*Article* **Assessment of the Relations for Determining the Profitability of Dairy Farms, A Premise of Their Economic Sustainability**

**Rodica Chetroiu 1,\*, Ana Elena Cis,mileanu 2, Elena Cofas 3, Ionut Laurentiu Petre 1,4,\*, Steliana Rodino 1, Vili Dragomir 1, Ancut,a Marin <sup>1</sup> and Petrut,a Antoneta Turek-Rahoveanu <sup>1</sup>**


**Abstract:** The profitability of dairy farms is a broadly addressed issue in research, for different farming systems and even more so now, when it comes to the issue of sustainability in different agricultural fields. The present study presents an evaluation of the relations used for the determination of profitability of various categories of dairy farms, in terms of size, geographical area, and total milk production. In order to analyze the associated influence exerted on the level of profitability by the selected technical and economic indicators, regression functions were applied. The TableCurve program was used to determine the ideal equation that describes the data entered in a two- or three-dimensional representation. The research results showed that the size of farms and the level and value of milk production are directly correlated with profitability, and the unit cost is inversely correlated with it.

**Keywords:** farm profitability; milk production; regression functions

#### **1. Introduction**

Economic efficiency is one of the key prerequisites for ensuring the competitiveness of any business regardless of the economic sector of production or position in the value chain [1]. Kingwell R. (2011) [2] showed that profitable farming systems are often large, complex, highly technologized, and involve time-consuming activities even for high-skilled managers. The farm productivity derived from production technology properly adapted to given conditions determines the financial results, and these influence strategic decisions regarding further development or, in some cases, to cease operations [3].

Previous studies [4] have demonstrated that higher intensification of agricultural activities significantly increases production efficiency. Profitability of the farm can be achieved by improving the input–output ratio and also by increasing income based on expanding production capacity, thus aiming to achieve competitive agricultural systems [5].

The modern farmer must be a skilled manager, selecting different investment opportunities so as to obtain as high a profit as possible, while fully developing human capital and observing environmental protection rules, all at the same time [6]. The available resources and the existing capacities of a given farm determine its development plans [7]. In order to be competitive, farmers need to be constantly aware of changing circumstances and have the ability to adapt to changes in the economic environment [8]. Proper management strategies can only be implemented based on detailed analysis of farm indicators.

**Citation:** Chetroiu, R.; Cis,mileanu, A.E.; Cofas, E.; Petre, I.L.; Rodino, S.; Dragomir, V.; Marin, A.; Turek-Rahoveanu, P.A. Assessment of the Relations for Determining the Profitability of Dairy Farms, A Premise of Their Economic Sustainability. *Sustainability* **2022**, *14*, 7466. https://doi.org/10.3390/ su14127466

Academic Editor: Giuseppe Todde

Received: 18 May 2022 Accepted: 13 June 2022 Published: 18 June 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

The economic sustainability of milk production, as in other economic activities, is measured using the net profit indicator [9,10]. A good understanding of the influence of each cow's contribution to farm profitability can lead to improved dairy farm management [11]. There are a variety of interconnected factors that affect the efficiency of dairy farms, including management decisions, genetic factors, feed self-sufficiency, and animal welfare [12].

The profitability of dairy farms also depends on the efficiency of feeding, associated with the milk production obtained [13]. Farm profitability is influenced by fluctuations in prices for various inputs, especially feedstock, which have the highest share on expenditure, as well as the volatility of finished-product prices [14]. Low-performing farms have low milk production, unbalanced feed ratios, and low forage area. Large farms have higher turnovers and are more productive because they use better technology; at the same time, they are more specialized in production activity [15].

Economies of scale are one of the factors influencing the economic efficiency of milk production and economic sustainability [16]. Small farmers have limited bargaining power, so in order to become more competitive in the market a change of scale and the development of innovative capacity are needed [17].

Economic sustainability can also be achieved by limiting the number of dairy cows to those that can be fed mainly with forages from the farmer's own farm [18]. Another important factor influencing the economic performance is the labor force and its productivity [19].

The aim of this study was to evaluate the relations for determining the profitability of dairy farms of various sizes, with different levels of milk production, with different allocations of expenditure categories, and located in different areas.

#### **2. Materials and Methods**

Data from 54 farms from 20 counties located in all 8 development regions of Romania were used. Most of the farms (23) were located in the South-Muntenia Region of Romania.

The total sample of dairy cows from the 54 farms was 3966 heads, calculated as the average number of milking cows for the end of the years 2018, 2019, and 2020, without taking into account other age and production categories of cattle, which were not the subject of the study. A share of 51.41% were located in the South-Muntenia Region, included in the largest plain area in Romania. The rest of the livestock composition was as follows: South-West Oltenia Region—4.56% West Region—1.52%, North-West Region—4.83%, Central Region—7.72%, North-East Region—18.01 %, South-East Region—1.24%, and Bucharest– Ilfov Region—10.70%. The average farm size calculated for the period 2018–2020 was 73.44 heads, with a minimum of 5.0 cows and a maximum of 568.3 cows.

Total milk production from the 54 farms (calculated as an average of 2018, 2019, and 2020) was 264,465 hectoliters, distributed by development regions as follows: in the North-East Region 59,080.9 hL (22.34%), in the North-West Region 7261.6 hL (2.75%), in the West Region 3753.8 hL (1.42%), in the South-West Oltenia Region 6248.7 hL (2.36%), in the South-West Region Muntenia 143,477.8 hL (54.26%), in the Central Region 17,416.1 hL (6.59%), in the South-East Region 1925.7 hL (0.73%), and in the Bucharest–Ilfov Region 25,280.4 hL (9.56%) (Figure 1). The average milk production on the farm in the period 2018–2020 was 4554.94 L/cow, with a minimum of 2600 L/cow and a maximum of 9633.3 L/cow.

Data collection from farms encountered some difficulties, primarily due to the fact that it took place during the COVID-19 pandemic with restrictions on mobility and social distancing, so that the originally planned interviews could not be conducted directly on farms, but were conducted mostly by phone. Another challenge was related to the availability of farmers to provide information on different categories of expenditure or delivery prices of products, even though their identity was anonymized. The questions in the questionnaire referred to the landform of the area where the farm was located, the livestock number for the 3 years, milk production, maintenance system, farm equipment, feed rations, different categories of expenses, sale of production, etc.

**Figure 1.** Milk production from case studies, by counties. Source: authors' illustration, using map chart on geographical regions in Excel.

For each of the 54 farms, the annual estimates of expenditures and the annual budgets of revenues and expenditures of the farms were calculated. The average estimate and the average budget for the 3-year period were calculated.

The structure of the estimated variable expenses included the following elements: feedstock, biological material (heifer value), energy and fuel, medicines and medical supplies, other material expenses, supply quota, and insurance. In addition, the fixed costs included labor costs, general costs, interest on loans, and depreciation.

Based on the elements of the estimate and the data provided by the farms on the capitalization of the main production (milk) and secondary production (calf, manure, and animal slaughtering), the revenue and expenditure budget was prepared.

