VARIABILITY OF FARM INCOME IN PLANT PRODUCTION FARMS IN THE PERSPECTIVE OF COMMON AGRICULTURE POLICY REFORM

Ewa Kołoszycz, Artur Wilczyński
Department of Management, Faculty of Economics, West Pomeranian University of Technology Szczecin, Poland

ABSTRACT

The research was concerned with assessment of the level of income from family farms specialized in arable crops in the years 2014–2020. The analysis assumed a stochastic nature of yields of chosen crops and their prices on the basis of historic data. Using data and measures of the Polish FADN, five models of farms were created and simulations were carried out addressing also the problem of the influence of implementation of agricultural practices beneficial for the climate and the environment on the economic situation of farms. The obtained results indicate that nominal income from the family farm in the majority of farms will remain at the level similar to the income in the year 2014. Agricultural practices beneficial for the climate and the environment, as the new instrument of Common Agricultural Policy, may influence both the structure of crops and the economic situation only in the case of farms of the biggest acreage.

Key words: risk, variability of yields, variability of prices of agricultural products, greening.

INTRODUCTION

Making predictions about the future is a complex process. Typically, it consists in gathering and analyzing historic data on a chosen system and then extrapolating the observed regularities to the future. In this manner it is possible to research reactions of a system to changes in its structure and/or its environment. The process becomes more complicated with the increase of instability and complexity of the environment. These problems are also present in preparing forecasts about production-economic situation of farms. As an established knowledge says, agricultural production has a certain specificity resulting from its specific features, both primary and secondary [24]. This specificity gives rise to the risk that – at a distance of years – any decision taken in the farm may prove wrong.

In the process of making predictions about the future, quantitative methods are widely used, mainly of mathematical-statistical nature, including those based on the statistic-economic models. Employment of the deterministic and stochastic computer simulation methods becomes increasingly common. In deterministic simulations a small number of combinations of variables with probabilities a priori is admitted. Experimental results obtained in this manner give fragmentary or rather pointwise reflection of possible effects in the researched system, e.g. a farm. For that reason stochastic simulation methods using random or pseudorandom number generation are employed. In the particularly popular Monte Carlo method it is assumed that chosen input data are of random nature and generate a certain scope of achievable results. These methods allow simultaneous analysis of the influence of many random variables on the obtained results.

Forecasts based on stochastic models are widely used in agriculture. Such models are employed mainly to research impact of changes in farm risk management strategies on the incomes gained [2, 7], and to analyze factors influencing the level of economic results achieved by farms [3, 4, 6, 8, 17, 18, 20]. Analyses covered different types of farms and a considerable share of research concerned plant production, where random factors (and climatic factors in particular) significantly influence the production-economic results. Such studies are carried out especially in periods of market changes, mainly due to desire to explore of the impact of applying new mechanisms affecting the economic situation of farms. The reform of the Common Agricultural Policy creates the need to assess its impact on the farms. The future agricultural policy should be based on three objectives: viable food production (income support and safety net mechanisms for producers), sustainable management of natural resources and climate action (improved the integration of environmental requirements) and balanced territorial development (support for rural development across the EU) [12]. The principal instruments used in the new-CAP are affected by economic, environmental and territorial factors [1, 10, 15, 23]. From 2015 onwards, the CAP introduces new policy instrument in Pillar 1, the Green Direct Payment (greening). This payments will be linked to the respect of three obligatory agricultural practices, namely maintenance of permanent grassland, ecological focus areas and crop diversification [19]. The existing literature provides only limited information on the impact of greening on farm income. Although a few studies have examined link between new-CAP instruments and profitability [9, 22, 26, 27], little empirical evidence exists to inform about details of the economic situation of different types of farms. Similar attempts have been made for arable farms in Italy [11, 26] and Germany [14]. The results of these studies indicate the negative greening effect – decreasing gross profit margin according to the farms organization, their specialization and location, with stronger impacts for the intensively operating farms.

For these reasons it was decided in this research to analyze the influence of variability of yields and prices of produced crops on economic results of plant production farms, taking account of changes of the Common Agricultural Policy regulations in the years 2014–2015. The objective of the research was to determine the risk of not gaining any income from family plant production farms of different production size in the years 2014–2020.

