, in turn, is the sum of the values of standard outputs achieved from agricultural activities subject to the Community Typology of Agricultural Holdings [3]) [3,41].

Studies in the literature indicate a multiplicity of farm development factors, which can be classified, e.g., into natural and non-natural factors [42], as well as external (exogenous) and internal (endogenous) factors [26,43]. Among the natural factors influencing farm development are the relief of the land, the quality of the soil used, and the climate. Among the non-natural factors, shaped by man, agrotechnology deserves attention, including, among others, crop rotation, land reclamation carried out, basic soil cultivation, mechanical and chemical plant care, as well as soil fertilisation through organic and mineral fertilisation. The non-natural factors influencing the development of agricultural holdings include many so-called socioeconomic factors, such as the size of agricultural holdings (area and economic), the specialisation of production carried out, its commodity nature and, as a result, its competitiveness on the market, the level of mechanisation and technicalisation of production, the condition of owned farm buildings, the amount of human labour input, or the level of human capital [42]. The influence on the development of agricultural holdings of factors such as rural infrastructure, the efficiency of advisory services and institutions disseminating knowledge in the field of agriculture, the impact of various factors related to the agricultural market, as well as the economic and agricultural policy of a given country cannot be overestimated, especially in small agricultural holdings [44]. External (exogenous) farm development factors are those that directly or indirectly affect agriculture from outside, i.e., macroeconomic factors. These include technological factors, e.g., new scientific discoveries, political and legal factors, e.g., government policy and international cooperation, economic factors such as the income of farming families, the rate of inflation, or the economic situation. They also include natural and socio-cultural and demographic factors [43]. Internal (endogenous) factors, on the other hand, are primarily those that result from the production potential, i.e., the resources of production factors: land, labour, and capital and their quality, interrelationships, and ways of use [43,44]. Some of the most frequently described internal factors of farm development in the literature are those related to land resources, labour resources, production intensity—crop and livestock, and budget transfers to farms, i.e., subsidies [26,45,46,47]. A review of these factors indicates the multitude of challenges small farms face in order to grow and function. Due to the fact that supporting the development of small farms is a difficult task, as it touches upon resolving the dilemma of their low economic efficiency, the identification of factors determining the development of small farms has always been at the centre of attention of both policy-makers and researchers. Research carried out in this area indicates that factors that may favour the development of small farms include increased productivity [48], easier access to credit [49], external innovation [50], or agricultural advice and training [51]. This is confirmed by a study by Szymańska et al. [52], which indicates that preferential loans, access to aid from EU programmes dedicated to, among other things, farm development, as well as the amount of long-term liabilities and family income are important parameters influencing the level of investment. In contrast, research by Grzelak [53] found that farmers’ assets do not show a strong relationship with income. The issue of the impact of various factors on the income of small farmers in Enugu State Nigeria was addressed by Mukaila et al. [54]. They included education, farm size, access to farm advisory services, income from non-farm activities, and access to credit among the factors affecting income growth. The only factor inhibiting the income of small farmers was the age of the farmers.

In light of the above information, the study of factors influencing income levels and reproductive activity on small farms in selected EU countries seems to be justified. The aim of this study was to assess the impact of endogenous agricultural factors on farm development as measured by farm income and fixed asset reproduction in four selected European Union (EU) countries, namely Greece, Portugal, Lithuania, and Poland.

2. Materials and Methods

The research period covered 2017–2021, and the subject of the analysis was small individual farms located in EU countries (Greece, Portugal, Lithuania, and Poland). The figures for this study were taken from the FADN (Farm Accountancy Data Network) system database. The main criterion for the selection of farms for this study was the lower threshold for the economic size of farms collecting data in the FADN system, which was EUR 4000 standard output (SO) for 2017–2021 [41,55,56,57]. In the FADN database, only 12 countries had such a minimum economic size threshold during the analysed period. On this basis, 4 countries were selected, and the criteria for the selection of these countries for this study were the date the country joined the EU structures, the climate zone, and the Human Development Index (HDI). Taking these criteria into account, Greece and Portugal were selected for this study, representing the so-called ‘old’ EU countries that have been members since 1981 and 1986, respectively. Most of their territory is located in a humid subtropical climate zone called the Mediterranean. Two further countries, Lithuania and Poland, joined the EU in 2004, and most of their territory is located in the warm temperate transitional climate zone [58,59]. In addition, all selected countries are characterised by a similar high degree of socioeconomic development, which was determined using data from Human Development Reports. Between 2017 and 2021, the HDI averaged about 0.87 for the selected countries [60]. The farm groups qualified for this study came from countries with different farm area structures. As a result, in the first two countries, i.e., Greece and Portugal, the analysed farms represented in 2021 29.8 and 35.6%, respectively, of the total number of farms represented within the FADN system. On the other hand, in the remaining countries (Lithuania and Poland), they represented 72.7 and 63.9%, respectively [61].

