Printer Friendly

Farm productivity and profitability: a comparative analysis of selected new and existing EU member states (1).


The 1990s witnessed widespread changes in farm structures, government policies and agricultural markets in the Central and East European Countries (CEECs) that are New Member States (NMS) of the EU. These changes have resulted in a more differentiated set of farming systems that have to deal with the effects of international trade liberalisation and competition in an enlarged single European market. In view of the enlargement of the EU, there has been a growing interest in the competitiveness, productivity and profitability of farming in the CEECs. More specifically, farm performance has been seen as critical in several debates about the implementation of the CAP in the NMS, such as how farm structures and the agricultural labour force will evolve in the region, whether collective farms and their successor forms will survive in mature market conditions and how adoption of the CAP will affect farm profitability.

This paper presents an overview of the key findings of research undertaken on the performance of CEEC farms, which sought to shed light on these issues. The research investigates the private profitability and total factor productivity (TFP) of farms classified by several variables including size, legal type and agri-environmental region. This is used as the basis for a discussion of the overall survivability of farms in different countries and, thus, the likelihood of future restructuring. The study focuses on three CEECs, namely the Czech Republic, Hungary and Poland and concerns both corporate and individual farms. In order to compare the findings from the CEECs with the situation in existing EU Member States, similar analysis was conducted for two contrasting regions, the region of Navarra in Spain and South-East England. These areas were chosen to reflect the diversity of agricultural regions that already exist within the EU: Southern and mountainous agriculture in Navarra and North European lowland agriculture in England, and small family farms in Navarra and much larger farms in England using a high proportion of hired labour.


At the outset of transition, agriculture in the CEECs was widely perceived as inefficient (Brooks et al., 1991). Three main factors were believed to account for inefficiency: (a) inappropriate farm sizes, (b) the weakness of state- and collective-owned farms as an organisational type and (c) central planning. The centrally planned environment insulated farms from market signals and by providing enterprises with soft budget constraints created a disincentive to improve efficiency (Lerman et al., 2001). Dismantling the command economy, liberalising prices and instituting hard budget constraints were therefore seen as mechanisms for improving productivity in the agricultural sector of the CEECs. However, improvements in efficiency were seen as achievable not just from macroeconomic reform but also from 'micro-level' changes in farm size, organisational type and other agency factors (Lerman et al., 2001).

As the main specific features of the CEECs at the outset of transition were perceived to be the large size of commercial farms and the dominant share of agricultural land managed by cooperative and state farms, size and organisational type have been the most frequent factors considered in investigating variations in performance (Hughes, 2000a). While other factors are clearly also important in determining performance (such as agri-environmental characteristics, endowments of human capital and the nature of up- and downstream markets) (Mathijs and Vranken, 2001), these issues have received comparatively less attention (for a review, see Gorton and Davidova, 2004). In this section of the paper, the CEEC-specific debate on size and organisational type is introduced and previous analyses of variations in agricultural performance in the Czech Republic, Hungary and Poland discussed. This provides a basis for the cross-national comparative analysis presented in Sections 'Methodology' and 'Data Sets'.

Farm size

Former Czechoslovakia and Hungary were characterised by a bi-polar farm structure. Agricultural land use in both states was dominated by large state and cooperative farms supplemented by small-scale auxiliary plots. By the end of the 1980s in the Czech Republic, cooperative and state farms accounted for 61 and 25 percent of the total agricultural area, respectively. For the same year in Hungary, cooperative and state farms accounted for 75 and 15 percent of the total area, respectively. In both countries, the mean farm size of state and cooperative farms was in excess of 4,000 and 2,000 ha, respectively. In contrast, Poland never extensively collectivized, and during the 1980s, private farms accounted for approximately 75 percent of agricultural land and a similar share of agricultural output (GFA, 1997). These private farms were small by Western standards (over 2 million units with approximately 1.4 million ha). Regarding the socialised farms that were established in Poland, state farms were in size and employment levels similar to those in Hungary and Czechoslovakia but the cooperatives were smaller (mean of 297 ha in 1985) (GUS, 1990).

The size of state and cooperative farms was widely perceived as sub-optimal (Brooks et al., 1991). Such large farms, it was argued (Schmitt, 1993), suffered from high transaction costs due to the large number of workers employed in absolute terms and per hectare (Lerman et al., 2001). With such a large hired labour force, the costs of monitoring effort were seen as a cause of diseconomies of scale. In Poland, the persistence of small-scale peasant farms was also seen as sub-optimal in that economies of scale were not realised (GFA, 1997). In short, the bi-polar structure, it was argued (Lerman et al., 2001), meant that agricultural land use was dominated by collective farms that were too large and peasant farms in Poland or auxiliary plots in the Czech Republic and Hungary that were too small.

The relationship between efficiency and farm size in the post-communist era has been extensively studied and analysis for the Czech Republic, Hungary and Poland is reviewed below. Hughes (2000a), through a comparison of total factor productivity, found evidence of economies of scale for arable farms in the Czech Republic for up to 750ha and little evidence of diseconomies of scale above this threshold. For Hungary, in contrast, he found that diseconomies of scale did appear to set in above 500 ha, supporting the view that the communist era collective farms were excessively large. In Poland, empirical work has concentrated on individual farms. In an early study using data for 1993, van Zyl et al.'s (1996) TFP analysis indicated that individual farms that were relatively large by Polish standards (above 15 ha) were, on average, less efficient than farms below 15 ha in size. This result was surprising for the authors in that it appeared to show that the smallest peasant farms were not less efficient. However, given the absence of significant numbers of individual farms of above 25 ha, it could not be concluded that economies of scale did not exist for sizes outside the sample range. In addition to estimating TFP, van Zyl et al. (1996) also applied data envelopment analysis (DEA). From the DEA analysis, they discovered no significant differences in scale efficiency between farm sizes. Latruffe et al. (2005), using more recent data and applying DEA analysis, found, however, a significant difference in terms of scale efficiency with the smallest farms (between 1-2 and 2-5 ha) being the least efficient for both crop and livestock production.

