The Territorial Impact of the Farmers' Early Retirement Scheme

The Territorial Impact of the Farmers’ Early Retirement Scheme Zografia Bika Abstract This article considers the causal relationships between the European farmers’ early retirement policy instrument and structural and social outcomes. The member states of Greece, Ireland and France have been among the main beneficiaries of this policy scheme, which encourages farmers aged between 55 and 66 to retire and transfer their land to a younger farmer, with multiple effects. The analysis in this article suggests that the accrued pension income is mostly higher than applicants’ previous earnings from farming, while the policy instrument is only moderately successful in enlarging farms or encouraging new entrants, with the majority of transfers occurring between members of the same family. The farmers’ early retirement scheme appears to be connected more to the regional traditions of inheritance than to farm structures per se. The impact on an unbalanced European age–farmer structure is evaluated as differing little from what would have happened anyway, with only short-term effects reported in relation to the time scale. In summary, the scheme is subject to manipulation in terms of regionspecific circumstances and thus fulfils social objectives rather than the purpose of structural cohesion. Introduction T he territorial aspect of farm succession has been found to be neglected in the recent body of English language work, which emphasises the pathways of farmer retirement and the impact of the farm family cycle on farmers’ decisionmaking and land use and the intergenerational exchanges/relations (Hutson 1987; Keating and Munro 1989; Potter and Lobley 1992, 1996; Gasson and Errington 1993; Kimhi 1994; Pesquin et al. 1999; Wallace and Moss 2002; Pietola et al. 2003). To this extent, previous knowledge on farm succession emerges as being dependent on farm-specific parameters and, by paying little attention to regional variation, it is thus partial and potentially detrimental to relevant agricultural policy making, such as the EU farmers’ early retirement scheme (ERS). Intergenerational farm transfer is not only a facet of farm business management but also a significant element in understanding how the geography of rural society and the rural economy itself, is © 2007 The Author. Journal Compilation © 2007 European Society for Rural Sociology. Published by Blackwell Publishing, 9600 Garsington Road, Oxford OX4 2DQ, UK Sociologia Ruralis, Vol 47, Number 3, July 2007 ISSN 0038–0199 Impact of the farmers’ early retirement scheme 247 changing. ‘Keeping the name on the land’, that is, ‘to ensure that the family continues to farm in a particular locality’, in Errington and Lobley’s words (2002, p. 1), has territorial implications. In the example of Wales (Hutson 1987, p. 219), it was observed that the ideal of ‘generational fragmentation’ – the establishment as far as possible of each child in a viable but separate farm or career outside farming – has changed to one of consolidation and generational incorporation to form an extended unit. Potter and Lobley’s survey research in England and Wales (1992, p. 1996) has provided evidence of elderly farmers’ land management decisions being affected by the presence of a successor. In particular, British farmers without successors are much more likely to simplify their enterprise structure, extensify production, reduce farm capacity, develop another income source and to leave dairying, in comparison to those who have successors (Potter and Lobley 1992). Such literature on farm succession, transfer and entry consists of household-based analysis with determinants related to the family farm cycle, such as the single marital status of elderly farmers and the absence of successors, or general issues of timing being treated as responsible for low levels of early retirement. The impact of the regional context on retirement patterns is largely ignored. As Hutson (1987, p. 215) also informs us – on the basis of farming research material in Wales, ‘the practical focus of kinship has narrowed from emphasising links between farm families to concentrating on relationships within individual family businesses’. In a similar framework, the territorial implications of elderly British farmers’ decision making are often given no other consideration than a simply remark that ‘twice as many elderly farmers with successors are in dairying compared to those without’ (Potter and Lobley 1992, p. 324). It becomes obvious that the scarcity of English language literature on farmer retirement is offset by discussions on farm succession as the counterpart of farmer retirement. In contrast to that, it is argued here that succession and retirement patterns are influenced less by the changing forces of capitalist agriculture but more by regional social organisation factors external to the family itself. Burton and Walford’s (2005) recently published work acknowledges that farm size (and its capitalist rationality) has been less important than social aspects of the family unit, such as co-operation and planning issues, in determining land division in the largest farms of south eastern England since 1960, albeit without critically investigating macro and meso structural influences, other than suggesting that agricultural policy uncertainties increase the probability of division. Interestingly, the comparative Irish literature revolving around inheritance (Commins and Kelleher 1973; Rogers and Salamon 1983; Kennedy 1991), that is, the transfer of property rather than status, appears to be more informative regarding the historical conditions surrounding farmer retirement and the regionally determined socioeconomic practices affecting it. The geographical diversity in the patterns of intergenerational farm transfers is well reported at the European level (Lamaison 1988; Tracy 1989; Perrier-Cornet et al. 1991; Gasson and Errington 1993; Errington and Lobley 2002). As Blanc and Perrier-Cornet also point out (1993, p. 322), inheritance practices ‘cover well-defined geographical areas that rarely correspond to national units’. They often vary among © 2007 The Author. Journal Compilation © 2007 European Society for Rural Sociology. Sociologia Ruralis, Vol 47, Number 3, July 2007 248 Bika classes (Cooper 1976; Le Roy Ladurie 1976; Thompson 1976; Fertig 2003), between regions (Habakkuk 1955; Homans 1969) and over time (Howell 1976; Kennedy 1991). Even if one goes back to pre-industrial Europe with the help of historical demographers (Kaser 2002; Fertig 2003) there is little evidence of an inescapable chain between inheritance and marriage that balances resources and population (in a non-partitioning sense). This chain is instead seen as an ideal and in practice is variable both across regions and time. There have been numerous attempts to summarise inheritance strategies (PerrierCornet et al. 1991; Blanc and Perrier-Cornet 1993; Ross Gordon Consultants 2000). In particular, ‘the combination of share-out in kind (possible or actual) and egalitarian practice’ is common in many Mediterranean regions and ‘it fosters the development of pluriactivity, retirement farming and even survival agriculture for the unemployed’ (Blanc and Perrier-Cornet 1993, pp. 322–323). However, diversity remains paramount with some Greek Aegean islands (such as Karpathos and Nisiros) even preserving a system of inheritance whereby ‘only the first son and first daughter receive substantial endowments of real estate’ (Herzfeld 1980) or other islands (such as Kalymnos) where inheritance patterns had privileged women (Bernard 1976). Similarly, ‘in Italy, primogeniture is still followed in the autonomous Tyrol province of Alto Adige’ (Gasson and Errington 1993, p. 196). In a different fashion, regions in The Netherlands and Germany are characterised by inheritance strategies based upon the need to preserve the unity of the holding and thus favouring unequal shares and full-time employment. However, the concept of splitting farms equally among all heirs (Realteilung) has been the prevailing inheritance system in the southwest and some areas of north Germany (Ross Gordon Consultants 2000). By contrast, in the UK, with the exception of East Anglia (see Habakkuk 1955; Homans 1969) and Ireland, with the exception of Tory Island, (see Fox, 1978), single successor and non-compensatory systems of inheritance are predominantly encountered. A similar inheritance pattern of ‘keeping the name on the land’ is found in Finland (Abrahams 1991). The French inheritance traditions stand in the middle of the two extremes with the farm successor paying ‘compensatory sums to the coheirs, but the land is on average under-valued by half compared with open market prices’ (Blanc and Perrier-Cornet 1993, p. 324). Danish and Belgian regions follow the French inheritance pattern of equal shares and a single successor, but with parents rather than coheirs being the transferors. However, such regional variations are superseded when, as David Symes (1990) points out, inheritance and succession issues are no longer fundamental in an agricultural world that is increasingly incorporated in a wider capitalist economy. However, in areas where the capitalist integration of agriculture is incomplete, offfarm employment affects opportunity costs and partible inheritance remains prevalent, for example, in southern Europe, where a great degree of intergenerational conflict and early retirement policy imperatives are still registered. In different historical contexts, the territorial impact of traditional inheritance systems is stronger than globalisation would suggest. As Kennedy points out about the west of Ireland in the pre-Famine period (1991, p. 480), ‘more complete incorporation into market society may tilt the balance decisively against partible inheritance’, which has been dominant until this point. © 2007 The Author. Journal Compilation © 2007 European Society for Rural Sociology. Sociologia Ruralis, Vol 47, Number 3, July 2007 Impact of the farmers’ early retirement scheme 249 Surprisingly, Symes (1990, p. 289) notes that only ‘in the past, traditional inheritance systems have had a profound impact upon the social geography of rural areas’. But are inheritance systems really a thing of the past? An exploratory attempt to quantify the highly differentiated inheritance traditions and subsequent farmer retirement tendencies at the European level should be useful in provisionally answering such a question and in terms of conceptualising agricultural policy making. The rate of farmers’ early retirement, signifying the frequency of their withdrawal from land ownership, will be under examination here, rather than ‘the series of transitions’ (Gasson and Errington 1993, p. 237) and the retirement paths or plans in which they have engaged. Rogers and Salamon’s (1983) comparative analysis of two French farming communities and two communities of Illinois grain farmers (German–American and Irish–American) indicated that in the communities favouring multiple-heir systems, regardless of their ethnic origins, early retirement was the preferred farm exit option and matched delicate population/land balances. Partible inheritance was seen as responsible for low celibacy rates, small family sizes and low out-migration