In this research, an attempt was made to ensure that the research results were as comparable as possible; hence, data averaged for the period under study from the publicly available FADN accounting database (FADN Public Database) were used for analysis, which ensured the selection for the research of objects classified according to the same methodology (this includes the economic size of farms or production type). For this purpose, for the research samples of farms from individual countries, only those were selected which, in accordance with the EU Community Typology of Farms, in addition to the same economic size (EUR 4–25 thousand SO), belonged to the same agricultural types. As a result, small holdings from the following five agricultural types were included in this study: those specialising in field crops, dairy cattle (cows), grassvorous animals (without dairy cows), granivorous, and mixed (plant–animal) production holdings. To such agricultural types in the considered period, a dominant part (over 90%) of Polish small economic farms belonged [62], which—depending on the year of research—was generally reflected in the whole population of Polish farms covered by research of the FADN system [54]. After selecting the study period and the countries in which the farms were located, the results of the groups of farms of a certain economic size and a certain agricultural type were considered separately for each country but as an average for the entire study period (years 2017–2021). Finally, the results of farm collectivities from each country composed of collective study groups (multiple groups of farms separated by economic size and agricultural type) were examined. The analysed collective groups included 39 such farm groups located in Greece, 42 farm groups in Portugal, 47 groups located in Lithuania, and 49 groups located in Poland. For each of the individual groups included in the studied collective, the average results of farms representing a certain (consisting, according to the requirements of the FADN system, of at least 15 objects) number of such farms were presented. Source data from the FADN Public Database were available in Excel sheets. A software package called Statistica 9.0, used for statistical data analysis, was used for the calculations.

According to the literature sources analysed, the income from the family farm and the reproduction rate of farm fixed assets were taken as measures that could indicate the development of the farms under consideration [3,37,63,64]. Farm income is understood as the economic surplus that remains to pay for the farmer’s and their family’s labour inputs and to cover the cost of own capital involved in the farm’s operational activities [65]. The fixed asset reproduction index, on the other hand, is calculated as the percentage ratio of net investment to the value of the farm’s fixed assets (net investment is the value of gross investment less depreciation calculated for a specific accounting year; according to the FADN methodology, the value of gross investment is the value of purchased and manufactured fixed assets for a given accounting year minus the value of assets sold and transferred free of charge and then increased by the difference in the value of the basic herd [3]). It is assumed that when the magnitude of this indicator is between −1.0 and 1.0%, farms are characterised by a simple reproduction of fixed assets, i.e., we assume that within these limits, the level of assets does not change, it is restorative; when it is above 1.0%, known as expanded reproduction, the value of assets is increasing; and when it is below −1.0%, known as narrowed reproduction, the value of the asset decreases [66].

Endogenous development factors were selected using an expert method based on previously analysed literature sources and database. Nine factors that are most often indicated as those that potentially affect farm development were selected for preliminary analysis. Their calculation was made possible by using the available FADN database. Of the internal development factors analysed, the article presents, above all, those that were statistically significant in at least one country, and these were qualified for this study. These factors included the total utilised agricultural area [UAA] (in ha), the share of leased UAA in total UAA (in %), the number of full-time employees on the farm (in AWU), the number of full-time employees (total) per 100 ha of UAA (in AWU/100 ha of UAA), the share of hired labour inputs in total labour inputs (in %), the intensity of crop production, which was measured by the level of direct costs (in EUR) incurred for crop production per 1 ha of UAA, livestock production intensity, measured by the level of direct costs (in EUR) incurred for livestock production per LU (livestock unit), livestock density per 100 ha UAA (in LU/100 ha UAA), and the share of total subsidies (allocated jointly to operational and investment activities of the farm) in farm income.

In order to determine the value of each of these 9 indicators, as well as the level of farm income and the size of the reproduction rate of fixed assets of the farms under consideration for each of the adopted countries, 13 different variables (a total of 52 variables) found in the database used were used, such as variables determining the area of agricultural land, labour inputs of hired workers, the value of farm income, the value of net investment, or the value of the fixed assets of the farms under study. Their average values were calculated as averages for the considered farm groups and weighted by the number of farms for which the results of these farm groups presented in the database were representative. The results calculated in this way are presented in both

Table 1. More important factors in the development of small farms in selected European Union countries, on average, in 2017–2021.

Specification	Greece	Portugal	Lithuania	Poland
Total utilised agricultural area (UAA), ha	8.05	14.50	19.78	11.59
Share of leased land in total utilised agricultural area, %	57.4	16.0	41.6	17.5
Number of full-time employees (total) on the farm, AWU	0.79	1.23	1.21	1.24
Number of full-time employees (total) per 100 ha of UAA, AWU/100 ha of UAA	9.9	8.5	6.1	10.7
Share of hired labour inputs in total labour inputs, %	8.1	5.7	2.1	1.6
Share of total subsidies (i.e., total subsidies for operating and investment activities) in family farm income, %	78.9	68.3	149.1	102.3

Source: Own calculations based on FADN Public Database.

and

Table 2. Family farm income and the fixed assets reproduction rate of small farms in selected European Union countries, on average, in 2017–2021.