The empirical evidence on the relationship between farm size and efficiency is less clearcut than many supposed at the outset of transition. There is both mixed evidence on the view that the state and collective farms were too large (some support for Hungary but not the Czech Republic) and that the peasant farms in Poland were too small (only for very small farms under 5 ha is there evidence of clear inefficiencies of scale). These findings point to the importance of factors other than size in determining efficiency and that there is no clear cross-national optimal farm size.

Organisational type

In addition to inappropriate size, state and cooperative farms were widely seen as an inefficient organisational form (Schmitt, 1993). Institutional economists, in particular, have argued that family farms are a superior organisational type in agriculture in that they minimise transaction costs and are therefore more efficient than cooperative, state-owned or corporate enterprises (Schmitt, 1993; Hagedorn, 1994). Family farms, it is argued (Pollak, 1985), do not suffer from a principal-agent problem in that incentives for workers are internalised as the family provides both management and labour. As the costs of supervising and monitoring hired labour in agriculture can be high:

'All production cooperatives based on member labor suffer from shirking and free riding, which are induced by strong moral-hazard behavior among the members. In socialist agriculture, these weaknesses were further aggravated by monitoring and enforcement difficulties associated with size (also a well-known universal factor) and by the evils of the administrative command system (a unique feature in the socialist countries)'.

(Lerman et al., 2001, p. 33).

As with the assumptions on size, a number of empirical studies have analysed the relationship between organisational type and efficiency. These studies have looked at the differences between individual and corporate farms, where the latter category includes production cooperatives, joint stock companies and limited liability firms. Most corporate farms have their origins in the state and cooperative farms of the communist era and a review of the transformation of collective farms in the Czech Republic and Hungary is presented in Hughes (2000b) and Csaki and Lerman (1998), respectively.

Using data for the mid-1990s and applying TFP and DEA analysis, both Hughes (2000b) and Mathijs and Vranken (2001) found that, when other factors were controlled for, family farms in Hungary did appear to be more efficient than their corporate counterparts. In the Czech Republic, a similar significant difference was found for livestock farming but not crops. Curtiss (2002), who applied Stochastic Frontier Analysis (SFA) for analysing arable farms in the Czech Republic, found that cooperatives performed better than individual farms in wheat and rapeseed production but that the latter were superior for sugar beet cultivation.

As with farm size, the evidence on the relationship between efficiency and farm type is therefore not clearcut. While there is support for the initial propositions for Hungary, in the Czech Republic the results are more complex. Moreover, even where the average corporate farm is less efficient than the average family farm, some cooperatives and companies perform well. This suggests that at least some corporate farms can solve the governance problems discussed by Lerman et al. (2001).

While empirical work to date has been informative, further research can be justified on two counts. First, the research to date has been based mainly on data for the early and mid-1990s. This was a period of immense change in agriculture with a widespread and significant drop in output throughout the CEECs (OECD, 2000). In some countries, reforms were delayed and changes in ownership and management recorded in official statistics masked the developments at the farm level (Hughes, 2000b). It is therefore important to see if the trends identified for the early and mid-1990s reflect merely the characteristics of initial transition or more durable phenomena. Second, the studies reported above analyse variations in relative efficiency on a country-by-country basis, identifying the characteristics of farms that are relatively more efficient in a particular sample. However, this offers little insight into the comparative cross-country picture and the profitability of farms and returns on resources employed. An analysis of profitability is important to understand the probable nature of future restructuring in the sector, especially when compared against existing EU Member States. These objectives guide the methodology presented below.



Farm profitability is analysed through the estimation of ratios between the costs and revenues for each farm. A ratio smaller than one indicates a profitable farm and vice versa. The use of ratios has been preferred as it simplifies cross-farm and cross-country comparisons. Costs include labour, land, capital (depreciation and interest) and intermediate consumption. The revenue side includes proceeds from the sale of agricultural products, the value of non-marketed agricultural output, proceeds from other activities and net current subsidies. Revenues from other activities refer to proceeds from gainful activities that are inseparable from the main farm accounts (Tanton and Williams, 2000). As the share of the revenue from these other activities is rather small, for example, 3.9 percent in Navarra and 8.4 percent in Hungary, and as farm accounting records do not split the costs of these activities but treats them as integrated into total farm costs, the overall revenue and costs for each farm were taken into account in the analysis. The most important source of 'other revenue' in the CEECs is from renting out agricultural land.

Three cost-revenue ratios for each farm have been calculated. The central ratio used as a reference is the private cost benefit ratio (P_CB). For the ith farm, the P_CB is taken to be:

P_C[B.sub.i] = ([C.sup.t.sub.i] + [C.sup.f.sub.i])/[R.sub.i] (1)

where [C.sup.t.sub.i] is the cost of tradable inputs, [C.sup.f.sub.i] is the cost of non-tradable factors of production (based on private prices or estimates for non-paid land and labour input) and [R.sub.i] is revenue excluding current subsidies net of taxes. This ratio provides a framework for evaluating profitability when the full opportunity costs of all factors are assessed. The initial data did not include a notional rent for owned land and wages for non-paid labour input. For this reason, for non-paid land and labour input, a set of shadow prices were estimated using regional averages. Family labour was valued using the average regional farm unit labour costs. Farm labour costs were used as it was assumed that most of the farmers in the studied CEECs had low opportunity costs and their second best alternative would be to become farm workers. Given significant spatial variations in wage rates, adjustments were made at the regional level. As far as land was concerned, if a farm had a mix of rented and owned land, the rent paid was imputed to the owned land, as it was assumed that rented and own land were in close proximity, and thus, were of a compatible quality. If a farm did not rent land, then the average regional rent was applied to the owned land.

Two other profitability ratios were also calculated. The first, cost-revenue plus subsidies (C_Rs), exactly matches the entries in the EU's Farm Accountancy Data Network (FADN) that was transposed to the CEECs and, therefore, [C.sup.f.sub.i] does not include estimates for non-paid labour and land, and [R.sub.i] includes the net current subsidies. This ratio is used as an indicator of farm survivability. If farmers can cover their paid costs, they may continue farming even though the returns to their own factors might be very low or zero (Ellis, 1988). The final ratio, cost-revenue without subsidies (C_R), does not include estimates for non-paid labour and land, and also excludes direct subsidies. The rationale for calculating this last ratio is to give an insight into the effect of the direct budgetary transfers on different farm types and between countries. It is expected that the ranking of farms will change with the use of different ratios depending on their integration into factor markets, namely the role of hired labour and rented land in the farming process. Such effects are deliberately sought in order to provide a complete assessment of the economic profitability, survivability and dependence on government transfers of farms in different countries and with varying characteristics. As the approach is static, little, however, can be said about dynamic adjustments to changes in policy.