SAMPLE AND METHOD

With the objective of determining distribution of values of income from the family farm the Monte Carlo simulation method was used. This method has its specific procedure, which was carried out in this research in four basic stages:

1. Creating a model of farm. In this research the TIPI-CAL model was employed. TIPI-CAL (Technology Impact and Policy Impact Calculation) is a multi-period recursive model allowing deterministic and/or stochastic simulations of changes in farms [16]. Five models of farms were created with the use of data and measures from farms participating in the Polish FADN in the year 2011 as published by the IERiGŻ (Institute of Agricultural and Food Economics) [13]. Each model of farm reflected the production-economic capacity of a group of farms specialized in arable crops, separated according to the farmland area size categories used in the FADN (Tab. 1). Data and measures of model farms from the year 2011 were the starting point for further analyses. Basing on these data and measures for each model farm there was established, inter alia, the structure of crops and the value and structure of capital. It was assumed that investments in fixed assets (not including farmland) will be only simple replacement investments allowing continuation of farming without changes in production technology. Value and costs of production after the year 2011 were determined on the grounds of price indices of individual products and inputs (calculated in relation to the level from the preceding year) using to this end statistical data published by the GUS (Polish Statistical Office) and forecasts of the World Bank [28] and OECD-FAO [21]. Designed models farms have some limitations. Firstly, the model fails to optimize production. The structure of resources and the production volume does not change in the model throughout the period of analysis. The model used a nominal values, not taking into account the inflation. Another limitation is that due to the limited availability of historical data from farms in the model used in part of projections historic data from the general statistics.
2. Defining random variables in the model and their distributions. In this research it was assumed that among many variables shaping economic results of plant production farms yields of crops and their prices are of random nature. These are variables which cannot be influenced by a farmer or an influence of a farmer on these variables is very limited. Both variables often occur in research concerning income risk in farms [2, 3, 5–7, 17]. Analysis covered yields and prices of basic farm crops: wheat, rye, barley, oats, rapeseed, sugar beets and potatoes. With the objective of determining parameters of distributions of yields and prices data from the GUS on average historic yields in voivodships in the years 1999–2012 were used. In the case of crops’ yields normal distribution was used assuming that a yield cannot be lower than zero. In the case of prices there was used triangular distribution characterized by three parameters: the minimum value, the maximum value and the most probable value, i.e. the modal value. The minimum value was assumed to be the chosen crop’s lowest price observed in all voivodships in the years 2005–2012 and the maximum value – the highest price from the same set of data. The most probable value in the years 2014–2020 was established on the grounds of forecasts of the World Bank [28] and OECD-FAO [21] for particular years with the use of chain indices, thus preserving the differences in prices obtained by each of the model farms. On the grounds of data from the GUS there was also determined the correlation between yields and prices of crops using to this end Pearson linear correlation coefficients. Correlation coefficients were of deterministic nature throughout the analyzed period.
3. Carrying out simulations. @Risk 6.0 software was used for simulations. For each model farm ten thousands iterations were made, which allowed to determine precisely the distribution of probability of income from each of the analyzed family farms for each year from the period 2014–2020.
4. Determining the distribution of probability of income from the family farm and interpretation. To present and compare the obtained results box plots were used containing information on the location, dispersion and shape of the distribution of income from family the farm.

The research made it also possible to assess the impact of changes of requirements for the farms applying for subsidies to operational activity taking effect from the year 2015. According to the adopted new regulations of the Common Agricultural Policy, besides meeting cross-compliance requirements, farmers will be also required to undertake agricultural practices beneficial for the climate and the environment (referred to as ‘greening’). The relevant scenario was created to assess the influence of this regulation on economic results of farms. Analysis of farmland structure of farms as to meeting the said requirements was conducted and if a farm was not meeting the requirements, then setting aside was assumed in the required acreage. In addition, it was assumed that after the year 2017 the European Commission would increase the required share of ecological focus areas in farms from 5% in the year 2015 to 7% in the year 2018.

RESEARCH RESULTS

Simulations encompassed the determination of future shape of the distribution of probability of income from the family farm in two scenarios of activity of the model plant production farms. The first scenario was the reference scenario, which did not assume undertaking agricultural practices beneficial for the climate and the environment and which was the baseline for the other scenario. The second scenario assumed meeting by the farms all the Common Agricultural Policy requirements adopted for the years 2014–2020, which demand, beginning with the year 2015, providing the ecological focus area in farms of the area of more than 15 hectares of arable land. Such an approach allowed to indicate the expected profitability differences resulting from the introduction of the new Common Agricultural Policy mechanisms determining considerably operation of plant production farms. Detailed statistics presented results are given in Appendix A.

Results obtained for model farms M-8 and M-15 do not include the scenario requiring creation of the ecological focus area because the area of arable land in these farms does not exceed 15 hectares (Fig. 1, 2). In the case of the smallest farm in terms of the area of arable land (M-8) it can be observed, that the expected value, determined with the mean value of the possible results obtained for the distribution of probability, is continuously on the decrease. In the year 2014 the mean value for the distribution of income from the family farm was around PLN 14 K and in the year 2020 the expected value decreased to the level close to PLN 1.5 K.