The technical–economic indicators calculated were: value of production, value of main production, total costs, costs for main production, variable costs, fixed costs, unit cost, profit or loss per unit of product, taxable income rate, threshold in value units, threshold in physical units, and exploitation risk rate, using the following relationships:

*Value of production VQ = VQm + VQs*, in which: *VQm*—value of main production, *VQs*—value of secondary production.

*Total costs TC* = *VC* + *FC*, in which: *VC*—variable costs, *FC*—fixed costs.

*Costs for main production MC* = *TC* − *VQs*

*Variable costs VC* = *FoC* + *EnC* + *MedC* + *OC* + *SupC*, in which: *FoC*—forages costs, *EnC* energy and fuel costs, *MedC*—medicines costs, *OC*—other material costs, *SupC*—supply costs.

*Fixed costs FC* = *LabC* + *GC*, in which: *LabC*—labor costs, *GC*—general costs.

*UC* = *MC/MP*, in which: *MP*—main production

*Total profit TPr* = *VQ* − *TC*

*Profit or loss per unit of product Pr/l* = *TPr/MP*, in which: *TPr*—total profit.

*Net profitability rate NPrR* = (*TPr/MC*) × 100

*Margin on variable costs MgVC* = *VQ* − *VC*

*Margin on variable costs MgVC*% = *MgVC/VQ* × 100

*Profitability threshold in value units PrThv* = (*FC/MgVC*%) × 100

*Profitability threshold in physical units PrTrph* = *PrThv/UP*, in which: *UP*—unit price. *Exploitation risk rate ERR* = *PrThv/VQm*

In order to analyze the associated influence of different technical and economic indicators of dairy farms on the results regarding profitability, the TableCurve program was used, which can determine the ideal equation, and, respectively, the representative regression, which describes the data entered. Thus, the relationships between two calculated indicators were illustrated by the resulting curves, and the relationship that includes three indicators was integrated into a spatial model.

#### **3. Findings and Discussions**

#### *3.1. Distribution of Farms in Case Studies*

In order to study the dairy farm size distribution, to compare them with the normal distribution (Gaussian curve) and to highlight the strength of dairy farm size, a graphical representation of the sample was performed, as well as statistical analysis of data.

As can be seen from Figure 2, the physical size of dairy farms in the sample analyzed in the case studies showed a different distribution than normal, with most farms measuring herds between 5 and 100 heads.

**Figure 2.** Distribution of the physical size of dairy farms. Source: authors' own elaboration.

Figure 3 shows the clustering of farm size in the sample.

**Figure 3.** Clustering the size of farms in the sample.

By farm size segments, the average milk production was as follows: in the category below 20 heads, in which 25 farms were included, the average milk production was 3910.67 L/cow, at 21–50 heads (13 farms) was 4471.79 L/cow, at 51–100 heads (10 farms) it was 4328.33 L/cow, and in the category over 100 heads (6 farms), it was 7797.22 L/cow. The smallest size segments, below 100 heads, with yields below 4000 L, generally had the lowest values of profitability indicators, high operating risk rates, and negative safety indices. They also had among the highest unit costs and the lowest labor productivity.

Data related to the size of the farm were analyzed and interpreted with descriptive statistical indicators. Thus, were determined in Table 1 these indicators related to the data string, and, respectively, physical size of the farms.

**Table 1.** Determination of descriptive statistical indicators for farm size.


Source: authors' own elaboration.

Regarding the average of the farm segment taken into analysis, it was of 73.4 heads per farm, with a standard error of 17.7. However, the median was 24.5 heads. Regarding the homogeneity of the data, they were not homogeneous, with a standard deviation of ± 130 heads, which caused very large variation. However, the study aimed to cover as many classes of farm size as possible.

The indicators that study the data distribution, the vaulting (Kurtosis) and the asymmetry (Skewness), were aligned, and at the same time confirmed the graphical distribution in Figure 4. The vaulting coefficient showed a positive value, well above the zero value of 6.98, which describes a leptokurtic distribution. Similarly, the symmetry coefficient confirmed the graphical representation, reaching a value of 2.81, which causes asymmetry to the left.

**Figure 4.** Farm size in the case studies. Source: authors' own elaboration.

#### *3.2. Centralized Data Analysis*

Following the analysis of the 54 dairy farms, it was possible to centralize the technical and economic indicators with the help of the simple arithmetic mean, as well as the standard deviation (Table 2).

**Table 2.** Determining the averages of technical–economic indicators.


Source: authors' own elaboration. Note: AVRG—average, L—liter.

The size of the farms in the analyzed segment varied between 5.0 heads per farm and 568.3 heads per farm, registering an average of 73.44 heads per farm, with a variation of 130.2 heads (Figure 4).

In terms of per capita yield, there was an average milk production of a minimum 2600 L of cow's milk per head and 9633.3 L of cow's milk per head, with an average of all the farms in the study of 4554.94 L/cow, and a standard deviation from this average of 1809.3 L.

Differences in the prices obtained from the sale of milk relate both to milk sold to the dairy processing industry [20] and to milk marketed directly on the market, as drinking milk, as cheese, or through milk dispensers. The value of milk production, determined per unit of product, ranged between 0.27 USD (1.10 RON)/L and 0.88 USD (3.67 RON)/L, with an average value of 0.37 USD (1.56 RON)/L, and a standard deviation of 0.12 USD (0.5 RON)/L.

Analyzing the expenses, there was a variation between 0.27 USD (1.13 RON)/L and 0.71 USD (2.94 RON)/L. On average, the level of expenses was 0.37 USD (1.55 RON)/L, with a deviation of 0.09 USD (0.4 RON). Thus, it was possible to identify an increase in the lower limit of expenditures compared to the value of production, exceeding the latter. Farms with the lowest production values run the risk of not being economically sustainable. Comparing the standard deviation for the value of production (indicator related to price) and the standard deviation for the expenses related to a liter of milk (indicator related to cost), it was found that there were no significant differences, with the deviation for the value of production being ±0.12 USD/L, and in the case of expenses being ±0.09 USD/L. Thus, even if the price varied quite a bit (±32%), unfortunately the costs also varied similarly by ±25.8%, which indicated that the production technologies were influenced fairly high by both external factors and by the cost elements, and the cost was also influenced by the level of production, being in an almost linear relationship with it [21]. Nutrition strategies

and good breeding practices can also contribute to increasing the efficiency of animal production [22].

The structure by elements of expenditures, depending on the farm size—small-, large-, and medium-sized farms—is illustrated in Figure 5.

**Figure 5.** Expenditure structure according to the minimum, maximum, and average size of the farm. Source: authors' own elaboration.

When analyzing the structure of costs, it could be observed that, for the smallest farm in the sample (five dairy cows), the share of variable costs represented 94% of total costs. On the other hand, for the largest farm in the sample (568 dairy cows), the share of variable expenditures was 66% of total expenditures. Management costs for large farms were much higher than for small farms. On average, which was 73 dairy cows, the share of variable expenditures per farm and per unit of product was around 77% of total expenditures and the share of fixed expenditure was 23%.

