Scottish Agricultural Tenure Evidence Review

6 Tenure And Farm Performance

6.1 Surprisingly little empirical evidence is available on the relative performance of farms held under different tenure arrangements. This partly reflects a lack of suitable data for analysis, but also the confounding effects of other influences on farm performance. Indeed the economic literature stresses the context-specific nature of optimal tenure arrangements, highlighting inter-linkages between land tenure decisions and other farming decisions (Schickele, 1941; Currie, 1986; Otsuka et al., 1992; Alan and Lueck, 2003) and policy inter-linkages between tenure legislation, agricultural subsidies and R&D support plus wider rural development topics such as off-farm employment opportunities, housing and pensions (EC, 1982; Swinnen et al., 2013).

6.2 Published analysis of farm tenure is dominated by attention to situations in developing (e.g. Africa, Asia) and transitional (e.g. Eastern Europe) economies where agriculture remains a significant sector and changes to tenure arrangements have been dramatic (e.g. widespread redistribution of land). In such cases, demonstrable efficiency gains are generally reported - but are often relative to a low base and at least partly attributable to basic improvements in legal recognition and protection of private property rights (Ladejinsky, 1964; USAID, 2013).

6.3 For industrialised economies in Western Europe and North America, changes to farm tenure arrangements are typically less dramatic, involving marginal changes to terms and conditions rather than radical reforms. Moreover, agriculture is often subject to other government interventions (e.g. production subsidies) that make it difficult to isolate the impact of modest tenure changes. Indeed Hill (1974 and 1985) suggests that tenure effects have largely been neglected by economists due to a pre-occupation with production and trade aspects of agricultural policies.

6.4 This neglect extends to data collection, with information on tenure often being incomplete and/or of poor quality. For example, few countries appear to routinely collect information on the nature and prevalence of different types of landlord and lease arrangements. Yet landlords take several possible forms - including central and local government, charities and Churches, NGOs, financial institutions, landed estates, other farmers and non-farming rural residents - and the motivations and behaviours of each are likely to be different. Equally, farms can vary considerably in their tenure mix, with the simple extremes of complete owner-occupation and complete tenancy masking the multitude of forms that mixed tenure may take through leases of different types and duration with different landlords (some of whom may be family members).

6.5 In addition, statistical quality assurance is often absent and information is typically lacking on productivity and on how management responsibility relates to stated tenure arrangements.^[14] Consequently, although changes to headline figures of (e.g.) freehold and leasehold land may be reported, these may contain inaccuracies and are typically too coarse to infer much about specific changes or any associated efficiency changes.

6.6 Where attempts have been made to compare farm performance under different tenure arrangements using farm surveys in industrialised countries, the results are somewhat variable and ambiguous. For example, notwithstanding hypotheses that tenanted farms will deploy more working capital and farm more intensively over larger areas, little difference was found in early American studies (e.g. Miller, 1959), although later ones found tenants more focused on short-term profitability (e.g. Garcia et al., 1982).

6.7 In the UK context, Britton and Hill (1978) suggest that tenanted farms may deliver better returns, but that any performance differential relates also to farm size and farm type. Hill (1974) and Gasson and Hill (1984) suggest that mixed tenure farms may deliver superior results, but that farmer characteristics (e.g. attitude, education) are a more important determinant of performance. Moreover, results are sensitive to how finely tenure categories are defined and to how (imputed) family labour and (especially) land costs for owner-occupiers are treated.

6.8 Similarly, comparisons of investment levels between owner-occupied and tenanted farms are equally open to interpretation. For example, although both Bonthron (1969, for Scotland) and Harrison (1975, for England) report higher average levels of investment in fixed assets by owner-occupiers relative to landlords, both note wide variation within a given class of tenure and (moreover) that it is not possible to judge the appropriateness of the investments made (i.e. whilst lower levels of investment could be too low, higher levels could be unnecessarily high). They also highlight the significance of grant-aid in overall investment levels.

6.9 Casual inspection of recent data from the Scottish Farm Accounts Survey (FAS) suggests that average levels of working capital are similar across owner-occupiers and tenants, but that owner-occupiers are carrying significantly more debt in absolute terms - reflecting their (rather than landlord) financing of land and building assets, although rising land values inevitably mean that such debt is proportionately smaller than for tenants when expressed relative to assets. However, there appears to be considerable variation within each type of tenure and across different farm types and sizes.

6.10 More rigorous analysis of viability (based on short and long-run income levels) suggests that tenanted farms may have some performance advantages, but that other factors - notably the degree of diversification into different farm and non-farm activities - have a stronger influence (pers. comm., Barnes, 2014). Further research may be merited, but is beyond the scope of this project.