lable at ScienceDirect

Journal of Rural Studies 32 (2013) 240e250

Contents lists avai

Journal of Rural Studies

journal homepage: www.elsevier .com/locate/ j rurstud

Farmers’ motives for diversifying their farm business e The influenceof family

Helena Hansson a,*, Richard Ferguson a, Christer Olofsson a, Leena Rantamäki-Lahtinen b

aDepartment of Economics, Swedish University of Agricultural Sciences, P.O. Box 7013, SE-75007 Uppsala, SwedenbMTT Agrifood Research Finland, Finland

Keywords:DiversificationFamily businessFarmsFactor analysisMotivesRural policySweden

* Corresponding author. Tel.: þ46 18 67 17 14; fax:E-mail address: Helena.Hansson@slu.se (H. Hansso

0743-0167/$ e see front matter � 2013 Elsevier Ltd.http://dx.doi.org/10.1016/j.jrurstud.2013.07.002

a b s t r a c t

This study examined the motives underlying decisions by farmers to start new ventures outside con-ventional agriculture and assessed the importance of the farm family situation in formulating thesemotives. The study was based on quantitative data obtained from 309 Swedish farmers who self-reported that they ran other ventures in addition to their agricultural production enterprises. Factoranalysis revealed a structure with two underlying motives for starting a venture outside conventionalagriculture, namely ‘business development for reasons to reduce risk and to use idle resources’ and‘business development for social and lifestyle reasons’. The factor scores obtained were related tomeasures of involvement of the farmer’s spouse in planning and managing the new venture and theimportance of family considerations at a more general level. The results showed that the motives forstarting new ventures were dependent on the situation of the farm family, but that the two groups ofindicators identified influenced these motives in opposite ways. These findings are of obvious interest forrural development policy makers.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Diversification of the rural economy is one of the most highlyprioritised themes in rural development policy in European Unionmember states (Council Regulation (EC) No. 1698/2005). Diversifi-cation is considered essential to improve rural economic growth,employment and migration. One recommended strategy fordiversifying the rural economy is to help farmers find new orcomplementary commercial ways of using existing resources (suchas farm buildings, labour, equipment, pasture and forests) that werepreviously employed in conventional agriculture. For the farmbusinesses, this would imply diversification to activities outsidemainstream agriculture. The scientific literature reports great in-terest in farm diversification, its occurrence and determinants (e.g.Ilbery, 1991; McNally, 2001; Chaplin et al., 2004; Gorton et al.,2008; Barbieri and Mahoney, 2009; Maye et al., 2009; Brandthand Haugen, 2011; Grande, 2011; Vik and McElwee, 2011). Manystudies also focus on pluriactivity by farmers, covering not only thedevelopment within the farm business but also incorporating otherincome-generating strategies such as off-farm employment ormultiple business holdings (e.g. Alsos et al., 2003; Serra et al., 2004;McNamara and Weiss, 2005; Lagerkvist et al., 2007).

þ46 18 67 35 02.n).

All rights reserved.

In order to formulate successful policy, policymakers have to takethe behaviour of the target population into close consideration.Pietola and Lansink (2001) emphasised that successful agriculturalpolicy should build on in-depth understanding of the behaviour offarmers. Indeed, several authorshave recentlypointedout theneed toclearly understand the motivation for entrepreneurs in order to un-derstand their behaviour (e.g. Delmar 1996; Baum et al., 2001; Shaneet al., 2003; Schjoedt and Shaver, 2007; Kirkwood, 2009) and thestart-up of new ventures in agricultural businesses is an example ofentrepreneurial activity (Alsos et al., 2003; McElwee, 2008).

There has long been an interest in the literature in relatingreadily observable farm and farmer characteristics, such as existingagricultural enterprises, farm size, farmer’s age and gender, etc. tothe occurrence of farm diversification and pluriactivity (e.g.McNally, 2001; Chaplin et al., 2004; Serra et al., 2004; McNamaraand Weiss, 2005; Lagerkvist et al., 2007; Maye et al., 2009). How-ever, there is now a growing interest in also understanding themotives underlying decisions by rural actors, such as farmers, todiversify their businesses (Polovitz Nickerson et al., 2001; Alsoset al., 2003; McGehee and Kim, 2004; McGehee et al., 2007;Barbieri and Mahoney, 2009; Northcote and Alonso, 2011; Vikand McElwee, 2011). For instance, Barbieri and Mahoney (2009)found that a desire to reduce risk and uncertainty was the mostimportant goal underlying decisions by farmers and ranchers inTexas to start diversified ventures, followed by a desire to grow and

H. Hansson et al. / Journal of Rural Studies 32 (2013) 240e250 241

provide market services, enhance financial conditions, fulfil per-sonal aspirations and pursuits, enhance revenue and family con-siderations. Similarly, Northcote and Alonso (2011) found that riskconsiderations, market opportunities, access to resources and life-style factors influenced diversification decisions among olive-growing farmers in Western Australia. Vik and McElwee (2011)found social motives to be more important than additional in-come in explaining diversification. Alsos et al. (2003) suggestedthere are three types of diversifying farmers; the pluriactive farmer,the resource-exploiting entrepreneur and the portfolio entrepre-neur. The pluriactive farmer is said to start the new business ac-tivities “in order to be able to sustain farming or expand the farm tobe the workplace for more family members, which in both casesdemands more income-generating activities” (Alsos et al., 2003, p.439); the resource-exploiting entrepreneur is motivated by the goalof utilising existing resources; and the portfolio entrepreneur isdriven by the goal of exploiting a business opportunity.

The consensus from this literature is that farmers’ motives fordiversifying their businesses are complex and include severalconsiderations other than economic aspects. Hence, rural policythat is built on too simplistic a viewof the goals underlying farmers’decision making, e.g. that farmers only undertake activities thatmaximise their economic return at a given level of risk, is unlikelyto be successful in influencing the behaviour of farmers in theintended way. Furthermore, the motives underlying farmers’ de-cision making are likely to be context-dependent, i.e. embedded in,and dependent on, the context in which the farm business is situ-ated. Given the ownership structure of farms in most Westerncountries, where farms are run as family businesses, one obviousand important consideration is the family situation of the familyowning the farm. Literature on family business management hasrecently called for a deeper understanding of how the family in-fluences the strategic management of family-owned firms (Sharmaet al., 1997; Chua et al., 1999; Chrisman et al., 2003; Chua et al.,2003; Westhead and Howorth, 2006). A presupposition in thepresent study is that the family situation will influence the motivesunderlying farmers’ decisions to diversify their farm businesses andthat this needs to be explicitly studied in order to understand howthe farm sector can develop in the future. Indeed, Peltola (2000, p.248) points out that pluriactivity by farmers may considerablybalance out variations in farm income and provide a more stablelivelihood. From this perspective, on-farm diversification may beviewed as a survival or adaptation strategy for the farm family andthe motives for this behaviour should be dependent on the farmfamily. However, this context-dependency of motives behinddiversifying farm businesses outside conventional agriculture hasnot been researched in previous literature.

Accordingly, this study focused on the motives underlying de-cisions by farmers to start new or complementary ventures outsideconventional agriculture, with the aim of identifying how thesemotives are dependent on the farm family situation. The aimwas toprovide information on how family involvement in the creation(planning and management) of the new venture and how familyconsiderations in general affect the motives underlying farmers’decisions to diversify outside conventional agriculture. The studywas based on empirical evidence relating to a sample of 309 farms inSweden that had reported starting a newor complementary ventureoutside conventional agriculture, or were planning to do so.