In addition to the ratios used in the profitability analysis, some standard financial ratios (Debt to assets, Leverage, RENGO and RENGM) were constructed on a farm-by-farm basis. The Debt to assets ratio represents the proportion of assets owed to creditors. Leverage is the relationship between the total loans and the net worth of the farm. It measures the degree to which debt is used to finance the farm business. RENGO is the ratio of rental costs (rents plus interests paid) to gross output and RENGM is the relationship between such rental costs and the gross margin. The last two indicators are measures of 'financial stress'--the pressure placed on a farm by the repayment of rent and interest (Franks, 1998). All value units are expressed in euros for ease of international comparison.


Productivity differences were estimated by the construction of a Tornqvist-Theil TFP index for all farms in the sample relative to a base case 'average farm' with results interpreted relative to the sample mean, showing groups of farms having above or below average TFP scores. The Tornqvist-Theil TFP index is recognised as a measure of technical efficiency and is considered to be an acceptable alternative to econometric estimation in cases where the data do not permit an underlying production function to be estimated (Capalbo and Antle, 1988). The Tornqvist-Theil TFP index applied here is a relative measure of productivity, comprised of the difference between an aggregated output index and an aggregated input index. Supposing there are two firms i and b that produce n outputs Qj (j = 1, ... n) using m inputs Xk (k = 1, ... m), then the index t can be defined as:


where for firm i, [R.sup.i.sub.j] represents the share of the value of the jth output in the total value of all n outputs, and [S.sup.i.sub.k] represents the share of the costs of the kth input in the total input costs of all m inputs. Two TFP indices have been calculated, one including estimated costs for own land and labour (TFP1) and one with paid costs only (TFP2).


Data were extracted from FADN surveys, which are implemented in all EU Member-States and some New Member States of the EU. Derived from national surveys, FADN is an important source of micro-economic data and is widely used for farm-level analysis. It is broadly representative for commercial agricultural holdings. (2) As the survey does not cover all agricultural holdings except only those that are of a size that can be considered as commercial, FADN is biased towards larger holdings in comparison to the total farm population. Therefore, the FADN sample excludes subsistence producers and this sector is not discussed in the scope of the paper. Although the subsistence sector is important in some CEECs, studying larger farms and excluding the purely subsistence ones is adequate for a comparison with the EU, as the very smallest 'farms' are likely to continue to produce for self-consumption, and to be less integrated into the market and exposed to competitive pressures even after accession (Kostov and Lingard, 2002).

FADN is still being piloted in Poland and at present the only national, annual, farm survey is conducted by the Polish Institute of Agricultural and Food Economics (IERiGZ) (3), and this has been used as the main source of data. In the UK and Spain, FADN data collection is not organized on a national basis but through a network of regional surveys. While aggregations of a limited range of variables are available at national level, it is difficult to collate individual farm data at a national scale. For this reason, one region was selected from Spain and the UK in order to reflect some of the variations in farm characteristics and natural conditions existing in the EU-15. Table 1 details the main characteristics of the datasets used in the analysis.

In the Czech Republic, the initial sample included 1,087 agricultural enterprises of physical and legal persons, which collectively managed 887,026 ha. After checking the individual data, 264 farms were excluded due to missing or inconsistent information, so the analysed sample included 823 farms. Considering management form, the largest group in the sample are individual farms, numbering 513 (62 percent) with an average size of 134 ha. Producer cooperatives are the second largest group, 154 (19 percent). The rest of the sample is made up of 95 joint stock companies (12 percent) and 61 limited liability companies (7 percent). The average size of corporate farms (cooperatives and companies) is 1,526 ha.

Hungary's FADN provides useable information on over 1,100 agricultural enterprises (individual and corporate farms). The 233 corporate farms that were included in the sample were made up of 21 partnerships, 66 limited liability companies, 10 joint ventures, 71 cooperatives and 65 other legal forms. For both Hungary and the Czech Republic, the main difference between the FADN and Agricultural Census returns is the lack of 'farms' below 1 hectare in size.

The Polish sample included 1,001 observations of only individual farms, as corporate farms have played a far more limited role than in the Czech Republic and Hungary. While it would have been beneficial to cover the corporate farm sector, such enterprises are not included in the IERiGZ database and as individual farms account for approximately 80 percent of Utilised Agricultural Area (UAA) in Poland, the survey does cover the backbone of Polish agriculture. A close examination of the data brought about the removal of 22 farms and the analysis was carried out with 979 observations. Classified by UAA, the highest proportion of the sample farms was between 10 and 25 ha (41 percent). Farms above 100 ha accounted for only 4 percent of the sample.

Finally, regarding the data available, it should be noted that although the conclusions are affected by the sample size, the existing samples were large enough to justify quantitative analysis. However, attention should be paid to the differences in data collection procedures between countries (especially in the allocation of fixed costs). While most New Member States are harmonising their own surveys with FADN procedures, this is still an on-going process. The cross-national analysis of data should therefore be seen as a way of highlighting broad trends and differences rather than giving pinpoint results.

As mentioned above, in the Czech Republic and Hungary, there are two major management types, individual farms and corporate farms, and different organisational types within the corporate group. In theory, there should be substantive differences between different corporate forms, particularly in their decision-making process (Hughes, 2000b). However, in the CEECs such differences are frequently far from clearcut. For example, often cooperatives do not apply the 'one man one vote' principle and their operation is similar to those of companies. In several joint stock companies, managers are also the largest shareholders. In addition, the pace of change in organisational structures is still strong in the CEECs. For these reasons, and having in mind the relatively small numbers of different corporate types, the emphasis is not on explaining variations in the performance of different corporate forms but mainly on comparisons between individual and corporate farms, and cross-country comparisons.


