‘My Grandfather Would Roll Over in His Grave’: Family Farming and Tree Plantations on Farmland*

By Neumann, Pamela D; Krahn, Harvey J; Krogman, Naomi T; Thomas, Barb R

ABSTRACT

In this paper we hypothesize that farmers with a stronger valuation of family farming will be more resistant to converting farmland to tree plantations. Our survey data analysis from 106 farmers in northern Alberta reveals that general opposition to trees on farmland is the strongest predictor of farmers’ resistance to the establishment of poplar tree plantations on privately-owned land. Valuation of family farming is the strongest determinant of resistance to trees on farmland. Among the potential intervening variables influencing support for tree plantations, including county, age, gender, number of children, and percent of income from farming, number of children and percent of income from farming had significant direct effects on valuation of family farming. This study suggests that economic incentives alone are unlikely to influence farmers’ willingness to convert their land to non- traditional uses, and that intergenerational transfer of land, and its relationship to valuation of family farming, deserves further attention in rural sociological scholarship.

Despite vast changes over the past century in how agricultural goods are produced and marketed in North America, the family farm continues to be seen as the ideal production unit by many farmers, agricultural policy makers, and even urbanites (Pfeffer 1989). Rural sociologists have documented the influence of family farm values on farmers’ willingness to adopt new farming techniques (Abd-Ella, Hoiberg, and Warren 1981; Anosike and Coughenour 1990; Sontag and Bulboz 1985). Among these studies, sociologists have not examined the role of family farm values in attitudes toward a major change facing family farms in North America today: farmland conversion to non-traditional uses.

One proposed farmland conversion of increasing interest globally is the planting of trees on previously cleared farmland for fiber and timber production (Boyle 1999; Sedjo 1999). In Canada and the United States the intensive management of tree plantations is currently a widespread practice (Adams et al. 2005; Zasada et al. 2004; Sedjo 2001), but not without controversy. In this paper we examine the influence of valuation of family farming on support for conversion of farmland, and hypothesize that farmers with a stronger valuation of family farming will be more resistant to converting farmland to a nontraditional use. Using the establishment of tree plantations as a general example, and Alberta-Pacific Forest Industries Inc.’s Poplar Farm Program as a specific change instrument, we model responses to land use change as a function of valuation of the family farm. Relationships between attitudes toward family farming and support for hybrid poplar plantations are described, and the implications of the receptivity (or lack thereof) of landowners to conversion of farmland to trees are discussed.

Understanding Family Farming as a Set of Values

Less than one hundred years ago, most farms in the United States and Canada were essentially subsistence operations, with the majority of the produce consumed by the immediate and extended family, and a small minority of products used for trading for other goods. In this social and economic context, all farms were “family farms,” small in scale, with all the management, capital, and labor contributed by the family. Throughout the twentieth century in Canada and the United States, family farming, while largely defining rural society, has declined as an economically viable way of life (Smithers and Johnson 2004).

Currently both family farms and rural communities in Canada are changing in response to a myriad of regional, national, and global forces. The need to compete in an expanding worldwide market with other agricultural producers and shrinking profit margins has forced many family-run farms out of business (Lind 1995). Each year from 1941 to 1991 over 8,000 Canadian farms, most of which were family- owned and -operated, went out of existence (Dasgupta 2001:112). Since then, another 33,000 farms have disappeared (Statistics Canada 200Ic). This abandonment of farming as a “household livelihood strategy,” termed the “agricultural transition” by Lobao and Meyer (2001:104), has typically involved large farms buying-up or leasing much of the land previously farmed by small and medium-sized operations.

It is clear that the family farm as a social and economic entity in North American rural communities is a declining institution, having less of an impact on rural economies as agriculture becomes increasingly commercialized and specialized in response to worldwide competition. However, the family farm as a cultural concept appears to be alive and well in the minds of rural peoples. Sociologists have referred to family farming as an important cultural symbol (Sinnema 2005; Taylor 1954), encompassing an influential set of values (Fink 1986; Pfeffer 1989; Rohwert 1951), where valuation of family is central (Fink 1986). Fink (1986) argues that the values associated with “good farming” are linked to family farming, in stark contrast to all negative aspects of corporate, in particularly impersonal and highly commercial, big business farming. The value of private ownership of land also appears central to the value set of family farming (Strange 1988), where private ownership is strongly associated with stewardship. Not only does a personal investment in one’s family-owned land deter negligence, but it follows that the person who is the most dependent on the land for survival and the most conscious of its eccentricities will be its most conscientious manager (Comstock 1989:19).

An important component of stewardship is that owners care for land in perpetuity. Intergenerational transfer of the farm goes beyond the passing down of capital assets such as land, buildings, and equipment. The next generation also inherits an occupation and a way of life, closely tied to familial relations and a legacy of a family coherence (Keating 1996; Keatingand Munro 1989). Wester- Herber (2004) argues that this continuity of connection to particular parcels of family-owned land is often central to farmer identity. When farm families worked side-by-side on the traditional family farm and the younger generation was taught the lay of the land, opportunities were created for shared dreams, satisfying communication, and intimacy (Zimmerman and Fetsch 1994:125). For people who were raised by farmers and taught practices and values that had been passed down through the generations, the values connected to family farming are likely to be extremely important to their sense of self, and to their opinions of more concrete farming options and policies.

Though the importance of family values in farming decisions are readily acknowledged by sociologists, there is still disagreement over how these values translate to decisions on the farm. Bennett and Kohl’s (1963) study of Canadian ranchers found that the most agriculturally innovative individuals were those who established their operation without any parental assistance. They suggest that when a farmer is connected with the previous generation (usually via the father), his or her own way of doing things is anchored in established, traditional methods, and thereby limited in the potential for innovation. Intragenerational connections (e.g. between siblings) may also serve to maintain traditional values because of the reinforcement of familial ideals among peers (Carlson and Dillman 1983). This preservation of values maythen lead to reduced innovation, as shown in Bennett and Kohl’s (1963) study.