amongst the community members. In general terms, equal shares did not appear to provide the multiple heirs with employment and viable holdings but they helped them to avoid permanent out-migration and retain social relationships in the farming community as valuable land-owning members. In this way, multi-heir systems are seen to encourage stronger geographic immobility, more densely populated rural areas and richer social interaction. Very few researchers have dealt explicitly with farmers’ retirement decisions, which were found in the case of Maryland ‘to be mainly determined by personal considerations, and only secondary importance is given to succession considerations’ (Kimhi and Lopez 1999, p. 154). It was also Fennel (1981) who first argued that there is a correlation between the level of urbanisation in a region and farmer retirement, with the latter being resistant to control, planning and policy schemes. The marked regional variation was originally brought into the light when she stated that ‘basically retirement is an urban concept and farmers in some regions and countries are more immediately affected by urban values’ (Fennel 1981, p. 32). Another comparative study (England, France, Ontario, Quebec and Iowa) has also reported geographical variation in the patterns of retirement from farming (Errington and Lobley 2002), which was attributed more to different social norms and/or institutional practices than country and other farm and farmer characteristics. This variation is matched with the fact that different western European socio-occupational groups proceeded separately in obtaining access to modern pension schemes with civil servants being the first to succeed (in the 1850s), as part of an exchange for their loyal services (Conrad 1991). In response to this literature, the main hypothesis developed here is that farmer retirement tendencies are fundamentally linked to the other possible uses of elderly farmers’ time and thus are more dependent on their region-specific surroundings (historical, institutional and cultural) than on sectoral considerations, farm business management issues and the demands of productivism. The analysis of the interplay of individual, local and regional factors through the detailed examination of the uptake of the farmers’ ERS aims to integrate farm exit experiences from different European contexts. It is based on a review of the national evaluation reports of the © 2007 The Author. Journal Compilation © 2007 European Society for Rural Sociology. Sociologia Ruralis, Vol 47, Number 3, July 2007 250 Bika ERS, which show how such policy has affected the ways in which European farms are passed from one generation to another. Complementary to this material is case-study work that has been undertaken at a regional level and is used here to draw together comparable case study findings from different EU member states. Additional evidence considered includes statistical analysis of Nomenclature of Units for Territorial Statistics, Level 3 (NUTS3) data. ERS objectives and regulations The ERS aims to address the perennial structural problems of the age profile and poor viability of farm holdings (farm size structure), a main feature of a number of member states. The ERS provides a pension for elderly farmers to retire on and an opportunity for young farmers to take over holdings and practise farming. Member states are required to ensure that all farms involved in the scheme are managed in a manner that protects the countryside, although the details of how this should be achieved are not specified, and there are no guidelines on environmental management. Historically speaking, the ERS origins can be traced back to the 1972 Directive 72/160/EEC (EEC Council Directive 1972) that originally concerned ‘measures to encourage the cessation of farming and the re-allocation of agriculture for the purposes of structural improvement’ (O’Shaughnessy 2004). However, the EU farmers’ ERS developed fully fledged as one of the three ‘accompanying measures’ (plus agri-environment and afforestation) of the CAP reform of 1992 (so-called Mac Sharry) through the Regulation 2079/92 (EEC Council Regulation 1992) and is thus an integral part of emerging rural development plans. Alongside the other Mac Sharry changes in the mechanisms for assigning aid (the price floors are lowered but farmers are compensated by ‘production neutral payments’ such as direct arable area or livestock-headage payments), the introduction of the ERS was connected to sustaining the rural world of the EU in economic, social and environmental terms (rural development) and more particularly, dealt with fears that CAP reform would reduce dramatically the number of young farmers. Most recent eligibility criteria, through the Regulation 1257/99 (2000–2006) (EC Council Regulation 1999) included limitations in relation to age (the transferor should be not less than 55 years old but not yet of normal retirement age at the time of transfer, whilst the transferee should not exceed a maximum age), occupation (transferors must have practised farming for the preceding 10 years and they also have to cease all commercial farming activity and release their land to suitable transferees, who must practise farming on the holding for not less than five years), economic viability (measured in terms of an obligatory increase in the size of the transferee’s agricultural holding) and farming skills on the part of the successor (testified either by attending the Certificate of Farming course or other adequate farming experience). In short, the ERS 2000–2006 offered incentives for farmers aged between 55 and 66 to transfer their farms (by gift, lease or sale) to qualified young farmers by providing them with a pension of up to an annual maximum of €15,000. It was paid monthly per person retiring for a maximum of 15 years, but in any case could not continue beyond the applicant’s 75th birthday or exceed €150,000 in total value. Where payments extended beyond the normal state retirement age, the early retirement © 2007 The Author. Journal Compilation © 2007 European Society for Rural Sociology. Sociologia Ruralis, Vol 47, Number 3, July 2007 Impact of the farmers’ early retirement scheme 251 pension was granted as a supplement to the state pension, taking into account the amount of early retirement support. Considerable variation is reported between countries in the levels of subsidies that were paid to recipient farmers. In the post-2006 rural development proposals of the EU Commission a higher rate of pension (€180,000 total amount per transferor) is combined with the requirement of ERS transferees to be set up in farming through an installation aid grant, or to be existing farmers who must want to enlarge their lands (no viability criterion). To this extent, the post-2006 ERS does not operate solely as a land supply mechanism but also includes an incentive for the younger farmer (Castle 2004). National and regional applications of the ERS The uptake of the various versions of ERS in different European countries is indicative of the structural effects (age and scale), costs and efficiency gains of the instrument itself. Overall, it appears to have allowed a number of elderly farmers to retire in a dignified way. However, the scheme’s potential to accelerate structural adjustment in terms of radical changes in the age profile has not been achieved (Naylor 1982; Paniagua 2000), since more than 30 per cent of the most targeted farmers in Mediterranean countries are still aged over 65 years old at NUTS3 level (Eurostat Regio 19971), whilst the respective EU-15 national average (1999/2000) has lingered at 29 per cent (CIAS 2004). This is also in agreement with the conclusion of the ex-post evaluation (Agra CEAS Consulting 2003, p. 3) of measures under Regulation (EC) No. 950/97, where it is stated that ‘relatively little land was released as a result of the Early Retirement Scheme’ (EC Council Regulation 1997). On the other hand, the examination of the ERS at a regional level allows for a distinct spatial pattern of adoption to gradually come into view. Its use as a social policy to reduce the financial difficulties of elderly farmers has been deemed insufficient because of the relatively low pensions on offer, the lessening of other entitlements (medical cards, social security benefits or subsidies), and regional imbalances in its uptake (as relatively prosperous farming areas benefit the most): Some stakeholders argued that the beneficiaries of the ERS usually held small and marginal farms that have been excluded from the market due to their low level of competitiveness, and which were therefore not attractive for young farmers to take over. (Agra CEAS Consulting 2003, p. 80) Nevertheless, it has facilitated a partial transfer of resources from older farmers to younger ones for the purposes of setting up or enlarging farms, which primarily supported the process of farm household adaptation to CAP reform and restructuring (Allaire and Daucé 1996). In line with an integrated rural development approach, which puts emphasis on the process as part of the outcome by bringing together different actors in the farming community, it can thus be claimed that the ERS has achieved a certain degree of success. To this extent, the ERS ‘has proven to be most popular in France, Ireland and latterly Greece, and these countries in aggregate accounted for 88 per cent of total spending between 1992 and 1999’ (Caskie et al. 2002, p. 12). Spanish farmers were © 2007 The Author. Journal Compilation © 2007 European Society for Rural Sociology. Sociologia Ruralis, Vol 47, Number 3, July 2007 252 Bika much more modest participants of the ERS, whilst Danish farmers showed that it was only their second-best option (525 participants) and a great majority preferred to use a more generous national early retirement package available across all sectors of the economy (Caskie et al. 2002). Considerable synergistic effect with the young farmers scheme was reported in Finland (Agra CEAS Consulting 2003). The ERS was not mandatory, and hence was not implemented in some countries, including Italy, Luxembourg, Sweden and the UK. In the latter case, early attempts to encourage such retirement were also deemed unsuccessful with only 0.1 per cent of UK farmers taking up ‘the pension annuities or lump sums offered in the EC Outgoers Scheme between 1975 and 1977’ (Gasson and Errington 1993, p. 224). The template of French farmers’ early retirement In France, a supplementary retirement grant, the indemnité viagére de départ (IVD), to the state old age pension for full-time farmers over 65 years old has been in operation since 1962. A spatial dimension was introduced in one of its later versions, with an early retirement premium (non complément de retraite, IVD) being available from 1968 to 1974 in problem areas (such as Brittany and the Massif Central) to farmers between 60 and 65 years of age (Naylor 1982). By the 1970s the early retirement premium started to be associated with farm enlargement objectives, or ‘the installation of suitably qualified young farmers’ (Naylor 1982, p. 28). ‘Between 1963 and 1978, a total of 562,000 farmers have received the IVD pension and one-third of the total agricultural area of France had been transferred through the scheme’, although ‘the objective of achieving a shift in land holding towards younger farmers has not been fully achieved’ (Naylor 1982, p. 29). To this extent, IVD constituted the institutional template on which the EU ERS was built in 1992. However, it was admitted even then that the IVD mainly advanced the farm transfer by a few years rather than increased the rate itself (Naylor 1982). A variety of ERS regulations have been tried in France in three successive periods: 1992–1994, 1995–1997 and 1998–2006. Cost considerations, social objectives and structural aims determined their operational usefulness. Flat-rate payments (€5,335 per year for farms up to a maximum of 10 ha with a supplement of €76 per ha for each additional ha up to a maximum of 50 ha) in the period 1992–1994, gave way in 1995–1997 to differential pension payments for participants depending on the status of the transferee (the setting up of a young farmer, the enlargement of a young farmer’s existing holding or just the enlargement of a farmer’s holding). The limited contribution of the first French ERS to the promotion of new entrants to farming (15 per cent) and its unfocused design led to a targeted but also more expensive second national ERS. This was finally transformed into an instrument of social policy for elderly French farmers in dire straits or poor health (third ERS 1998–2006). The Council Regulation 1257/99 (EC Council Regulation 1999) covers the ERS 2000– 2006 in France, which has attracted 1,300 farmers in the first two years of implementation (2000–2001). 