As viability and economic sustainability indicate the ability of the farm to operate longer and to grow, labor productivity indicators in relation to output are also important [23]. Directing funds to investments that improve labor productivity encourages sustainable practices on dairy farms [24]. Labor productivity in dairy farms is determined by a number of factors, including, for example, the volume of manual labor and the degree of mechanization. Large-scale dairy farms have higher labor productivity than other farms [25]. The indicator can be expressed in physical units of product, or in value units. The productivity of work in physical expression ranged between 0.01 man-hours per liter and 0.17 man-hours per liter, with an average working time to obtain a liter of milk of 0.06 man-hours. The productivity of labor in value terms ranged between 8.27 RON/man-hours and 208.37 RON/man-hours, but, on average, in one hour of work a worker produced milk in value of 43.56 RON. The size of the farms in the analyzed segment varied between 5.0 heads per holding and 568.3 heads per holding, registering an average of 73.44 heads per farm, with a variation from the average of 130.2 heads (Figure 4).

In order to ensure economic sustainability in conditions of market competition, a proper decision making plays a key role [26]. Economic sustainability can also be determined on the basis of the costs related to the value of the main production. In this situation, there are three indicators, shown in Figure 6.

Expenses per 1000 RON main production characterizes more strongly the degree of economic sustainability. This indicator shows the share of expenditure in the value of production, the rest representing the share of profit. Labor costs ranged from 3.07% to 48.66%, with an average of 19.9%. The high shares of this indicator were affected by the extreme data from certain case studies in the sample, in which the average production was only 2600–2700 L/cow, with farm sizes below 12 heads.

**Figure 6.** Determining the economic sustainability of farms based on costs and the value of production. Source: authors' own elaboration.

Cheng, S., Zheng, Z., and Henneberry, S. (2019) [27] showed that, compared to large farms, smaller farms consume more labor force, and for higher yields, more labor efforts, inputs, and precision technology are necessary. Productivity changes are more important for smaller farms and require further modernization of technology, with a certain balance between own and borrowed capital [28].

Analyzing the expenses with materials, they oscillated, with weights between 46% and 91.5%. A key indicator associated with maximizing farm-level profitability is the proportion of forages purchased [29], as the forages accounts for the largest share of material costs. An increase in feed prices increases the cost of milk [30], and thus profitability will be negatively affected.

Finally, analyzing the total expenses related to 1000 RON main production, it was observed that the most efficient farm registered a level of expenses of 786 RON to obtain a value of milk production of 1000 RON, which can be concluded as having an added value of 21.4%. On the other hand, the most economically inefficient farm was the one that had to make a financial effort of 1242 RON to produce milk worth 1000 RON, which obviously led to a loss for that farm. In general, on average, it was observed that the level of expenses incurred to obtain a milk production of 1000 RON was higher than this threshold by 7 RON, which suggested that, on average, the farms studied do not make a profit per unit of product, being at a slight loss, mainly due to low levels of milk production.

#### *3.3. Correlation of Farm Size with Production, by Landforms*

In order to determine the influence that dairy farm size may have on total production, a regression equation can be applied between these two variables, with the farm size being the independent variable and total production as the dependent variable. Thus, following the graphical representation of data and the point cloud, the regression line and the corresponding equation can be identified. This correlation was made for each geographical area included in the case study farms (plain, hill, mountain).

Regarding the influence that the farm size can have on the milk production for the 24 farms located in the plain area, it was observed that the Pearson correlation coefficient between variables was very high, being 0.97, and the coefficient of determination was 0.949 as can be seen from Figure 7. This suggested that the dependent variable (milk production) is explained in a proportion of 94.8% by the independent variable (farm size in the plain area).

Analyzing the regression equation, it can be observed that the value of the independent variable coefficient is 8228.5 units. Thus, it was estimated that at an increase of one unit in the independent variable, the dependent variable will increase by 8228.5 units. In other words, for farms located in the plain area, an increase in the size of the farm by one cow results in an increase in total production by 8228.5 L of milk.

**Figure 7.** The correlation between farm size and total production for the 24 farms located in the plain area Source: authors' own elaboration.

Regarding the influence that the size of the farm can have on the milk production for the 14 farms located in the hill area, it was observed that Pearson correlation coefficient between variables was very high, being 0.99, and the coefficient of determination was 0.986, as can be seen in Figure 8. This suggested that the dependent variable is explained in a proportion of 98.5% by the independent variable.

**Figure 8.** The correlation between farm size and total production for the 14 farms located in the hill area. Source: authors' own elaboration.

In the regression equation, the value of the independent variable coefficient was 9624.6 units. It can be estimated that at an increase of one unit in the independent variable, the dependent variable will increase by 9624.6 units. In other words, for the farms in the hilly areas, an increase in the farm size by one cow results in an increase in total production by 9624.6 L of milk.

Regarding the influence that the farm size can have on the milk production for the 16 farms located in the mountain area, it was observed that, between the variables, the Pearson correlation coefficient was very high, 0.99, and the coefficient of determination

was 0.987, as can be seen from Figure 9, suggesting that the dependent variable is 98.7% explained by the independent variable.

**Figure 9.** The correlation between farm size and total production for 16 farms located in the mountain area. Source: authors' own elaboration.

The regression equation in this situation presented the value of the coefficient of the independent variable of 6137.7 units. This means that at an increase of one unit in the independent variable, the dependent variable will increase by 6137.7 units for this model.

The influence of the main production value on the farm profit level was illustrated using the applications in the TableCurve program, in which a nonlinear regression was used (Figure 10), described by the ideal equation:

$$y = \frac{a + b\mathbf{x} + c\mathbf{x}^2 \ln\_{\mathbf{x}} + d\mathbf{x}}{\ln\_{\mathbf{x}} + e\mathbf{x}^{0.5}} \tag{1}$$

with 95% confidence limits. The value of the coefficient of determination (r2) was very high, given the objective of the program, namely, to identify the function that passes through most points, so this coefficient was 0.94, and r2 adjusted of 0.93 assumes, in this case, that the dependent variable (profit) is explained by the independent variable (the value of the main production) in a proportion of at least 93%. Such a high coefficient of determination determines a very strong correlation coefficient (r) of 0.969, indicating a strong relation between variables (Figure 10). The value of the statistical parameter Fstat is approximately 194.9, being much higher than the value of the parameter Fcritical, in this case F0.05; 1; 53 being 4.023. Therefore, the null hypothesis of equal means between variables is rejected, the quadratic mean intergroup being higher than the quadratic mean intra-group, and it can be concluded that there is a statistically significant difference between the means of the sample.