2. Theoretical framework

2.1. How to define a diversified farm business

From a theoretical, strategic management point of view, diver-sification of businesses can be analysed in what is called the Ansoff

product market growth matrix (originally due to Ansoff, 1957),where a firm can adopt three diversification strategies: i) Devel-oping new products for existing markets; ii) bringing existingproducts into new markets; and iii) serving newmarkets with newproducts (e.g. Johnson et al., 2011). According to this system, a farmbusiness would be diversified if following any one or more of thesethree strategies. However, in most of the farm business literature,the concept of a diversified farm business is less straight-forward.

While there is no generally agreed standard definition of what ismeant by a diversified farm business, a review of definitions used inthe relevant literature (e.g. Ilbery,1991;McNally, 2001; Turner et al.,2003; Barbieri and Mahoney, 2009) suggests that to be considereddiversified, a farmwould have to use its farm resources for activitiesother than production of conventional crops and livestock togenerate income, or would have to add value to raw materialsoriginating from primary production, e.g. by running a small-scaleprocessing plant. Hence, the definition of a diversified farm busi-ness is somewhat different to that used in strategic management, inthat such farms would have to use their farm resources to generateincomeoutside conventional farmproduction, e.g. a farmproducingand selling milk and grain would not be considered diversified.Another difference is that what is considered to be vertical inte-gration (further processing of a primary product or on-farm mar-keting and retailing) in the strategic management literature (e.g.Johnson et al., 2011) is considered to be diversification in the farmdiversification literature (e.g. McNally, 2001; Turner et al., 2003). Onthe other hand, vertical integration as normally undertaken onfarms, such as on-farm processing or marketing and retailing, ofteninvolves adding value to the products and satisfying the needs of aparticular type of customer (often customers valuing locally pro-duced food, traceability and/or organic production), and cantherefore be considered diversification in the strategicmanagementsense. Furthermore, because the definition of a diversified farmbusiness is alsomade in relation towhat is considered conventionalfarm production, the definition is likely to change over time, asnotedbyTurner et al. (2003). Those authors provided the example oforganic farming, which was initially considered farm diversificationbut is now part of mainstream farming. While this imprecise defi-nition of a diversified farm business may be problematic from atheoretical point of view, it is convenient from a practical point ofview since it is adaptable to the changing terms and conditionsunder which farming is conducted.

This study takes its starting point in the farm business, with thefarmer as the manager, and defines diversification as involvementin ventures outside conventional agriculture in which farm re-sources are used, or ventures based on further on-farm processingand/or marketing and retailing of products. This definition of farmdiversification closely resembles that used in previous studies (e.g.McNally, 2001; Turner et al., 2003; Barbieri and Mahoney, 2009;Vik and McElwee, 2011), and is based on the distinction that farmdiversification is viewed as activities taking place outside conven-tional agriculture ewhich in turn is viewed as one enterprisee butwhere farm resources are used. Hence, a farm that is engaged in adiverse set of agricultural productions, such as livestock, forestryand crops, is not considered to be diversified with the definitionused here. Furthermore, a farm engaged in on-farm processing ormarketing and retailing aiming at meeting the needs of a particulartype of customers is considered diversified with this definitioneven though its production may be highly specialised.

It should be noted that this definition of a diversified farmbusiness concerns the activities undertaken at the farm, using thefarm’s resources. A parallel concept, which should not be confusedwith diversification, is pluriactivity, which focuses on the activitiesof the farmer and is based on the ‘diversified’ sources of income forthe farm household (as opposed to the farm business). Hence, while

H. Hansson et al. / Journal of Rural Studies 32 (2013) 240e250242

pluriactivity includes off-farm income and other independent(non-agricultural) businesses that the farmer may run, the interesthere is limited to the income-generating activities conducted at thefarm, using farm resources.

2.2. Farmers as entrepreneurs and motives behind farmdiversification

Following the definition of entrepreneurship developed byShane and Venkataraman (2000) and Shane et al. (2003), whichconsiders the creative process whereby an entrepreneur discoversnew opportunities and uses resources (existing or new) to evaluateand exploit these, in this study the start-up of new ventures outsideconventional farm businesses has been considered an entrepre-neurial activity. Although a number of farmers may start similartypes of new ventures, the start-up of these ventures is stillconsidered a creative process since it requires contracts with newcustomers and/or suppliers, new marketing channels and re-organisation of the management in the business. McElwee (2008)suggests a taxonomy of farmers where those engaged in variousnon-farming activities are considered entrepreneurial; and Carterand Rosa (1998) suggest that farmers, like other small businessowner-managers, adjust their firms to changing conditions. Thus,diversification of the farm business outside conventional agricul-ture can be seen as a creative response and a way of implementingnew innovations devised by the farmer or other family members.

The motives behind entrepreneurial activities in general (notparticularly farm businesses) and why people take the step ofstarting ventures have long been of interest in the entrepreneurshipliterature (e.g. Brockhaus, 1980; Segal et al., 2005; Schjoedt andShaver, 2007; Kirkwood, 2009). Although this literature is primar-ily interested in what motivates employed people to leave theiremployment to start ventures, i.e. nascent entrepreneurs, similarterminology is used in the present study. The general consensus isthat entrepreneurs are either pushed into entrepreneurial activitiesby dissatisfaction with their previous job situations, or pulled intothem by job and life satisfaction (see Schjoedt and Shaver, 2007 foran overview of the literature). However, this view of entrepre-neurshipwas challenged by Schjoedt and Shaver (2007), who foundthat job dissatisfaction could not explain start-up of new ventures.They concluded that people who start new ventures are relativelyhappy with their existing jobs, so pull factors could be expected toprovide more motivation for entrepreneurial activities.

The push/pull literature would suggest that diversificationoutside conventional agriculture is “opportunity driven” (pull factor)or “necessity driven” (push factor). The termopportunity driven couldbe used in a situation in which new non-agricultural activities arestarted because the farmer has seen a business opportunity, wants toimplement a good business idea, or wants to reallocate existing re-sources and/or gain business growth. The term necessity drivencouldbeused to refer to a situation inwhich farmerhas todiversify inorder to become self-employed, secure family income or decreaserisks caused by changes in the market situation (see Bowler, 1999, p.297; Barrett et al., 2001, p. 315; United Nations, 2001, p. 8).

The push/pull literature should be distinguished from otherliterature describing farmers’ goals and values at a more generallevel (Gasson, 1973; Willock et al., 1999). The interest in the push/pull framework is in understanding the motives for moving fromone stage to another, rather than the general goals and values thatmay more continuously underlie decision-making.

2.3. The farm, the farm family and farm diversification

It has been argued that many rural entrepreneurs are in factlifestyle entrepreneurs, who emphasise aspects such as quality of

life, customer-centred development, good leadership skills, internalmarketing and a good reputation, as well as prioritising successionof the next generation over financial success (Potter and Lobley,1992; Gasson and Errington, 1993; Komppula, 2004; Reijonen,2009). This is especially the case in farming. For instance, manyfarmers highly value aspects of their work such as the ability tomaintain their lifestyle and good family life (Cuykendal et al., 2002).Likewise, Duffy and Nanhou (2002) claim that analyses basedpurely on a profit maximisation assumption are not valid forfarming. A farm often has two interconnected dimensions: the farmand the farmer’s family living on the farm. Making a profit is oneelement of the utility function of the farm. Other elements mayinvolve job and family satisfaction, power, social prestige, or adesire for a quiet life. While in some cases profit maximisation canreinforce other elements of the utility function, in many casesfarmers are likely to face trade-offs between profit maximisationand these other elements.