On the other hand, a few studies have found increased valuation of family farming to have a positive influence on agricultural adoption. Abd-Ella et al.’s (1981) study of Iowa farmers found that long-term family goals acted as an incentive for adoption of innovative farm techniques, where farmers who were more family goal- oriented were more likely to try new techniques. Carlson and Dillman’s (1983) U.S. study also found that individuals who were actively farming with relatives were more likely to pursue innovative soil conservation practices, the planting of alternative varieties of traditional crops, the use of new chemicals, or the purchase of machinery, and suggested that farmers connected to family are more likely to make capital and physical farming decisions that reflect long-term family goals.

Conversion to Non-Traditional Uses: Trees on Farmland

Trees on farmland can take many forms. Some farms in North America contain natural stands of trees, or woodlots, which can be harvested for profit or set aside for environmental, aesthetic, and recreational purposes. North American farmers have also long participated in state-sponsored tree planting initiatives for windbreaks around fields and homesteads (Droze 1977). Agroforestry techniques have been implemented by some farmers in the form of alley cropping (rows of harvestable trees planted between rows of traditional grain crops), silvopasture (trees planted on pasture land for shade and shelter for livestock), and riparian buffers (trees planted along waterways for environmental benefits) (Raedeke et al. 2003:64).

All of th\ese tree-farm interactions involve trees, natural or planted, growing alongside the production of traditional agricultural products, in contrast to plantation forestry, which involves the complete replacement of traditional crops. The establishment of tree plantations on previously cleared farmland typically involves the planting and managing of relatively small areas of land (e.g., 30 to 50 hectares) with fast growing trees that are selected for their growth characteristics and subjected to both mechanical and chemical (e.g., fertilizers, herbicides, pesticides) growth enhancers (Boyle 1999). Plantation forestry, in its various forms, is undertaken for habitat restoration, aesthetic purposes, hunting potential, and income generation and can be realized with tree plantings of various types. Trees can be harvested for Christmas trees, lumber, pulp, or wood chips and may potentially become a source of widespread biomass fuel for home use and for commercial sale (Berndes, Hoogwijk, van den Broek 2003). Finally, the Canadian government is currently exploring the benefit of reforestation on previously cleared land to sequester carbon in an effort to offset the effects of climate change (Smith et al. 2005).

Despite the fact that many governments currently laud the benefits of converting farmland to trees, and even go so far as to implement it into policy (Haverstock 2005), the social perceptions of trees as alternative crops are poorly understood. Economic studies describe the financial benefits of converting ‘marginal’ farmland to trees in one form or another and suggest financial concerns as the key constraints to adoption of these practices (e.g. van Kooten, Shajkh, and Suchanek 2002). Not considered are the social concerns associated with the widespread planting of trees on previously cleared farmland. As much of the farmland in North America was historically forested, an immense amount of effort was required to make the land suitable for traditional agriculture. Generations of farmers meticulously cleared the land, and continue to keep it clear, in order to pursue a traditional farming lifestyle. In this historical context, planting of trees may be seen as “a reversal, of letting the country go back to how it was before White settlement came” (Barlow and Cocklin 2003:516).

In addition to recognizing the history of trees on farmland, it is important to consider the fature effects of planting trees on farmland. By planting trees back onto cleared land, farmland is taken out of traditional crop production. In Wester-Herber’s (2004:114) review of the sociology of land use conflicts, she argues that “if farmers in the area have been dependent on the land for farming and cattle grazing, and this land has been changed physically…it can no longer provide the means for an everyday existence.”

Some studies also suggest farmers have a contentious relationship with foresters when forestry operations are adjacent to agriculture (Gourlay 1986; Raedeke et al. 2003). In Raedeke et al.’s (2003) study of farm-forestry interactions, many farmers felt that forestry workers show little regard for the land, and that they tend to exploit farmers when woodlots are managed in partnership. Farmers who have managed their land for generations often view traditional agriculture as the “correct use of land” (Le Heron and Roche 1985:216).

Alberta-Pacific Forest Industries Inc. and the Poplar Farm Program

In 1999, Alberta-Pacific Forest Industries Inc. (Al-Pac), a forestry company operating in the boreal forest of north-eastern Alberta, began to implement plantation forestry practices within a 200-kilometer radius of its mill (see Figure 1.). The mill was completed in August, 1993 and is the largest single-line kraft pulp mill in North America (AlbertaPacific Forest Industries 2001). The mill processes aspen from the company’s 58,000 square kilometer FMA (i.e., forest management agreement, a 20 year renewable lease) which is located on Crown (public) land in northern Alberta and harvested through an agreement with the provincial government. The mill is currently producing an average of 1,800 air-dried metric tons of pulp per day.

Through the Poplar Farm Program the company is establishing plantations consisting of hybrid poplar trees on local privately- owned land. The Alberta government has mandated that no foreign- owned company can lease private land for more than 20 years and that foreign leases are non-renewable. However, an exception was made for Al-Pac, and the company has been given approval to lease land for a maximum period of 30 years with the option of renewal. The land Al- Pac leases is restricted by soil class (the company cannot lease the most valuable agricultural areas for greater than 32 hectares contiguous) as well as size (the company requires at least 20 hectares to establish a plantation). The company hopes to establish a total of 25,000 hectares of hybrid poplar plantations by 2020 (Alberta-Pacific Forest Industries 2002). Landowners have the option of being paid to manage the plantations on their land, which would involve weeding, discing, and/or some chemical applications (Al-Pac 2002). Al-Pac plans to harvest the plantation trees in approximately 20 years, a strategy which will potentially provide 15 to 17 percent of its mill’s fiber needs.