The resulting curve illustrated that as the value of the main production increases, so does the size of the farm's profit. In any agricultural activity, farmers pursue the efficient use of factors of production in order to maximize profits [31,32]. Furthermore, the welfare conditions of cows, associated with a higher level of milk production, are reflected in higher economic margins for the farm [33]. However, technical conditions are not the most important determinant of the level of profitability and price fluctuations also influence farm profits [34]. Prices are the main contributor to income risk, along with the level of milk production [35].

**Figure 10.** The equation of the value of main production influence on the level of farm profit. Source: authors' own elaboration.

As the net income of the farm is also influenced by its size, the comparison of farms of different sizes can be problematic if this aspect is not taken into account [36]. The influence of farm size on the level of financial results, namely, profit or loss, was described by the ideal equation:

$$y = a + b\mathbf{x}^{0.5} + c\mathbf{x} + d\mathbf{x}^{1.5} + e\mathbf{x}^2 + f\mathbf{x}^{2.5} + g\mathbf{x}^{2.5} + h\mathbf{x}^{3.5} \tag{2}$$

with 95% confidence limits. The value of the coefficient of determination (r2) was very high, given the objective of the program to identify the function that passes through most points, so this coefficient was 0.867 and r<sup>2</sup> adjusted of 0.84, which means, in this case, that the dependent variable (profit) is explained by the independent variable (farm size) in a proportion of at least 84%. Such a high coefficient of determination results in a very strong correlation coefficient (r) of 0.931, which indicates a strong link between the variables.

Yan, J., Chen, C., and Hu, B. (2019) [37] found that the relation between farm size and profit efficiency in agricultural production is illustrated by a U-shaped curve. In the present study, the curve of this equation indicates that the profit of the farm is in a directly proportional relationship to the size of the farm (Figure 11). In fact, large dairy farms have higher economic sustainability. Therefore, they are more likely to operate for medium and long periods of time [38]. However, in the case studies, there were also smaller cow farms which obtained comparable profits to larger farms [39], which indicates that the farm size is not the sole factor in determining the level of profitability.

Ferrazza, R.A., Lopes, M.A., Prado, D.G.O., Lima, R.R., and Bruhn, F.R.P. (2020) [40] concluded that the intensification of activities is the main determinant of economic results, milk production per cow being the most positive indicator correlated with profitability. In addition, the above-mentioned authors pointed out that the profitability of milk production depends in particular on the price of milk, so that it is particularly important to allocate inputs efficiently, thus contributing to the economic sustainability of dairy farms.

**Figure 11.** The equation of the influence of farm size on the level of profit. Source: authors' own elaboration.

Illustrating the correlation between the total milk production of the farm and its profit, the curve of the regression equation alternates two convex segments with two concave segments, but on an ascending path, according to the relation:

$$y = a + b\mathbf{x}^{0.5} + c\mathbf{x} + d\mathbf{x}^{1.5} + e\mathbf{x}^2 + f\mathbf{x}^{2.5} + g\mathbf{x} + h\mathbf{x}^{3.5} + i\mathbf{x}^4 \tag{3}$$

with a probability of 95% (Figure 12). The value of the coefficient of determination (r2) was very high, given the objective of the program, namely, to identify the function that passes through most points, so that this coefficient was 0.907 with an r<sup>2</sup> adjusted of 0.88, which means, in this case, that the dependent variable is explained by the independent variable in a proportion of at least 88%. Such a high coefficient of determination results in a very close correlation coefficient (r) of 0.952, which indicates a strong link between the variables.

Hadrich, J.C. and Olson, F. (2011) [41] demonstrated that a single indicator may not capture the aspects of farm size and performance and that several indicators should be used. Therefore, studying the concomitant influence of two variables, namely, farm size and total milk production, on the farm profit level, a three-dimensional illustration of the regression equation is obtained as:

$$z = a + b\mathbf{x} + c\ln\_x + d\mathbf{x}^2 + e(\ln\_y)^2 + f\mathbf{x}\ln\_y + g\mathbf{x}^3 + h(\ln\_y)^3 + ix(\ln\_y)^2 + j\mathbf{x}^2\ln\_y \tag{4}$$

with r<sup>2</sup> calculated of 0.92, r2 adjusted of 0.90, and 95% probability, indicating that farm profit increases in direct proportion to farm size and total milk production (Figure 13). The value of the statistical parameter Fstat is approximately 57.86, being much higher than the value of the parameter Fcritical, in this case F0.05; 2; 52 being 3.18. Therefore, the null hypothesis of equal means is rejected and it can be concluded that there is a statistically significant difference between the means of the sample.

**Figure 13.** The equation of the influence of farm size and total milk production on the level of profit. Source: authors' own elaboration.

The judicious use of production management factors, such as farm size and milk production, has a positive impact on farm profitability [42].

The application of the TableCurve program to highlight the correlation between farm size, unit cost, and profit level produces a three-dimensional illustration of the regression equation:

$$z = a + bc + cy + dx^2 + ey^2 + fxy + gx^3 + hy^3 + ixy^2 + jx^2y \tag{5}$$

with r<sup>2</sup> calculated of 0.94, r<sup>2</sup> adjusted of 0.93, and 95% probability, indicating that farm profit increases in direct proportion to farm size and is inversely related to unit cost (suggested by the concavity of the graphical representation) (Figure 14). The value of the statistical parameter Fstat is about 80, being much higher than the value of the parameter Fcritical, in this case F0.05; 2; 52 being 3.18. Therefore, the null hypothesis of equal means between variables is rejected, the quadratic mean inter-group being higher than the quadratic mean intra-group. Thus, we conclude that there is a statistically significant difference between the means of the sample.

**Figure 14.** The equation of the influence of farm size and unit cost on the level of profit. Source: authors' own elaboration.

Lukas Kiefer, Friederike Menzel, and Enno Bahrs (2014) [43] have shown that efficiently managed milk production creates the potential to optimize farm income. The calculation of efficiency in milk production should account for unit costs [44] and their minimizing. Dairy farms need to find ways to ensure that their production cost is lower than the market price of milk, and that the strategy to increase the farm size allows reduction in production costs [45]. It is necessary for farmers to periodically analyze milk production, production costs, and profit in order to identify those favorable factors that may contribute to increasing the profitability of their activities [46]. The exact knowledge of the cost of production by the farmer is a management tool [47]. In terms of unit cost of production, large farms have much lower costs, on average, than smaller farms [48].

The difference in production technology and inputs could be an explanation for the difference in productivity between large and small farms, given the same prices relative to inputs [49]. Studies by Yu Sheng, Alistair Davidson, Keith Fuglie, and Dandan Zhang (2016) [50] show that farmers who respond to changing technologies and prices by replacing different inputs thus gain "income effects". In order to ensure economic sustainability, managerial effort and technological investment is needed to increase the daily average of milk production without increasing the average variable cost [51]

#### **4. Conclusions**

Analyzing from the perspective of profitability, there are rates of return between about −20% and +10%, and in the sample analyzed, thus, it can be concluded that several dairy farms were not profitable in the analyzed period.