Specification	Greece	Portugal	Lithuania	Poland
Family farm income, EUR/farm	7.223	8.826	5.028	3.795
The fixed assets reproduction rate of farms, %	−2.3	−0.1	8.7	−1.5

Source: Own calculations based on FADN Public Database.

.

The next stage of the research was to assess the strength of the correlation relationships between the defined measures of farm development (income and asset reproduction) and selected potential factors influencing the amount of these measures. For this purpose, the non-parametric Spearman’s rank correlation test was used. This makes it possible to determine the strength and type of relationship between two statistical characteristics, X and Y, in a situation where the distribution of the statistical community of at least one of these characteristics is not a normal distribution. Rank correlation also has the advantage of being insensitive to outlier observations that have occurred in the study population. The result of this test is the Spearman rank correlation coefficient (R_s) [67,68], calculated as follows:

$$R_s=1-{\displaystyle \frac{6\sum _{i=1}^n{d}_i^2}{n\left(n^2-1\right)}},$$

where

$R_S$—the result of the Spearman rank correlation test;

$d_i$—the difference between the ranks that are assigned to the i-th observation of the first and second characteristics;

$n$—the number of objects of the set of one of the features.

The Spearman rank correlation coefficient (R_s) is a number in the interval [−1, 1]. In contrast, the absolute value of the correlation coefficient [R_s] reports the strength of the correlation. The ranges of this value are described as follows:

0.0–0.2—very weak correlation (virtually no relationship);

0.2–0.4—weak correlation;

0.4–0.6—moderate correlation;

0.6–0.8—strong correlation;

0.8–1.0—very strong correlation.

Due to the fact that the research concerns agriculture, it was assumed that a correlation is statistically significant when the test probability (p-value) is below the assumed significance level of 0.1.

After analysing the correlation results of the 9 selected factors with farm income and the rate of reproduction of fixed assets, 6 factors were finally qualified for further analysis, i.e., those for which the results of correlation with at least one of the aforementioned development measures and generally in at least one of the countries considered were statistically significant. These factors included the area of agricultural land, the share of rented UAA in total UAA, the number of full-time employees, the number of full-time employees per 100 hectares of UAA, the share of hired labour input in total labour input, and the share of total subsidies in farm income.

Horizontal analysis was used to assess the variation in the economic situation of the farms by comparing the analogous results of farms from selected countries [69].

The obtained results of the research, however, should not be directly identified with the average results of all small economic farms belonging to the considered agricultural types in the countries adopted to research. However, these results may, with a fairly high probability, indicate certain dependencies and trends relating to these types of farms in general.

3. Results and Discussion

The analysis of statistically significant correlations between the factors selected for this study and the amount of income from the farm and the reproduction of its fixed assets allowed for the following factors to be selected for further analysis: the utilised agricultural area (total) of the surveyed farms, the share of leased agricultural land in the total utilised agricultural area, the number of total full-time employees, the number of full-time employees per 100 ha of utilised agricultural area, the share of hired labour input in total labour input, and the share of total subsidies in farm income.

The basic characteristics of the farms analysed in the research are presented in Table 1. On average, in the period 2017–2021, among the countries of the European Union considered, the group of economically small individual farms located in Lithuania (19.78 ha) had the largest UAA, and Greece had the smallest (8.05 ha). In Portugal, the average UAA of the analysed farms was 14.50 ha, and in Poland, it was 11.59 ha. This means that the Polish farms included in this study used, on average, 41.4% less UAA acreage compared to Lithuanian farms and were 44.0% larger in area than Greek farms (Table 1).

Greece, however, was characterised by the highest share of leased land in agricultural land, at 57.4%. Lithuania also had a relatively high share at 41.6%. In contrast, in Poland, it was lower in relation to Greece and Lithuania by 39.9 and 24.1 percentage points, respectively, and 1.5 percentage points higher than in Portugal. On average, between 2017 and 2021 in Portugal, Lithuania, and Poland, the average number of full-time workers was slightly more than 1.2 AWU on the farms in question, and only in Greece was it more than 30% lower, accounting for 0.79 AWU. In this case, this could have been the result of a shortage of workers, but also the result of the lower labour demand on these farms, which could have resulted from the noticeably smaller area of their agricultural land than in the other countries analysed. The average number of full-time employees per 100 ha of UAA of the analysed farms was comparable in Poland, Greece, and Portugal. The share of hired labour inputs in total labour inputs incurred in the analysed farms was the lowest in Poland and amounted to 1.6% on average, and in comparison with the analogous indicator calculated for farms in Lithuania, Portugal, and Greece, it differed by 0.5, 4.1, and 6.5 percentage points, respectively. Such a low share of hired labour outlays in economically small Polish farms was mainly associated with their real low demand for hired labour. In particular, the number of employed persons in these farms was often higher than the real needs would indicate, as hidden unemployment has been recorded in these farms for many years. This was pointed out by, among others, Kowalski [70] and Kołodziejczak [71]. However, it sometimes happens in other regions of the world that farms are short of labour, which happens mainly as a result of an insufficient supply of workers in the local market. This is particularly true for farms that are larger in area and specialise in livestock production, such as dairy farming [72]. Such farms, therefore, have to use hired workers, sometimes foreigners, quite often [73].