Family business management literature suggests that businesseswith strong connections to the owners’ family distinguish them-selves from other types of businesses and that the influence of thefamily needs to be explicitly understood and studied (Sharma et al.,1997; Chua et al., 1999; Chrisman et al., 2003; Chua et al., 2003;Westhead and Howorth, 2006). Sharma et al. (1997) defined a fam-ily business as “a business governed and/or managed on [a] sus-tainable, potentially cross-generational, basis to shape and perhapspursue the formal or implicit vision of the business held bymembersof the same family or a small number of families” (p. 2). This accu-rately describes farms in general inmanyWestern countries. Sharmaet al. (1997) andChrismanet al. (2003) claim that the influence of thefamily is especially pronounced in strategicmanagement of thefirm;in the goals set by familymembers and therefore also in the need forthe firm to have an implicit or explicit strategy to implement thosegoals and for a control mechanism to monitor progress towards thegoals. The start-up of new ventures outside conventional agricultureis an example of a strategic decision on the farm and therefore adecision with underlying motives that is likely to be significantlyinfluenced by family considerations. Indeed, non-financial goals andfamily agendas are significant in family businesses (Westhead andHoworth, 2006). Hence, family agendas are likely to drive thedevelopment of the farm business. At the same time, the embedd-edness of the farm business in the farmer’s family might in itselfmotivate the creation of new ventures outside agriculture. Forinstance, unsatisfied needs of family members as regards employ-ment and the desire for a stable net family income might work aspush factors for the start-up of new ventures outside agriculture.

3. Data and methods

The unit of analysis in this study is the farmer in his/her role asmanager of a farm business, and we are interested in the motives ofthe farmers in diversifying their farm businesses outside conven-tional agriculture, and how these motives are dependent on thefamily situation of the farmer. The study is based on data obtainedin a questionnaire sent to the 929 farmers who participated in theSwedish Farm Economic Survey in 2008. The purpose of the FarmEconomic Survey is to serve as a basis for Sweden’s participation inthe EU Farm Accounting Data Network, for which the target pop-ulation is farms comprising at least eight European size units. In thecase of Sweden, this size restriction means that the approximately30 000 largest farms are the target population in the Farm Eco-nomic Survey. The sample used in the Farm Economic Survey isstratified according to geographic location, production type andeconomic size, measured in standardised gross-margins.

Considering that official Swedish statistics reports 72 609 farmsin Sweden in 2007, the size limit used in the Farm Economic Survey

H. Hansson et al. / Journal of Rural Studies 32 (2013) 240e250 243

may seem unnecessarily restricting. However, the definition of afarm used in official Swedish statistics is rather broad, classifying aholding as a farm if consists of at least two hectares of arable land,has a horticultural production with at least 2500 m2 indoor culti-vation or 200m2 outdoor cultivation, or has large animal stock. Thismeans that very small holdings are included and counted as farms,while they e in practice in Sweden e are more likely to be countryside homes of people who are working full-time in nearby towns.Adding a criterion of on-farm labour requirement of at least 800 h,which represents about 50% of a full-time position in Sweden andensures significant commercial farm activities, leaves a populationof about 30 000 farms e about the same size as in the farm eco-nomic survey. Thus, excluding very small farms from the goalpopulation used in the present study limits possible bias fromagricultural holdings that are not likely to be commercial busi-nesses. Furthermore, in Swedish agriculture, it seems that thepropensity to diversify outside conventional agriculture is greaterin the group of larger farms, suggesting that it is this group of farmsthat currently drive rural economic development in this respect.Indeed, the Swedish Board of Agriculture (2007) reported that 30%of all holdings registered as farms in Swedenwere to be categorisedas diversified, having additional ventures outside conventionalagriculture, whereas Hansson et al. (2010) reported that as many as75% of the farms in the Farm Economic Survey in 2007 werediversified in the sense that they obtained revenue from activitiesthat used existing farm resources outside of the conventional farmbusiness and that 5% of the sample was diversified in the sense thatthey provided the market with value-added products.

The questionnaire used in this study was sent out in November2009 and after two reminders a final response rate of 73% wasachieved. The questionnaire sought to collect data for a largerresearch project, ofwhich this study is one part. Farmerswere askedto indicate whether they alone, together with their spouse, or theirspouse alone run any kind of venture apart from conventionalagriculture on their farm, or planned to do so. The questionnairespecified that ventures not yet yielding revenue should be includedand should be considered an on-going venture outside conventionalagriculture. Farmers were also asked to indicate what type of ven-ture they ran or planned to start running, with the following alter-natives given: contract work, farm building rentals, housing rentalsto permanent tenants, agri-tourism ventures such as farm stays orsummer cottage rentals, farm shops or other direct sales outlets, on-farm processing of farm products, horse-related businesses, energyproduction and “other”. In the last alternative, respondents wereasked to indicate what type of venture they ran. Some mentionedforestry, but since this is typically viewed as a conventional farmactivity, these farmers were re-coded and considered non-diversified unless they had also indicated that they ran some othertype of venture outside conventional agriculture. In total, 309farmers reported that they ran ventures in addition to their con-ventional agricultural enterprises and were thus included in thestudy. Of these,1 63% reported contract work, 9% farm buildingrentals, 23% housing rentals to permanent tenants, 11% agriculturaltourism, 8% farm shops, 7% on-farmprocessingof farmproducts,11%horse-related businesses, 9% energy production and 33% “other”.

3.1. Methodological considerations when measuring farmers’motives for diversification

In this study, we considered the motives for entrepreneurialbehaviours to be theoretical constructs that cannot be measured

1 Some respondents ran several ventures outside conventional agriculture andhence the figures add up to more than 100%.

directly. Instead, empirical indicators that were considered toreflect the underlying theoretical constructs were used. A review ofthe literature examining the motives for entrepreneurial behav-iours suggests that both single- andmultiple-item approaches havebeen used (e.g. Polovitz Nickerson et al., 2001; McGehee and Kim,2004; Schjoedt and Shaver, 2007; Barbieri and Mahoney, 2009;Vik and McElwee, 2011). In this study, a multiple-item approachand psychometric methods were used to estimate the underlyingtheoretical constructs. The advantages of a multiple-item approachare that so-called latent constructs are captured by explicitly takingthe construct validity and reliability into consideration.

When estimating latent constructs based onmeasurement items,an important decision is whether to use a reflective or formativemeasurement model (e.g. Podsakoff et al., 2003). Jarvis et al. (2003)provided a set of criteria for deciding this, namely causality be-tween construct and measurement indicators, interchangeabilitybetween indicators, covariation between indicators and whether ornot indicators have the samenomological network.Whenmeasuringthe motives for entrepreneurial behaviour, we concluded that areflective measurement model should be used for the followingreasons. i) Causality: it is plausible to assume that changes in themotives for entrepreneurial behaviour cause changes in the mea-surement items and not the other way around; ii) measurementitems for one motive for entrepreneurial behaviour can be assumedto be interchangeable without fundamentally changing the theo-retical meaning of the construct; iii) indicators of one motive forentrepreneurial behaviour can be assumed to co-vary; and iv) mea-surement items are assumed to come from the same nomologicalnetwork, i.e. that they share antecedents and consequences.

3.2. Scale design

In order to capture farmers’ motives for developing venturesoutside agriculture, ameasurement scalewas adapted fromBarbieriand Mahoney (2009). The exact wording used for measurementitemswas developed to fit the context of our study and for the samereason a fewmeasurement itemswere dropped and two added, butin all essential respects the measurement scale used here corre-sponds to that of Barbieri and Mahoney (2009). A similar mea-surement scale was used by Polovitz Nickerson et al. (2001) andMcGehee and Kim (2004) to investigate farmers’ motives for start-ing agri-tourism. By reducing the measurement items by principalcomponent factor analysis (PCA) to their underlying components,Barbieri and Mahoney (2009) found that the components underly-ing farmers’ decisions to diversify related to: reducing uncertaintyand risk, growing and servicing markets, improving finances, per-sonal aspirations and pursuits, increasing revenue and family con-nections. Polovitz Nickerson et al. (2001) and McGehee and Kim(2004) also reduced their measurement scales with PCA, produc-ing underlying components that they categorised as: Social reasons,economic reasons and external influences, and formal motivations,formal-substantive mix motivations and substantive-formal mixmotivations, respectively. For the reasons stated above, we questionthe rationale behind using PCA, which is based on a formativemeasurement model, instead of a principal axis factor analysis,which is based on a reflective measurement model, to find the un-derlying motivational constructs. Therefore, we do not expect ourmaterial to be reduced to the sameunderlying factors (or constructs)when applying a reflective measurement model.