The increase in the physical size of the farm, no matter the geographical area, positively influenced the milk production. However, in the mountain area the increase in production was slower than for plain and hill areas.

A graphical representation of the profitability of dairy farms was elaborated. The farms with a low value of main production had a small increase in profit, while when the value of main production increased, the profit growth became slower. Further, as the value of production increases, the curve indicates an exponential evolution of the profit.

In determining the farm's profit equation based on the farm size, it was found that in the case of small farms, the increase in livestock leads to a relatively small increase in farm profit, and subsequently, once the size of 400–450 cows is exceeded, the increase in numbers will lead to an exponential increase in farm profits.

The statistical analysis that describes the farm profit equation according to the total milk production led to an almost sinusoidal graph, actually formed of several connected Gaussian curves. Therefore, the profit of the farms increased with the increase in production, up to the moment when the increase in production involved a high level of costs to support it, so that the profit turned into a loss when the level of expenses exceeded that of income. Subsequently, the situation replicated, at a higher level of total production and profit, and so on.

The graphical representation of the multiple regression of farm profit indicated that the highest profit values were recorded when the farm size and milk production were as high as possible. This situation is usual for large and very large farms, but it must be pointed out that most farms in this study owned between 5 and 100 cows. Furthermore, most of the small farms had a fairly high unit cost, being in a situation of economic inefficiency, but the highest profit was recorded in terms of a low unit cost and a high physical size of the holding (ideal case, encountered in the case of large and very large holdings). At the same time, there are quite high profits in the case of medium-sized farms with the lowest possible unit costs.

**Author Contributions:** Conceptualization, R.C. and I.L.P.; methodology, R.C., I.L.P., A.E.C. and E.C.; software, I.L.P. and R.C.; validation, E.C., A.E.C., S.R. and V.D.; formal analysis, I.L.P., R.C., S.R., A.M. and P.A.T.-R.; investigation, S.R., A.M., V.D. and P.A.T.-R.; resources, R.C. and A.M.; data curation, E.C. and A.E.C.; writing—original draft preparation, R.C., I.L.P. and V.D.; writing—review and editing, R.C., I.L.P. and S.R.; visualization, R.C., I.L.P. and S.R.; project administration, R.C. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by ADER 24.1.2. project: "Research on the economic efficiency of raising sheep, goats, dairy cows, cattle and buffalos".

**Institutional Review Board Statement:** This study did not require ethical approval.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** The data presented in this study are available on request from the corresponding author. The data are not publicly available due to privacy.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


**Jurgita Paužuoliene˙ 1, Ligita Šimanskiene˙ <sup>1</sup> and Mariantonietta Fiore 2,\***


**Abstract:** The article analyses the problems of food waste and responsible consumption that include taking into account environmental-social-health and economic impacts of products and services. The study raises the research question related to whether people consume food responsibly. Analysis of research literature sources, systematization, synthesis, generalization, quantitative research and data processing methods were used in the article. The questionnaire was arranged on the pollimill.com website, and the link was shared with selected possible respondents. The survey was carried out in Lithuania and in European countries. The survey sample is equal to 1080 respondents (566 respondents from Lithuania and 514 from Italy, Poland, Latvia, Germany and France). A simple random sample was used in this research. The survey highlighted that the majority of respondents in the survey state that food is not often wasted. In addition, findings show that the population of Lithuania emits slightly less food than the population of the European countries participating in the survey. These findings could be crucial for the future green directions from the side of policymakers.

**Keywords:** sustainable production; responsible consumption; food waste; Europe

#### **1. Introduction**

The EU and the EU countries have to reach by 2030 their Sustainable Development Goal 12.3 target to halve per capita food waste at the retail and consumer level, and reduce food losses (according to the EU actions against food waste). Around 88 million tons of food waste are delivered yearly in the EU. According to preliminary calculations, every EU citizen throws away about 173 kg of food every year that could still be consumed [1]. For example, in Lithuania, the amount of food waste in the mixed municipal waste is about 15%, with an average of 41 kg of food waste per person per year; on the other hand. 75 million tons of bio-waste from municipal waste is created every year across Europe. It is crucial that recycling of bio-waste has to take place in order to meet the overall recycling target of 65% of municipal waste by 2035 [2].

The problem of food waste is relevant throughout the food supply chain, from the production of agricultural products to storage, processing, transport, trade and consumption [1]. Food waste poses environmental, ethical, and economical questions, and shows the need to change our food system.

Food waste prevention is included in the EU's plan for a circular economy, which the European Commission defines as where "the value of products, materials and resources is maintained in the economy for as long as possible, and the generation of waste [is] minimised". This strategy aims to improve competitiveness, promote sustainable growth, and create new jobs [3].

The revised 2018 Waste Framework Directive adopted on 30 May aims to reduce and monitor food waste and report back regarding progress made. Therefore, Member States have to:

**Citation:** Paužuoliene, J.; ˙ Šimanskiene, L.; Fiore, M. What ˙ about Responsible Consumption? A Survey Focused on Food Waste and Consumer Habits. *Sustainability* **2022**, *14*, 8509. https://doi.org/10.3390/ su14148509

Academic Editor: Gioacchino Pappalardo

Received: 7 June 2022 Accepted: 8 July 2022 Published: 12 July 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).


The crucial starting point of responsible consumption is awareness (behavior and attitude) of the impacts of consumption. Therefore, this paper aims to investigate whether people consume food responsibly.

The paper is presented in the following structure: the next section presents scientific analysis on the food waste issue. Then, research methods are described, and later main results and findings are presented. Finally, discussion and conclusions, with some suggestions for future research, close the paper.

#### **2. Literature Review**

Food waste is a global problem and it happens when food is left unused due to poor commercial appearance, or leftovers from uneaten food that are not composted. A common cause of food wastage is improper food storage—it spoils. Another is consumer shopping habits, and in some countries oversupply. Waste of food means any food lost due to spoilage or waste [4,5]. Thus, the term "wastage" encompasses both food loss and food waste [6]. Food waste is defined as food lost in any food supply chain. The food is then discarded and not used for any other productive use, e.g., animal feed or seeds. The FUSIONS framework defines food waste as "food and inedible parts of food removed from the food supply chain" that is to be disposed of (e.g., crops ploughed back into the soil, left unharvested or incinerated, food disposed of in sewers or landfill sites, or fish discarded at sea) or used for nutrient recovery or energy generation (e.g., through composting, or anaerobic digestion and other bioenergy pathways) [7]. Food is wasted in many ways, for example [8]:


Food losses and waste also impact on other natural resources, many of which are scarce. Three key related resources are freshwater, cropland, and fertilizers [9]. The problem of food wastage is multifaceted, ranging from the misuse of arable land, the financial loss to restaurants and hotels and the discarding of prepared meals by households, while counting the working time of employees in cooking [10]. Another problem with food waste is that if food waste is not composted, it emits a lot of methane in landfills—more powerful greenhouse gases than even CO2. Huge amounts of food emissions contribute to global warming and climate change. With agriculture accounting for 70 percent of global water consumption, food waste is also a huge waste of fresh and groundwater resources [11].