The analysis of the share of total subsidies (granted jointly for operating and investment activity of farms) in the analysed farms showed that on farms from the so-called “old” EU countries, it constituted 70–80% of the income from a farm, and in the countries included in the EU in 2004, it constituted over 100% of the value of this income. It is possible to surmise that in Poland and Lithuania, subsidies were used to cover economic losses generated at the level of income from the farm. However, it is worth noting that on small farms, in addition to subsidies, income from non-agricultural activities plays a large role, which, to some extent, helps to reduce economic dependence on subsidies at the farmer’s household level.

Pieroni [74] wrote about earning off-farm income, noting that most Italian farming families also obtain income from wage labour outside of agriculture. The same problem was analysed by Giller et al. [20], who studied households from sub-Saharan African countries and pointed out that a small proportion of households earn sufficient subsistence income, in connection with which they pointed out the need for the simultaneous development of off-farm employment. Similar observations, but using households from Latin America, Africa, and Asia as examples, were made by Ellis and Allison [75], claiming that at the end of the 20th century and the beginning of the 21st century, farm families from those areas of the world derived 40–50% of their household income from off-farm activities and cash transfers from urban areas or abroad. This opinion was also confirmed by Zhang [76], describing Chinese farms, and Pfeiffer et al. [77], who studied farms from Mexico. An analogous phenomenon was also noted by Zegar [78], considering the bi-professionalism of many Polish farm owners, as well as Dzun [79], considering the development of entrepreneurship in Polish rural areas. Some researchers also mentioned that farming families used part of the income earned outside the farm to invest in the farm. Such a situation was noted, for example, by de Janvry et al. [80], who gave the example of farms from China, as well as Babatunde [81], describing farms from Nigeria.

For the farm groups in question, on average between 2017 and 2021, the highest farm income was obtained in Portugal (8826 EUR/farm) and the lowest in Poland (3795 EUR/farm), a difference of 57.0%. Compared to Greek and Lithuanian farms, income on Polish farms was lower by 47.5 and 24.5%, respectively. Analysing the average level of the reproduction index of the fixed assets of farms from the groups under consideration, it was found that on average, only Lithuanian farms increased the value of fixed assets in the adopted years of the research, and this index was 8.7%. It should be noted that Lithuanian farms were also characterised by relatively high land resources. This is a relationship described in the literature, which implies that higher reproduction rates are accompanied by higher land and often labour and capital resources [82]. On Portuguese farms, there was a simple reproduction of fixed assets over the period, indicating that these farms did not increase the value of the fixed assets driving their development. In contrast, the narrowed reproduction of fixed assets was found on farms from the other countries analysed. This means that on farms located in Greece and Poland, economically small farms belonging to selected agricultural types generally did not reproduce fixed assets, which may not have been conducive to their development. In Polish farms, however, the level of the indicator in question (−1.5%) was more favourable (higher by 0.8 percentage points) than in Greek farms (Table 2), which does not mean, however, that Polish farms were undertaking investment activities. Small farms in Poland, when undertaking investment activities, most often use their own capital, the importance of which is the greater the capital, the smaller the scale of production [83,84]. Similar conclusions were also reached by Enjolras and Sanfilippo [85], who, giving the example of French farms, showed that farmers first finance investments were with their own funds, followed by loans or long-term loans, and then short-term liabilities. Hertz [86], on the other hand, analysing Bulgarian farms, showed that owners of small- and medium-sized farms quite rarely invest in the farm, and if they do, it is most often with funds from additional sources of income. The above results also indicate that small farms rarely undertake investment activities, which is also confirmed by the results of the European Commission report, which indicates that small farms invest less than large farms and are less dependent on external sources of financing [87].

The analysis of the strength of the correlations between farm income and the level of selected internal factors potentially influencing farm development shows a large variation in the strength of this correlation across countries (

Table 3. Correlation between family farm income and the results of important indicators that may affect the development of small farms in selected European Union countries, on average, in 2017–2021.