The measurement items were categorised as being of push orpull type, although in the original forms of the measurement scalethis typology was not used. Items associated with responding to amarket opportunity or business idea, desire to reallocate existingresources (including working hours offered by family members),independence and lifestyle were categorised as pull motivators,

H. Hansson et al. / Journal of Rural Studies 32 (2013) 240e250244

whereas items associated with a poor economic situation in thebusiness (including debt) and undesirable fluctuations in businessincome over the year were categorised as push motivators. In-dicators associated with continuing the business, keeping thebusiness within the family, making the business attractive prior to ageneration change and attracting labour were categorised as pushmotivators, because of their origin in necessity.

Measurement items were posed as statements and the re-spondents were asked to indicate on a Likert scale of 1e5 the degreeto which they agreed with the proposed statements. In order toensure that respondents coulddistinguishbetween responseoptionsin ameaningful way, the anchors ‘completely unimportant’, ‘neutral’and ‘very important’were used for the lower,middle and upper scalepoints respectively. No opt-out alternatives were provided.

3.3. Measuring family influence

According to Sharma et al. (1997) and Chrisman et al. (2003), theinfluence of family on the business is especially pronounced in thestrategic management of family firms. The family members set upgoals that thefirmneeds an implicit or explicit strategy to implementand a control mechanism to monitor progress. Therefore, in order toanalyse the influence of the farm family on the underlying motivesfor the start-up of ventures outside conventional agriculture, weassessed the magnitude of family considerations in general whenrunning the business and the involvement of the farm family in thecreation of the new venture. Our starting hypothesis was that farmswith stronger family considerations and with active involvement ofthe farmer’s spouse have different motives for diversification thanfarms where family considerations are less pronounced and wherethe spouse is not involved in the creation of the new venture.

The magnitude of family considerations in general whenrunning the business was measured by developing four measure-ment items. These, which were posed as statements on a 1e5 Likertscale, were: “When I run my business it is important to. 1. Workwith family members. 2. Hand over awell-kept business to the nextgeneration. 3. Keep the business within the family. 4. Manage afamily farm.” A factor analysis (not shown) of these indicatorssuggested unidimensionality of the scale and a Cronbach’s alpha of0.790 showed reliability. The summed scale of these measurementitems was used in the subsequent analyses.

The involvement of the farm family in the creation of the newventure was measured by considering the spouse’s participation inthe planning process that led to the setting up of the new ventureoutside conventional agriculture and in the actualmanagementof thenew venture. In particular, the spouse’s involvement in the planningand actual management of the new venture was measured withdummy variables with the following response options: 1. My spousewas the main planner of the new venture. 2. My spouse and I tookequal parts in the planning of the newventure. 3.My spouse took lesspart thanme in the planningof the newventure. 4.Myspouse did nottake part at all in the planning of the new venture. For the questionabout the spouse’s involvement in the actualmanagement of thenewventure, the following response options were used: 1. My spouse isthe main manager of the new venture. 2. My spouse and I take equalparts in themanagement of the new venture. 3. My spouse takes lesspart than me in the management of the new venture. 4. My spousedoes not take part at all in the planning of the new venture. For bothquestions an alternative response option was provided to fit unmar-ried respondents and respondents not living together with someone.

3.4. Statistical methods

The data obtained were analysed in two separate steps. In thefirst of these, an exploratory common factor analysis was used to

develop measures of the motives driving the development of newventures outside conventional agriculture by summarising the re-spondents’ statements into a number of factors representing theunderlying theoretical constructs. The exploratory approach waspreferred to the confirmatory approach (see e.g. Podsakoff et al.,2003; Hair et al., 2010) because, for reasons stated above, wecould not assume from previous literature the underlying motiveswe might find.

To ensure the factorability of the matrix of measurement items,Kaiser’s overall measure of sampling adequacy (KMO) was evalu-ated. The overall KMO and that of all individual measurement itemsshould be above 0.5 for a factorable matrix (Hair et al., 2010). Bothconditions were satisfied. The number of factors retained in thefinal analysis was guided by Kaiser’s criterion, which is based on theeigen-values of the factors, by the suggestions of screeplot andparallel analysis, and by the ambition to obtain a factor solutionthat could be interpreted in a theoretically sound and reliable way.

Factor loadings were considered significant if they were above0.35, which is the recommended cut-off value for a sample sizebetween 250 and 350 (Hair et al., 2010). Measurement items withlower factor loadings were deleted one at a time, starting with thatwith the lowest communality, until only significant factor loadingsremained. The factor solution was rotated to facilitate interpreta-tion, so that for each item its loading on a single factor was maxi-mised and for each factor its number of highly loading measureditems was minimised. The oblique (oblimin) rotation was used forthis purpose, since it is plausible to assume that the motives forstarting new ventures outside conventional agriculture are corre-lated and the oblique rotation was therefore expected to generatetheoretically more valid factors.

Cronbach’s alpha, item-to-total and item-to-item correlationswere assessed once the final factor structure had been establishedto evaluate the reliability of the measurement scales obtained (Hairet al., 2010). Factor scores were predicted to obtain individualmeasures of the motives.

In a second step, the measures used to capture family influencewere regressed on the factor scores. The seemingly unrelatedregression techniquewas used for this, because the same regressorswere used in both models. The factor scores were tested fornormality, but since normality had to be rejected statistical infer-ence was based on bootstrapped standard errors.

All statistical analyses were conducted using STATA version 10.1and 12.0.

4. Results

4.1. Description of measurement items

Average scores and standard deviations of the measurementitems used to capture farmers’ motives for starting new venturesoutside conventional agriculture are shown in Table 1. The stron-gest motive of those suggested appeared to be the desire tostrengthen the economic situation of the business, followed by theperception of a market demand and reallocation of resources thatcould not be fully used on the farm. The least important of thesuggested motives was making the business more attractive toemployees. This may be a reflection of the fact that many Swedishfarms are still too small to have employees. Table 1 also categorisesthe measurement items into the push/pull typology.

4.2. Common factor analysis

The results of the common factor analysis are displayed inTable 2. Kaiser’s criterion suggested that there were two factors,whereas the parallel analysis and screeplot suggested there might

Table 1Descriptive statistics of suggested motives for starting new ventures outside conventional agriculture.

Suggested motive Push/pullcategorisation

Averagescore

Standarddeviation

I/we wanted to test a new business idea Pull 3.103 1.051The business did not fully employ all family members who wanted to work

in the businessPull 2.926 1.266

Unacceptable economic situation in the business Push 2.937 1.258The business had resources (e.g. machinery and/or buildings) that could not

be fully used on the farmPull 3.586 1.204

I/we perceived a market demand Pull 3.589 1.072I/we wanted to reduce fluctuations in the business income over the year Push 3.159 1.288I/we wanted to reduce the impact of negative aspects such as falls in producer prices Push 3.125 1.208I/we wanted to get the business less dependent on external developments

such as business cycle, political decisions etc.Push 3.244 1.129

It was a way of being able to continue the business Push 3.111 1.298I/we wanted to increase income at times of the year when the farm

generates little or no revenuePush 3.289 1.301

I/we wanted to strengthen the economic situation of the business Push 3.809 1.127I/we wanted closer contact with customers Pull 2.629 1.193I/we wanted to reduce debts in the business Push 2.826 1.309I/we wanted to improve my/our quality of life Pull 3.360 1.188I/we wanted to capitalise on a hobby Pull 2.486 1.255It was a way of keeping the business within the family Push 2.493 1.290It was a way of making the business attractive prior to a generation change Push 2.397 1.293It was a way of making the business more attractive to employees Push 2.108 1.186

Note: The statements weremeasured on a Likert scale ranging from 1 to 5 (completely unimportante very important). Theminimum score of all suggested goals was 1 and themaximum score was 5.