Comparing the extent of food waste according to the level of development of countries, more unused food is discarded in economically stronger countries [9]. However, there is also a significant amount of food wastage in developing countries, especially in the supply of food to retail chains. Researchers do not have exact data on how much unused food is lost to smallholder farms. Also, in developing countries, especially in Africa, storage losses on farms can be significant, although the exact nature of such losses is much debated [9]. But, some research shows that consumers are becoming more socially conscious and are including ethical considerations in their purchase decisions [12], as well as becoming increasingly interested in various forms of responsible consumption [13]. Consumers have more product choices and, therefore, have more opportunities to reveal their social preferences when making purchase decisions.

Analyzing food waste by different food groups, the authors found that vegetables (24%) and fruit (22%), followed by cereals (12%), meat (11%) and oil crops (10%) accounted for the largest share of food waste. The fish and eggs food groups, which make up the smallest parts of the food supply chain, also generate the lowest quantities of food waste in absolute terms, despite the fact that much of these food groups (50% and 31%, respectively) go to waste [14].

Based on the Lithuanian State Food and Veterinary Service [1], unreasonable food waste is promoted by:


Nowadays, the concept of sustainable consumption is becoming more and more a major interest of the population and the latter make more conscious food purchasing decisions [15]. Indeed, increasing awareness towards environmental issues and climate change led society to the formation of sustainable consumption habits [16]. However, Ganglmair-Wooliscroft and Wooliscroft [17] argue that also external factors such as government regulations, business initiatives, and geographic characteristics determine consumers' behavior, including food consumption as well as recycling.

On the other hand, Block et al. [18] prove that consumers are often mistaken in estimating the consumption that is the basis of raising food waste. In the line with this, numerous initiatives have been launched in order to benefit from the remainder of these products. For instance, food waste valorization to hydrogen on the one hand reduces the harmful impact on nature, decreasing the quantity of spoiled food in the environment and, on the other hand, the alternative energy source is generated by transformation of biogas that can replace fossil fuel or produce electricity [19,20]. Additionally, it is economically feasible [21].

According to [22], the right tools for reducing food loss and waste have the potential to increase the sustainability of food supply chains. For this reason, authors suggest government to finance the relevant infrastructure for recycling disposed products and consumers' education for shifting towards responsible consumption including earlier food donation. Similarly, Sundin et al. [23] prove the environmental feasibility of food donation calculating a double of the benefit comparing to anaerobic digestion. Kumar and Dholakia [24] see the huge role of the firms to change consumers' behavior. Authors argue that firms have a power to promote innovative thinking, address consumers' environmental identity as well as brand assurance, and edit consumers' choices.

It is noteworthy that the COVID-19 pandemic has influenced consumers' food purchase decisions, their management and consumption that, in turn, has reduced the household waste [25,26]. Nowadays, 81% of consumers make a list before shopping and check the expiration date of the product that is almost double of the number before pandemic [27]. The increased awareness about food waste and its impact on the environment lead the reduction in the quantity of spoiled and thrown products even if purchases have increased during the COVID-19 pandemic [28–31].

In this context, as claimed by a recent research [32], a holistic 4Es Ethical, Equity, Ecological and Economic approach can be useful for better handling food loss issues along the agri-food chain from upstream to consumers by changing the entrepreneur and consumer approaches. Finally, the spread of the pandemic has been leading society to re-think the manner in which we produce and consume food by facing new future green global challenges [32,33].

#### **3. Research Methodology**

The quantitative research method was used in the research. The questionnaire was prepared on the pollimill.com website, and the link was sent to respondents. Regarding the criteria, only those respondents who had an internet connection could participate in the study. The research was guided by ethical principles: the principle of goodwill is ensured by the statements of the questionnaire, which are presented in a respectful style, without creating preconditions for respondents to lose privacy; applying the principle of respect to the individual, the purpose of the study was explained to the respondents; volunteering is the free will of study participants to participate or not to participate in a study; research participants were guaranteed anonymity and data confidentiality. The collected empirical data were processed using the SPSS 20.00 (Klaipeda university, Klaipeda, ˙ Lithuania) (Statistical Package for the Social Sciences). The data processing descriptive statistics were used, such as percentiles, mean, mode, and standard deviation. The data were also processed by independent samples *t*-*test* where significant differences are when *p* ≤ 0.05. To assess the reliability, or internal consistency, of a set of scale, Cronbach's alpha coefficient was used.

The research population. The questionnaire items are based on the analysis of scientific literature and EU strategy on sustainable consumption [28,31,34,35]. The survey was done in February and March 2022. The respondents were reached during the third pandemic period by means of internal research mailing lists of the University of Klaipeda and Foggia [34,35]. The goal was to get as many responses as possible from different European countries, but in this study we were only able to collect data from these countries. The study selected this kind of online research to survey consumers in a fast manner, thus assuring safety and security under pandemic conditions [34,35]. The items of the questionnaire were corroborated by a virtual focus of experts in the agri-food-sustainable field. The survey sample is composed of 1080 respondents. In this survey, 566 respondents from Lithuania and 514 from other European countries (Italy, Poland, Latvia, Germany and France) participated. A simple random sample was used in the research. This kind of sample is a subset of a statistical population in which each member of the subset has an equal probability of being chosen. Preferably, using random sampling, the sample size should be larger than a few hundred in order to allow a simple random sampling to be applied correctly. This method has been selected in order to get as much information as possible on the analyzed topic.

Principles of compiling the questionnaire. The questionnaire consisted of five-point ranking scale questions [33]. First, we asked respondents where they usually buy food. The second and third questions in the questionnaire were designed to find out the respondents' buying habits, how often they pay attention to certain aspects when shopping and how often they throw away certain types of uneaten food. Respondents rated the questions on a five-point ranking scale from 1 to 5, with 1—very often and 5—never. The four questions explain whether respondents compost the waste, and the last one explains responsible food consumption habits of respondents. For this question, respondents rated the statements on a five-point ranking scale from 1 to 5, with 1—strongly agree and 5—strongly disagree [33].

The last four questions were designed to find out the demographics of the respondents, gender, age, monthly income and country of residence.

Demographic characteristics. The demographical data are provided in Table 1. From the table we can see that 52.4% of the respondents were from Lithuania, and 47.6% of the respondents (from Latvia 4.6%, from Poland 12.7%, from Germany 2.3%, from Italy 18.5%, from France 9.4%) were from other EU countries (see Table 1). As the number of respondents from different countries is quite different, the data will usually be analyzed together for other European countries, comparing the data with the Lithuanian data.



#### **4. Results**

The research evaluates the responsible food consumption habits in the daily life of Lithuanian and other European countries' respondents. Respondents of the research had to evaluate items about shopping habits with a range scale, where 1 means that respondents very often do that, and 5—they never do that. For the assessment of the question scale internal consistency, Cronbach's alpha coefficient was used, for a properly composed question scale should be greater than 0.7. In our case, Cronbach's alpha coefficient value ranged from 0.733 to 0.946 (see Table 2). A high Cronbach's alpha coefficient means that the items in the questionnaire are highly correlated.