Agriculture (cereal crops and cattle) is one of the core industries in the region within which Al-Pac operates. The Poplar Farm Program represents the first time that hybrid poplars have been considered for plantations, as compared to windbreaks, in this region of the province. Neumann, Krogman, and Thomas (2007) conducted exploratory interviews about the Poplar Farm Program with key informants in the region, but there has been no systematic quantitative investigation of local residents’ responses and attitudes toward the Poplar Farm Program or any tree plantation program in North America.

Valuation of Family Fanning and Trees on Farmland: A Model of Support for a Tree Plantation Program

Figure 2 highlights a theoretical model of support for a tree plantation program, specifically Al-Pac’s Poplar Farm Program. Following the literature on family farming values and agricultural innovation adoption, and with the understanding that symbols and values can be powerful determinants of behavior, we hypothesize that high valuation of family farming will increase resistance to trees on farmland and will have a negative effect on support for the program. This hypothesis mirrors Bennett and Kohl’s (1963) findings that farmers with stronger connections to past generations were less likely to support new innovations. In addition, planting trees on farmland may be “perceived as erasing an important symbol of previous generations and of a person’s family heritage” (Raedeke 2003:73) because it is a move away from traditional land uses. In many Alberta farming communities several generations of farmers labored for years felling trees, chopping stumps, and picking roots in order to clear land for farming. The idea of planting trees on land meticulously cleared and kept clear by parents, grandparents, and great-grandparents could be seen as an affront to the sacrifices those generations made to provide for their families. In addition, the land planted with trees may be perceived as land that is no longer farmable, in the traditional sense, by future generations. Hence, we expect that farmers with strong valuation of family farming would be less likely to support a tree plantation program (Neumann et al. 2007).

In addition to the prior effects of family farm values, we also hypothesize that resistance to trees on farmland will have a direct negative effect on support for the Poplar Farm Program because of the stigma farmers associate with forestry (Raedeke et al. 2003) and the strong connections farmers have to cleared land. Furthermore, since planting trees on farmland is the defining characteristic of Al-Pac’s Poplar Farm Program, we expect that resistance to planting trees on farmland will be a more influential determinant of support for the tree plantation program than our control variables, namely, location (county), age, gender, family, or financial considerations.

County is included as a control variable because the specific characteristics of a farming community or region may influence farmers’ responses. Westlock County, the more southern of the two counties, is located just north of Edmonton, the capital of the province of Alberta, and is less isolated than the more northern county of Athabasca. Land in Westlock is more productive for traditional crops and has fewer trees, with the total farm market value in the area estimated at $866 million (Canadian dollars, or approximately $740 million U.S.) spread out over 1,000 farm operators in 2001 (Statistics Canada 200Ib) as compared to an estimated total farm value of only $544 million (Canadian dollars, or approximately $465 million US) in Athabasca with approximately the same number of operators (Statistics Canada 200Ia). In addition, Athabasca County is situated next to an expanse of public land managed for forestry, and the Al-Pac pulp mill operates within its limits. Athabasca residents have more experience with Al-Pac, and the establishment of the mill itself was controversial (Sherman 1997). Though the high value of land in Westlock County may lead to decreased support for the program, the controversy regarding the original establishment of the mill in the County of Athabasca (Gismondi 1997) may also have a detrimental effect on support for the program. Given these elements, we do not hypothesize the direction of a county effect on support for the Poplar Farm Program.

We include age as a control variable in order to capture any generational differences. Older farmers might have stronger ties to past generations, and may also be at the point of considering retirement and passing on the farm. Both of \these factors may lead to stronger feelings about the family farm. In addition, older people have been found to be more traditional in their opinions of farming, seeing a “moral responsibility to sustain land resources” (Walter 1997:48). Elder farmers’ value of stewardship and continuity of past practices contrasts with the more managerial values embraced by younger generations (Walter 1997). Hence, we expect that older farmers would be more resistant to changing from traditional uses of land and less likely to support the Poplar Farm Program, especially if they have been personally involved in clearing the family land of trees in order to make it farmable.

Gender is the third control variable in our model. Traditionally, farm women’s values were viewed as complementary and subordinate to those of their husbands (Moore 1989), so few studies focused on gender differences. However, several recent studies examining the role of gender in adoption of agricultural innovations, specifically those associated with sustainable farming practices, found women to have less involvement in the selection of new practices (Meares 1997; Rogers and Vandeman 1993). In addition, Moore’s (1989) study of farm families in Wisconsin showed that men expressed more interest in farm management and change than women. Moore (1989) also found that women tended to view farming as a component of family life, whereas men were more likely to view farming as a business enterprise. These findings suggest that men would be more likely to view the Poplar Farm Program as an opportunity to supplement the farm business, and women would likely have a stronger valuation of family farming.

We also expect that the number of children a farmer has will have a positive effect on the strength of family farming values. Presumably a farmer with more children will have a stronger sense of family than a farmer with fewer children, and more inclination to view fanning as a family matter. In addition, the concept of intergenerational transfer of the business and way of life would be more salient for a farmer considering his or her children’s future.

Our final control variable, percent of income from farming, is a relative measure of how much of the family’s money comes directly from fanning. This distinguishes full-time farmers from part-time or hobby farmers. We hypothesize that farmers who rely on farming for a greater proportion of their income will be more resistant to planting trees on farmland and the Poplar Farm Program because they may see these practices as a threat to their business. Alternatively, farmers who rely more on off-farm work income may see trees as an investment on their land that requires minimal effort and allows them to continue pursuing other means of gaining income.