43,000 French farmers participated in the first ERS (a rate of 30 per cent), which was heavily criticised for its inadequate representation of young transferees (6,600). On the other hand, one-third of those who adopted the first ERS were found to be © 2007 The Author. Journal Compilation © 2007 European Society for Rural Sociology. Sociologia Ruralis, Vol 47, Number 3, July 2007 Impact of the farmers’ early retirement scheme 253 women, an incidence attributed by Allaire and Daucé (1996) to the lowering of the normal retirement age to 60 years old (they replaced their spouses who had previously retired early). Moreover, as Brangeon et al. (1996) discuss, a pension income higher than farmer’s existing earnings was found to be the main explanatory variable for those who decided to participate in the first ERS (Caskie et al. 2003). The participation rate fell considerably in the second ERS (16 per cent) that appealed to only 18,000 farmers, with a high cost (€600 million) alongside a significant increase in the numbers of young farmers setting up (8,000) and a sectoral attraction for dairy and beef farmers (Caskie et al. 2002). 700,000 ha were released in the period 1995–1997, while 1.3 million ha had been released in 1992–1994 due to the simpler nature of the rules governing the first ERS and its higher absorbency rate of multiple generations waiting to exit farming. An evaluation of the second ERS in France (Daucé et al. 1999) has pointed out its short-term orientation, in that it only managed to bring forward farmers’ retirement decisions by a few years and to instigate by only 15 per cent the establishment of young farmers that would not have taken place otherwise. Modelling its impact against underlying trends in farmer retirement has shown that it was strongly supportive of the young farmers’ setting up in farming families (Daucé et al. 1999). Moreover, it was argued that speeding up the numbers retiring as a result of the ERS would be followed by a below-trend retirement rate. To this extent, aggregate retirement levels would remain the same in the long term and thus current changes in retirement have a break-even point, something that would have happened anyway. It was concluded that the second ERS (in operation from 1995 to 1997) became an expensive policy because it ended up promoting only a modest number of young farmers’ setting up (Daucé et al. 1999). Contrasting French dairy and cereal regions In the French case, the highest levels of adoption of the IVD grant (1963–1978) were reported to be in areas of least need (Naylor 1982). It was also noted that ‘government support, through the CAP, for the maintenance of agricultural prices at levels which encourage small farmers to remain in business also conflicts with retirement policy’ (Naylor 1982, p. 33). Moreover, areas of part-time farming appeared to have had lower than average levels of IVD adoption (Naylor 1982, p. 31). As Allaire and Daucé (1996) point out, the national average rate of early retirement observed for the period 1992–1994 obscures relatively high regional disparities in uptake in France. There was a strong regional contrast in the participation rates between the Paris basin and the littoral Mediterranean areas (around 15 per cent of the eligible population) as opposed to those in the geographical crescent that includes Bretagne, Burgundy and Lorraine, passing from the heart of the Massif Central (30 per cent). To this extent, the dairy regions affected the most by restructuring were the most eager participants in the ERS, whilst the cereal regions alongside those of intensive crops remained relatively indifferent to the scheme (Allaire and Daucé 1996). Historically speaking, this crescent is also an area that follows distinct versions of western French customary inheritance law (as opposed to south France and its anti-egalitarian legacy of written law), where the parental right to show preference for © 2007 The Author. Journal Compilation © 2007 European Society for Rural Sociology. Sociologia Ruralis, Vol 47, Number 3, July 2007 254 Bika one child could not be exercised, exemplified by the practice in Bretagne of total equality among heirs (Le Roy Ladurie 1976; Derouet 1996). On the other hand, the rate of early retirement was not found to decrease with increasing farm size (Allaire and Daucé 1996). Results observed for the first two years of application of the French agricultural policy of early retirement (1992–1994) show that the ERS rate, 21 per cent on average, was higher for farmers with lower incomes (Brangeon et al. 1996). In France, regional discrepancies were also found in the second ERS rates of farm transfers to young farmers; these were largely explained by differences in average regional agricultural incomes and/or positive attitudes to the relevant institutions (Daucé et al. 1999). The case of Greek non-compliance with ERS In Greece, the first version of the ERS was implemented through Regulation 1096/88 (1989–1990) (EEC Council Regulation 1988), which proved to be an immature plan to foster early retirement. Farm transfers were made by gift, sale or lease on the basis of an informal written agreement between the transferor and the young transferee, which meant in practice that the majority of holdings continued to be cultivated by those who opted for early retirement (Koutsomiti 2000). This first version allowed 61,500 Greek farmers to participate in 1989 and 9,500 in 1990. Regulation 3808/89 then introduced stricter eligibility criteria with the transferor having to cease all farming activity rather than just share a joint unit with a young farmer (Louloudis et al. 1993, p. 470). In any case, only two per cent of the farms were transferred on a permanent basis in the period 1989–1990, while 41.38 per cent of the total number of applicants were later found to be ineligible (Louloudis et al. 1993, p. 470). The objective of the second version of the ERS in Greece through Regulation 2079/92 (EEC Council Regulation 1992) was an uptake by 50,000 farmers. This was not achieved, with only 26,623 participants recorded at the end of the period 1995– 1999, and problems of non-compliance with the requirement to cease farming continuing to be present. The average holding size of the participating transferors at the country level was less than 3 ha (or 1.14 ha of non-irrigated land, 0.90 ha of irrigated land and 0.71 ha planted with trees) (Koutsomiti 2000). Although Greece has very severe structural problems in agriculture at NUTS3 level (Eurostat Regio 1999), ‘the Scheme was not viewed primarily as an aid to structural adjustment but as a social policy for the alleviation of low farm incomes’ (Caskie et al. 2002, p. 17). Only 63,726 ha were transferred to young farmers between 20 and 40 years old during 1995–1999 in Greece. Modelling the factors affecting farmers’ decisions whether or not to continue their agricultural activities in less favoured areas (LFAs) has attributed primary importance to age and succession, and much less to the LFA payments (Spathis and Kaldis 2003). To this extent, the ERS was considered to be instrumental in the combat against rural depopulation in LFAs, with any delays in the farm transfers to younger farmers being potentially ‘fatal’. In other words, elderly farmers in LFAs who choose to remain active in farming for longer maximise the likelihood of pushing their potential successor out of farming altogether. © 2007 The Author. Journal Compilation © 2007 European Society for Rural Sociology. Sociologia Ruralis, Vol 47, Number 3, July 2007 Impact of the farmers’ early retirement scheme 255 ERS participation rates in the North Aegean islands In Greece, 2,500 farmers (out of 8,151; National Statistical Service of Greece 1991) participated in the first version of the ERS on the island of Lesvos, part of the North Aegean Region of particular disadvantage. However, this did not constitute evidence of a changing balance in the island’s age–farmer structure because of extensive fraud (Koutsomiti 2000). Anecdotal evidence also suggests that many farmers accepted participation without being aware of the accompanying regulations, as a result of which they lost their price subsidies (Koutsomiti 2000, p. 54). This created considerable hesitation on the part of the Lesvian farmers towards successive versions of the ERS. Remaining in Lesvos, the second version of the ERS (EEC Council Regulation 1992) attracted a smaller number of participants (96 participating farmers, or 1 in 908 inhabitants) than the neighbouring islands of Chios (325, or 1 : 242 inhabitants) and Samos (173, or 1 : 160 inhabitants) despite Lesvos’ bigger size and total population. ERS-participating farmers represent only 1.1 per cent of the total number of farmers in Lesvos (National Statistical Service of Greece 1991); however, the former were found to be less than 60 years old because of the massive exodus of the older generations during the first ERS. Moreover, the total area of holdings attracted under the ERS 1995–1999 was only 0.4 per cent of the total utilisable agricultural area (UAA) in Lesvos (Koutsomiti 2000). The location of the participating farmers was of considerable importance, with more than half of them coming from the olive-growing areas of the island. The tourist areas in the island, and those with a high share in pasture, are hardly represented in the ERS. The greater part of the agricultural land transferred under the ERS has been planted with trees, especially olives (18.19 per cent), which is considered normal for a district dominated by this type of cultivation (91.99 per cent). The comparison of the average holding size transferred by the early retirees (2.37 ha) with that received by young transferees (1.73 ha) showed that the ERS has contributed to a 72.5 per cent increase in