Specification			Greece	Portugal	Lithuania	Poland
Correlation between family farm income, EUR, and the following:	total utilised agricultural area (UAA), ha	Spearman’s rank correlation coefficient value (Rs)	0.598	0.795	0.850	0.691
		Interpretation of correlation result	MPC	SPC	VSPC	SPC
		Test statistic probability (p-value)	0.002	0.000	0.000	0.000
	share of leased land in total utilised agricultural area, %	Spearman’s rank correlation coefficient value (Rs)	0.221	0.575	0.585	0.609
		Interpretation of correlation result	WPC	MPC	MPC	SPC
		Test statistic probability (p-value)	0.289	0.004	0.000	0.000
	number of full-time employees (total) on the farm, AWU	Spearman’s rank correlation coefficient value (Rs)	0.847	0.389	0.401	0.624
		Interpretation of correlation result	VSPC	WPC	MPC	SPC
		Test statistic probability (p-value)	0.000	0.066	0.017	0.000
	number of full-time employees (total) per 100 ha of UAA, AWU/100 ha of UAA	Spearman’s rank correlation coefficient value (Rs)	−0.017	−0.696	−0.866	−0.516
		Interpretation of correlation result	VWNC	SNC	VSNC	MNC
		Test statistic probability (p-value)	0.936	0.000	0.000	0.001
	share of hired labour inputs in total labour inputs, %	Spearman’s rank correlation coefficient value (Rs)	0.579	0.043	0.458	0.261
		Interpretation of correlation result	MPC	VWPC	MPC	WPC
		Test statistic probability (p-value)	0.002	0.847	0.006	0.104
	share of total subsidies in family farm income, %	Spearman’s rank correlation coefficient value (Rs)	−0.167	−0.160	−0.425	−0.839
		Interpretation of correlation result	VWNC	VWNC	MNC	VSNC
		Test statistic probability (p-value)	0.425	0.466	0.011	0.000

Interpretation of correlation result: VWPC—very weak, positive correlation; VWNC—very weak, negative correlation; WPC—weak, positive correlation; WNC—weak, negative correlation; MPC—moderate, positive correlation; MNC—moderate, negative correlation; SPC—strong, positive correlation; SNC—strong, negative correlation; VSPC—very strong, positive correlation; VSNC—very strong, negative correlation. Source: Own calculations based on FADN Public Database.

). Even greater variation is also seen in the results of the correlation between the fixed asset reproduction rate of the analysed farms and the factors selected for analysis (

Table 4. Correlation between the fixed assets reproduction rate of farms and the results of important indicators that may affect the development of small farms in selected European Union countries, on average, in 2017–2021.

Specification			Greece	Portugal	Lithuania	Poland
Correlation between the rate of reproduction of fixed assets of a farm, %, and the following:	total utilised agricultural area (UAA), ha	Spearman’s rank correlation coefficient value (Rs)	0.130	0.457	0.102	0.482
		Interpretation of correlation result	VWPC	MPC	VWPC	MPC
		Test statistic probability (p-value)	0.536	0.029	0.560	0.002
	share of leased land in total utilised agricultural area, %	Spearman’s rank correlation coefficient value (Rs)	0.006	0.000	0.179	0.395
		Interpretation of correlation result	VWPC	VWPC	VWPC	WPC
		Test statistic probability (p-value)	0.977	1,000	0.304	0.012
	number of full-time employees (total) on the farm, AWU	Spearman’s rank correlation coefficient value (Rs)	0.462	−0.222	0.105	0.228
		Interpretation of correlation result	MPC	WNC	VWPC	WPC
		Test statistic probability (p-value)	0.020	0.309	0.550	0.156
	number of full-time employees (total) per 100 ha of UAA, AWU/100 ha of UAA	Spearman’s rank correlation coefficient value (Rs)	0.104	−0.650	−0.130	−0.443
		Interpretation of correlation result	VWPC	SNC	VWNC	MNC
		Test statistic probability (p-value)	0.621	0.001	0.458	0.004
	share of hired labour inputs in total labour inputs, %	Spearman’s rank correlation coefficient value (Rs)	0.182	0.021	0.318	0.003
		Interpretation of correlation result	VWPC	VWPC	WPC	VWPC
		Test statistic probability (p-value)	0.383	0.923	0.063	0.987
	share of total subsidies in family farm income, %	Spearman’s rank correlation coefficient value (Rs)	−0.295	0.064	−0.205	−0.013
		Interpretation of correlation result	WNC	VWPC	WNC	VWNC
		Test statistic probability (p-value)	0.152	0.771	0.237	0.938

Interpretation of the correlation result: as in Table 3. Source: Own calculations based on FADN Public Database.

). Moreover, in this case, differences are noted not only in the strength of the correlation relationships considered but also in their direction (positive or negative).

The results of this study indicate that, on average, there was a statistically significant positive correlation between family farm income and the area of the farms in the years 2017–2021 in the economically small farms selected for this study in all of the countries considered. Similar results were also obtained by examining these correlations at the level of the countries analysed; nevertheless, studies of this relationship indicated differences in its strength, as shown by the values of the Spearman correlation coefficients (Table 3). In the cases in question, the test probability (p-value) was in the range of 0.000–0.002, so it was much lower than the assumed significance level of 0.1. This research, therefore, indicated that as the area of farmland increased, there was an increase in farm income, and vice versa. This is also confirmed by the study of Kambo et al. [88], which found that farm income generally increases as farmland area increases. However, they noted that the exception was the very small farms, using less than 5 ha of UAA, where the value of income was higher than on small farms. This means, on the other hand, that the classification of farms according to land resources does not always reflect the differentiated income situation of the farms. This has already been recognised by Wiatrak [89], who emphasised that the differentiation of agricultural income—both globally and within area groups—depends on the area of the farm, but in combination with the level input of living and materialised labour.