Data and Methods

We test the model described above with survey data collected from a sample of private landowners in two counties of the province of Alberta in February of 2005. County maps were used to prepare a list of private landowners within fifteen kilometers of the towns of Westlock and Athabasca. An exhaustive sample was then drawn from this list. Landowners were first phoned to request their participation. In Westlock County owners of 478 eligible plots of land were called in order to achieve a list of 110 willing participants. In the County of Athabasca owners of 411 eligible plots of land were called to obtain a list of 110 willing participants. The first-phase response rates were, consequently, 23 percent and 26.8 percent for Westlock and Athabasca, respectively.

The questionnaires were then dropped off at the landowners’ homes and picked up within three days by a team of senior undergraduate students. A total of 191 questionnaires were completed, 89 in Westlock and 102 in Athabasca, resulting in second-phase response rates of 81 percent and 93 percent respectively (or 18.6% and 24.8% of the original sampling frame). These second-phase response rates are high, likely due to the intensive ‘drop-off/pick-up’ survey administration method used.

A comparison of sample age and gender distributions with the Farm Operators 2001 Census results for both communities (Statistics Canada 2001a, 200Ib) demonstrated that final sample members did not differ from the population from which they were drawn in terms of age and gender. Consequently, though our first-phase response rates were low, we are confident that our sample was representative of the farm operator population in the two counties at the time of the survey. However, as is typically the case with survey data, not all respondents answered all the survey questions.1 Missing data was most problematic for our income question which was answered by only 72 percent of respondents (see Appendix 2). Consequently, in the following test of our theoretical model, our analytic sub-sample consists of 106 respondents (56% of the total sample of 191) who provided complete information on all the variables used in our analysis.2

Focusing on the analytic sub-sample (n = 106), 31 percent of respondents were female, and the average age of the sample was 53 years. There was no significant difference between counties for these two variables. Respondents owned an average of 609 acres and, again, there was no difference between counties. Of the participants included in this analysis, 58 percent identified themselves as cattle farmers (no difference between counties). However, a significantly higher proportion of Westlock farmers identified themselves as grain farmers (84% compared to 64% of Athabasca farmers). There was also a significant difference between counties for reported percentage of income from farming (an average of 39% in Athabasca compared to 60% in Westlock).

Table 1 displays descriptive statistics for all the variables in our model (for the analytic sample of 106 respondents, as well as for the components of the intervening variable indices used in our analysis). The mean response (3.20) for support for the Poplar Farm Program (“How do you feel about the Poplar Farm Program?”) was slightly above average on the fivepoint (very negative-very positive) response scale, suggesting overall neutral to positive inclinations toward the program in both counties.

A qualitative thematic analysis (using Qualitative Research Software NVivo) of all participants’ responses to five individual open-ended questions in the questionnaire was initially employed to develop the “resistance to trees on farmland” and the “valuation of family farming” indices. Rponses to these questions ranged from singleword answers to numerous paragraphs. The majority of all participants (92%, 174 participants) responded to at least one of the questions. This thematic analysis, in tandem with a statistical analysis of inter-item reliability of possible index components, generated two indices with high Cronbach’s Alpha values (see Table 1).

Examination of the open-ended responses revealed that trees were mentioned as either a preferred land-use on farmland or as a detriment to the landscape by 58 percent (101) of all respondents to the openended questions. The inter-item reliability analysis also demonstrated that many of the quantitative items were measuring this concept. Consequently, “resistance to trees on farmland” was measured using an index composed of eight items that tapped reactions to the planting of a variety of tree types on farmland as well as general feelings about changes to the appearance and function of farmland.

Family farming was also a salient theme in the open-ended responses. Over 70 percent of those who answered these questions (122 respondents) mentioned the Poplar Farm Program as either an asset to family farming in the form of an alternative crop or a hindrance to family farming. Inter-item reliability analysis of quantitative responses also revealed the centrality of the family farming concept. Hence, valuation of family farming is measured in our model with a six-item index addressing themes of individual valuation of inheritance, continuity (intergenerational farming), and family ownership.

Table 2 displays bivariate correlations (for the analytic sample of 106 respondents) among the variables in our theoretical model that are generally consistent with our hypotheses. Resistance to trees on farmland is much more highly related to support for the Poplar Farm Program (r = -0.707) than is valuation of family farming (r = -0.197). Valuation of family farming and resistance to trees on farmland are strongly related, in the predicted positive direction (r = 0.423). Correlations as high as 0.7 might be interpreted by some as evidence that we have simply measured the same concept with different indicators. However, high correlations can also demonstrate strong causal connections, and we argue that in this data set we are observing powerful attitudinal determinants of support for the Poplar Farm Program. Furthermore, our thematic analysis of responses to openended survey questions revealed that, in contrast to direct expressions of resistance to trees on farmland, other negative perceptions of the program including environmental concerns and mistrust of the company were more salient.4 In other words, support for and resistance to the Poplar Farm Program is complex and multifaceted.

Findings

Path analysis was used to determine the empirical validity of our theoretical model. The first step involved regressing support for the Poplar Farm Program on resistance to trees on farmland, valuation of family farming, and the five exogenous control variables (county, age, gender, number of children, and percentage of income from farming). Resistance to trees on farmland was the dependent variable in the second multiple regression equation, with valuation of family farming and the five control variables serving as predictors. The final equation positioned valuation of family farming as the dependent variable. Thus, valua\tion of family farming and resistance to trees on farmland serve as intervening variables in this causal model. Table 3 contains the regression (path) coefficients for each of the three estimated equations.