Table 2 compares the sample farms according to four sets of variables: size, rented factors, intensification and income variables. Size variables are the utilised agricultural area (UAA) per farm, the value of output and total assets per farm, and labour input measured in annual work units (AWU). Rented factor variables are the shares of hired labour and rented land in total labour input and land utilized, respectively. Also two measures of relative factor use are shown in Table 2. The first one is the amount of land per AWU with larger scores being an indicator of less intensive agriculture. The second measure is the value of depreciation per annual work unit (DEPAWU), in which case higher values are used as a proxy for greater capital per worker employed. The returns to hired labour are expressed as wages paid per hired AWU.

Table 2 shows that according to the four size measures (average UAA per farm, average output, total assets and labour input), the countries fall into three groups. These three groups are: the largest farms (the Czech Republic), medium size farms (Hungary and South-East England, although measured by assets, the South-East English farms are the largest) (4) and small farms (Poland and Navarra).

The main differences among the CEECs in farm size stem from the existence, or the lack of, corporate farms. Although corporate farms are widespread in both the Czech Republic and Hungary, there is a considerable difference between the two countries. The land area available to successor farms in Hungary decreased substantially during transition as a result of the adopted procedures of land reform and farm restructuring, especially because a large area of land had to be set aside for compensation purposes under the Compensation Act (OECD, 1994). Thus, although in the Hungarian FADN sample there were more than 200 corporate farms, 55 percent of them were below 300 ha. As mentioned, in the Czech Republic, the average size of the corporate farms was above 1,000 ha. In addition, in comparison to Hungary, corporate farms use a much higher share of UAA in the Czech Republic. These differences are reflected in the larger average area, output and assets per farm in the Czech Republic compared with Hungary.

The Polish farms are the smallest by all size measures. However, according to size they appear closer to farms in the EU (Navarra, Spain) than their counterparts in the Czech Republic. When assets are measured per ha, then the Polish farmers seem to be much better capitalised than farms in the Czech Republic or Hungary. This results from their longer history of independent farming. Not surprisingly, South-East England has the most capitalised farms, having assets per ha more than 60 percent higher than in Navarra and nearly three times that of Poland.

The differences between the countries regarding the use of land rental and labour markets are striking. The Polish farms rely almost entirely on their own resources. Only 6 percent of labour input is accounted for by hired labour and only 17 percent of total land is rented. Thus, they are dependent on the initial family endowment of resources and familial human capital. This lack of integration into factor markets is a clear indicator of the peasant character of Polish agriculture. Most of the farms in Navarra are located in marginal areas (Less Favoured Areas [LFAs] or former Objective 5b areas) (5). Less favoured areas include mountainous areas, which are in danger of abandonment due to low land productivity and difficult cultivation, but where the conservation of the countryside is a priority, and other areas where agriculture is affected by a specific handicap. Farmers in EU LFAs receive compensation payments to help ensure that they continue farming, maintain the countryside and promote a viable rural community. In addition, with an aim of achieving regional economic and social cohesion, in the 1990s the EU used Structural Funds to promote rural development in regions with a low level of socio-economic development which had a high share of agricultural employment, poor agricultural incomes or were subject to rural depopulation (Objective 5b). Although most of the sampled farms in Navarra are located in these marginal areas and rely on own labour, in contrast to Poland they also rent land in order to achieve a reasonable size and generate an acceptable income: family farms in Navarra have three times as much land per AWU than in Poland. The Czech Republic and Hungary, due to their corporate farms that depend almost fully on rented land and hired labour, are nearer to the English case of large family farms in terms of the extensive use of land rental and labour markets.

Another striking difference concerns the pay of hired labour. In this case, the clear divide is between existing and new EU Member States. Although theoretically accession to the EU may accelerate the equalisation of product and factor prices, the order of magnitude of the differences in agricultural wages is such that most probably a large gap will persist for a long time post-accession. Incentives for agricultural labour from the CEECs to move to work, at least seasonally, in West European farms, are likely to persist and this is a phenomenon that currently occurs.


The differences in farm structural characteristics bring about important consequences for the profitability of farming. Table 3 presents the average farm profitability for the sample farms in each of the analysed countries according to the three profitability ratios.

The private cost-benefit ratio is sensitive to the shadow prices applied to non-paid labour and own land. This is particularly important for individual farms that mainly rely on own resources. However, as shown in Table 3, even in regions where farming uses mainly own resources, as in Navarra, if the resources are effectively used, farms can be near the break-even point according to the private cost-benefit ratio.

The most profitable farms are in Navarra and Hungary. The fact that in this group there is one existing and one new EU Member State tends to undermine any easy generalisations. The profitability of farms in Navarra cannot be solely attributed to CAP headage and acreage payments and transfers received because of their location in LFAs or objective 5b areas. It is true that direct payments account for 13.6 percent of the gross output of Navarra's farms and that almost all sample farms receive direct payments (Table 4). However, the importance of direct payments in South-East England is not substantially different but the English farms are unprofitable on both the private cost-benefit and cost-revenue without subsidies ratios.

The Hungarian farms have the best prospects among analysed New Member States according to their profitability. They are near the break-even point on the private cost-benefit ratio and are profitable according to the other two ratios (Table 3). They achieve this profitability with more modest direct payments than in existing EU states. Net current subsidies account for slightly more than 5 percent of the gross output (Table 4).

For the Czech Republic and Poland, agricultural profitability is a major problem. The average scores for each of the three profitability ratios are above 1. Even without accounting for the opportunity costs of own resources, and including net current subsidies, 52 percent of the sample farms in the Czech Republic and 40 percent in Poland are unprofitable. While Polish farmers do not benefit from direct payments, the Czech farms receive more net current subsidies in relative terms than the Hungarian farmers, but nevertheless their private profitability is low.

Undoubtedly, agri-environmental conditions play an important role in farm performance. For example, the Czech farms are classified into five agri-environmental regions that reflect different conditions for farming, notionally called the maize, sugar beet, cereal-potato, potato and mountainous-forage regions. The best for agriculture is the first zone (maize region) and they are listed in descending order. In terms of agri-environmental regions, the worst results were recorded, not surprisingly, in the mountainous forage region where, on the basis of the private cost-benefit measure, no farms were profitable (Table 5). However, even in the best agri-environmental regions (maize and sugar beet) the majority of farms were loss making. In the cereal and potato regions, only 22 and 13 percent of farms were profitable, respectively, according to the private cost-benefit ratio.