The analysis of the results of the Portuguese, Lithuanian, and Polish farms shows a statistically significant, at least moderate, positive correlation between farm income and the share of leased UAA in total UAA. Only in Greek farms was the correlation between these parameters weak and statistically insignificant (Table 3). This situation may indicate that in countries where leased land constitutes a relatively large share of the structure of the land used for agriculture (as was the case in Greek farms), the relationship between income and leased land ceases to be significant. This may be influenced by other factors, e.g., the legal conditions for the operation of tenancies in a country, the cost of tenancies, the stability of tenancies, etc. Furthermore, as Jenkins’ [90] research on Irish farms indicates, country-scale land leasing, as a strategy to increase farm profitability, appears to be more common on larger entities than on smaller farms. Similar conclusions were reached by Dzun and Józwiak [91,92], who showed that the highest proportion of farms using leasing (around 70%) as a source of income enhancement is maintained on farms with the largest area of UR. Jenkins [90] also identified the four leading farm production types in Ireland in terms of propensity to lease land, i.e., specialised dairying, specialised beef production, mixed livestock grazing, and specialised arable farming.

When considering the correlation between farm income and the number of persons fully employed on the farm, it was found that in all of the analysed farm groups, the correlation was statistically significant, and positive, but only in Greek farms was the correlation very strong, while on farms from other countries, the strength of the relationship was much weaker (Table 3). The results of this analysis may indicate that income is closely dependent on the number of people working on the farm and may indicate that a large number of people working on the farm generally favours the generation of high income. This is also confirmed by research by Wiatrak [89], who noted that farm income increases as the area of the farm increases, which is a consequence of both increasing the scale of production (larger acreage) and increasing the level of labour inputs expended. It should be added, however, that in the case of the farms considered in this study, this income is relatively high on farms with a relatively small number of full-time workers. This research also shows that a relatively large number of persons employed on the farm translates into a lower strength of the abovementioned correlation relationship.

This is also indicated by the results of the correlation relationships between family farm income and the number of full-time (total) employees per 100 ha UAA. This research shows that in three out of the four farm groups included in the analysis, namely the facilities located in Portugal, Lithuania, and Poland, an increase in farm income was associated with a decrease in the number of full-time employed persons per 100 ha utilised agricultural area with a concomitant increase in demand for hired labour. In the abovementioned countries, the values of the Spearman correlation coefficients between income and the number of full-time employees per 100 hectares of UAA considered were −0.696, −0.866, and −0.516, respectively, and the p-value was between 0.000 and 0.001. Only in Greece was the correlation very weak and statistically insignificant. The value of the aforementioned coefficient in this case was −0.017, and the p-value (0.936) was very much above the assumed significance level of 0.1.

When analysing the results of the correlation between income from the farm and the share of hired labour inputs in total labour inputs incurred in the discussed groups of farms, it was found that in Greek and Lithuanian farms, the correlation was statistically significant, moderately strong, and positive. This means that in economically small farms from these countries, as farm income increased, there was an increase in the share of hired labour inputs in total labour inputs. However, on farms located in Poland and Portugal, the mentioned correlation was statistically insignificant, which means that the share of hired labour does not always translate to income growth. The values of the correlation coefficients for these countries were 0, 0.261, and 0.043, respectively. These results indicate that the relationship of the income of the analysed farms with labour is shaped by the need to hire non-full-time staff, since there is not enough labour on the analysed farms. However, it should be clearly emphasised that in order to encourage people to work on the farm, it is necessary to provide them with adequate wage rates. The problem of labour shortages on farms is widespread and is related to demographic problems in rural areas, which is particularly evident in Portugal and Poland.

The study of the relationship between farm income and the share of subsidies in total income (obtained jointly for operating and investment activities of the farm) showed that in the “old” EU countries selected for this study, this correlation practically did not occur, while in Lithuania and Poland, it was statistically significant and negative. Such a situation may paradoxically indicate that subsidies in small farms have a practically limited effect on their development, and in some cases, this development may be less effective, as exemplified by Lithuanian and Polish farms (Table 3). This is confirmed by the research of Kłodziński [47], who noted that although public support creates strong incentives for the growth of individual and collective activity, in some cases, it may weaken this activity. Similar conclusions were reached by Wilkin [93] and Żmija [22].

Summarising the above, among the analysed factors which significantly affect the level of generated income, regardless of the location of farms, we should mention the area of agricultural land and the number of full-time employees on a farm, which is also confirmed by sources in the literature. The share of leased land is also important; nevertheless, its influence on income level may be conditioned by the impact of other factors, which can be indicated by the results of research in Greek farms. Income is generally less influenced by the share of hired labour inputs, which probably shows a strong relationship with the agricultural type and specialisation of farms. Subsidies represent a large share of the income of small farms but do not always favour their development (for example, this is indicated by studies of farms located in Lithuania and Poland). Such a situation may indicate that subsidies, especially in economically small farms, are a social support for farm families and are generally not spent on investments, which is confirmed by the analysis of the relationship between fixed assets reproduction and subsidies. However, according to Grzelak [82], subsidies play an important role in the revival of reproduction processes, which in the period of worse economic conditions in agriculture act as a kind of stabiliser, counteracting the weakening of asset reproduction processes. He adds, however, that the production profile and, above all, economic size are important in asset reproduction processes and points out that in specialised farms, subsidies ensured extended reproduction, while in units without specialisation and those that are economically small, there was a clear decapitalisation of assets.