As Table 3 reveals, the first two multiple regression equations have only one variable with a significant partial coefficient while the “family farming” equation has two significant predictors. Beginning with the exogenous control variables, the standardized coefficients show positive effects of total number of children in the family (β = 0.219) and farm income (β = 0.226) on valuation of family farming. Valuation of family farming, in turn, has a significant effect on resistance to trees on farmland. As expected, this effect is positive (β = 0.365), indicating that higher valuation of family farming increases resistance to planting trees on farmland. In the final stage of the model, the only significant predictor of support for the Poplar Farm Program is resistance to trees on farmland (β = -0.753). This extremely powerful effect largely accounts for the large amount of explained variance in this equation (R^sup 2^ = 0.53). The greater the resistance to trees on farmland, the less the support for the Poplar Farm Program, Figure 3 models these significant direct effects (non- significant paths are estimated, but not shown).

The very large path coefficients linking the two intervening variables (valuation of family farming and resistance to trees on farmland) with the dependent variable (support for the Poplar Farm Program) give cause for concern about possible multicollinearity in the equations. However, a systematic analysis of changes in the equations as the highly correlated variables were added one by one allays our concerns, since we did not observe any substantial changes in path coefficients or standard error estimates. In other words, despite these strong relationships, these equations are stable. Essentially they reveal a very basic but powerful storyline about farmers with increased financial dependence on farming and more children placing greater value on the family farm. This sentiment, in turn, increases resistance to trees on farmland and, ultimately, results in opposition to the Poplar Farm Program. The significance of this basic story line is reflected in the fact that when the two intervening variables (valuation of family farming and resistance to trees on farmland) are removed from the analysis, only 7 percent of the variation in support for the program with the five exogenous control variables is explained (analysis results not shown).

Our discussion so far has focused only on the direct effects on each of the predictor variables. However, by calculating indirect and total effects of the control variables and the first of the intervening variables (valuation of family farming) on support for the Poplar Farm Program we can learn more about the modeled relationships (Ryan et al. 2005). Indirect effects (via another variable) are calculated by multiplying the direct effect of a prior variable on the intervening variable by the direct effect of the intervening variable on the ultimate dependent variable (Lait and Wallace 2002:415). The total effect represents the sum of the direct and indirect effects of each determinant on the dependent variable.

When only direct effects are taken into account, resistance to trees on farmland is the only variable appearing to have a substantial effect on support for the Poplar Farm Program (Figure 3). However, when indirect and total effects are calculated, we see a more complex picture. Table 4 shows that, because of the substantial size of several core direct effects in the model, a number of the indirect effects are also of substantial size.5

The effect of county on support for the Poplar Farm Program changes substantially when indirect effects are taken into account (Table 4). The total effect of County on support for the program (- 0.156) is more than twice the size of its direct effect (-0.070). This demonstrates not only that valuation of family farming and resistance to trees on farmland are important intervening variables, but suggests Westlock residents are more supportive of the program than Athabasca residents. Compared to direct effects only, the effect of gender on support remained virtually the same when total effects were calculated (Table 4), providing support for the hypothesis that men are more supportive of the Poplar Farm Program. Percentage of income from farming has a more pronounced effect on support when its total effect is calculated (total effect = -0.217, compared to β = -0.131). Thus, as hypothesized, part-time farmers support the program more than fulltime farmers.

Table 4 also shows that resistance to trees on farmland is a very important intervening variable between valuation of the family farm and support for the Poplar Farm Program. The total effects of valuation of family farming (-0.144) is only marginally larger than its direct effect (β = 0.131), but the sign (of the total effect) is reversed. This suggests that, other things being equal, farmers with higher valuation of the family farm exhibited greater support of the program (Table 3). However, when valuation of family farming’s effect is mediated by resistance to trees (Table 4), farmers with higher valuation of the family farm are less supportive of the program.

Summary and Interpretation of Findings

This study examined the effect of valuation of family farming on farmers’ orientations to farmland conversion to non-traditional uses, specifically the establishment of tree plantations. This is an especially timely study as tree plantations are becoming more widespread in Canada and the United States, and their establishment on previously cleared farmland is a commonly proposed technique (Adams, Hobbs, and Johnson 2005; Sedjo 2001; Zasada et al. 2004). We began with a testable theoretical model of an internal value (valuation of family farming) and an external opinion (resistance to trees on farmland) that were hypothesized to influence support for a tree plantation program initiated in northeastern Alberta, Canada.

Percent of income from farming and number of children had the strongest direct effects on valuation of family fanning, supporting the arguments that the concept of family farming is more salient with those having greater financial dependence on farming, and that attitudes about family heritage and intergenerational transfer are more prominent among farmers who have children of their own (Keating and Munro 1989). However, as the R^sup 2^ value of this equation was relatively low (14%), it is clear that the determinants of valuation of family farming are complex and in need of more study. Place identity (Wester-Herber 2004), religion (Miller and Luloff 1981), and cultural background (Salamon 1985) are only a few of the social characteristics that may influence valuation of family farming. Both quantitative and qualitative sociological analysis of this concept would contribute a great deal to the family farming literature, and in so doing, would also shed light on the factors underlying land use controversies in agriculture areas.

Valuation of family fanning is clearly the strongest determinant of resistance to trees on farmland. Our results showed that a greater commitment to the notion of the family farm led to increased resistance to trees on farmland. This finding is consistent with previous research showing strong intergenerational des to the clearing and maintenance of farmland. Following Wester-Herber’s (2004) “continuity” concept, farmers who see intergenerational transfer and intergenerational farming as important would likely see planting trees on previously farmed land as a break in that tradition.