The comparison of average profitability scores by legal form for the Czech Republic and Hungary (Table 6) reveals that Czech farms are uniformly loss making with the exception of individual farms on the cost-revenue plus subsidies ratio. The opposite is true for Hungary with only one ratio above 1, the private cost-benefit measure for individual farms.

When the results by region and legal type are considered together, the individual farmers in the Czech Republic register the best results in the maize region but they have one of the worst returns in the mountainous forage regions according to P_CB and C_R ratios (Table 7). Only when subsidies are accounted for in the revenue (C_Rs), can individual farms be identified as having the highest profitability in all agri-environmental regions.

The poor economic performance of the Polish and Czech farms is also clear from Table 8, where the net value added (NVA) per ha and AWU is presented. Their labour productivity is also low, in Poland NVA per AWU is nearly 12 times less than in South-East England and 14 times less than in Spain. This is in line with Pouliquen's estimates that the value added per worker in Poland is equal to only 8 percent of the EU level (Pouliquen, 2001). The best performers according to NVA are the Spanish farms. Hungarian farms again record good results.

The lack of profitability makes the long-term viability of a large number of Czech and Polish farms questionable unless they manage to restructure. The issue is even more serious in the Czech Republic due to the high level of farm indebtedness (Table 9). Czech farms are funded by debt and average debts are higher than the net worth of the farms (leverage above 1). However, their financial stress is not as high as would have been expected by their level of indebtedness, in fact it is less than in the two EU regions. This is because most of the debt is in the form of non-bank liabilities, held by the successor farms either to individual owners of the assets for producer cooperatives or to the state for limited liability companies. As a result of the adopted reform legislation, these farms did not need to repay these debts for several years after their establishment. (6) For this reason, the financial stress is lower than it would have been under similar situations in Western Europe.

Polish farmers do not rely on external financing either due to external constraints (access to credit) or personal choice.


TFP scores are expressed in relation to the sample mean that has been normalised to unity. While one is able to identify farms which are relatively more efficient with a higher TFP index score in a particular sample for one country, this might bear little relationship to what may be considered internationally productive. Therefore, what it is possible to compare internationally is the share of farms that have high TFP scores in each sample and whether they produce the predominant portion of output and to what extent they depend on net current subsidies. The ranking of productivity scores between different management types can also be compared.

Table 10 presents the country results according to TFP1 (including estimated costs for own resources).

In all countries, farms with TFP scores above 1 are in a minority and in the case of Navarra they constitute only 29 percent of the sample farms. At first glance, it seems that the results for Navarra are contradictory: too high a percentage of profitable farms and too low a share of farms with higher than the average productivity. However, a more detailed analysis shows that all the farms that have a TFP score above unity are also profitable according to all profitability ratios. Productive farms account for 62.7 per cent of all Navarra farms that are profitable according to P_CB ratio. Thus, productivity and profitability are related ([chi square] coefficient significant at the 0.01 level).

Two important features stem from the productivity analysis. With the exception of Spain, the minority of productive farms produces a majority of the total output. From this point of view, once again Hungary has the best performance with 85 percent of the output produced in farms having technically efficient input-output combinations. The results for South-East England indicate that the productive farms tend to rely less heavily on net direct subsidies. In South-East England, 47 percent of productive farms absorb only 26 percent of the total net current subsidies of the sample. This, however, is not the case in the other analysed countries.

In Hungary and the Czech Republic, according to management type, corporate farms have higher TFP scores than individual farms (Table 11).

Family farms are less productive despite the high expectations at the outset of the reform process that better incentives involved in individual farming would boost their efficiency. The reasons for this result are complex, including the long-standing tradition of farming in association in the NMS and a high share of hired labour in corporate farms allowing them to recruit labour with necessary skills for technical agricultural and management positions. In some cases, former collective farm managers were able to siphon off the most attractive parts of the business into new corporate farms that yield good returns. The argument that corporate farms benefit solely from economies of size does not seem to hold, at least for Hungary. When for the present data set the size has been controlled for, individual farms still appeared as less productive than their corporate counterparts.


The 1990s witnessed extensive restructuring that created a more complex pattern of farming in Central Europe. As a result, there is no neat divide in profitability between Western and Central Europe. The estimated profitability ratios indicate that farms in Hungary and Navarra fared the best, with the worst problems being in parts of the Czech Republic and Poland. The main difference between the two West European cases and Central Europe is not in terms of the number of farms that are profitable but rather in terms of capital intensity and wage rates.

In explaining the relatively poor profitability of farms in the Czech Republic, Poland and South East England, a number of factors can be cited. Comparing farms in the Czech Republic with their Hungarian counterparts, the former appear overmanned. In the Czech case, land per AWU is significantly lower than in Hungary (38ha compared with 53 ha) despite average wage rates being higher. Agri-environmental conditions also play a part with profitability in less attractive areas (eg potato and the mountainous forage regions) being significantly worse. As a result, Net Value Added per AWU is over four times greater in Hungary compared to the Czech Republic.

Comparing the family farms of Navarra with those in Poland, it is apparent that in the former case farms rent in more land to generate a reasonable return. In the Polish case, farms are relying to a far greater extent on family-owned land. The Spanish farms are also better capitalised. Polish farms have high average values for total assets per ha by Central European standards, but they are still significantly less than in the Spanish case. In addition, the quality of Polish capital has been questioned (Latruffe et al., 2005). Latruffe et al. (2005) in their efficiency analysis identify that many Polish farmers have purchased an extensive range of machinery and equipment irrespective of their farm's size and the potential efficiency with which such capital could be used. The maintenance costs for old and obsolete capital inherited from the communist era are high. While farmers in a better financial state and with larger farms have invested in modern and more expensive equipment, the bulk of the smaller farms have invested in old, second-hand machinery. The smallest farms as a result allocate the highest percentage of depreciation in comparison to the original costs of capital (IERiGZ, 1998, 2002). The relative superiority of farms in Navarra compared to Poland is apparent for all three ratios and cannot just be reduced to the effect of direct payments, although the latter play their role in supporting private profitability.