the size of the transferor’s holdings in Lesvos. To this extent, the ERS succeeded in quadrupling the size of the Lesvian transferor’s holdings, an efficiency gain much higher than that intended, i.e. 15 per cent for the period 1995–1999 and 10 per cent for the period 2000–2006. However, only the small-sized farms have benefited from the ERS, with only 3.1 per cent of the participating farmers owning more than 5 ha. Moreover, almost half of the transferee’s holdings (42.3 per cent) are between 1 and 2 ha, which shows the extent of land fragmentation in Lesvos (Koutsomiti 2000). Another important feature of the Lesvos case study was that 70.8 per cent of the ERS participating farmers were women, whilst only 29.2 per cent were men. This reflects the ways in which the average farm household operated in Lesvos in 1995– 1999, the period under examination (Koutsomiti 2000). Men were involved the most in off-farm employment and thus were excluded per se from the ERS-eligible population (EEC Council Regulation 1992), and this created the necessary conditions for women to take on the occupational status (in official terms) of farmers. The same trend was reported for the young transferees (62.5 per cent were women and 81.3 per cent were less than 34 years old) (Koutsomiti 2000). This is in striking contrast with © 2007 The Author. Journal Compilation © 2007 European Society for Rural Sociology. Sociologia Ruralis, Vol 47, Number 3, July 2007 256 Bika the census data (National Statistical Service of Greece 1991) where 70 per cent of the agricultural holdings’ owners in Lesvos were reported to be male (Koutsomiti 2000). The explanation for these contradictory characteristics can once again be found in the use of the ERS as part of the farm household’s strategy to increase its income as a whole by ‘bending’ regulations (or misreporting in census returns), maximising pluriactivity and ‘juggling’ resources. In support of the argument of family bonds interfering with the ERS application, 72.9 per cent of the land transfers in Lesvos were found to be intergenerational (from parents to children). Irish ERS time gains In Ireland, the first initiative to encourage farmers aged 55–65 to retire early came into operation between 1974 and 1985 (EEC Council Directive 1972), but only 600 farmers participated. ‘This level of adoption was much less than an exploration of the potential attitudes towards the scheme of elderly farmers in the west might have suggested’ (Gillmor 1999, p. 80). Unwillingness to retire early, farm succession issues, sentimental bonds with the land and low pensions were the reasons for deterring the great majority from participating in the scheme. The first round of the Irish version of the EU ERS (EEC Council Regulation 1992) commenced in 1994 and was linked to both farm transfer to eligible transferees and subsequent farm enlargement, with a variable rate paid depending on the transferor’s farm size. A total of 9,380 farmers participated during the period 1994–1999 (23 per cent on average), out of about 40,000 who were believed to be eligible, resulting in the release of 283,800 ha (AFCon Management Consultants 2003). As Murphy (1997, p. iii) points out: the majority of transfers under the scheme were to family members, and the average size of farm transferred was 31.37 ha (average farm size in Ireland is 26 ha), whilst the average size of the enlargement clause was 21.23 ha. Approximately 30 per cent of transferees were aged 35 or more. Since 1997 after changes in the Irish law, gift tax/inheritance tax has affected only very large farms valued at more than £1.8 million. Total expenditure on the scheme (1994–1999) was €315 million. However, it was argued that the scheme has been least successful in those parts of the country which might be deemed to need it most because of the old age structure of the farmers and the small size of holdings. (Gillmor 1999, p. 84). 52 per cent of the ERS Irish applicants (1994–1999) were women who applied either in their own right as landowners or as joint managers where a partner, usually a spouse owned the land (Murphy, 1997). In most cases, joint management was declared only when the female partner was the younger person retiring (10 years younger for the 17 per cent of the men over 55 years old, Central Statistics Office Ireland 1991) and the only eligible person for the ERS. However, ‘90 per cent of the farm holders in the 1991 agricultural census were found to be male’ (Murphy 1997, p. 11): © 2007 The Author. Journal Compilation © 2007 European Society for Rural Sociology. Sociologia Ruralis, Vol 47, Number 3, July 2007 Impact of the farmers’ early retirement scheme 257 Farm women in practice make a significant contribution to running the farm but their work has been to a certain extent invisible because they are not classified as ‘actual’ farmers as by and large they are not landowners. (Murphy 1997, p. 11) To this extent, Murphy continues (1997, p. 74), some dead weight loss (excess burden) was possibly involved in ‘certain transfers taking place under the system of joint management’ (Murphy 1997). The national evaluation of this first version of the Irish ERS (1994–1999) also included a sample survey of 94 beneficiaries, equally split between men and women. Forty per cent of the surveyed farms were below 30 ha, and only five per cent of the beneficiaries had farms less than 10 ha. The survey findings also showed that only a small proportion commenced a gradual hand-over of activities in the five years prior to retirement, while the majority intended to retire at the normal age (Murphy 1997, p. 60). The ERS was successful in persuading farmers to retire on average 6 years earlier than normal. The value of the pension was the main factor influencing farmers’ decisions to take early retirement (for 79 per cent of women and 70 per cent of men). More than half the retirees said that they were financially better off since retirement, and a further 35 per cent said that their income had remained about the same with only 10 per cent claiming that they were worse off. Furthermore: when people were leasing rather than transferring the holding outright the main reason for leasing was usually that they were not able to give it to their family at that particular point and so they were leasing it in the meantime. (Murphy 1997, p. 61) The second version of the ERS (2000–2006) in Ireland recognised that the previous enlargement requirement (EEC Council Regulation 1992) acted in favour of ‘the established and larger farmers who are more able financially to acquire land by lease or purchase’ (Gillmor 1999, p. 84). In response to this criticism, the new version gave leeway to young part-time farmers who were now able to qualify as ERS transferees. Scanlan’s remark (2002) in relation to the revised ERS’s ‘need for fine tuning’ provides additional information. The maximum available payment was thought to be less attractive than in the previous round, mainly because of increased living standards in Ireland and higher state pensions (Caskie et al. 2002; Scanlan 2002). Survey findings based on a sample of 89 of the 878 farmers who have taken early retirement in Limerick have shown that they were on the whole relatively young (their average age at retirement was 59) and married (70 per cent), while half of them had leased their land to non-family (though neighbouring) farming members (Lavery 2004). For the first two years of the Irish rural development plan 2000–2006 measures, 1,257 ERS participants (39,804 ha transferred) have been reported (AFCon Management Consultants 2003). A year later, at the end of 2003, there was still a total of 1,952 transferors, representing 23.5 per cent of the estimated participation (Castle 2004). The complication of the western Irish attachment to farming On a geographical basis, Irish ERS applicants (EEC Council Regulation 1992) were also distributed in a highly unbalanced fashion (Murphy 1997), with the majority of applicants coming from the traditionally more prosperous farming areas (1994– 1999: Leinster 32 per cent, Munster 49 per cent, Connacht 12 per cent and Ulster 7 © 2007 The Author. Journal Compilation © 2007 European Society for Rural Sociology. Sociologia Ruralis, Vol 47, Number 3, July 2007 258 Bika per cent). A distinct spatial pattern of involvement in the first ERS was reported, with more than half of the farmland transferred being in the south-west Irish dairy counties, characterised by medium-sized to large farms, a strong commercial orientation towards farming, and young Irishmen prepared to farm on a full-time basis. ‘This was even more evident in terms of the transfer of land than in relation to farmer involvement’ (Gillmor 1999, p. 81). The lowest rates of participation in the ERS were in the west and north-west Irish regions, which are characterised by a higher proportion of unmarried farmers and small-sized, poorer, dry stock farms associated mostly with low-income cattle and sheep grazing activities (Gillmor 1999). The age profile of family farm operators over 65 years old also exhibited regional variation (Munster: 20.6 per cent and Connacht: 29.4 per cent). The cost of acquiring additional land to expand holdings and title deficiencies might have been greater deterrents in the west than elsewhere (Gillmor 1999, p. 83). To this extent, the most common means of land acquisition in 1994 (CSO) was inheritance in Connacht, family transfer in Munster and purchase in Leinster (Gillmor 1999). Thus, the more traditional western part of the country showed signs of stronger attachment to the land and more willingness to transfer land only upon death. Murphy (1997) has also highlighted the presence in the west of Ireland of many part-time farmers alongside significant numbers of elderly farmers who were single and without a readily identifiable successor. These elderly farmers were either too attached to farming and disinterested in retirement and occupational role changes (Murphy 1997) or they were too old to qualify (under ERS) in their own right and did not have spouses who would be able to qualify on the basis of joint management. Moreover, the small size of part-time farms prevailing in the west did not allow their owners to qualify as an ERS transferee, as they could not claim that they practiced farming as main occupation (eligibility criterion, 1994–1999). The second ERS’s uptake was also slower than expected because of the foot-and-mouth disease crisis, while regional imbalances in uptake have remained resilient (Munster/South Leinster: 75 per cent of applicants; Connacht/North Leinster/Ulster area: 25 per cent) (Caskie et al. 2002; Scanlan 2002). The overly strict Spanish ERS eligibility criteria In Spain, the initial European package on early retirement (Reg. 1096/88 and 3808/89 [EEC Council Regulation 1988, 1989]) was linked to the restructuring objectives of young farmers and attracted only very small numbers of beneficiaries (868 in 1990 and 1991) (Paniagua 2000, p. 115). After the eligibility criteria became slightly more relaxed during 1993–1997 in the first round of the Spanish version of the ERS (EEC Council Regulation 1992), which attempted to strike a balance between social and structural objectives, 3,279 farmers participated (1.4 per cent of the eligible population). Its distinctive feature