County of residence had an impact on support for the Poplar Farm Program. Specifically, residents of Westlock County were, overall, more supportive of the program whereas residents from Athabasca County, where the Al-Pac mill is located, were found to be less supportive. This finding suggests that the contentious history of the establishment of the Al-Pac mill in this County may be negatively affecting support for AlPac’s tree farming program. The effect of gender on support, though small (total effect β = 0.065), is notable. The fact that men were more supportive of the Poplar Farm Program reflects previous research showing men to be more interested in alternative agriculture (Meares 1997; Moore 1989; Rogers and Vandeman 1993). Our hypothesis that age would have a negative effect on support for the program is not supported in the model (total effect β = -0.017), which may suggest that plans to transfer the land to children may be more important than the age of the farmer. Those farmers without children or with children who do not plan to take over the farm may find the Poplar Farm Program more appealing, and be more likely to view planting trees on farmland as a way to continue earning an income after they retire.

As expected, percent of income from farming had a negative effect on support for the Poplar Farm Program, suggesting farmers with less of their income from farming were more supportive of the program. Perhaps these farmers see the program as an opportunity to obtain supplemental income from their land while maintaining their off- farm work. In addition, some of these farmers may be of retirement age, receiving the majority of their income from pensions and savings. The Poplar Farm Program might then be seen as a hands-off way to manage the land. Valuation of family farming was also a strong negative determinant of support for the Poplar Farm Program when total effects were taken into account. When valuation is mediated by resistance to trees on farmland, a farmer’s beliefs in the family farm can have a strong dampening effect on his or her support.

Resistance to tree\s on farmland was shown to be the strongest determinant of support for the Poplar Farm Program, two to three times more influential than any of the other effects when total effects were considered. Farmers who were resistant to the general idea of planting trees on farmland were also much less likely to support the Poplar Farm Program. It is important to consider the strength of the relationships revealed in our study. Though many land use decision-making models are based solely on economic factors, it is clear from our results that a general aversion to trees on farmland (based on concerns about changes in appearance, non-traditional use, and a non-food crop) is the strongest determinant of support for the Poplar Farm Program, much more so than income.

The results from this study should remind sociologists and policymakers to consider farm values as a key predictor of receptivity to agricultural land use conversions. Government agencies and private sector firms often focus primarily on rental fees and subsidies as incentives for land use change, without addressing how residents’ core values are often embedded in their land use preferences and choices. Future studies should attempt to capture more of the values and opinions determining support for alternative land use programs, including the study of collective notions of what constitutes farming and of appropriate conversions of farmland to other uses.

As the number of North American farmers dwindles, and as farmland is increasingly converted to non-agricultural uses, it is important to understand the motivations behind farmland conversion to non- agricultural uses (Zollinger and Krannich 2002:442). Family farms as economic entities are of decreasing importance to rural economies. Yet this study has shown that family farming as a rural symbol and set of values still profoundly impacts farmers’ opinions about agricultural change programs and policies. This study also suggests that resistance to the type of land use change is an important predictor of support for programs like the Poplar Farm Program.

Increased knowledge about the interrelationships between farming values and practices is necessary in order to “develop public policy, educational programs, appropriate extension activities and substantive theory directed to creating viable small farm systems and enhancing the quality of life in rural communities” (Sontag and Bubolz 1985:49). Future studies of farming practices and policies would be enhanced by research on the impact of farming values on opinions for and against farming options, especially those options that are not easily reversed, such as putting agricultural land back into trees.

1 See Appendix 1 for a description of the exact wording of each measure used in our analysis.

2 see Appendix 2 for comparison of the analytic sub-sample (the 106 farmers providing data for the test of our theoretical model) and the excluded respondents. White the two groups do not differ significantly on most variables, including our ultimate dependent variable (support for the Poplar Farm Program), the respondents in our analytic subsample are less likely to be pan-time fanners and are younger than excluded respondents. As our discussion of findings below reveals, percent of income from farming plays a significant predictive role in our final model. Thus, if more of the excluded respondents had been included in the analytic sample, the greater variation in this predictor variable might, if fact, have led to its effects in the final model being somewhat stronger.

3 We also note that a correlation of 0.7 translates into shared variance of 49%, leaving 51% of the variance unexplained.

4 Unfortunately, our survey did not include direct quantitative measures of these themes, so we were unable to incorporate them into our theoretical model as control variables.

5 Significance tests are not provided for the analysis of indirect and total effects since these calculations were completed by hand. We consider an indirect effect of 0.05 or larger to be substantively significant.

References

Abd-Ella, M.M., E.O. Hoiberg, and R.D. Warren. 1981. “Adoption Behavior in Family Farm Systems: An Iowa Study.” Rural Sociology 46:42-61.

Adams, W.T., S. Hobbs, and N. Johnson. 2005. “Intensively Managed Forest Plantations in the Pacific Northwest: Introduction.” Journal of Forestry 103:78-82.

Alberta-Pacific Forest Industries. 2001. History. Retrieved March 20, 2004, (www.Al-Pac.ca).

_____. 2002. Poplar Farms. Alberta-Pacific Forest Industries Inc.

Anosike, N. and C.M. Coughenour. 1990. “The Socioeconomic Basis of Farm Enterprise Diversification Decisions.” Rural Sociology 55:1- 24.

Barlow, K. and C. Cocklin. 2003. “Reconstructing Rurality and Community: Plantation Forestry in Victoria, Australia. “Journal of Rural Studies 19:503-19.

Bennett, J.W. and S. Kohl. 1963. “Two Memoranda on Social Organization and Adaptive Selection in a Northern Plains Region.” Plains Anthropologist 8:238-48.

Berndes, G., M. Hoogwijk, and R. van den Broek. 2003. “The Contribution of Biomass in the Future Global Energy Supply: A Review of 17 Studies.” Biomass and Bioenergy 25:1-28.

Boyle, R. 1999. “Planted Forests: Views and Viewpoints.” New Forests 17:5-9.

Carlson, J.E. and D.A. Dillman. 1983. “Influence of Kinship Arrangements on Farmer Innovativeness.” Rural Sociology 48:183-200.