H. Hansson et al. / Journal of Rural Studies 32 (2013) 240e250 245

be as many as five factors. Therefore, solutions with five, four, threeand two factors were evaluated. A factor solution with two factorswas retained on the grounds that the solution was sound from ascale-development point of view (significant loadings on eachfactor, no double-loading items and reliable measurement scalesfor each factor) and hence the factors could be interpreted on asound basis. Three statements did not load significantly on anyfactor in the final solution and were therefore removed. KMO of thefinal matrix was 0.900, with KMO of individual measurement in-dicators ranging from 0.812 to 0.953, showing that the matrix washighly factorable.

Table 2Factor solution for suggested motives.

Suggested motive Pc

I/we wanted to increase income at times of the year when the farm generateslittle or no revenue

P

I/we wanted to reduce fluctuations in the business income over the year PI/we wanted to strengthen the economic situation of the business PI/we wanted to get the business less dependent on external developments

such as business cycle, political decisions etc.P

Unacceptable economic situation in the business PIt was a way of being able to continue the business PI/we wanted to reduce the impact of negative aspects such as

falls in producer pricesP

The business did not fully employ all family members who wantedto work in the business

P

I/we wanted to reduce debts in the business PIt was a way of making the business attractive prior to a generation change PIt was a way of keeping the business within the family PI/we wanted to capitalise on a hobby PI/we wanted closer contact with customers PIt was a way of making the business more attractive to employees PI/we wanted to improve my/our quality of life PCronbach’s alphaa

Range item-to-item correlationa

Range item-to-total correlationa

Note: Significant factor loadings in bold. Three items did not load significantly on any faa Based on significant loadings.

The two factors in the solution were considered to be inter-pretable in a meaningful and valid way. The first factor mainlycomprised measurement items related to the economic situation inthe business, fluctuations in business income and producer pricesand dependency on external developments. We interpreted this asreflecting a desire to develop the business to reduce economic riskin the business and to use idle resources.

The second factor mainly comprised measurement itemsrelating to the development of the situation surrounding thefarmer and covering both his/her life situation in general (‘I/wewanted to improvemy/our quality of life’, ‘I/wewanted to capitalise

ush/pullategorisation

Factor 1 business developmentfor reasons to reduce risk andto use idle resources

Factor 2 businessdevelopment for socialand lifestyle reasons

ush 0.836 �0.081

ush 0.830 0.007ush 0.760 �0.075ush 0.749 0.082

ush 0.738 �0.054ush 0.737 0.040ush 0.717 0.126

ull 0.423 0.066

ush 0.485 0.236ush �0.062 0.835ush 0.165 0.685ull �0.117 0.606ull 0.093 0.577ush 0.031 0.531ull 0.255 0.420

0.902 0.8120.221e0.752 0.234e0.7460.556e0.844 0.664e0.817

ctor and were removed from the analysis.

H. Hansson et al. / Journal of Rural Studies 32 (2013) 240e250246

on a hobby’, ‘It was away of keeping the business within family’ and‘It was away of making the business attractive prior to a generationchange’) as well as motives more directly related to the business (‘I/we wanted closer contact with customers’ and ‘It was a way ofmaking the business more attractive to employees’) of the farmerand the business. This factor was interpreted as reflecting an un-derlying motive of business development for social and lifestylereasons.

While the Spearman rank correlation between the factors is0.599, showing that the factors are strongly related to each other,the factors differ conceptually in that the first factor represents amotive more oriented towards economic considerations whereasthe second factor represents a motive more oriented towards socialand lifestyle considerations.

Item-to-total correlations based on measurement items withsignificant factor loadings and considered for each factor were allabove the recommended cut-off value of 0.5 (Hair et al., 2010). Allitems representing one factor with significant loadings correlatedgenerally to each other, with a correlation coefficient above thesuggested cut-off value of 0.3 (Hair et al., 2010), although the in-dividual correlation coefficients were somewhat lower on threeoccasions. For both factors, Cronbach’s alpha, based on factors withsignificant loadings, was well above the suggested cut-off value of0.7 (or even 0.6 in exploratory settings such as this) (Hair et al.,2010). Overall, these figures show that the obtained measurementscales are reasonably reliable.

4.3. Influence of the farm family on motives

4.3.1. Bivariate analysis for spouse’s involvement in creation of thenew venture

The results of a bivariate analysis of the influence of the spouse’sinvolvement in creation of new ventures on motives underlyingdecisions to develop ventures outside conventional agriculture areshown in Table 3. The sample was divided into groups dependingon the degree of involvement of the spouse in the planning processand the actual management, respectively, and average factor scoresfor the two motives were calculated for all groups. Group-belonging is indicated on the rows and average factor scores arereported in the columns. Interestingly, the data generally showedthat the lowest average factor scores were obtained by the group ofrespondents who indicated that the spouse was the main planner/main manager of the new venture and the group who indicatedthat the spouse did not take any part in the planning process or inthe management. In both the planning and the management case,

Table 3Bivariate analysis of the influence of the spouse’s involvement in the creation of new ve

Percentageof sample

Spouse’s involvement in planning processMy spouse was the main planner of the new venture 6My spouse and I took equal parts in the planning of the new venture 23My spouse took less part than me in the planning of the new venture 30My spouse did not take part at all in the planning of the new venture 32Spouse’s involvement in the actual managementMy spouse is the main manager of the new venture 10My spouse and I take equal parts in the management of the

new venture10

My spouse takes less part than me in the management of thenew venture

29

My spouse does not take part at all in the management of thenew venture

40

Note: Average factor scores differ significantly within the groups (p-value between 0.097populations. Response option “I am single” not reported. The sample is represented twi

the larger average factor scores were obtained by the group of re-spondents who indicated that the spouse did take part in the cre-ation of the new venture, but to equal or smaller extent than therespondent.

4.3.2. Correlations between factor scores and general familyconsiderations

A correlation analysis between the summed scale for familyconsiderations in the business in general and the two motivationalconstructs was performed. This analysis showed that family con-siderations in the business in general were strongly and signifi-cantly correlated to ‘business development for reasons to reducerisk and to use idle resources’ and to ‘business development forsocial and lifestyle reasons’, with correlation coefficients of 0.272(p-value 0.000) and 0.379 (p-value 0.000), respectively.

4.3.3. Estimation of the modelAn analysis (not shown) of the spouse’s involvement in the

planning process and in the management of the new ventureshowed that respondents had answered these questions verysimilarly. For instance, if a spouse was the main planner of a newventure, he or she was also the main manager of the new venture.Therefore two regression models were run: In Model 1, the impactof the spouse’s involvement in the planning of the new venturewasconsidered, whereas in Model 2, the impact of the spouse’sinvolvement in the management of the new venture was consid-ered. In bothmodels, the summed scale for the magnitude of familyconsiderations in the business in general was included. Further-more, the following control variables were used: distance (in kil-ometres) to the nearest town with at least 20 000 inhabitants (aproxy for spouse’s prospects of obtaining employment outside thefarm business), number of members in the household, share ofrevenue in 2009 estimated to originate from diversified activities,existence of resources that cannot be fully used in the conventionalagriculture, and whether all household members who are able towork can support themselves in the conventional agriculture.