Small farms in Poland, due to their small area, low profitability, and weak links to the market, fulfil production functions to a small extent. Dzun [79] indicates that the vast majority of these entities are oriented only towards survival. This phenomenon is particularly evident in the reduction in the level of investment outlays. This is also indicated by the present study, which shows that in most of the countries analysed, the surveyed farms do not undertake investment activities. The exceptions, however, are the small Lithuanian farms, where investment activity may have occurred, as indicated by the reproduction rate of their fixed assets, which was positive at 8.7%, and the relatively large share of subsidies in farm income.

The analysis of the correlation results between the rate of reproduction of fixed assets of the farms in question and the share of total subsidies in farm income made it apparent that in the economically small farms adopted for this study, in all the countries considered in this study, there was practically no relationship between these variables. Thus, it can be concluded that the change (+/−) in the share of subsidies in farm income had no effect on the change in the reproduction rate of the fixed assets of the farms in question. However, it should be emphasised that agricultural subsidies play an important function in the development of agriculture; nevertheless, their importance depends on the market activity of farms. This is also confirmed by Grzelak’s research [82].

The results of the correlation between the reproduction rate of the fixed assets of the surveyed farms and the area of their agricultural land showed that, on average, the correlation between these parameters was practically non-existent in the Greek and Lithuanian farms in 2017–2021. According to the results presented, it was statistically insignificant. On farms located in Portugal and Poland, the correlation was statistically significant, positive, and moderately strong. This means that, on Polish and Portuguese farms, an increase in the reproduction rate was associated with an increase in the acreage of farmland owned (Table 4). Such a situation may indicate the growing importance of agricultural land as capital for small farms in these countries. This is confirmed by Grzelak’s study [53], which also indicates that farmers in EU countries are increasing the value of their assets (including land), which, however, does not translate directly to their income. Investment, as a factor influencing income growth and improvement in the quality of life on farms, was mentioned by Marks-Bielska and Lizinska [94], who showed that investments carried out to achieve these goals are usually the purchase of machinery and equipment for agricultural production and the construction of farm buildings, and less often, the purchase of land. However, Bański [45] mentioned that relatively larger farms can more often afford large investments (e.g., construction of livestock buildings or purchase of land). This is because they are in a better financial condition and have more opportunities to obtain a bank loan for this purpose. Indirectly, this recently cited fact was also pointed out by Fenyes et al. [95], and earlier by Michaelas et al. [96], who noted that small agricultural economic entities may face obstacles when applying for loans, as lenders do not want to finance small business units, which they consider to have a high risk of bankruptcy.

Taking into account the results of correlation between the reproduction index of fixed assets and the share of leased UAA in the total area of agricultural land, it should be stated that only in Poland was the correlation between these parameters statistically significant and positive. This indicates that land leasing generally does not affect the increase in the level of investment; nevertheless, it can be a stimulus for the increase in income on farms, through, for example, an increase in the scale of production. However, as sources in the literature indicate, this situation is more often the case for large farms [86,92,97].

This research also showed no relationship between the reproduction rate of the fixed assets of the farms in question and the number of full-time workers on them, especially on Portuguese, Lithuanian, and Polish farms. A moderately strong, statistically significant correlation was recorded only on Greek farms. In these units, a strong correlation between income and the number of persons employed was also shown. It is worth noting that, on average over the period 2017–2021, Greek farms were characterised by the smallest—among the farm groups considered—number of full-time employees and the largest share of hired workers, as well as a high income from agriculture, which may ultimately indicate an efficient use of labour inputs. The reasons for this situation can be seen, among other things, in the production structure of Greek farms, in which livestock production plays a significant role, as confirmed by data from the FADN system.

The analysis also showed, at best, weak, statistically insignificant relationships between the rate of fixed asset reproduction and the number of full-time workers per 100 ha of UAA (in Greece and Lithuania) or hired labour, which may indicate that these factors do not influence the uptake of investment activity. Thus, no direct relationship was shown between labour and the reproduction of the fixed assets of these farms (Table 4). The situation was slightly different for the objects located in Poland and Portugal.

As a rule, subsidies do not provide an impulse in small farms to undertake investment activities, as indicated by the results of the correlation between the reproduction index of fixed assets of the surveyed farms and the share of total subsidies in farm income (Table 4). Therefore, it can be concluded that the subsidy supplements insufficient income but is usually not used to develop economic activity.