Comstock, G. 1989. “Is the Family Farm a Thing of the Past- Ethics in Agriculture.” Managing Agricultural Technology for Profit March 5-8, 1989, The Lodge at Kananaskis.

Dasgupta, S. 2001. Rural Canada: Rural Sociological Perspectives. Queenstown, Ontario: The Edwin Mellen Press.

Droze, W.H. 1977. Trees, Prairies, and People: A History of Tree Planting in the Plains States. Denton, TX: Texas Woman’s University.

Fink, D. 1986. “Constructing Rural Culture: Family and Land in Iowa.” Agriculture and Human Values 2:43-51.

Gismondi, M. 1997. “Sociology and Environmental Impact Assessment.” Rural Sociology 22:457-79.

Gourlay, S. 1986. “Motivation: What Farmers Want.” Pp. 125-28 in Farming and Forestry: Proceedings of a Conference held at Loughborough University 26-28 September 86, edited by G.R. Hatfield. Surrey, UK: Forestry Commission.

Haverstock, L. 2005. Speech from the Throne 2005. Government of Saskatchewan. Retrieved March 2, 2006 (www.gov.sk.ca).

Keating, N.C. 1996. “Legacy, Aging, and Succession in Farm Families.” Generations 20:61-5.

Keating, N.C. and B. Munro. 1989. “Transferring the Family Farm: Process and Implications.” Family Relations 38:215-18.

Lait, J. and J.E. Wallace. 2002. “Stress at Work: A Study of Organizational-Professional Conflict.” Relations Industrielles 57:463-87.

LeHeron, R. and M. Roche. 1985. “Expanding Exotic Forestry and the Extension of a Competing Use for Rural Land in New Zealand.” Journal of Rural Studies 1: 211-29.

Lind, C. 1995. Something’s Wrong Somewhere. Halifax, Nova Scotia: Femwood Publishing Company Limited.

Lobao, L. and K. Meyer. 2001. “The Great Agricultural Transition: Crisis, Change, and Social Consequences of Twentieth Century U.S. Farming.” Annual Review of Sociology 27:103-24.

Meares, A.C. 1997. “Making the Transition from Conventional to Sustainable Agriculture: Gender, Social Movement Participation, and Quality of Life on the Family Farm.” Rural Sociology 62:21-47.

Moore, K.M. 1989. “Agrarian or Non-Agrarian Identities of Farm Spouses.” Rural Sociology 54:74-82.

Miller, M.K. and A.E. Luloff. 1981. “Who is Rural?” Rural Sociology 46:608-25.

Neumann, P.D., N.T. Krogman, and B.R. Thomas. 2007 (Forthcoming).. “Public Perceptions of Hybrid Poplar Plantations: Trees as an Alternative Crop.” International Journal of Biotechnology.

Pfeffer, M.J. 1989. “Values and Policy Conflict in West German Agriculture.” Agriculture and Human Values 6:59-69.

Raedeke, A.H., J.J. Green, S.S. Hodge, and C. Vadivia. 2003. “Farmers, the Practice of Fanning and the Future of Agroforestry: An Application of Bourdieu’s Concepts of Field and Habitus.” Rural Sociology 68:64-86.

Rogers, D.M. and A.M. Vandeman. 1993. “Women as Farm Landlords: Does Gender Affect Environmental Decision-Making on Leased Land?” Rural Sociology 58:560-8.

Rohwert, R.A. 1951. “Family Farming as a Value. Rural Sociology 16:330-39.

Ryan, V.D., K.A. Agnitsch, L. Zhao, and R. Mullick. 2005. “Making Sense of Voluntary Participation: A Theoretical Synthesis.” Rural Sociology 70:287-313.

Salamon, S. 1986. “Changing Family Farms and Rural Community Relationships.” Pp. 245-52 in New Dimensions in Rural Policy: Building Upon Our Heritage, edited by R.C. Wimberley, D. Jahr, and J. Johnson. Washington, D.C.: Joint Economic Committee of Congress.

Sedjo, R.A. 2001. “The Role of Forest Plantations in the World’s Future Timber Supply.” Forestry Chronicle 77:221-25.

_____. 1999. “The Potential of High-Yield Plantation Forestry for Meeting Timber Needs.” New Forests 17:339-59.

Sherman, J. 1997. “Some Consequences of Cheap Trees and Cheap Talk: Pulp Mills and Logging in Northern Alberta.” Ecologist 27:64- 8.

Sinnema, J. 2005. “How the West was Hyped: U of A Student Explores Myth, Reality of Small Farm.” Edmonton Journal October 17, p. Al.

Smith, R.A., B.L. McFarlane, J.R. Parkins, and P.A.M. Pohrebnuik. 2005. Landowner Perspectives on Afforestation for Carbon Sequestration in Canada’s Prairie Provinces-Information Report NOR- X-401. Edmonton, Alberta, Canada: Natural Resources Canada.

Smithers, J. and P. Johnson. 2004. “The Dynamics of Family Farming in North Huron County, Ontario. Part I. Development Trajectories.” The Canadian Geographer 48:191-208.

Sontag, M.S. and M.W. Bubolz. 1985. “Research in Progress: Case Studies of Family Adaptation to Changing Resources and Environments.” Agriculture and Human Values 2:48-51.

Statistics Canada. 2001a. Agriculture Community Profiles: Athabasca, Alberta. Retrieved October 15, 2005 (www.statcan.ca).

_____. 2001b. Agriculture Community Profiles: Westlock, Alberta. Retrieved October 15, 20\05 (www.statcan.ca).

_____. 2001c. Total Area of Farms, Land Tenure and Land in Crops, by Provinces (1981-2001 Censuses of Agriculture). Retrieved October 15, 2005 (www.statcan.ca).

Strange, M. 1988. Family Farming: A New Economic Vision. Lincoln, NE: University of Nebraska Press.