The results are shown in Tables 4 and 5. In both models, thesummed scale for family considerations in the business in generalemerged as a strongly significant predictor. However, the spouse’sinvolvement in planning and management of the new ventureseemed to have less influence on the motives underlying decisionsto develop the new venture. In both models, the spouse’s involve-ment in the planning and management process was statisticallysignificant only when he/she was the main planner/manager andonly for ‘business development to reduce risk and to use idle

ntures.

Average factor score for motive: businessdevelopment for reasons to reduce riskand to use idle resourcesAverage (st.d.)

Average factor score for motive:business development for socialand lifestyle reasonsAverage figure (st.d.)

�0.93 (1.200) �0.28 (0.897)�0.0003 (0.980) 0.04 (0.945)

0.18 (0.860) 0.24 (0.872)�0.03 (0.891) �0.22 (0.866)

�0.68 (1.032) �0.12 (0.799)0.11 (0.760) 0.31 (0.850)

0.14 (0.925) 0.12 (0.909)

�0.03 (0.942) �0.15 (0.926)

and 0.012) according to the non-parametric KruskaleWallis rank test of equality ofce; once for the planning question and once for the management question.

Table 4Regression results for spouse’s involvement in planning, Model 1. Seemingly unrelated regression model.

Dependent variable; motive: ‘Businessdevelopment for reasons to reducerisk and to use idle resources’

Dependent variable; motive: ‘Businessdevelopment for social and lifestylereasons’

Estimated coefficient Estimated coefficient

Intercept �0.904c �1.020b

My spouse was the main planner of the new venturea �1.116b �0.447My spouse and I took equal parts in the planning of the new venturea �0.327 �0.312My spouse took less part than me in the planning of the new venturea �0.269 �0.127My spouse did not take part at all in the planning of the new venturea �0.221 �0.363Family considerations 0.069b 0.088b

Distance to nearest town with at least 20 000 inhabitants �0.001 �0.002Number of members in household 0.012 �0.001Share of revenue in 2009 originating from diversified activities (%) 0.014b 0.006c

Resources that cannot be fully used in conventional agriculturea 0.013 �0.084All household members who can work can support themselves

in the conventional agriculturea�0.466b �0.267d

Fit statistics Chi square: 88.38 (p-value 0.000),R-square 0.271

Chi square: 63.26 (p-value 0.000),R-square 0.210

Statistical significance based on bootstrapped standard errors because the dependent variables were not normally distributed, n repetitions ¼ 1000.a Dummy variable: 1 if yes; 0 if no.b Statistically significant at the 1% level or less.c Statistically significant at the 5% level or less.d Statistically significant at the 10% level or less.

H. Hansson et al. / Journal of Rural Studies 32 (2013) 240e250 247

resources’. Because of high correlations between the respectivevariables describing the spouse’s involvement in the planning andmanagement process, backward manual elimination was used toreduce themodels to only include significant variables (not shown).The results obtained from this procedure showed that also thevariable ‘My spouse did not take part at all in the planning of thenew venture’ had a statistically significant negative impact on‘business development for social and lifestyle reasons’ (p-value0.052). Hence the results show that family influences on motivesunderlying decisions to start new ventures outside conventionalagriculture may work in opposite directions. Among the controlvariables, the variable measuring share of revenue in 2009 esti-mated to originate from diversified activities had a significantlypositive impact on both motives whereas the variable consideringwhether all household members who are able to work can support

Table 5Regression results for spouse’s involvement in management, Model 2. Seemingly unrela

Dedevris

Est

Intercept �0My spouse is the main manager of the new venturea �0My spouse and I take equal parts in the management of the new venturea �0My spouse takes less part than me in the management of the new venturea �0My spouse does not take part at all in the management of the new venturea �0Family considerations 0Distance to nearest town with at least 20 000 inhabitants �0Number of members in household 0Share of revenue in 2009 originating from diversified activities (%) 0Resources that cannot be fully used in conventional agriculturea 0All household members who can work can support themselves

in the conventional agriculturea�0

Fit statistics ChR-s

Statistical significance based on bootstrapped standard errors because the dependent vaa Dummy variable: 1 if yes; 0 if no.b Statistically significant at the 1% level or less.c Statistically significant at the 5% level or less.

themselves in the conventional agriculture had a significantlynegative impact on motives to diversify outside conventionalagriculture.

In terms of model performance, 20.2e27.5% of total variation inthe dependent variables was explained by the regression models.While these percentages show that there are other aspects thatexplain farmers’ motives to develop ventures outside conventionalagriculture, they also show that a notable share of the variance isexplained by the variables considered in this study.

5. Discussion and conclusions

This study examined the motives underlying decisions byfarmers in Sweden to start new ventures outside conventionalagriculture, in particular how these motives are dependent on the

ted regression model.

pendent variable; motive: ‘Businesselopment for reasons to reduce

k and to use idle resources’

Dependent variable; motive: ‘Businessdevelopment for social and lifestylereasons’

imated coefficient Estimated coefficient

.969b �1.186b

.889b �0.267

.057 0.004

.163 �0.077

.189 �0.246

.070b 0.090b

.001 �0.002

.008 0.006

.013b 0.007b

.002 �0.080

.497b �0.293c

i square: 89.85 (p-value 0.000),quare 0.275

Chi square: 60.12 (p-value 0.000),R-square 0.202

riables were not normally distributed, n repetitions ¼ 1000.

H. Hansson et al. / Journal of Rural Studies 32 (2013) 240e250248

farm family situation. Building on the definition of entrepreneur-ship developed by Shane and Venkataraman (2000) and Shaneet al. (2003), the start-up of such new ventures was consideredan entrepreneurial activity.

Since the results are based on a sub-set of a sample of farmersdrawn from a target population representing about the 30 000largest farms (those considered commercial farm businesses) inSweden, the study results can be generalised to the group offarmers that run farms to earn at least part of their living. Thisgroup can be contrasted to hobby-farmers with small-holdings,which are not large enough to allow the owner to earn his/herliving out of the farm activities. The target population used in thestudy arguably represents farms in Sweden that have the economicand financial means to drive development in Swedish rural areasbased on diversification outside conventional agriculture. Inclusionof small-holdings in the analysis would probably bias the results byreflecting the decision making processes of people who live onfarms for personal and/or hobby reasons rather than to run abusiness to earn at least part of their living. As there are a signifi-cant number of hobby-holdings in rural Sweden, future studies toidentify possible differences in motives for developing diversifiedactivities outside conventional agriculture between commercialfarmers and owners of small-holdings, as well as possible differ-ences in how these motives are dependent on the family situation,might contribute to a more general rural development discussion.The findings in this study, however, can be useful for policy makerswishing to exploit the motives that trigger commercial farmers tostart new ventures outside conventional agriculture. In particular,findings show how these motives are dependent on the situation ofthe farm family.

The common factor analysis revealed two underlying motives,which were interpreted as i) business development to reduce riskand to use idle resources and ii) business development for socialand lifestyle reasons. These findings differ from those reported inrelated quantitative studies (e.g. Polovitz Nickerson et al., 2001;McGehee and Kim, 2004; Barbieri and Mahoney, 2009), bothwith respect to number of underlying motives and the nature ofthese motives. While some of the discrepancy may be due tocultural differences in the empirical study areas, there is also ev-idence that differences in the domains of underlying motivesdepend on the method used to extract these motives. We used areflective measurement model, which assumes causality goingfrom the construct to the indicator.2 In particular, our findingssuggest that it may not be feasible to consider motives for diver-sification outside agriculture at the disaggregated level suggestedby previous literature (especially Barbieri and Mahoney (2009),who identified six underlying motives). Instead, diversificationseemed to be driven by more overarching motives related to themanagement and development of the business and the situation ofthe farmer and his/her family. Thus, diversification outside con-ventional agriculture was viewed by the respondents as: i) Abusiness development strategy to reduce risk and to use idle re-sources; and ii) a business development strategy for social andlifestyle reasons.