Results of the research obtained for the M-8 farm demonstrate also that, beginning with the year 2018, 50% of observations include the results, where there exists a probability of gaining negative income from the family farm. In the first year of the analysis the interquartile range (H-spread) of the researched income was always above zero but in the year 2020 around 40% of the obtained results of simulations were in the area below zero.

The distribution of probability of future income in the farm M-15 indicates that between the years 2015 and 2020 the income can be expected to be stable despite of the risk related to fluctuations of yields of crops and sale prices of crops (Fig. 2). In the case of the farm M-15, the area of which is nearly two times bigger than the area of the farm M-8, results remaining between first and third quartile indicate positive values of income from the family farm. Examination of the dispersion of distribution of profitability in the two above analyzed farms allows to observe that in the case of the farm with smaller area of arable land such dispersion is clearly lower.

According to the rules of the Common Agricultural Policy the model farm M-25 will have to maintain a share of its arable land as the ecological focus area. Such situation will arise as from the year 2018, when the decision, if any, as to the increase of such an area from 5% to of 7% will be taken.

From the distribution of probability of income from the family farm for the farm M-25 it follows that its expected value in the scenario including the agricultural practices beneficial for the climate and the environment will be characterized by nearly identical levels throughout the years 2015–2020 (Fig. 4). It should be also observed that in the situation where in the analyzed farm a share of land will be set aside for the ecological focus area the expected income will be lower in comparison to the reference scenario (Fig. 3, 4). For the expected value the difference will be very small and it will be around PLN 1.5 K.

Comparing the dispersion of the results obtained for the farm M-25 with the results for the farms M-8 and M-15 it can be indicated that the diversity of the values obtained in the simulations is bigger in the case of the farm M-25. It means that this farm is much more susceptible to changes of the income driven by fluctuations of the individual risk factors, which in this case are yields and procurement prices of crop products.

Just as in the case of results obtained for the farm M-25 setting aside a share of land for the ecological focus area by the farms M-40 and M-116 and the increase, if any, of this share beginning with the year 2018 will influence the decrease of the expected value of income from the family farm (Fig. 5–8). The difference between the mean value for the distribution of income between the researched scenarios will amount to around PLN 2 Kin the case of the farm M-40 and in the case of the farm M-116 it will reach the amount of PLN 10 K. An analysis of distribution of probability of future profitability in farms M-40 and M-116 allows to indicate that variability of yields and prices of crop products and introduction of the new rules of the Common Agricultural Policy will influence the deterioration of profitability in these farms. In the year 2014 the value of the expected income is higher than in the next years of operation of these farms.

As it was observed above production size influences the dispersion of results of the conducted simulations concerning the probability of future shape of profitability in the researched farms. In the distribution obtained for the model farm M-40 the interquartile range in the years 2014–2020 remains within the limits from PLN 28 K to PLN 36 K and in the farm of more than three times bigger acreage of arable land it ranges from PLN 98 K to PLN 112 K..

CONCLUSIONS

Future shape of the Common Agricultural Policy introducing new rules governing agricultural markets and making farm support conditional on meeting additional requirements will significantly determine future economic situation of farms. Agricultural practices beneficial for the climate and the environment will be invested with special importance. Results of this research indicate that the requirement of crop diversification will be of little importance for Polish farms. Basic premises for this state of affairs are the present structure of crops and the legislation treating spring and winter crops of the same species as separate types of crops. Analysis of mechanism related to maintaining the ecological focus area (EFA) in farms indicates that strength of this mechanism’s influence on future profitability should not be significant either. The number of elements counted as ecological focus areas should also facilitate meeting this requirement. In the biggest of the researched farms, which works around 115 hectares of arable land, the increase of the ecological focus area to 7% in 2018 would mean the additional 3 hectares of land allotted for the area.

The research on future distribution of probability of income from the family farm demonstrated that only in the model farm of less than 10 hectares of arable land the constant decrease of profitability may be expected in the researched period. The expected value in the year 2020 will be close to zero and the obtained simulation results remaining within first quartile indicate the loss on farming. It could be presumed that in the future such farms might be eliminated from the market as they will be unable to make necessary investments allowing for development of production.

In the case of other farms the distribution of probability of income from the family farm indicates similar levels of the income in the years 2015–2020. The results show that the implementation of practices related to greening will not increase the dispersion of income in farms.  However, such situation cannot be considered advantageous. The forecasts of changes of yields and procurement prices of crop products combined with the instruments of the Common Agricultural Policy provide the stabilization of profitability of the nominal nature only. Adjusting the results by inflation would produce lower values of the income. Results of simulations indicate also that in the year 2020 it could be expected that income from the family farm would be close to the level from the year 2014 with lower values of the income in the years 2015–2019.

Responses to this article, comments are invited and should be submitted within three months of the publication of the article. If accepted for publication, they will be published in the chapter headed 'Discussions' and hyperlinked to the article.