Taylor, C.C. 1954. “The Family Farm in the New Society.” Rural Sociology 19:271-80.

van Kooten, G.C., S.L. Shajkh, and P. Suchanek. 2002. “Mitigating Climate Change by Planting Trees: The Transaction Costs Trap.” Land Economics 78:559-72.

Walter, G. 1997. “Images of Success: How Illinois Fanners Define the Successful Farmer.” Rural Sociology 62:48-68.

Wester-Herber, M. 2004. “Underlying Concerns in Land-Use Conflicts-The Role of Place-Identity in Risk Perception.” Environmental Science and Policy 7:109-16.

Zasada, M., J.P. Siry, and C.J. Cieszewski. 2004. “Intensive Forest Management in the U.S. South.” Sylwan 148:61-72.

Zimmerman, T.S. and R.J. Fetsch. 1994. “Family Ranching and Farming: A Consensus Management Model to Improve Family Functioning and Decrease Work Stress.” Family Relations 43:125-31.

Zollinger, B. and R.S. Krannich. 2002. “Factors Influencing Farmers’ Expectations to Sell Agricultural Land for Non- Agricultural Uses.” Rural Soaology 67:442-63.

Pamela D. Neumann

Department of Rural Economy

University of Alberta

Harvey J. Krahn

Department of Sociology

University of Alberta

Naomi T. Krogman

Department of Rural Economy

University of Alberta

Barb R. Thomas

Alberta-Pacific Forest Industries Inc.

Department of Renewable Resources

University of Alberta

* The authors would like to thank SSHRC (Social Sciences Humanities Research Council of Canada) for their financial support and Alberta-Pacific Forest Industries Inc. for their financial and in-kind support of this project and production of the Alberta map by M. Smith. Also, we would like to thank all of the participants in the interviews for their time and hospitality, and Drs. Bonnie McFarlane and Debra Davidson for their input into this paper. Direct correspondence to: Naomi Krogman, Associate Professor, Department of Rural Economy, 5-15 General Services Building, University of Alberta, Edmonton, Alberta, T6G 2H1. Email: naomi.krogman@ualberta.ca.

Appendix 1: Measurement of Theoretical Concepts

Support for the Poplar Form Program:”How do you feel about the PoplarFarm Program?”: 1 = ‘very negative’ to 5 = ‘very positive’

Resistance to Trees on Farmland measures

Land Appearance: “(My) land stays the same in appearance”: 1 = ‘not at all important’ to 5 = ‘extremely important’

Land Farmed Traditionally: “(My) land is farmed in a traditional way”: 1 – ‘not at all important’ to 5 = ‘extremely important’

Cleared Land to Bush: “When thinking about previously cleared farmland in your county, in YOUR opinion, how acceptable are the following uses? – let it go back to bush”: 1 = ‘completely unacceptable’ to 5 = ‘completely acceptable’ (reverse coded)

Cleared Land to Native Trees: “When thinking about previously cleared farmland in your county, in YOUR opinion, how acceptable are the following uses? – grow native trees for pulp and paper”: 1 = ‘completely unacceptable’ to 5 = ‘completely acceptable’ (reverse coded)

Cleared Hybrid Trees: “When thinking about previously cleared farmland in your county, in YOUR opinion, how acceptable are the following uses? – grow hybrid trees for pulp and paper”: 1 = ‘completely unacceptable’ to 5 = ‘completely acceptable’ (reverse coded)

Cleared Land to GM Trees: “When thinking about previously cleared farmland in your county, in YOUR opinion, how acceptable are the following uses? – grow genetically modified trees (GMOs) for pulp and paper”: 1 = ‘completely unacceptable’ to 5 = ‘completely acceptable’ (reverse coded)

Tree Plantations on Farmland: “When thinking about tree plantations, how concerned are you about growing trees on farmland?” 1 = ‘not at all concerned’ to 5 = ‘very concerned’

Good Farming and Food Production: “In YOUR opinion, how important are the following to good farming? – production of food”: 1 = ‘not at all important’ to 5 = ‘extremely important’

Valuation of Family Farming measures

Responsibility to Keep Land in Family: “Do you feel a responsibility to keep (your) land in the family?”: 1 = ‘yes’; O = ‘no’ (receded to 4 and 2, respectively)

Importance of Land Inheritance: “How important to you personally are the following characteristics of your land? – the land is inherited by your family”: 1 = ‘not at all important’ to 5 = ‘extremely important’

Land is a Family Heritage: “How do you feel about your land? – my land is a family heritage”: 1 = ‘strongly disagree’; 2 = ‘disagree’; 3 = ‘neutral’; 4 = ‘agree’; 5 = ‘strongly agree’

Importance of Family Ownership of Land: “HoW important to you personally are the following characteristics of your land? – the land is owned by your family”: 1 = ‘not at all important’ to 5 = ‘extremely important’

Importance of Family Farming Land: “How important to you personally are the following characteristics of your land? – the land is farmed by your family”: 1 = ‘not at all important’ to 5 = ‘extremely important’

Importance of Next Generation Farming Land: “How important to you personally are the following characteristics of your land? – the land is farmed by the next generation”: 1 = ‘not at all important’ to 5 = ‘extremely important’

County:County of residence: 1 = Athabasca; O = Westlock

Gender: “Sex”: 1 = male; O = female

Age: “What year were you born?”: open-ended; converted to age using 2005 (year survey administered) as reference

Number of Children: “How many children do you have? – living at home – living away from home”: open-ended responses added together

Percent of Income From Farming: “Approximately what percentage of (your total net household income before taxes in 2003) is from farming?”: open-ended

Copyright Rural Sociological Society Mar 2007

(c) 2007 Rural Sociology. Provided by ProQuest Information and Learning. All rights Reserved.