While the majority of Polish farms are unprofitable when the opportunity costs of own land and labour input are accounted for, 60 percent break-even if only paid costs are considered. If self-exploitation (accepting low returns to owned labour and land) occurs, as in many peasant societies (Ellis, 1998), the survivability of small-scale farms in Poland is likely to be greater than economic cost benefit analyses would predict.

In Hungary and the Czech Republic, when the opportunity costs for own labour and land are accounted for and farms operate without current subsidies (private cost-benefit ratio), corporate enterprises are the most profitable. Corporate farms in both countries, however, do suffer from relatively high debts, often due to the nature of the reform process. They also rely almost exclusively on hired labour and rented land so that unlike most small individual farms they cannot rely on self-exploitation as a strategy to cope with a downturn in agricultural fortunes. As a result, the farms that are the most competitive during an era of good returns may not be those best placed to weather a period of poor agricultural profitability.

In South-East England, high factor costs are important determinants of its relatively poor profitability. For example, NVA per AWU in South-East England was 23,665 euros compared to an average wage rate per AWU of 18,790 euros. In Navarra, NVA per AWU was 29,185 euros compared against an average paid wage rate of 12,312 euros. Thus, wages account for the equivalent of 79.4 percent of NVA per AWU in South-East England compared against only 42.2 percent in Navarra.

As measured by the P_CB ratio (full cost-benefit), the greatest structural problems lie in Poland. The returns on own labour and land are exceptionally low and the figures on poor private profitability mirror the findings of research on the international competitiveness of Polish agriculture (Gorton et al., 2001). The majority of individual farms persist through a lack of other employment options and a degree of self-exploitation--too many people are trying to earn a living out of too small farms. To deal with this problem, the stimulation of the non-farm rural economy is paramount. At present, the latter is underdeveloped in Poland (Chaplin et al., 2004) and this hinders structural adjustment. Chaplin et al. (2004) identify that diversification (both enterprise and/or off-farm employment) is linked to the level of general education and the availability of public transport. In Poland, the educational attainment of farmers is low and infrastructural issues are poorly addressed in current EU-led initiatives for rural development. Dealing with structural problems in rural Poland will thus require a greater emphasis on improving educational attainment and mobility.

For the years analysed, direct payments in Poland were insignificant. In Hungary and the Czech Republic about four-fifths of commercially oriented farms received direct payments but these were much less in absolute terms and as a percentage of gross revenue than in existing EU Member States. In December 2002, the Copenhagen European Council concluded the accession negotiations with countries from Central Europe. It decided that direct payments for acceding countries should be 'phased-in' over a period of 10 years from an initial level of 25 percent of the direct payments granted to farmers in the current EU Member States. However, national governments can 'top-up' the direct payments. Subject to authorisation by the European Commission, they can top-up by up to 30 percent or to a maximum of 10 percent above the level that farmers received under pre-accession national schemes. The introduction of these direct payments in the NMS will have a significant impact. However, it is important to note that there is a core of farms in Western and Central Europe that could potentially survive without direct payments and that are not strongly dependent on policy protection. Moreover, Chaplin et al. (2004) found that increases in agricultural price support and the introduction of direct payments lowers the propensity of farmers to diversify and vice versa. Therefore, while direct payments could improve the private profitability of agriculture in Central Europe, they are likely to impede restructuring in countries like Poland where structural change and the movement of labour out of agriculture are critical.


Brooks, K, Guasch, JL, Braverman, A and Csaki, C. 1991: Agriculture and the transition to the market. Journal of Economic Perspectives 5(4): 149-161.

Capalbo, SM and Antle, JM (eds) 1988: Agricultural productivity, measurement and explanation. Hopkins University Press: Washington, DC.

Chaplin, H, Davidova, S and Gorton, M. 2004: Agricultural adjustment and the diversification of farm households and corporate farms in Central Europe. Journal of Rural Studies 20(1): 61-77.

Csaki, C and Lerman, Z. 1998: Land reform and farm restructuring in Hungary during the 1990s. In Wegren SK (ed). Land Reform in the Former Soviet Union and Eastern Europe. Routledge: London. pp. 224-261.

Curtiss, J. 2002: Efficiency and structural change in transition. Institutional Change in Agriculture and Natural Resources, Vol. 12. Shaker Verlag: Aahen.

Ellis, F. 1988: Peasant economies: Farm households and agrarian development. Cambridge University Press: Cambridge.

Franks, J. 1998: Predicting financial stress in farm businesses. European Review of Agricultural Economics 25(1): 30-52.

GFA. 1997: Analysis of Polish agriculture and food economy's competitiveness and complementarity in regard to EEC/EU over the time span 1990-1995. Final workshop report, PHARE-Project P9312-05-7.

Gorton, M, Danilowska, A, Jarka, S, Straszewski, S and Zawojska, A. 2001: The international competitiveness of polish agriculture. Post Communist Economies 11 (4): 445-457.

Gorton, M and Davidova, S. 2004: Farm productivity and efficiency in the CEE applicant countries: A synthesis of results. Agricultural Economics 30(1): 1-16,

GUS. 1990: Biuletyn Statstyczny/statistical yearbook. GUS: Warsaw.

Hagedorn, K. 1994: Changing Organisation of Agriculture as a Result of the Transformation Process. Paper presented at workshop 'The Impact of CEE Agricultural Reform on Domestic and International Markets', Leuven, 10-11 June.

Hughes, G. 2000a: Total productivity of emergent farm structures in Central and Eastern Europe. In: Banse, M and Tangermann, S (eds). Central and Eastern European Agriculture in an Expanding European Union. CABI Publishing: Wallingford. pp. 61-87.

Hughes, G. 2000b: Agricultural decollectivisation in Central Europe and the productivity of emergent farm structures. PhD Thesis, Wye College, University of London, unpublished, Mimeo.

IERiGZ. 1998: Produkcyjno-Ekonomiczna Sytuacja Gospodarstw Prowadzacych Rachunkowosc rolna w latach 1994-1996. Instytut Ekonomiki Rolnictwa I Gospodarki Zywnosciowej: Warsaw.

IERiGZ. 2002: Produkcyjno-Ekonomiczna Sytuacja Gospodarstw Prowadzacych Rachunkowosc rolna w latach 1998-2000. Instytut Ekonomiki Rolnictwa I Gospodarki Zywnosciowej: Warsaw.