It should be noted that the questionnaire used to collect mea-surement indicators on farmers motives were mailed to the personregistered as the owner of the farm. While the owner and the

2 For comparison, a principal component factor analysis was run on the dataset(results not shown). This analysis suggested three underlying components in thedata. The first two resembled the first two factors obtained in the common factoranalysis, the third comprised a factor not part of the solution to the common factoranalysis. This shows that different assumptions about the underlying theoreticalmodel, and hence extraction method, have a notable influence on results andconclusions.

spouse are likely to realise similar motives for a decision to developdiversified activities outside conventional agriculture, theremay beinstanceswhere this is not the case and this may bias the analysis ofhow motives are dependent on the family situation. In futurestudies, it would therefore be interesting to evaluate the consis-tency in underlying motives and values between different decisionmakers in the same farm business.

The entrepreneurship literature suggests that entrepreneurs arepushed into entrepreneurial activities, based on dissatisfactionwith a previous situation, or pulled by job and life satisfaction (e.g.Schjoedt and Shaver, 2007). In this context it is interesting to notethat the two factors obtained in our factor solution both comprisedpush and pull elements and hence there is no evidence of under-lying motives to represent either push or pull motivational ele-ments. While our data are retrospective in that we asked whatmotivated the respondents to diversify outside conventional agri-culture only after at least some of them had implemented diver-sification, and may therefore be flawed by people’s falserecollections, our results suggest that underlying motives consist ofa mix of push and pull elements and that it appears to be combi-nations of push and pull factors that motivate entrepreneurial ac-tions. However, more quantitative empirical research should bedevoted to investigating whether more general motives for entre-preneurial activities always consist of mixes of push and pull ele-ments or if it is possible to distinguish general level motivesunderlying entrepreneurial behaviour that have a clear push or pullcharacter.

The novel contribution of this study lies in its determination ofthe context-dependency of the motives underlying farmers’ de-cisions to diversify outside conventional agriculture, more specif-ically as affected by the situation of the farm owner’s family. Thisissue is largely ignored in previous studies on farmers’ motives fordiversifying (Polovitz Nickerson et al., 2001; McGehee and Kim,2004; Barbieri and Mahoney, 2009; Northcote and Alonso, 2011).We drew on insights from family business management (Sharmaet al., 1997; Chua et al., 1999; Chrisman et al., 2003; Chua et al.,2003; Westhead and Howorth, 2006) and found that family influ-ence measured both at the level of more general family consider-ations and spouse’s involvement in the creation in the new ventureinfluenced the motives underlying decisions to diversify. The in-fluence of the spouse in the creation of the new venture wasespecially pronounced when the spouse was the main planner andmanager of the new venture, as could be expected.

Interestingly, the influence of family measured in the two waysconsidered in this study affected the motives for diversification inopposite directions, with higher scores on the two motivationaldimensions associated with higher family considerations of moregeneral type and lower scores associated with involvement ofspouse in the creation of the new venture. These findings suggest itis important to distinguish between different types of family in-fluence when analysing the context-dependency of motives todiversify and not to use them interchangeably. It should also benoted that when the farmer’s spouse was the main planner and/ormanager of the new venture, business development to reduce riskand to use idle resources was a significantly less important motivefor diversification outside agriculture. This suggests that otherunderlying motives for decisions to diversify were not captured inthis study. For instance Anthopoulou (2010) found in his case studyin Greece that rural women’s businesses are often such that theyprovide them with expected income within the context of thefamily’s strategic plan. These kinds of motives may relate to thespouse’s employment situation, a desire to raise children in a homeenvironment etc. and should be further explored in future research.

It is also possible that the desire to provide a family householdwith a larger and more stable income, a dimensions of family

H. Hansson et al. / Journal of Rural Studies 32 (2013) 240e250 249

consideration not explicitly studied here, may be the origin of boththe identified underlying motives and thus, that family consider-ations in farmers’ decisions to run diversified businesses may beeven stronger than findings indicate.

The findings hold clear interest for rural development policy,which would benefit from more in-depth insights into factors un-derlying farmers’ decision-making (e.g. Pietola and Lansink, 2001)Traditionally, in both Sweden and Europe, farms have strong con-nections with the owning family, and with farm diversification as aprioritised area in Swedish and European agricultural and ruralpolicy, these findings offer policy makers valuable insight in un-derstanding how underlying motives for farm diversification aredependent on the situation of the farm family. The results clearlyshow that farmers are driven by a complex set of motives todiversify their farms outside conventional agriculture, and thatthese are dependent on the family situation in which the farmexists.

From a policy point of view, it would be interesting to furtherexplore what would happen to farm businesses if the influence offamily considerations at a more general level were to be reduced orremoved. Our findings suggest farmers’ motives for diversifyingwould be affected, but more research is needed to understand whatstrategies theywould then pursue. There is alsomore to learn aboutthe decision making in farm businesses that have not diversified,whether they are driven by similar motives and whether the familysituation exercises a similar influence. Such issues are particularlyrelevant in light of the current development in Swedish agriculture,where farm businesses are becoming increasingly larger, withincreasing needs for financial resources that are triggering neworganisational and legal forms. These changes are apt to lead to adecrease in the traditionally strong connection between family andfamily considerations and the farm business, creating a differentpolicy context.

Acknowledgements

This study was financed by the Swedish Research Council For-mas, which is gratefully acknowledged.

References

Alsos, G.A., Ljunggren, E., Pettersen, L.T., 2003. Farm-based entrepreneurs: whattriggers the start-up of new business activities? Journal of Small Business andEnterprise Development 10, 435e443.

Ansoff, I., 1957. Strategies for diversification. Harvard Business Review 35, 113e124.Anthopoulou, T., 2010. Rural women in local agrofood production: between

entrepreneurial initiates and family strategies. A case study in Greece. Journal ofRural Studies 26, 394e403.

Barbieri, C., Mahoney, E., 2009. Why is diversification an attractive farm adjustmentstrategy? Insights from Texas farmers and ranchers. Journal of Rural Studies 25,58e66.

Barrett, C.B., Reardon, P., Webb, P., 2001. Nonfarm income diversification andhousehold livelihood strategies in rural Africa; concepts, dynamics and policyimplications. Food Policy 26, 315e331.

Baum, J.R., Locke, E.A., Smith, K.G., 2001. A multidimensional model of venturegrowth. The Academy of Management Journal 44, 292e303.

Bowler, I., 1999. Modelling farm diversification in regions using expert and decisionsupport systems. Journal of Rural Studies 15, 297e305.

Brandth, B., Haugen, M., 2011. Farm diversification into tourism e implications forsocial identity? Journal of Rural Studies 27, 35e44.

Brockhaus, R., 1980. The effect of job dissatisfaction on the decision to start abusiness. Journal of Small Business Management 18, 37e43.

Carter, S., Rosa, P., 1998. Indigenous rural firms: farm enterprises in the UK. Inter-national Small Business Journal 16, 15e27.

Chaplin, H., Davidova, S., Gorton, M., 2004. Agricultural adjustment and the diver-sification of farm households and corporate farms in Central Europe. Journal ofRural Studies 20, 61e77.