remained strict eligibility criteria regarding the economic farm size (a minimum of four European size units (ESUs) in 1993, which went down to 2–3 ESUs in 1995 for some regions), the age group (young transferees had to be less than 45 years old and transferors to be between 60 and 65 years old) and the kin relationship with the transferee (only first degree for the period 1991–1995). Not all social agents with an influence in the Spanish agricultural sector supported the © 2007 The Author. Journal Compilation © 2007 European Society for Rural Sociology. Sociologia Ruralis, Vol 47, Number 3, July 2007 Impact of the farmers’ early retirement scheme 259 restructuring process brought forward by the ERS, which has mainly been defended by government bodies rather than farmers’ unions or the farmers themselves (Paniagua 2000, p. 120). The geographical distribution of Spanish ERS farmers (1990–1994) was also concentrated with high yield regions and higher than average numbers of young farmers attracting the majority of aid granted throughout the country, such as with the example of Castile and León (1,031 beneficiaries, or 31.4 per cent of total aid) (Paniagua 2000). Castile and León is favourably characterised by ‘employment shares in small communities close to the national average for each of agriculture, manufacturing and services’ (Hoggart and Paniagua 2001, p. 66). However: the farmers attracted to the retirement programme have holdings of insufficient economic size (average: 14.9 ha), considering that the average area of holdings in the region is 31.1 ha, or 23 ha taking into account their UAA alone (Paniagua 2000, p. 116). The great majority of the holdings under the ERS were small in size (half were less than 10 ha), were owned (75.8 per cent against 59 per cent of the region’s average) and were multiply fragmented (each holding is made up of 12.7 plots on average) in Castile and León. To this extent, most of the holdings under ERS for the region (10,876 ha) showed very little intensive farming, although many are situated in high-yield districts ... Only four per cent of these holdings have a production capacity equal to or exceeding the average regional capacity (Paniagua 2000, p. 118). The farm transfer process (mainly by leasing) resulted in holdings which were double their original economic size but still unable to achieve economic viability (40 per cent of them had still less than 5 ha) (Paniagua 2000). The great majority of early retirees were married (74.2 per cent), while only 19.8 per cent were single and 5.8 per cent were divorced (Paniagua 2000). The probability of Finnish farm exits In Finland, 5,569 ERS participants were reported in 1995–1999 (Ministry of Agriculture 2001). Up to the end of 2000, a total of FIM 960.3 million was paid out in farmers’ early pensions (in existence since 1974), of which the EU’s share was FIM 368.3 million. Pietola et al. (2003) point out that early retirement schemes used singly as structural adjustment tools jeopardise their long-term capability because they create uncertainty over their continuation. This has obliged Finnish farmers in recent years to complement their obligatory retirement plan with market-led early retirement ones. Pietola et al.’s (2003) analysis of a stratified random sample of 963 Finnish farmers born between 1929 and 1943 and active in 1993 has shown that the overall probability of farm exit (with or without farm transfer through ERS) decreases with farmer age, an unfavourable agricultural output market, the farmland area, a small farm size and single marital status. In 1993–1998, the period under investigation, the marginal utility of early retirement benefits was also found to be higher for low income Finnish farmers (Pietola et al. 2003) whose pension levels depend on the © 2007 The Author. Journal Compilation © 2007 European Society for Rural Sociology. Sociologia Ruralis, Vol 47, Number 3, July 2007 260 Bika compulsory insurance fee they have paid during their working life. As Pietola et al. argue (2003, p. 114): price and subsidy changes in the period 1994–1998, which were caused by Finland’s entry into the EU (1995), jointly decreased the probability of transferring the farm to a new entrant by two-thirds. The territorial impact of the ERS in Finland has taken the form of a north–south divide, according to Pietola et al.’s (2003) study of the likelihood of an early exit from farming. Farmers located in northern and central Finland were more likely to retire early and transfer their farm to a new entrant (a 55 per cent increase in probability) than those in the south, where EU income subsidy programmes were employed substantially less and the opportunity cost of entering farming remained higher (given off-farm employment opportunities). Using the same Finnish farmlevel panel dataset for the period 1993–1998, farm exits are once again found to be predominantly among northern and non-livestock farmers, while the off-farm income of elderly farmers is explicitly found to reduce the probability of farm succession (Väre 2006). Most importantly, however, farmers are not found to co-ordinate their early retirement decisions with their spouses’ retirement under other pension schemes (as opposed to the ERS and its Finnish change-of-generation clause). Thus, Väre’s findings (2006) show not only how ERS dead weight losses can be manageable, but also substantiate how individual family cycle and timingspecific parameters, such as spousal retirement, (unexpectedly) matter less than contextual factors. Avoiding ERS dead weight loss in Norway In Norway, as part of the annual negotiations between the farmers’ unions and the government, an early retirement pension scheme was introduced in 1999. The aim of the scheme was to ease the transfer of farms to the younger generation where agriculture and forestry have been the main source of income. The early retirement pension scheme was introduced nationwide without allowances for regional differentiation. The restrictive regulations involved the transferors’ age (62–67 years old) and their income from agriculture and forestry for the last 15 years. In case of early retirement pension payments for two people (the farmer and partner), different restrictions applied regarding age (the partner needed to be at least 57 years old), occupation (partner had to have worked on the farm for the last five years) and income (the couple’s total farm income had to be at least 50 per cent of their total income). Another important condition of the Norwegian scheme was that the new landowner (transferee) could not be the transferor’s partner, thus avoiding the dead weight losses under joint management encountered in other countries. An annually fixed pension could be paid for a maximum of five years (from 62 to 67 years old). Norwegian farmers, numbering 5,409 in all, participated in the scheme during the period 1999–2003. Based on statistics regarding the numbers of Norwegian applicants for agricultural subsidies and the number of early retirement pensioners born between 1935 and 1941, the Norwegian Agricultural Authority estimated that approximately 30 per cent of the farmers between 62–67 years old entered the scheme © 2007 The Author. Journal Compilation © 2007 European Society for Rural Sociology. Sociologia Ruralis, Vol 47, Number 3, July 2007 Impact of the farmers’ early retirement scheme 261 (Statens landbruksforvaltning 2003). Most interestingly, one-third of the farmers who participated in the sample survey (88 young farmers and 155 elderly ones) of Norwegian farm transfers between 1986–1998 (Stubberud and Samseth 2000), opined that the scheme would have encouraged them to make the farm transfer earlier than otherwise. The evaluation report concluded that the scheme worked as intended (Statens landbruksforvaltning 2003). Norway has also displayed a spatial pattern in the uptake of its early retirement scheme, 1999–2003 (Statens landbruksforvaltning 2003). The NUTS3 regions SørTrøndelag and Nord-Trøndelag (in the middle of Norway), Rogaland (in the southwest) and Oppland (in the inland of the region Østlandet) had a higher uptake than the national average. These are the ‘strong’ agricultural counties with a higher than average employment in agriculture, animal husbandry and farm size (ha or number of animals). By contrast, the regions in the central part of south eastern Norway (in the Oslo region) characterised by cereal production, off-farm employment opportunities and part-time farming, had a lower than average uptake of the Norwegian early retirement scheme. It is likely that the early retirement pension scheme had been more attractive for elderly full-time farmers who had problems finding a job outside agriculture. The very low uptake of the scheme in Vest-Agder and Telemark (southern Norway) is harder to explain; small-sized farms, part-time farmers and a less vibrant labour market prevail here, with the low uptake being more associated with the traditional agricultural communities. Population density, regional inheritance practices and farmers’ early retirement The social significance of work on the farm and the impact of social relationships on the farmer’s decision to retire early have been shown in the above discussion to acquire a territorial dimension and can be characterised by regional imbalances. An attempt to synthesise case-study findings across the EU-27 leads to a clear-cut territorial conclusion which contributes to the existing literature that farmers’ decisions to retire early in response to policy incentives is region-specific, with those in relatively prosperous farming regions being more likely to exit than those located elsewhere. This article’s argument is that the regional imbalances in ERS uptake are consistent with suggestions that a connection exists between urbanisation levels and early retirement (Fennel 1981), demography and inheritance (Rogers and Salamon 1983) and, most importantly, population density and farming regions’ ERS participation rate (as shown below). If regional demographic patterns tend to correspond to the types of inheritance strategy preferred, as Rogers and Salamon claim (1983), then the ERS’s territorial impact might also be more correlated to demographic indicators rather than the family farm cycle. Comparative regional analysis at the NUTS3 level was carried out for the purposes of substantiating such an argument. Of the countries with the highest ERS rates two contrasting NUTS3 regions were selected for each on the basis of minimum and maximum uptake levels (Table 1). Eurostat-REGIO, FADN and Corine Land Cover databases were the sources of the NUTS3 data collected, whilst the methods of calculation were consistent for all variables included in the analysis with the exception of ERS rate. The numbers of ERS participants and hectares of farmland transferred © 2007 The Author. Journal Compilation © 2007 European Society for Rural Sociology. Sociologia Ruralis, Vol 47, Number 3, July 2007 262 NUTS3 level Spain Burgos a Population density, 1999 Meso accessible to markets, 1997b Population change, 1989–1999 (%) Unemployment rate, 2001 (%) Unemployment change, 1998–2001 (%)c GDP per head, 2000d GDP per head, 1995–2000 (%) Agricultural employment, (%) 1995e Average farm size (ESUs), 1999 Non-irrigated landf (%) Irrigated arable land (%) Vineyards (%) Olive groves (%) Pastures (%) Complex cultivation patternsg (%) Natural vegetation (%) Farmers over 65 years old, 1997h (%) Hotels (nos) 2000i Farmland (ha) France Soria Vendée Greece Marne Ireland Lesvos Lefkada West Ireland SW Ireland 24.00 345.57 8.90 387.17 80.50 300.87 69.30 287.33 44.80 360.58 59.50 390.83 25.60 357.74 45.20 321.54 -5.68 -6.93 6.62 1.58 -4.64 0.95 6.30 4.41 7.90 -51.23 4.60 -45.88 6.50 -24.42 8.30 -24.55 – – – – 4.00 -39.39 3.80 -51.90 20,996 30.57 11.46 18,085 24.65 16.25 18,670 25.85 9.22 23,916 25.43 9.61 15,952 37.07 – 13,056 61.81 – 18,943 53.88 20.90 28,970 68.49 14.34 13.48 65.87 5.43 0.16 0.00 0.86 7.78 13.48 74.75 3.69 0.11 0.00 1.02 3.73 41.94 26.43 0.00 0.19 0.00 19.62 46.51 76.58 83.95 0.00 3.70 0.00 3.07 4.08 4.33 – – – – – – 6.08 0.85 0.00 3.37 31.84 0.00 10.63 19.69 0.37 0.00 0.00 0.00 81.47 2.34 19.69 9.44 0.00 0.00 0.00 84.82 1.80 19.68 21.42 16.68 21.42 7.19 15.13 5.19 11.08 – 46.96 52.62 47.39 15.82 21.89 3.93 21.89 234 1,430,900 114 1,028,700 208 671,960 124 429,860 126 215,380 66 35,600 973 1,428,300 1,311 1,230,680 Bika © 2007 The Author. Journal Compilation © 2007 European Society for Rural Sociology. Sociologia Ruralis, Vol 47, Number 3, July 2007 Table 1: NUTS3 regional data 1,966.5 20,065 186 0.93 276.7 8,150 26 0.32 – 10,373 1,867 18.00 – 14,800 156 1.05 304.75 16,042 142 0.89 405.7 3,606 94 2.61 12,472 32,633 408 1.25 33,863 22,749 1,107 4.87 ERS, early retirement scheme; –, no available information at NUTS3 level. a No. of inhabitants per km2. b The accessibility variable is an indicator developed by the ESPON project 2.1.1 (‘Territorial impact of EU transport and TEN policies’). In this, ‘the average time to market of a region is used in order to measure disparities in accessibility beyond the market potential’. More specifically, time to market measures the expected average time a firm or household in region r would need in order to reach the market. It takes into account that larger markets must be visited more often, and that firms or households try to bias their interactions with the market in favour of regions that are near by in order to save travel time. The indicator is standardised, such that the reference is a hypothetical region that can reach all markets with zero travel time. That means an indicator equal to 100 minutes for a certain region r, say, means that from that region one would need 100 minutes on average to reach market partners for buying a good, signing a contract, or making use of other opportunities offered in a destination and measured by GDP. In the case of accessibility indicator on meso scale, the intensity of spatial interaction is halved every 25 minutes of travel time to market (ESPON project 2.1.1). c Percentage change in the unemployment rate 1998–2001, calculated as follows: ([UNRT01-UNRT1998] *100/UNRT1998). d GDP 2000 purchasing power standards per inhabitant. GDP per head 1995–2000 is calculated as follows: ([GDP2000PH-GDP1995PH] *100/GDP1995PH). e Percentage employed in agriculture, fishing, forestry at NUTS3 level. f Percentage of total agricultural area (defined as the sum of all the agricultural land use variables). The method of estimation is calculation of km2 value by area tabulation of Corine land use values and NUTS3 GISCO 1 Mio set and redistribution on REGIO land area value (Epson meta-data, 1986–1996). g Percentage of total agricultural area under complex cultivation patterns (that is, juxtaposition of small parcels of diverse annual crops, pasture and/or permanent crops). h Percentage of all farmers. i Data indicate off-farm employment in the respective regions. Data sources: Eurostat – REGIO, Corine Land Cover database, FADN database, ESPON database. Impact of the farmers’ early retirement scheme © 2007 The Author. Journal Compilation © 2007 European Society for Rural Sociology. Sociologia Ruralis, Vol 47, Number 3, July 2007 ERS farmland (ha) Farms (nos) 2000 ERS farmers ERS rate (ERS farmers/farms) 263 264 Bika were sourced from different material available at the NUTS3 regional level. In the French case, the ERS data refers to the period 1992–1994 (Allaire and Daucé 1995). In Greece, the data period was 1995–1999 (Koutsomiti 2000), while the numbers of Irish ERS participants referred to was for 1994–1997 (Murphy 1997), and the Spanish was 1990–1994 (Paniagua 2000). The number of farms for each different NUTS3 region is sourced from national census data (2000). Closer examination of Table 1, showing the comparison of the selected NUTS3 regional data and the respective ERS participation rates, suggests that the latter increase proportionally in relation to population density. The more dispersed the population of a region, the lower the number of participating farmers to the ERS or, using OECD’s demographic definition of a rural area (i.e. one below 150 inhabitants per km2), the more urban the region, the higher the propensity of the farming populace towards early retirement. Such observation required further statistical investigation, which was obtained by adding 65 NUTS3 regions into the analysis (Greece: Ionian Islands and North Aegean islands, Ireland, France: ChampagneArdenne, Basse Normandie, Pays de la Loire, Bretagne, and Languedoc-Rousillon, Norway and Spain: Castile-León). To provide additional data for this statistical investigation, the distribution of the ERS rate between different types of regions in Europe was also analysed. For such a purpose three different typologies were used: population density, OECD and LFA (Table 2). The results show that, as anticipated, densely populated, leading and non-LFA regions attract the highest numbers of ERS participants. In summary, NUTS3 analysis suggests that regions with high population density, used as an aggregate proxy of both their inheritance systems (Habakkuk 1955; Hallam 1958) and degree of urbanisation (OECD) provide a cushion for farmers exiting farming. Population density emerges as a territorial indicator of farmers’ propensity for early retirement. However, it would be too simplistic to regard regional variation in the ERS uptake as a sole product of population densities without further exploring the causal or co-incidental character of the relationship. In this context, it is more appropriate to state that demography and social organisation appear, at the regional level, to assume greater influence than the economy in addressing farmers’ unbalanced age profile and/or production structure. Such an assertion has previously found support in quantitative research carried out on the regional (county-level) propensity of US farmers to cease farming, which is found to be higher if they reside in densely populated counties, among other factors (Goetz and Debertin 2001). Nevertheless, this is an argument that invites in-depth ethnographic or anthropological research beyond the limitations of the present enquiry, which remains an heuristic provocation for future study. To further substantiate the claim that farmers’ retirement patterns should be placed within a broader historical enquiry about the socioeconomic contexts in which they operate, an illustration of the utility of the inheritance cultures in explaining territorial variation was sought in the literature. To this extent, impartible inheritance customs have been deemed to be largely responsible for the following: the concentrated geographic source of the sixteenth century modern, or national, western European state structures in feudal zones (Hechter and Brustein 1980); the higher propensity to emigrate in mid-nineteenth century Germany (Wegge 1999); the © 2007 The Author. Journal Compilation © 2007 European Society for Rural Sociology. Sociologia Ruralis, Vol 47, Number 3, July 2007 Impact of the farmers’ early retirement scheme 265 Table 2: Regional typologies Types of regions Population density types of regions Less than 25 inhabitants/km2 25–50 inhabitants/km2 50–100 inhabitants/km2 100–150 inhabitants/km2 More than 150 inhabitants/km2 Total population density LFA types of regions Less than 25% 25–50% 50–75% 75–100% Total LFA percentage OECD types of regions Predominantly rural + leading Predominantly rural + lagging Intermediate + leading Intermediate + lagging Predominantly urban + leading Predominantly urban + lagging Total OECD type Mean ERS rate No. of regions SD 1.92 2.77 5.14 5.36 5.87 3.59 20 17 18 7 3 65 1.89 2.30 4.63 4.11 6.59 3.46 8.51 4.73 2.26 1.76 4.19 14 6 4 23 47 4.38 2.12 1.78 2.18 4.17 5.38 3.26 6.24 2.44 4.58 – 3.59 9 33 5 17 1 – 65 5.64 3.03 5.12 2.78 – – 3.69 ERS: early retirement scheme. LFA: less favoured area. SD: standard deviation. The OECD regional typology is based on various stages. The first stage identifies rural communities according to their population density. A community is defined as rural or urban if its population density is below or above 150 inhabitants per km2. The second stage classifies regions according to the predominant percentage of population living in rural communities (>50%, 15–50%, or <15%). In a final step of analysis, population and employment changes were used and the split was achieved by distinguishing between leading and lagging regional performances compared with the respective national averages. Data sources: Eurostat – REGIO, Corine Land Cover database, FADN database, ESPON database. tendencies of non-inheritors’ towards earlier transitions out of the Spanish parental home in contemporary Galicia (Holdsworth 1998), especially for women in postfamine Ireland (Guinnane 1992); and even for the short-lived, frequently replaced and only partially used nicknames in south-west Ireland (Breen 1982). By contrast, partible inheritance practices facilitated another set of historical occurrences, for it: condemned the Muscovite nobility to a servile status before the Tsar (Kivelson 1994); set-up long-lasting nicknames and kinship groupings above the level of household in Tory Island (post-famine Ireland) (Breen 1982); offered Germans access to cash and collateral services in mid-nineteenth century Hesse-Cassel (Wegge 1999); co-existed with seasonal (as opposed to permanent) migration and customs of local domestic industry in nineteenth century European peasant families (Habakkuk 1955); destroyed the eighteenth century English small farmer (yeoman) as a social class through a surfeit of mortgages/debts incurred to provide for the non-inheritors (Thompson 1976); led to the liquidation of English family businesses in early © 2007 The Author. Journal Compilation © 2007 European Society for Rural Sociology. Sociologia Ruralis, Vol 47, Number 3, July 2007 266 Bika industrial Stockport after the death of their Victorian middle-class proprietors whose wills were mostly concerned with providing a secure (possibly rentier income but also mobile and specialised capital) income for their offspring (Owens 2002); linked to lower co-residence rates in present-day southern Spanish regions. (Holdsworth et al. 2002). The link between partible inheritance and the European experience of preindustrial population growth (in the sense of multiplication of local nuclear households) in particular, has been explicitly endorsed by a number of historians (Habakkuk 1955; Hallam 1958; Parish and Schwartz 1972; Berkner and Mendels 1978; Wegge 1999). Partible inheritance practices put the continuation of the social unit of