**Table 2.** Cronbach's alpha coefficient for different questions groups.


Analyzing respondents' shopping habits, we submitted mean and *p* value. *p* value shows the significant differences between countries (significance level is *p* ≤ 0.05) (see Table 3). The standard deviation for the analyzed items ranges from 0.290 to 1.010.

**Table 3.** Respondents' shopping habits (mean).



#### **Table 3.** *Cont.*

Notes: Range scale where 1 means very often; 2 often; 3 rarely; 4 sometimes and 5 never. Significant differences are when *p* ≤ 0.05.

The results show that the mean between Lithuania and other European countries are very similar, but shopping habits are quite different. Respondents were asked to evaluate their buying habits, how often they pay attention to certain aspects when shopping. Respondents rated the statements on a five-point ranking scale from 1 to 5, with 1—very often and 5—never. Research results reveal that most of the respondents only rarely buy products with a promotion, even if it was not in their plan (2.98); buy frozen food (mean 3.19) or shop as much as they need for the day (mean 3.33). Most of the respondents sometimes choose the foods advertised (mean 3.44); buy fast food products (processed, semi-finished) (mean 3.45) or buy products that are about to expire (because they are usually cheaper) (mean 3.60). Most of the respondents often plan purchases by making a list (mean 2.21); choosing foods and pay attention to the composition of the product (mean 2.20); choose foods of local origin (mean 2.19); buy fresh (unprocessed and not frozen) food (mean 1.99) and buy for a longer period (4–5 days) (mean 2.03) (see table).

The independent T sample test discloses the mean difference between country groups, significant data is in bold. The obtained data show that respondents from EU countries try to buy foods that are packaged in recyclable containers (*p* = 0.001 < 0.05); respondents from Lithuania prefer organic foods (*p* = 0.000 < 0.05). Respondents from EU countries are more likely to buy products with a promotion, even if it was not in their plan (*p* = 0.039 < 0.05). For a bigger part of respondents from Lithuania, the appearance of the product does not matter (*p* = 0.001 < 0.05). The Lithuanian respondents are more likely to choose foods of local origin (*p* = 0.000 < 0.05) and buy imported food (*p* = 0.001 < 0.05). The EU country respondents are more likely to buy fresh (unprocessed and not frozen) food compared with Lithuanian respondents (*p* = 0.017 < 0.05). For other items, country does not have a significant impact as the *p*-value is higher than 0.05, which indicates that there is no statistical difference.

If we look at the overall averages, the data ranges from 1.99 to 3.6 (mean). This shows that respondents are more likely to agree with the options available for purchase. The majority of respondents in both groups choose to buy fresh products and the minority buy products that will soon expire. It is interesting to note that in both groups, respondents said that they rarely choose the advertised products (average 3.44), which is somewhat surprising. This shows that respondents in the survey have an opinion about what they need when they go shopping, and it is difficult to change their opinion at the store. It is needed to mention that respondents from both groups go to the stores with a shopping list, plan to do so, and probably do not throw away unused food. Our respondents also look at the composition of the product and look if the packaging is recyclable.

We searched for whether there is a statistical relationship between the income received by the respondents and the place of shopping. However, no statistical dependencies have been identified. The responses of some higher-income respondents do not differ statistically from those of lower-income respondents. Respondents usually shop in supermarkets, it does not depend on the amount of income they receive. There is no statistical link between low-earning respondents growing their own vegetables or fruits.

We were interested in how often respondents buy food. The results reveal that most of the respondents buy food on average two to three times a week (Lithuania 26.9%, other EU countries 20.5%) (Figure 1). Also, a lot of respondents from Lithuania buy food once a week (16.9%). It is likely that the majority of respondents actually give priority to fresh produce when shopping several times per week, as mentioned in previous responses.

**Figure 1.** Respondents' frequency of food shopping; Significant differences are when *p* ≤ 0.05. Chi-square test *p* = 0.000 < 0.05.

We searched for a statistical relationship between respondents' income and shopping frequency. However, no statistical dependencies have been identified. The responses of some higher-income respondents do not differ statistically from those of lower-income respondents. Respondents usually shop two or three times a week, which does not depend on the amount of income they receive.

The research data reveal that respondents sometimes or never throw away food. The mean ranges from 3.10 to 4.41. The obtained data show that there are statistically significant differences between respondents from EU countries and Lithuania (*p* = 0.000 < 0.05), analyzing the question of what products respondents throw away (see Table 4). Respondents from Lithuania more often throw away bread products, fruits and vegetables, while respondents from European countries more often throw away dairy products, meat, fish, pasta, eggs and sweets compared to respondents living in Lithuania. Both groups of respondents throw away eggs the least. These responses are also consistent with the responses where we asked respondents about their shopping habits, showing that respondents do not usually buy unplanned groceries.


**Table 4.** Data on food waste.

Notes: Range scale where 1 means very often; 2 often; 3 rarely; 4 sometimes and 5 never. significant differences are when *p* ≤ 0.05.

It is quite a big problem when people make unnecessary food, buy unplanned or order too much food in restaurants or cafés and do not consume it, and then they just throw it away. We asked respondents to evaluate some statements related with this. The research data reveal that the respondents are quite sustainable consumers, they disagree with most statements and the mean ranges from 2.57 to 4.10 (see Table 5). Significant differences between two groups of respondents are seen in four statements. Research results show that respondents from European countries are less likely to order too much food in cafes than respondents living in Lithuania, however, these differences are very small. The Lithuanian population is less likely to use food products that have an expiration date and are less likely to buy products at a discount, although they do not consume them later and discard food less often depending on the seasonality of the year.

**Table 5.** Food consumption habits.


Notes: Range scale where 1 means strongly agree, 2 agree, 3 partially agree, 4 disagree and 5 strongly disagree. Significant differences are when *p* ≤ 0.05.

The results revealed that only about 17 percent of respondents compost food waste (Lithuania 17.5%; Europe 16.6%) (Figure 2). About 13 percent of respondents threw it together with other waste. A total of 4.4 percent of respondents would not think about it. A total of 17.4 percent of Lithuanian respondents and 13.5 percent of respondents from Europe would like to do it, but do not have a chance.

**Figure 2.** Results of food composting of respondents; Significant differences are when *p* ≤ 0.05. Chi-square test *p* = 0.467 > 0.05.

Assessing the answers of both groups, it can be seen that a part of the respondents compost food; a large part of respondents would like to do it, but do not have the chance. In reality, only 34.3 percent of respondents are not worried about it and it is not relevant for them. We think the results are really promising. It is clear that it is important for the majority of respondents in the study not to waste food, while at the same time taking care of food waste disposal. It is likely that some live in apartments where composting is more difficult. Composting is an important element in sustainable waste management.