Chrisman, J.J., Chua, J.H., Sharma, P., 2003. Current Trends and Future Directions inFamily Business Management Studies: Toward a Theory of the Family Firm. In:Coleman White Paper Series.

Chua, J.H., Chrisman, J.J., Sharma, P., 1999. Defining the family business by behavior.Entrepreneurship Theory and Practice 23, 19e39.

Chua, J.H., Chrisman, J.J., Sharma, P., 2003. Succession and nonsuccession concernsof family firms and agency relationships with nonfamily managers. FamilyBusiness Review 16, 89e107.

Cuykendal, C., LaDue, E., Smith, D.R., 2002. What Successful Small Farmers Say. theResults of a Survey of Successful Small Farm Operators. R.B. 2002-01. Agricul-tural Finance and Management at Cornell, Ithaca, New York.

Delmar, F., 1996. Entrepreneurial Behavior & Business Performance (Dissertation).Stockholm School of Economics.

Duffy, M., Nanhou, V., 2002. Factors of Success of Small Farms and the Rela-tionship Between Financial Success and Perceived Success. Staff GeneralResearch Papers with Number 10349. Iowa State University, Dept. of Eco-nomics, Ames USA.

Gasson, R., 1973. Goals and values of farmers. Journal of Agricultural Economics 24,21e42.

Gasson, R., Errington, A., 1993. The Farm Family Business, first ed. CAP International,Oxon, UK.

Gorton, M., Douarin, E., Davidova, S., Latruffe, L., 2008. Attitudes to agriculturalpolicy and farming futures in the context of the 2003 CAP reform: a comparisonof farmers in selected established and new member states. Journal of RuralStudies 24, 322e336.

Grande, J., 2011. New venture creation in the farm sector e critical resources andcapabilities. Journal of Rural Studies 27, 220e233.

Hair, J.F., Black, W.C., Babin, B.J., Anderson, R.E., 2010. Multivariate Data Analysis e aGlobal Perspective, seventh ed. Pearson, New Jersey, USA.

Hansson, H., Ferguson, R., Olofsson, C., 2010. Understanding the diversification andspecialization of farm businesses. Agricultural and Food Science 19, 269e283.

Ilbery, B.W., 1991. Farm diversification as an adjustment strategy on the urbanfringe of the West Midlands. Journal of Rural Studies 7, 207e218.

Jarvis, C.B., Mackenzie, S.B., Podsakoff, P.M., 2003. A critical review of constructindicators and measurement model misspecification in marketing and con-sumer research. Journal of Consumer Research 30, 199e218.

Johnson, G., Whittington, R., Scholes, K., 2011. Exploring Strategy Text and Cases.Pearson Education, Essex, England.

Kirkwood, J., 2009. Motivational factors in a push-pull theory of entrepreneurship.Gender in Management: An International Journal 24, 346e364.

Komppula, R., 2004. Success Factors in Small and Micro Business. Keskustelua-loitteita/Joensuun yliopisto, Taloustieteiden laitos n:o 17. University of Joensuu,Joensuu, Finland.

Lagerkvist, C.J., Larsén, K., Olson, K.D., 2007. Off-farm income and farm capitalaccumulation: a farm-level analysis. Agricultural Finance Review 67, 243e257.

Maye, D., Ilbery, B., Watts, D., 2009. Farm diversification, tenancy and CAP reform:results from a survey of tenant farms in England. Journal of Rural Studies 25,333e342.

McElwee, G., 2008. A taxonomy of entrepreneurial farmers. International Journal ofEntrepreneurship and Small Business 6, 465e478.

McGehee, N.G., Kim, K., 2004. Motivation for agri-tourism entrepreneurship. Jour-nal of Travel Research 43, 161e170.

McGehee, N.G., Kim, K., Jennings, G.R., 2007. Gender and motivation for agri-tourism entrepreneurship. Tourism Management 28, 280e289.

McNally, S., 2001. Farm diversification in England and Wales e what can we learnfrom the farm business survey? Journal of Rural Studies 17, 247e257.

McNamara, K.T., Weiss, C., 2005. Farm household income and on- and off-farmdiversification. Journal of Agricultural and Applied Economics 27, 37e48.

Northcote, J., Alonso, A.D., 2011. Factors underlying farm diversification: the caseof Western Australia’s olive farmers. Agriculture and Human Values 28,237e246.

Peltola, A., 2000. Viljelijäperheiden monitoimisuus suomalaisilla maatiloilla (Plu-riactivity of farm families in Finland). MTTL:n julkaisuja 96/2000, p. 280 (inFinnish).

Pietola, K.S., Lansink, A.O., 2001. Farmer response to policies promoting organicfarming technologies in Finland. European Review of Agricultural Economics28, 1e15.

Podsakoff, P.M., MacKenzie, S.B., Podsakoff, N.P., Yeon Lee, J., 2003. The mismeasureof man(agement) and its implications for leadership research. The LeadershipQuarterly 14, 615e656.

Polovitz Nickerson, N., Black, R.J., McCool, S., 2001. Agritourism: motivations behindfarm/ranch business diversification. Journal of Travel Research 40, 19e26.

Potter, C., Lobley, M., 1992. Ageing and succession on family farms: the impact ondecision-making and land use. Sociologia Ruralis 32, 317e334.

Reijonen, H., 2009. Role and Practices of Marketing in SMEs (Doctoral dissertation).University of Joensuu, p. 79.

Schjoedt, L., Shaver, K.G., 2007. Deciding on an entrepreneurial career: a test of thepull and push hypotheses using the panel study of entrepreneurial dynamicsdata. Entrepreneurship Theory and Practice 31, 733e752.

Segal, G., Borgia, D., Schoenfeld, J., 2005. The motivation to become an entre-preneur. International Journal of Entrepreneurial Behaviour & Research 11,42e57.

Serra, T., Goodwin, B.K., Featherstone, A.M., 2004. Determinants of investments innon-farm assets by farm households. Agricultural Finance Review 64, 17e32.

Shane, S.S., Venkataraman, S., 2000. The promise of entrepreneurship as a field ofresearch. Academy of Management Review 25, 217e226.

Shane, S.S., Locke, E.A., Collins, C.J., 2003. Entrepreneurial motivation. HumanResource Management Review 13, 257e279.

Sharma, P., Chrisman, J.J., Chua, J.H., 1997. Strategic management of the familybusiness: past research and future challenges. Family Business Review 10, 1e36.

H. Hansson et al. / Journal of Rural Studies 32 (2013) 240e250250

Swedish Board of Agriculture, 2007. Jordbruksföretagets kombinationsverksamh-eter och jordbrukarhushållets inkomster (Other gainful activities on the agri-cultural holding and income of the agricultural household). Statistikrapport2007:3. Swedish Board of Agriculture.

Turner, M., Whitehead, D., Barr, D., Fogerty, M., Errington, A., Lobley, M., Reed, M.,2003. Farm Diversification Activities: Benchmarking Study 2002-Final Report toDEFRA. University of Exeter and University of Plymouth.

United Nations, 2001. Global Entrepreneurship Monitor. 2001 Executive Report.United Nations’ Association of the United States of America and the BusinessCouncil for the United Nations.

Vik, J., McElwee, G., 2011. Diversification and the entrepreneurial motivations offarmers in Norway. Journal of Small Business Management 49, 390e410.

Westhead, P., Howorth, C., 2006. Ownership and management issues associatedwith family firm performance and company objectives. Family Business Review14, 301e316.

Willock, J., Deary, I.J., Edwards-Jones, G., Gibson, G.J., McGregor, M.J.,Sutherland, A., Dent, J.B., Morgan, O., Grieve, R., 1999. The role of attitudesand objectives in farmer decision making: business and environmentally-oriented behaviour in Scotland. Journal of Agricultural Economics 50,286e303.