reproduction above the economic one in pre-industrial times. Habakkuk (1955, p. 6) suggested that for nineteenth century Europe the higher proportion of marriages in the equal-division inheritance regions was likely to produce in the aggregate more children than the fewer but more productive marriages of the single-heir regions. Kaser (2002) described how the equally partible male inheritance patterns of north east Europe (1500–1900) resulted in high household complexity and high population density. In this framework, impartible inheritance slows down population growth by lowering nuptiality rates. Hallam (1958) used thirteenth century census data to show quantitatively how practising partible inheritance in Lincolnshire (Midland UK) meant denser populations. However, in the feudal zones of western Europe, where the landlords resisted subdivision, impartible inheritance prevailed because ‘it kept the size of tenement constant’ and ‘arose from the lord’s interest in preserving his return from the tenement’, while ‘non-inheriting offspring were forced to follow other pursuits in the growing urban centres, the developing bureaucracy, or the armies of the feudal kings’ (Hechter and Brustein 1980, p. 1074). To this extent, inheritance practices are seen to have effects on both population density and mobility and, as Habakkuk explains in detail (1955, p. 8), ‘the population in single-heir areas may have been less dense, but it was more capable of permanent movement’. By investigating the geography of inheritance patterns as part of an effort to expand a research seam, it is suggested here that impartible inheritance regions, with a plethora of adult offspring who never married or emigrated, leads to lower population densities and raises the opportunity cost of the time of elderly farmers (and thus the value of giving up farming) if compared with the more social, multiheir regions. Or as Holdsworth et al. (2002, p. 1002), using present-day Spanish statistical data, argue, ‘the combination of high proportions of both married stayers and pre-marital leavers in the north’ does not correlate with economic determinants such as high regional unemployment but rather reflects regional kinship systems. Hence the statistical correlation also found here between population density and ERS participation rates is proved to be reinforced by historical region-based material that shows inheritance practices tying children to the land and thus affecting not only their occupational but also life-cycle choices such as age at marriage, farm entry and farm exit. This addresses the problem of identifying structural/ institutional constraints in strategic action such as farmers’ early retirement, among others (e.g. leaving home). Academic discourse thus needs to refocus attention on © 2007 The Author. Journal Compilation © 2007 European Society for Rural Sociology. Sociologia Ruralis, Vol 47, Number 3, July 2007 Impact of the farmers’ early retirement scheme 267 kinship, inheritance and custom-based interaction that has largely been deflected by the theoretical post-1980 shift from the ‘backwardness’ of local/regional level practices to global processes (Sutton 1997). Conclusions There are some important points that can be made in relation to the territorial impact of the ERS. In the countries with the highest rates of participation (France, Greece and Ireland), the structural effect was little different from that which would have occurred anyway, albeit over a slightly longer time scale. The ERS did not increase the rate of retirement in the long run and mostly did not encourage farm transfers outside the family. A considerable effect was reported only in relation to the LFAs, which are characterised by higher than average sensitivity to the timing of exits from farming (such as in Greece). The time gains offered by the ERS are important in relation to depopulation problems and the demographic scarcity of farm successors prevailing in LFAs, where the younger generation’s rejection of farming as a career due to delays in farm transfers leads not only to alternative employment but also to out-migration. Interestingly, however, in the post-decoupling era (and its guaranteed minimum level of direct payments irrespective of output), national applications of the ERS might emerge as ‘the only incentive scheme available for the release of lands’ against the option of just holding on to one’s land, semi-retirement and claiming entitlements, as Castle (2004, p. 11) also points out. A distinct spatial pattern of adoption of the ERS exists (France, Ireland, Norway, Finland and Spain). The highest levels of adoption were reported in relatively prosperous farming regions and those with a higher number of young farmers – areas least in need of the scheme. There is strong sectoral attraction for dairy/intensive farming regions (France, Ireland and Norway) and/or high yield regions (Greece and Spain). Cereal regions remain largely indifferent to the ERS (France and Norway). A pension income higher than existing earnings emerged as the main explanatory variable for those who decided to participate in the ERS (France, Ireland). The ERS adoption rate is higher amongst those farm households with lower than average incomes (France, Spain, Greece and Finland), while the absence of young successors and farmer’s single marital status decreases the likelihood of exit from farming (Finland, Spain and Ireland). Part-time farmers do not favour the ERS (Norway, Greece, Ireland and Finland). To conclude, the ERS has been used to achieve both social and structural objectives but it has failed to account for regional variations in conditions. Its design has varied by country and depends largely on national objectives. It has been more successful in ensuring the continuation of family farming and population stabilisation than in enhancing competitiveness and structural adjustment. Farmers’ early retirement has been more greatly associated with region-specific factors than with intra-family relationships (between parents and children), as it is somehow assumed in the farm succession literature. The ERS does not comprehensively assist in ‘bridging the generations in a changing world’ but rather reflects regional differences. Regional imbalances indicate that early retirement patterns are linked less to the profitability of farming and are associated more with demographic matters, social organisation and © 2007 The Author. Journal Compilation © 2007 European Society for Rural Sociology. Sociologia Ruralis, Vol 47, Number 3, July 2007 268 Bika the absence of young successors. Early retirement rates are more part of a regional pattern than solely a family farm process. Acknowledgements The research reported here was carried out within the ESPON 2006 programme as part of Project 2.1.3 ‘The Territorial Impact of the CAP/RDP’. The project partner institutions and personnel were the Arkleton Centre for Rural Development Research, University of Aberdeen (Zografia Bika, Andrew Copus, Yvonne Gourlay, Teresa Serra, Deb Roberts, Mark Shucksmith, Ken Thomson), the Federal Institute for Less Favoured and Mountain Areas, Wien (Thomas Dax, Ingrid Machold, Oliver Tamme), the Institute of Spatial Planning, University of Dortmund (Gunther Kroes, Martina Huelz) and the National Institute for Regional and Spatial Analysis, Maynooth ( Jim Walsh, Jeanne Meldon). Project special advisors are Isabel Bardaji Azcárate (Universidad Politecnica de Madrid), Tibor Ferencz (Budapest University of Economic Sciences and Public Administration) and Lars Olof Persson (NordRegio, Stockholm). The author is particularly grateful for the ideas and assistance of Ken Thomson, Deb Roberts and Mark Shucksmith. The views in this article are those of the author only, who bears full responsibility for any errors. The ESPON Managing Authority is not liable for any use that may be made of the information contained herein. Note 1 Regio is Eurostat’s harmonised regional statistical database. It covers the main aspects of economic and social life in the European Union, classified to the three levels of the NUTS. References Abrahams, R. (1991) A place of their own: family farming in eastern Finland (Cambridge: Cambridge University Press) AFCon Management Consultants in association with University College Cork (2003) CAP rural development plan 2000–2006: mid-term evaluation. Available online at http://www. agriculture.gov.ie Accessed 29 May 2007 Agra CEAS Consulting (2003) Ex-post evaluation of measures under regulation (EC) No. 950/97 on improving the efficiency of agricultural structures The European Commission DirectoratGeneral for Agriculture. Available online at http://www.ceasc.com Accessed 29 May 2007 Allaire, G. and P. Daucé (1995) Etude du Dispositif De Préretraite en Agriculture: Rapport Général (INRA Dijon et Toulouse: ENESAD) Allaire, G. and P. Daucé (1996) The farmer early retirement scheme 1992–1994: first balance and structural impacts. Economie rurale 232 (4) pp. 3–12 Berkner, L.K. and F.F. Mendels (1978) Inheritance systems, family structure and demographic patterns in Western Europe, 1700–1900. Pp. 209–223 in C. Tilly ed., Historical studies of changing fertility (Princeton, NJ: Princeton University Press) Bernard, R. (1976) Kalymnos: the islands of the sponge fisherman. Pp. 291–307 in M. Dimen and E. Friedl eds, Regional variation in modern Greece and Cyprus: towards a perspective on the ethnography of Greece (New York: Academy of Sciences) Blanc, M. and P. Perrier-Cornet (1993) Farm transfer and farm entry in the European Community. Sociologia Ruralis 33 (3/4) pp. 319–335 Brangeon, J.L., G. Jegouzo and M. Quiqu (1996) Early retirement scheme and low income in the agricultural sector. Economie rurale 232 (4) pp. 13–19 Breen, R. (1982) Naming practices in western Ireland. Man 17 (4) pp. 701–713 © 2007 The Author. Journal Compilation © 2007 European Society for Rural Sociology. Sociologia Ruralis, Vol 47, Number 3, July 2007 Impact of the farmers’ early retirement scheme 269 Burton, R.J.F. and N. Walford (2005) Multiple succession and land division on family farms in the south east of England: a counterbalance to agricultural concentration? Journal of Rural Studies 21 (3) pp. 335–347 Caskie, P., J. Davis, D. Campbell and M. Wallace (2002) An economic study of farmer early retirement and new entrant schemes for Northern Ireland. (Queen’s University Belfast: Department of Agricultural and Food Economics) Caskie, P., J. Davis, D. Campbell and M. Wallace (2003) The EU farmer early retirement scheme: a tool for structural adjustment. Conference Paper at the Agricultural Economics Society, 77th Annual Conference, 11–14 April 2003 (Plymouth: University of Plymouth) Castle, J. (2004) Expenditure review of the early retirement scheme (ERS2) 2000 (Dublin: Department of Agriculture and Food) National Statistical Service of Greece (1991) Agricultural and livestock census 1991 (Athens: National Statistical Service of Greece) Central Statistics Office Ireland (1991) Census of agriculture 1991 (Cork: Central Statistics Office Ireland) CIAS (2004) Compendium of Irish agricultural statistics (Dublin: Department of Agriculture and Food) Cooper, J.P. (1976) Patterns of inheritance and settlement by great landowners from the fifteenth to the eighteenth centuries. Pp. 192–327 in J. Goody, J. Thirsk, and