The reasons for the spatial differences in the analysed variables can be attributed to regional differences, differences in technology, and different environmental conditions, and other historical conditions, among others. This is confirmed, among other things, by the research of Ren et al. [10], who studied the relationship between farm size and productivity and noted that it can vary by region of farm location due to differences in technology and other conditions, such as the political context. The authors’ research also indicated that the structure of production can affect the variation in economic performance, as indicated by the results of Greek farms. Other factors, such as soil quality, available technology, and production assets such as sprinklers, also affect the variation in economic performance across countries. The historical background that has differentiated Western and Central and Eastern Europe may also have a large impact on the presented results, as confirmed by other studies in the literature. Differences are sought, among other things, in the complicated political situation of Central and Eastern Europe (loss of independence, uprisings, wars), in the different political and economic systems of each country, in the course of the first industrial and agrarian revolution, as well as in the process of the reprivatisation of agricultural land in Central and Eastern European countries, which contributed to the emergence of a large group of small farms [98,99,100,101,102,103,104,105].

In conclusion, the above analysis did not show clear relationships between the reproduction of the fixed assets and the considered potential development factors of small farms. This situation indicates that the reproduction of assets is strongly dependent on the on-farm income generated, which may provide the impetus for investment activity. This situation may be due to the fact that reproduction asset processes on a farm based on their own labour resources depend on the situation of the household, i.e., generational changes, satisfaction of consumption, accumulated savings, or external sources of income. This is emphasised by Grzelak [53], who notes that under conditions of domination of own resources in financing the functioning of the farm, there is a peculiar contradiction between the allocation of earned income: for accumulation or consumption (consumption). In his view, the realisation of the consumption of the farmer’s household is the priority, and their satisfaction allows only for investment. However, such a situation can only occur in the short term. This is pointed out by Grabowski [106], who emphasises that in the case of market-oriented households, the dynamics of reproduction processes in the long term are dependent on the size of resources, income, and thus the level of consumption. He also adds that the scale of reproduction processes is linked not only to the level of income and the factors creating it (prices and production costs) but also to the propensity to invest.

4. Conclusions

The results of the research carried out indicate that among the farm groups from the countries under consideration included in this study, a wide variation in the strength and direction of the correlations between the selected measures of farm development (family farm income and reproduction of its fixed assets) and selected endogenous factors of farm development is noticeable, which may be influenced by the specific characteristics of the farms in the selected countries.

The development factors identified during the analysis, i.e., the area of UAA (total), the share of leased UAA in total UAA, the number of total full-time employees, the number of full-time employees per 100 ha of UAA, the share of hired labour input in total labour input, and the share of total subsidies in family farm income, show a much higher correlation with farm income than with the reproduction of the farm’s fixed assets.

The analysis showed that, irrespective of the location of the farms in question, the factors significantly influencing the level of farm income included the farm’s utilised agricultural area and the number of full-time employees on the farm. The other factors were less important. Only in some countries was there a statistically significant correlation between farm income and the share of leased land, the number of full-time employees per 100 ha of UAA of the farm, the share of hired labour input, as well as the level of total subsidies received by the farm. Therefore, it can be assumed that the influence of these factors on the income from economically small farms of selected countries may be indirect and related to the impact of other factors not considered in this study, such as, for example, the location of the farm, the type of farming, the production specialisation of these farms, or historical background. Indeed, it should be noted that farm fragmentation still persists. This may be influenced by historical factors that have shaped the mentality of the rural population, including attachment to the land, and have also been a contributor to the creation of special systems directed to agriculture (e.g., pensions), which hinder structural change in the countryside.

An examination of the correlation relationships between selected development factors of the analysed farms and the reproduction of their fixed assets indicated a lack of or weak correlation. Relatively often in selected countries, the correlations were statistically insignificant. In none of the considered groups of farms located in the countries selected for this study was there a statistically significant relationship between the reproduction of the fixed assets of these farms and the total subsidies received for their activities. It can therefore be concluded that subsidies, although an important income support for farmers and their families, generally do not translate directly to the development of their farms. In view of the above, it may be legitimate to state that subsidies to farm families should be better targeted: some explicitly earmarked for social purposes (to ensure, first and foremost, multi-directional and sustainable rural development), others for the operation and economic development of economically viable farms.

The study of dependencies is the first stage of the analyses conducted, which allows us to conclude that there is a relationship between the phenomena under study. However, it is important for further analysis to conduct research to find out why this is the case and what are the effects of the measures taken. Further in-depth research on the causes and effects of the influence of various development factors on the economic situation of small farms may be justified. In addition, it will be reasonable to conduct more extensive research on the role of small farms in increasing the sustainability of agriculture and to study the relationship between the size of the farm and the amount of energy, fertilisers, fuels, etc., used in order to determine the impact of various groups of farms, especially small farms, on environmental sustainability. A limitation of this research, and especially of the interpretation of the results, is the inability to conduct research on individual data due to the lack of access to such data from individual countries, which limits comparative analysis between countries. In addition, due to a change in the rules of data collection in the FADN system, current research on data from this system may be limited due to an increase in the threshold for the entry of farms into the FADN system, which resulted in the elimination of small farms from the system, and makes it necessary to conduct individual organised research.