Kostov, P and Lingard, J. 2002: Subsistence farming in transitional economies: lessons from Bulgaria. Journal of Rural Studies 18 (1): 83-94.

Latruffe, L, Balcombe, K, Davidova, S and Zawalinska, K. 2005: Technical and scale efficiency of crop and livestock farms in Poland: Does specialisation matter?. Agricultural Economics 32(3): 281-296.

Lerman, Z, Csaki, C and Feder, G. 2001: Agriculture in transition: Land policy and changing farm structures in Central Eastern Europe and the former Soviet Union. World Bank: Washington, DC.

Mathijs, E and Vranken, L. 2001: Human capital, gender and organisation in transition agriculture: measuring and explaining the technical efficiency of Bulgarian and Hungarian farms. Post Communist Economies 15(1): 171-187.

OECD. 1994: Review of Agricultural Policies: Hungary. OECD: Paris.

OECD. 2000: Agricultural Policies in OECD Countries: Monitoring and evaluation. Vol. I Centre for Co-operation with Non-Members: Paris.

Official Journal. 1988: L144, 27.05. 1988.

Pollak, R. 1985: A transaction cost approach to families and households. Journal of Economic Literature 23: 581-608.

Pouliquen, A. 2001: Competitiveness and farm incomes in the CEEC agri-food sectors: Implications before and after accession for EU markets and policies Vol. I and II. European Commission, Directorate General for Agriculture: Brussels.

Schmitt, G. 1993: Why collectivisation of agriculture in socialist countries has failed: A transaction cost approach. In Csaki, C and Kislev, Y (eds). Agricultural Cooperatives in Transition. Westview Press: Boulder, CO. pp. 143-159.

Tanton, A and Williams, N. 2000: Farm business statistics for South East England. Wye College, University of London: London.

van Zyl, J, Miller, WR and Parker, AN. 1996: Agrarian structure in Poland: The myth of large farm superiority. The World Bank: Washington, DC.

(1) This paper is based on research conducted within the EU FP5 IDARA Project, QLRT-1999-1526. The authors are grateful for the financial support and the usual disclaimers apply. The authors also thank Barna Kovacs and Tamas Mizik for their collaboration.

(2) A commercial holding is defined as a farm that is large enough to provide the main gainful activity of a farmer and a level of income sufficient to support his or her family,

(3) Instytut Ekonomiki Rolnictwa i Gospodarki Zywnosciowej.

(4) Capital assets are valued at the cost of replacing them. Therefore, fixed capital items for South-East England are re-valued each year to reflect changes in their market value (Tanton and Williams, 2000).

(5) Objective 5 (b) areas were selected to receive special support from the EU's Common Agricultural Policy (CAP). According to the Council Regulation (ECC) No. 2052/1988 of June 1988 (Official Journal, L144, 27.05.1988), 'Areas eligible under Objective 5 (b) shall be selected ... taking into account in particular the degree to which they are rural in nature, the number of persons occupied in agriculture, their level of economic and agricultural development, the extent to which they are peripheral and their sensitivity to changes in the agricultural sector, especially in the context of reform of the Common Agricultural Policy'.

(6) Limited liability companies are to a large extent successors of the former state farms. Their assets had to be purchased and the new owners had to pay an initial instalment while the rest was recorded as long-term liabilities to the state. Cooperatives carry liabilities to former, currently non-farming, owners of assets (so-called eligible persons). The start of the repayment of these liabilities was delayed for 7 years. However, since January 2000 eligible persons have been entitled to claim their assets from cooperatives.


(1) Imperial College Wye Campus, AEBM, Wye, Ashford, Kent TN25 5AH, UK. E-mail:

(2) University of Newcastle, Newcastle, UK

(3) The Czech Institute of Agricultural Economics, Czech Republic

(4) CASE Foundation, Poland

(5) Public University of Navarra, Spain
Table 1: Characteristics of the data sets used in the analysis

Country Year(s) Type Useable
 analysed (a) number of
 farm records

Czech Republic 1998-1999 FADN 823
Hungary 2000 FADN 1,121
Poland 1999 IERiGZ 1,001
Navarra Spain 1996-1999 FADN 369
South-East 1999 FBS 183

Country Comments

Czech Republic
Poland Only individual farms
Navarra Spain
South-East The UK farm business survey (FBS) is
England organised at the request of the Department
 of Environment, Food and Rural Affairs
 (DEFRA). Only a proportion of farms surveyed
 contributes to EU FADN. FBS indicators are
 consistent with FADN indicators.

(a) For comparative purposes, the presentation of Czech and Spanish
results in this paper focuses on data for 1999.

Table 2: Background sample characteristics, 1999 (a)

 Czech Hungary Poland

Average UAA per farm (ha) 658 202 25
Average output (b) (EUR) 532,665 224,073 18,000
Average total assets (EUR) 870,542 204,484 86,000
Average total assets per ha (EUR) 1,450 1,977 3,440
Average AWU per farm 32 7.45 1.85
Land rented (%) 76 42 17
Hired labour input (%) 50 31 6
Land per AWU (ha) 38 53 13
DEPAWU (EUR) 2,421 2,427 1,294
Average paid wage (EUR per paid AWU) 3,552 3,490 2,308

 Navarra South-East
 Spain England

Average UAA per farm (ha) 50 141
Average output (b) (EUR) 97,000 399,753
Average total assets (EUR) 292,000 1,345,154
Average total assets per ha (EUR) 5,840 9,540
Average AWU per farm 1.49 6.35
Land rented (%) 45 34
Hired labour input (%) 10 53
Land per AWU (ha) 36 41
DEPAWU (EUR) 6,281 7,810
Average paid wage (EUR per paid AWU) 12,312 18,790

(a) For Hungary 2000 as the 1999 data set had many inconsistencies.

(b) Output includes net current subsidies.