E.P. Thompson eds, Family and inheritance – rural society in Western Europe (Cambridge: Cambridge University Press) Commins, P. and C. Kelleher (1973) Farm inheritance and succession (Dublin: Irish Farm Centre) Conrad, C. (1991) The emergence of modern retirement: Germany in an international comparison. Population: An English Selection 3 pp. 171–200 Daucé, P., F. Leturqu and M. Quinqu (1999) Impact of the second early retirement scheme on young farmers’ setting-up. Economie rurale 253 (10) pp. 51–57 Derouet, B. (1996) Nuptiality and family reproduction in male-inheritance systems: reflections on the example of the Franch-Comté (17th–18th centuries). The history of the family 1 (2) pp. 139–158 EC Council Regulation (1997) Improving the efficiency of agricultural structures, No. 950/97 of 20 May 1997 Official Journal of the European Communities, L 142, 02/06/1997 EC Council Regulation (1999) Support for rural development from the European Agricultural Guidance and Guarantee Fund (EAGGF) and amending and repealing certain Regulations, No 1257/1999 of 17 May 1999 Official Journal of the European Communities, L 160/80, 26/6/1999 EEC Council Directive (1972) Concerning measures to encourage the cessation of farming and the reallocation of utilised agricultural area for the purposes of structural improvement, No. 72/160/EEC of 17 April 1972 Official Journal of the European Communities, No L 096, 23/04/1972 EEC Council Regulation (1988) Establishing a Community scheme to encourage the cessation of farming, No. 1096/88 of 25 April 1988 Official Journal of the European Communities, L 110/1, 29/4/1988 EEC Council Regulation (1989) No. 3808/89 of 12 December 1989 amending Regulation (EEC) No. 797/85, (EEC) No. 1096/88, (EEC) No. 1360/78, (EEC) No. 389/82 and (EEC) No. 1696/71 with a view to expediting the adjustment of agricultural production structures. Official Journal of the European Communities, L 371, 20/12/1989 EEC Council Regulation (1992) Instituting a Community aid scheme for early retirement from farming, No. 2079/92 of 30 June 1992 Official Journal of the European Communities, L 215, 30/7/1992 Errington, A. and M. Lobley (2002) Handing over the reins: a comparative study of intergenerational farm transfers in England, France, Canada and the USA. Conference Paper at the Agricultural Economics Society, 8–11 April 2002 (Aberystwyth: University of Wales) © 2007 The Author. Journal Compilation © 2007 European Society for Rural Sociology. Sociologia Ruralis, Vol 47, Number 3, July 2007 270 Bika Fennel, R. (1981) Farm succession in the European Community. Sociologia Ruralis 21 (1) pp. 19–42 Fertig, G. (2003) The invisible chain: niche inheritance and unequal social reproduction in pre-industrial continental Europe. The History of the Family 8 (1) pp. 7–19 Fox, R. (1978) The Tory islanders: a people of the Celtic fringe (Cambridge: Cambridge University Press) Gasson, R. and A. Errington (1993) The farm family business (Wallingford: CAB International) Gillmor D.A. (1999) The scheme of early retirement from farming in the Republic of Ireland. Irish Geography 32 (2) pp. 78–86 Goetz, S.J. and D.L. Debertin (2001) Why farmers quit: a county-level analysis. American Journal of Agricultural Economics 83 (4) pp. 1010–1023 Guinnane, T.W. (1992) Age at leaving home in rural Ireland, 1901–1911. The Journal of Economic History 52 (3) pp. 651–674 Habakkuk, H.J. (1955) Family structure and economic change in nineteenth-century Europe. The Journal of Economic History 15 (1) pp. 1–12 Hallam, H.E. (1958) Some thirteenth-century censuses. Economic History Review 10 (3) pp. 340–361 Hechter, M. and W. Brustein (1980) Regional modes of production and patterns of state formation in western Europe. American Journal of Sociology 85 (5) pp. 1061–1094 Herzfeld, M. (1980) Social tension and inheritance by lot in three Greek villages. Anthropological Quarterly 53 (2) pp. 91–100 Hoggart, K. and A. Paniagua (2001) The restructuring of rural Spain. Journal of Rural Studies 17 (1) pp. 63–80 Holdsworth, C. (1998) Leaving home in Spain: a regional analysis. International Journal of Population Geography 4 (4) pp. 341–360 Holdsworth, C., D. Voas and M. Tranmer (2002) Leaving home in Spain: when, where and why? Regional Studies 36 (9) pp. 989–1004 Homans, G.C. (1969) The explanation of English regional differences. Past and Present 42 (1) pp. 18–34 Howell, C. (1976) Peasant inheritance customs in the Midlands, 1280–1700. Pp. 112–155 in J. Goody, J. Thirsk, and E.P. Thompson eds, Family and inheritance – rural society in Western Europe (Cambridge: Cambridge University Press) Hutson, J. (1987) Fathers and sons: family farms, family businesses and the farming industry. Sociology 21 (2) pp. 215–229 Kaser, K. (2002) Power and inheritance: male domination, property and family in Eastern Europe, 1500–1900. The History of the Family 7 (3) pp. 375–395 Keating, N.C. and B. Munro (1989) Transferring the family farm: process and implications. Family Relations 38 (2) pp. 215–219 Kennedy, L. (1991) Farm succession in modern Ireland: elements of a theory of inheritance. The Economic History Review 44 (3) pp. 477–499 Kimhi, A. (1994) Optimal Timing of farm transferral from parent to child. American Journal of Agricultural Economics 76 (2) pp. 228–236 Kimhi, A. and R. Lopez (1999) A note on farmers’ retirement and succession consideration: evidence from a household survey. Journal of Agricultural Economics 50 (1) pp. 154– 162 Kivelson, V.A. (1994) The effects of partible inheritance: gentry families and the state in Muscovy. The Russian Review 53 (2) pp. 197–212 Koutsomiti, E. (2000) Structural measures of common agricultural policy and sustainable development of insular districts: early retirement, new farmers and environment in the island of Lesvos (Lesvos: University of the Aegean) Lamaison, P. (1988) La diversité des modes de transmission: une géographie tencace. Etudes Rurales (110/111/112) pp. 119–175 © 2007 The Author. Journal Compilation © 2007 European Society for Rural Sociology. Sociologia Ruralis, Vol 47, Number 3, July 2007 Impact of the farmers’ early retirement scheme 271 Lavery, M. (2004) ‘Crazy rules’ – but positive verdict on farm retirement scheme. Irish Farmers Journal 24 January 2004 Le Roy Ladurie, E. (1976) A system of customary law: family structures and inheritance customs in sixteenth-century France. Pp. 75–103 in R. Forster and O. Ranum eds, Family and society (Baltimore, NJ: John Hopkins University Press) Louloudis, L., Maraveyas, N. and Martinos, N. (1993) The social dimension of the CAP. Pp. 463–478 in N. Petralias ed., Dimensions of social policy today (Athens: Sakis Karagiorgas Foundation) Ministry of Agriculture (2001) Performance audit report, information on farmers’ retirement support in budget bills (Dno: 205/54/01), Finland Available online at http://www.vtv.fi/ chapter_images/3842_062001en.pdf Accessed 2 June 2007 Murphy, S. (1997) Evaluation of the scheme of installation aid for young farmers: including an assessment of the effects of the scheme of early retirement from farming (Dublin: Department of Agriculture, Food and Rural Development) Naylor, E.L. (1982) Retirement policy in French agriculture. Journal of Agricultural Economics 33 (1) pp. 25–36 O’Shaughnessy, M. (2004) Europe at the margins: EU regional policy, peripherality and rurality. Paper presented at the Regional Studies Association Conference, 15–16 April 2004 (Angers: University of Angers) Owens, A. (2002) Inheritance and the life-cycle of family firms in the early industrial revolution. Business History 44 (1) pp. 21–46 Paniagua, A.M. (2000) Analysis of the evolution of farmers’ early retirement policy in Spain: the case of Castile and León. Land Use Policy 17 (2) pp. 113–120 Parish, W.L. and M. Schwartz (1972) Household complexity in nineteenth century France. American Sociological Review 37 (2) pp. 154–173 Perrier-Cornet, P. et al., eds (1991) La transmission des exploitations agricoles et l’ installation des agriculteurs dans la CEE: monographies nationales 11 volumes (Brussels: CCE DG VI and Dijon: INRA ESR) Pesquin, C., A. Kimhi and Y. Kislev (1999) Old age security and inter-generational transfer of family farms. European Review of Agricultural Economics 26 (1) pp. 19–37 Pietola, K., M. Väre and A.O. Lansink (2003) Timing and type of exit from farming: farmers’ early retirement programmes in Finland. European Review of Agricultural Economics 30 (1) pp. 99–116 Potter, C. and M. Lobley (1992) Ageing and succession on family farms: the impact on decisionmaking and land use. Sociologia Ruralis 32 (2/3) pp. 317–334 Potter, C. and M. Lobley (1996) The farm family life cycle, succession paths and environmental change in Britain’s countryside. Journal of Agricultural Economics 47 (2) pp. 172– 190 Rogers, C.S. and S. Salamon (1983) Inheritance and social organisation among family farmers. American Ethnologist 10 (3) pp. 529–550 Ross Gordon Consultants SPRL, Belgium (2000) The future of young farmers in the European Union Agriculture, Forestry and Rural Development Series, AGRI 134 EN (European Parliament L-2929 Luxembourg: Directorate General for Research) Scanlan, H. (2002) Retirement scheme needs fine tuning. Irish Farmers’ Journal 2 February 2002 Symes, D.G. (1990) Bridging the generations: succession and inheritance in a changing world. Sociologia Ruralis 30 (3/4) pp. 280–291 Spathis, P. and P. Kaldis (2003) Exploring factors conducive to maintaining Greek farmers in LFAs, Paper presented at the Conference ‘Less Favoured Areas and Development Strategies’, 20–23 November (Lesvos: University of the Aegean) Statens landbruksforvaltning (2003) Evaluering av tidligpensjonsordningen. (Evaluation of the early retirement pension scheme) (Oslo: Norwegian Agricultural Authority) © 2007 The Author. Journal Compilation © 2007 European Society for Rural Sociology. Sociologia Ruralis, Vol 47, Number 3, July 2007 272 Bika Stubberud, Kari V. and K. Samseth (2000) Slekter kommer, slekter går – om generasjonskifte i landbruket. (The times are changing – on the transfer of farms to the younger generation). NILF report no. 4 (Oslo: Norwegian Agricultural Economics Research Institute) Sutton, D.E. (1997) Local names, foreign names: family inheritance and national heritage on a Greek Island. American Ethnologist 24 (2) pp. 415–437 Thompson, E.P. (1976) The grid of inheritance: a comment. Pp. 328–360 in J. Goody, J. Thirsk, and E.P. Thompson eds, Family and inheritance – rural society in Western Europe (Cambridge: Cambridge University Press) Tracy, M. (1989) Government and agriculture in Western Europe 1880–1988 (Hemel Hempstead: Harvester Wheatsheaf ) Väre, M. (2006) Spousal effect and timing of retirement. Journal of Agricultural Economics 57 (1) pp. 65–80 Wallace, M. and J.E. Moss (2002) Farmer Decision-making with conflicting goals: a recursive strategic programming analysis. Journal of Agricultural Economics 53 (1) pp. 82–100 Wegge, S.A. (1999) To part or not to part: emigration and inheritance institutions in nineteenthcentury Hesse-Cassel. Explorations in Economic History 36 (1) pp. 30–55 Zografia Bika School of Management and Economics University of Edinburgh 50 George Square Edinburgh EH8 9JY, UK e-mail: [email protected] © 2007 The Author. Journal Compilation © 2007 European Society for Rural Sociology. Sociologia Ruralis, Vol 47, Number 3, July 2007