#### **5. Discussion**

Wasting food causes environmental and economic inefficiencies. It affects climate change, emissions, availability of natural resources, deterioration of land conditions, global hunger and can even be an underlying reason for an economic collapse [36]. The European Commission is taking the issue of tackling food waste very seriously. Reducing food waste has enormous potential for reducing the resources we use to produce the food we eat. Being more efficient will save food for human consumption, save money and lower the environmental impact of food production and consumption (EU actions against food waste).

This paper contributes to current debates on food waste management [10,14,37] by illustrating empirically what negative problems arise from unsustainable food waste.

To this aim, we conducted a study and assessed the food consumption habits of the European population. We compared the results with Lithuanian food consumption habits. This study allows to identify the respondents' shopping habits and the main problems of food consumption in the EU.

Excessive purchasing, over-preparation and unwillingness to consume leftovers are some of the main antecedents of food waste [11]. Part of the population does not even know that their actions are harmful to the environment and influence the economic circumstances negatively, which can be caused due to their cultural mindset, different traditions and certain everyday consumption routines. Therefore, having an educational intervention to increase consumers' awareness of the importance of green consumption enhances the general approach towards food management, its preparation and planning processes, which results in a remarkable decrease in food loss and waste levels [36,38]. Unreasonable food waste is promoted by improper planning of purchases and portions of food to be prepared, promotional shopping and so on [1]. Our study highlights that respondents are quite responsible, they do not make too much food and usually in a café/restaurant they order as much food as they can eat, but sometimes buy unplanned products at a discount.

In line with other research [11,14], we found that fruit and vegetables are the product group most commonly wasted. The fish and eggs food groups, which make up the smallest parts of the food supply chain, also generate the lowest quantities of food waste in absolute terms, despite the fact that much of these food groups (50% and 31%, respectively) go to waste [14,39]. The survey data also revealed that fish, eggs, grains and pasta are wasted less in the food supply chain.

Our findings also reveal that shopping habits of the respondents are quite different. Based on some research, customer shopping habits changed during the pandemic and unplanned shopping increased [34]. Our study revealed that the majority of the respondents go shopping for food two or three times per week, they plan purchases by making a list. It is very important to shop smartly and realistically [11].

Composting is an important element in sustainable waste management [40–42]. We find out that most of the respondents do not compost food, but would like to do that. In Lithuania, the amount of food waste in the mixed municipal waste is about 15%, with an average of 41 kg of food waste per person per year. Across the European Union, somewhere between 118 and 138 million tons of bio-waste arise annually, of which currently only about 40% (equivalent to 47.5 million tons per annum) is effectively recycled into high-quality compost and digestate [43].

#### **6. Conclusions**

The survey found that the majority of respondents state that food is not often wasted. This makes it a little more optimistic that global food waste and sorting problems will be addressed through people's awareness and real action efforts. The results of our survey show that the population of Lithuania emits slightly less food than the population of the European countries participating in the survey. Clearly, food wastage is not just a problem of family-specific intolerance, it is a global food security problem. This problem is directly linked to climate change, waste sorting and recycling, and other global ecological and economic or social problems. It is possible to notice the crucial role of educating and informing people. This should be the responsibility of national governments when allocating funds to educational programs. These programs should cover all age groups, from kindergarten to advertisements, flyers and conversations with adults. Adults would probably best understand and stimulate economic interest, with an emphasis on saving food and then composting food waste, because, unfortunately, not all adults are able to adequately assess the effects of climate change and their food supply needs. Consumers should also purchase food avoiding shopping routines and try to plan their food basket more so that they do not end up wasting edible food. On the other hand, generally, there are not many messages towards sustainable consumption in the majority of retailers; the most famous food retailers arrange communication strategies starting from their commercial goals rather than toward a zero-waste responsible behavior [44].

However, in line with recent research [25,26], it is possible to notice that the COVID-19 pandemic has affected consumer food habits, their management and consumption that, in turn, has reduced the household waste. Nowadays, most consumers try to define a meal list before shopping [27]. This increased awareness about food waste and its impacts on the environment helps reduce the quantity of spoiled and thrown products even if purchases have increased during the COVID-19 pandemic [28–31].

In line with [22], results highlight that adequate tools for reducing food loss and waste can become crucial to make green food supply chains. If, on one hand, it is relevant to implement infrastructures for recycling disposed products, on the other hand, training and education can shift habits towards responsible consumption as well as ethical consumption (i.e., by means of donations and food banks).

One of the best uses of discarded food is feeding livestock, saving precious resources that would have otherwise been used for producing commercial feed. If the food cannot be reused at all, we should at least try to recycle or compost it in a responsible manner instead of sending it to the landfills where it continues to rot [11]. The draft State Waste Prevention and Management Plan 2021–2027 by the Ministry of Environment of Lithuania defines tasks and goals for implementing separate collection of food and kitchen waste by 31 December 2023. High-quality compost used in agriculture has to be made from the separately collected bio-waste, and the restoration of areas for the preparation of energy plant media has to be vulnerable [45]. However, some findings show that people do not try to compost food and throw it away with other waste. The benefits of composting are significant: through composting, the quantity of garbage direct to the landfill is reduced, the organic matter is reused rather than dumped and it is recycled into a useful soil.

Composting can be defined as natural processes of recycling organic products such as leaves and food scraps into fertilizers that can enrich soil and plants. Recycling food and other organic waste into compost provides a range of environmental benefits, including improving soil health, decreasing greenhouse gas emissions, recycling nutrients, and mitigating the impact of climate changes. Composting can appear as much as an art as a science. Recent research and policies about managing wastes and producing food in an environmental way highlight a new interest in small-scale backyard composting as well as an interest in developing large-scale commercial and municipal composting systems.

Regarding research limitations, it can be noticed that only those respondents who use an internet connection and in the network of authors could participate in the study due to COVID-19 restrictions. The study makes uses also of a random selection of the respondents, so in the future it would be useful to do research that would cover all age groups and other demographical characteristics.

Future research direction aims to repeat the survey with as many European respondents as possible, to assess and understand food consumption habits and knowledge in the food waste chain and to make the widest possible range of consumers aware of the consequences of irresponsible food waste.

**Author Contributions:** Conceptualization, M.F., J.P. and L.Š.; methodology, M.F., J.P. and L.Š.; software, J.P. and L.Š.; validation, J.P. and L.Š.; formal analysis, J.P. and L.Š.; investigation, M.F., J.P. and L.Š.; resources, M.F., J.P. and L.Š; data curation, J.P.; writing—original draft preparation, M.F., J.P. and L.Š.; writing—review and editing, M.F., J.P. and L.Š.; visualization, J.P.; supervision, M.F. and L.Š.; project administration, L.Š.; funding acquisition, M.F. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** The study was conducted in accordance with the Declaration of Helsinki.

**Informed Consent Statement:** Informed consent was obtained from all subjects involved in the study.

**Data Availability Statement:** The data that support the findings of this study are available on request from the co-author J.P.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**