Table 3: Profitability ratios

 Czech Hungary Poland

Average private cost-benefit score 1.224 1.03 3.83
% of sample profitable on private 20 60.9 8.7

Average cost-revenue without subsidies 1.086 0.81 1.01
score of sample profitable on 38 81.5 60.4
cost-revenue without subsidies

Average cost-revenue plus subsidies score 1.003 0.76 1.01
% of sample profitable on cost-revenue 48 85.4 60.4
plus subsidies

 Navarra South-East
 Spain England

Average private cost-benefit score 1.098 1.374
% of sample profitable on private 45.0 14.8

Average cost-revenue without subsidies 0.714 1.095
score of sample profitable on 85.4 42.6
cost-revenue without subsidies

Average cost-revenue plus subsidies score 0.604 0.922
% of sample profitable on cost-revenue 92.7 74.9
plus subsidies

Table 4: Direct payments

 Czech Hungary Poland

Direct payments as % of gross output 6.4 5.2 0.03
of sample that receive direct payments 80 82 1.9

 Navarra South-East
 Spain England

Direct payments as % of gross output 13.6 14.0
of sample that receive direct payments 99.2 72.1

Table 5: Profitable and loss-making farms according to
agri-environmental region, Czech FADN sample, 1999, number and

 Maize Sugar beet Cereal-Potato
 region region region

Profitable 7 (35%) 70 (22%) 66 (22%)
Loss making 13 (65%) 253 (78%) 238 (78%)

Profitable 13 (65%) 137 (42%) 124 (41%)
Loss making 7 (35%) 186 (58%) 180 (59%)

Profitable 13 (65%) 160 (50%) 153 (53%)
Loss making 7 (35%) 163 (50%) 151 (50%)

 Potato Mountainous
 region forage region

Profitable 18 (13%) 0 (0%)
Loss making 118 (87%) 40 (100%)

Profitable 32 (24%) 3 (7%)
Loss making 104 (76%) 37 (93%)

Profitable 54 (40%) 19 (47%)
Loss making 82 (60%) 21 (53%)

Table 6: Profitability ratios for the Czech and Hungarian samples
according to management type


 Individual Ltd Coops Other
 farmers corporate

Private Cost-Benefit
Average score 1.09 0.86 0.78 0.93

Cost-Revenue without subsidies
Average score 0.82 0.86 0.78 0.79

Cost-Revenue plus subsidies
Average score 0.77 0.77 0.75 0.74

 Czech Republic

 Individual Ltd Joint Coops
 farmers stock

Private Cost-Benefit
Average score 1.26 1.20 1.19 1.14

Cost-Revenue without subsidies
Average score 1.05 1.19 1.19 1.13

Cost-Revenue plus subsidies
Average score 0.95 1.10 1.13 1.07

Table 7: Profitability according to agri-environmental region
and management form, Czech FADN sample, 1999

Class means Maize Sugar beet Cereal-Potato
 region region region

Individual farmers 1.046 1.166 1.293
Ltd companies 1.064 1.129 1.217
Joint stock comp. 1.178 1.214 1.104
Production coops 1.256 1.068 1.143
Regional averages 1.089 1.156 1.242

Individual farmers 0.901 0.997 1.052
Ltd companies 1.064 1.125 1.200
Joint stock comp. 1.178 1.207 1.096
Production coops 1.256 1.059 1.134
Regional averages 1.002 1.040 1.081

Individual farmers 0.880 0.961 0.978
Ltd companies 1.029 1.086 1.121
Joint stock comp. 1.152 1.167 1.038
Production coops 1.235 1.024 1.062
Regional averages 0.979 1.004 1.008

Class means Potato Mountainous Legal type
 region forage region averages

Individual farmers 1.366 1.694 1.2623
Ltd companies 1.274 1.270 1.2035
Joint stock comp. 1.244 1.429 1.1944
Production coops 1.178 1.341 1.1372
Regional averages 1.282 1.563

Individual farmers 1.089 1.391 1.0467
Ltd companies 1.263 1.256 1.1932
Joint stock comp. 1.240 1.429 1.1879
Production coops 1.173 1.336 1.1297
Regional averages 1.157 1.363

Individual farmers 0.895 0.898 0.9544
Ltd companies 1.134 1.022 1.0981
Joint stock comp. 1.170 1.071 1.1302
Production coops 1.095 1.198 1.0658
Regional averages 1.023 0.966

Table 8: Net value added (NVA) per farm, hectare and AWU (EUR)

 Czech Republic Hungary Poland Navarra South-East
 Spain England

NVA per farm 97,166 125,483 4,685 44,723 142,959
NVA per ha 116 974 199 1,961 1,436
NVA per AWU 3,472 16,481 2,044 29,185 23,665

Table 9: Financial ratios for the sample farms

 Czech Hungary Poland Navarra South-East
 Republic Spain England

Debt to assets 0.33 0.16 0.03 0.08 0.15
Leverage 1.53 0.39 0.04 0.11 0.25
RENGO 0.04 0.03 0.02 0.05 0.09
RENGM 0.09 0.04 0.04 0.12 0.36

Table 10: Farm productivity (TFP1 scores)

 Czech Hungary Poland

No of high productivity farms (TFP>1) 381 488 346
of high productivity farms 46 44 35
of sample UAA in TFP>1 farms 53 64 63
of sample output in TFP> 1 farms (a) 60 85 56
% of sample subsidies in productive farms 46 49 52

 Navarra South-East
 Spain England

No of high productivity farms (TFP>1) 106 86
of high productivity farms 29 47
of sample UAA in TFP>1 farms 29 38
of sample output in TFP> 1 farms (a) 37 69
% of sample subsidies in productive farms 33 26

(a) Output includes net current subsidies.

Table 11: Farm productivity by management type (TFP1 scores)


Management type Average
 TFP score

Family farms 0.96
Limited liability companies 1.16
Cooperatives 1.19
Joint ventures 1.42
Other 0.96
Total 1.00

 Czech Republic

Management type Average
 TFP score

Family farms 0.987
Limited liability companies 0.971
Cooperatives 1.033
Joint stock companies 1.035
Other N/A
Total 1.00
COPYRIGHT 2005 Association for Comparative Economic Studies
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2005 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Davidova, Sophia; Gorton, Matthew; Ratinger, Tomas; Zawalinska, Katarzyna; Iraizoz, Belen
Publication:Comparative Economic Studies
Geographic Code:4E
Date:Dec 1, 2005
Previous Article:Unemployment benefit systems in central and eastern Europe: a review of the 1990s (1).
Next Article:Individual farming as a labour sink: evidence from Poland and Russia.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters