Local Dependency, Land Use Attitudes, and Economic Development: Comparisons between Seasonal and Permanent Residents1

Rural Sociology 61(3), 1996, pp. 427-445 Copyright (j) 1996 by the Rural Sociological Society

Local Dependency, Land Use Attitudes, and Economic Development: Comparisons between Seasonal and Permanent Residents' Gary P. Green, David lJlarcouiller, * Steven Deller, ** Daniel Erkkila, *** N.R. Sumathi**** Department ofRural Sociology, Unioersity ofWlSconsin-lI!adison, lIfadison, WISconsin 53706 *Department of Urban and Regional Planning, University of WlSconsin-lIfadison, lIfadison, WISconsin 53706 **Department ofAgricultural Economics, University ofWlSconsin-lIfadison, lIfadison, WISconsin 53706 ***Tourism Center, University oflIfinnesota, lIfinneapolis, lIfinnesota 55455 ****Center for Community Economic Development, University ofWlSconsin-lIfadison, lIfadison, WISconsin 53706

For many rural communities in the United States, tourists and retirees make a major contribution to the local economy. The role of recreational home ownership in these communities is not well understood. We examine the attitudes toward land use controls and local economic development among seasonal and permanent residents. We compare growth machine and local dependency theory explanations for support of land use controls and growth activities. Based on focus groups and survey data collected from seasonal and permanent residents of a northern Wisconsin county, we find that full-time (permanent) residents are much more supportive of local economic development activities and less likely to favor land use planning than are seasonal residents. Socioeconomic differences between seasonal and permanent residents do not explain away the variation between the two groups in these attitudes. Among seasonal residents, support for land use controls declines as they spend more time at their lake homes. The results tend to support Cox and Mair's local dependency thesis. ABsTRACT

Introduction

Recreational or seasonal housing provides an important leisure opportunity for users and makes up a significant component of the housing stock in many rural communities, especially in the Upper Great Lakes region. The development and use of recreational housing contribute additional jobs and income to host communities, which are likely to be dependent upon natural resources for their 1 Paper presented at the North American Land Tenure Program conference, "Who Owns America? Land and Resource Tenure Issues in a Changing Environment," Madison, Wisconsin. This research is supported by a grant from the North Central Regional Center for Rural Development, with additional support by the University of Wisconsin-Extension and the Minnesota Extension Service. We wish to acknowledge the "local knowledge" of Sheila Landsverk and the technical assistance ofJeffery Alpi.

428

Rural Sociology, Vol. 61, No.3, Fall 1996

economic base and are likely to be unsuccessful in diversifying their economy," Recent research evaluating economic development strategies for rural communities has ignored the important role of recreational homes. Further, the research on tourism pays little attention to the contributions of recreational housing in local and regional economies (Jordan 1980). Most research on recreational housing has focused on the attitudes of either natives (Allen, Long, Perdue, and Kieselbach 1988; Lew 1989) or recreational homeowners (Gartner 1987; Girard and Garner 1993), but not on both. Other studies have focused on the social and economic contribution of retirees to local economies in rural communities (Deller 1995). The primary purpose of this paper is to examine the attitudes toward land use controls and local economic development activities among seasonal and recreational homeowners. We assess how various individual characteristics influence attitudes toward land use controls and local economic development. We are especially interested in assessing how local dependency shapes these attitudes. Local dependency and community

Growth machine theory has proven to be a powerful conceptual tool for examining the influence of land-based elites on local development policies, planning, and land-use decisions (Logan and Molotch 1987; Pfeffer and Lapping 1994). According to growth machine theory, land-based elites, such as local businesses and real estate developers, are active promoters of growth because they benefit from the increasing exchange value of land. Local residents, primarily homeowners, who are more interested in the use-value of land frequently are opposed to, or at least less supportive of, growth. Because of differential access to power and resources, the local growth machine is usually more successful than neighborhood organizations in advancing local development policies (Stone 1989). However, there are examples where the growth machine has not been successful in shaping the local development agenda (Swanstrom 1985). An important assertion of growth machine theory is that local residents are attached to their community in ways that link them through social, political, and economic ties (Wilkinson 1991). Moreover, these ties are frequently the basis for conflicts between resi2 One reviewer pointed out that second home development could also be viewed as an extractive industry in itself. Some would argue that seasonal residents are exploiting the natural environment as well through many of their activities on lakes (e.g., building lakefront property). Perhaps the crucial difference between seasonal home ownership and other extractive industries is that these natural resources are viewed in consumption rather than production terms. There may be differences in how residents relate to the natural environment if they view it as a resource to produce jobs versus something to preserve for future "consumption."

Local Dependency -

Green et al.

429

dents and land-based elites in localities where development projects take place. Growth machine theory is especially useful in understanding settings where conflicts in the interests of residents and land-based elites are acute, particularly over growth and land use (Capek and Gilderbloom 1992). Although growth machine theory recognizes that residents may not necessarily benefit from growth, it also acknowledges that residents may lose from economic decline in their localities through the loss of public services, tax increases to maintain services, or even environmental degradation. Economic growth contributes to the ability of local governments to generate revenues, and these revenues enable residents to pay lower taxes and improve public services (Peterson 1981). The loss ofjobs and businesses also may contribute to other signs of decay, such as vacant buildings, open dumps, or less environmental regulation. Because of the accommodation function of localities, local residents also need to maintain a viable local economy. This function helps explain why the ideology of growth may be attractive to many residents (Gottdiener and Neiman 1981). Much of the literature focuses on conflicts in localities over use and exchange value, but it often ignores the contradictory interests of residents who are interested in both accumulation and accommodation functions (Davis 1991). Home ownership represents a large investment for most families (Perin 1977). Consequently, home ownership as a crucial resource for wealth accumulation has been identified as an important social cleavage that has significant sociological consequences (Davis 1991). Although homeowners may be similar to land-based elites in some ways, they differ in others. In addition to an investment, homes provide use-value to its occupiers, which provides a degree of communality or ties with others. This contradiction may be less apparent for seasonal homeowners who may have less interest in the exchange-value of their homes." Relative to permanent residents, seasonal homeowners probably have less to gain and more to lose from growth and development. Our hypothesis is that most seasonal residents may be less interested in increasing the value of their second home, and they may be more concerned with maintaining environmental quality and controlling growth.' S Regarding this distinction between use- and exchange-value, it may be that both permanent and seasonal residents look to local growth and land use patterns policies to maximize exchange-value. But what drives recreational housing exchange value may be different than what influences market value in permanent housing stock. As a region grows, population density goes up driving both types of housing stock prices up, but recreational housing less so than permanent housing. Conversely, little or no growth, but heightened environmental quality will drive up recreational housing prices more than permanent housing. • It should be pointed out that, by maintaining environmental quality and control-

430

Rural Sociology, Vol. 61, No.3, Fall 1996

An alternative view is that concern with environmental quality and with controlling growth may vary by levels of local dependency among residents (Cox and Mair 1988). Rather than assuming that these attitudes toward land use and growth are a result of structural interests, Cox and Mair contend that support for growth attitudes is developed through local relationships and investments in immobile capital. Most businesses are dependent on the locality because of investments in the built environment. According to Cox and Mair, businesses promulgate a shared interest in the local community, and the local dependence of residents makes them receptive to these arguments. Moreover, residents are dependent on their locality due to their investments in home ownership, and social ties and relationships also may make individuals dependent on their local community for social support and resources (Stack 1974). Growth policies, according to this perspective, are not driven by local elites trying to maximize their exchange value, but are driven by dependency on local investments. As seasonal residents become more socially tied to permanent residents, we expect them to be more supportive of growth promotion and less supportive of land use controls. There is often a perception by permanent residents that growth and development will provide jobs for themselves and their children. Conversely, if seasonal residents do not become linked to their larger community, we expect them to emphasize more the use-value than the exchangevalue of their home. As a result, we expect that they will be less supportive of growth and more supportive of land use controls. The differences in orientation are largely a product of the differential social ties of the two groups. As permanent residents form social ties to their community, they develop a shared interest in growth and development. Seasonal residents may never become integrated into the larger community and thus would fail to develop this shared interest because they are less dependent on the local community for their social and economic life (Green 1996). Seasonal residents may become very attached to their community, through lakes associations and contacts with other seasonal residents; however, this interaction does not produce a shared interest in the larger community. In effect, full-time residents are defining community in production terms, and seasonal residents define it as collective consumption (Castells 1983). Thus, the growth machine theory and local dependency thesis have somewhat different hypotheses regarding why seasonal and permanent residents may differ in their attitudes toward land use ling growth, recreational home owners may be increasing the value of their property. This example illustrates how difficult it may be to separate use and exchange value as goals in this situation.

Local Dependency -

Green et al.

431

controls and local economic development. Growth machine theory would argue that the differences are primarily due to the class differences between the two groups. The local dependency thesis contends that ties to the community explain these differences. Research methods

Data for this analysis are drawn from focus groups and surveys conducted in Forest County, Wisconsin, in the fall of 1994. Forest County was selected as the research site because it has one of the highest proportions of seasonal homeowners in the region. Data from the 1990 Census of Population and Housing indicate that 50 percent of the housing units in Forest County are recreational homes, and this is the fourth highest ranking of any county in the state ofWisconsin. Forest County is located near the Michigan Upper Peninsula. Historically, the major industry in Forest County has been forest products, and its major raw material has been roundwood. In the late 19th century, railroad companies pushed service into the eastern part of the county and removed the old-growth softwood saw timber. Along with the development of a transportation system, sawmills sprang up throughout the region, and subsequent harvesting focused on the hardwood resources. By the 1930s, however, the old-growth timber supply waned, and the Depression closed most of the big mills. Since the 1950s, tourism has grown in importance. Much (75 percent) of the forest land is held by the U.S. Forest Service (Nicolet National Forest), which places limits on development in the county. We conducted four focus group sessions, one each among business owners, government officials, seasonal residents, and full-time residents. The focus group sessions were conducted to help identify hypotheses and to build the survey instruments to be used. We relied heavily on Krueger's (1994) suggestions for conducting focus groups. The group with business owners focused on the contributions recreational housing makes to the retail sector, as well as some of the problems businesses face in these communities. Government officials were asked to discuss the consequences of recreational homes for property tax values and real estate taxes. Both seasonal and full-time residents were asked to discuss the level of social contact between seasonal and full-time residents and various issues related to land use controls, growth promotion, and environmental protection. Each session lasted approximately 90 minutes. The local Cooperative Extension Agent, who was extremely familiar with the community, helped identify and recruit the participants," ~ Because participants in the focus group were not randomly selected, we may have injected some bias into the process. We were not using the focus groups to test hypotheses, but rather to help us develop survey instruments.

432

Rural Sociology, Vol. 61, No.3, Fall 1996

In addition to the focus group, we conducted mail surveys among seasonal and full-time residents in Forest County. We mailed the surveys during the fall of 1994 to avoid sending them during a time when seasonal residents may have been away from their permanent residence. Seasonal residents were identified by lake association membership lists. The local Cooperative Extension agent who works with the lakes associations is confident that almost all of the recreational homes in Forest County are on lakes and that almost all seasonal residents are members of these associations. Full-time residents were identified through county property tax records. This strategy excludes a small number of renters in the county," A random sample of approximately 550 households was chosen for each group. After three weeks, we mailed a postcard reminder to those who had not responded, and three weeks later a replacement questionnaire. This procedure yielded an excellent response rate of 79 percent among recreational home owners and 55 percent among full-time residents. There was a considerable overlap in the 1\\'0 questionnaires, but recreational homeowners were asked additional questions about their recreational property and their use of the home. Descriptive statistics for all of the variables included in the analysis are in the Appendix. The independent variables in the first stage of the analysis include property ownership (coded: l=full-time resident; O=seasonal resident), gender (coded: l=male; O=female), age (number of years), education (number of years schooling completed), household income, and years lived in the community. The 1\\'0 dependent variables are support for land use controls and local economic development activities. We created an additive scale using four questions to measure support for land use controls." The additive scale for local economic development activities included seven items, with respondents asked to indicate how important various economic development activities were in improving the quality of life in Forest County," Respondents answered on a five point scale, not important to very important. The questions in each scale were identified through factor analysis which revealed that the items were correlat6 Not including renters in the sample probably provides a clearer test of our thesis. Renters' level of suppon for growth and land-use policies are not as directly tied to land values. 'The response to these four statements was measured on a likert scale (0-6). The four statements were: (1) There needs to be land-use planning in Forest County; (2) There needs to be more cooperation among communities in planning for growth; (3) Use of private land should be based on what the owner wants rather than being restricted by zoning; and (4) Zoning restrictions hun more than they help an area. "The seven items were growing and harvesting trees, processing trees into wood products, extracting minerals, processing minerals, attracting manufacturing firms, development of retail and service industries, and supporting entrepreneurial activities.

Local Dependency -

Green et al.

433

ed around a single construct (Eigenvalue > 3.0). Other items that did not load on this construct were deleted from further analysis. We conducted a reliability check on these scales, and each had a Cronbach's Alpha of at least 0.7. To ease the interpretation of these scales, we recoded the variables so that the higher the score, the more support there was for land use controls and local economic development activities. Our regression analyses examine the effects of property ownership (seasonal homeowner versus full-time resident) and additional socio-demographic characteristics about attitudes toward land use and zoning. These socio-demographic characteristics are introduced into the model because factors such as gender, education, income, and age influence attitudes toward the environment and development (Buttel 1987; Van Liere and Dunlap 1980). We also include the length of time that the respondent has lived in Forest County. Length of residence was included because it influences the probability that individuals will have developed social ties and attachments to the community (Kasarda and Janowitz 1974). Our analytical strategy is to examine whether the relationship between property ownership (seasonal versus permanent resident) and support for land use controls and growth disappears when controlling for the sociodemographic characteristics, especially class. We use OLS regression to examine the relationships between the independent and dependent variables," Focus group session findings

The recreational homeowner focus group consisted of four "snowbirds" and three others who lived within a several hours drive from Forest County. One of the first community issues brought up by this group was a concern over public services delivery and cost. Recreational homeowners seemed willing to pay their share of taxes in Forest County. They recognized that a second home was a luxury and that Forest County had a low tax base. However, several focus group participants questioned the fairness of recent assessments and whether they received the level of public services in proportion to their share of taxes. Many of the individuals felt that seasonal homeowners had higher property assessments than did full-time residents, even when the value of their property was the same. Seasonal homeowners were not demanding significant increases in the level of services; rather, they were interested -in maintaining minimal levels of services. 9 Using the summated scales as dependent variables may violate some of the assumptions of OLS regression. However, given the robustness of OLS regression, we concluded that it was the best choice of analytical techniques.

434

Rural Sociology, Vol. 61, No.3, Fall 1996

Recreational homeowners recognized that the tax burden was placing a second home out of the reach of many people. Although they emphasized the effects of recent assessments on seasonal homeowners, there was no discussion of what this meant for permanent residents. One common concern among seasonal residents is their lack of influence on local public officials. As one permanent resident said: "If you're not a voter, you're shit ... And it's every time you go some place." Seasonal homeowners admitted that employment and population growth were important to increasing the tax base of the county, but they stressed the limits of growth and questioned whether the costs exceeded the benefits of growth. Recreational homeowners said that "locals" and recreational homeowners did not interact much, but as individuals, they had some long-standing relationships with people in Forest County. Overall, seasonal residents reported they rarely participated in local organizations other than lakes associations. None of the seasonal homeowners identified any specific sources of conflict between the two groups. Nine individuals were included in the focus group of full-time residents. Full-time residents identified a much different set of community issues and concerns. They were generally much more prodevelopment than the seasonal residents. Full-time residents saw high tech and low impact industries (other than wood-based industries) as holding the greatest benefits for the county. A major reason they focused on development issues was the amount of land that was owned by the federal government (national forest) or Native Americans. Federal land ownership restricted the amount of land that can be developed and, therefore, it restricted the tax base in the county. Another issue dividing seasonal homeowners from permanent residents in the county concerned the benefits and costs of a proposed copper mine in the area. Local residents were much more likely to see the benefits of job creation from this development, while seasonal homeowners questioned the environmental impact of the mine. Aside from development issues, full-time residents expressed concern over zoning. Their major criticism was that zoning tended to be arbitrarily enforced. Full-time residents tended to be much more critical of recreational homeowners than the reverse. During the focus group sessions, full-time residents talked about how recreational homeowners are "stand-offish." One individual voiced the following concern over seasonal homeowners: Kinda wondering about these seasonal homeowners here, and I'm getting the impression they are opposed to anything that will make a little noise, create a little dust, and

Local Dependency -

Green et al.

435

those kind of things that are going to give our people some work here, and that's probably the only problem I have with them. Throughout my life, I've worked for them, and I've made money off of them, but I think they have a little tendency to be a little stand-offish and try to take care of themselves, which there is nothing wrong with, but I think they are hurting the local people that are here. Most of the discussion centered on the differences in class position between the two groups. Seasonal homeowners were viewed as blocking additional growth and development in the community. In addition, full-time residents said that seasonal homeowners were demanding in terms of the types of services that were provided by the county. None of the permanent residents reported having much social contact with seasonal residents. Permanent residents expressed little concern with the equity of tax assessments. However, they were concerned that seasonal residents were driving up the value of land because they were willing to pay "outrageous" prices for lakefront property. As a result, property values throughout the community had increased, which made it more difficult for permanent residents to continue to live there. The focus groups' sessions revealed four major differences between recreational homeowners and full-time residents. First, there appeared to be major differences between the two groups with respect to the local issues they identified as the most important for the community. Recreational homeowners are primarily concerned with environmental issues and service provision; full-time residents are more concerned with increasing the tax base and development. Second, the two groups differ in their position about the benefits of growth and development. Recreational homeowners are more likely to question the benefits of growth and to emphasize growth limits. Third, the focus groups revealed interesting differences in the perceived relationships between the avo groups. Full-time residents are more likely to perceive tensions; they emphasize the class differences between the avo groups and their divergent attitudes toward growth, development, and the environment. Finally, the avo groups differ in what they perceive is the impact ofrecreational home ownership. Recreational homeowners view these impacts as almost entirely positive, particularly in terms of their contributions to the tax base. Full-time residents stress the additional demands recreational homeowners place on services, the increased cost of living and property values, and the class division between seasonal and permanent residents. Survey findings

The impression held by local residents, that there are substantial social class differences between seasonal and full-time residents, is

436 Table 1.

Rural Sociology, Vol. 61, No.3, Fall 1996 Attitudes about land use and zoning issues among full-time and seasonal residentsFull-time

Seasonal

t-value

- - - - - Mean - - - - - Land use planning implemented at the local level increases local control of local development issues. There needs to be land-use planning in Forest County. There needs to be more cooperation among communities in planning for growth. Use of private land should be based on what the owner wants rather than being restricted by zoning. Provision should be made for adequate public access to lakes. Zoning restrictions hurt more than they help an area.

2.06

1.91

1.50

2.07

1.55

5.30***

1.55

1.68

-1.48

2.48

3.63

-8.74***

1.86

2.83

-8.57***

3.29

3.96

-5.83***

*** P < .001. a These items were coded on a likert scale ranging from 0 (strongly agree) to 6 (strongly disagree). Thus, the lower score, the more favorable the item was rated by respondents. supported by the survey data. We find statistically significant differences in income, education, and occupation between seasonal and full-time residents. For example, almost 60 percent of the seasonal residents are professionals, while only about 34 percent of the fulltime residents are classified as professional. The annual household incomes of the two groups also were quite different. While roughly three percent of recreational homeowners had an annual household income of less than $15,000, roughly 22 percent of residents reported incomes in this category. On the other extreme, roughly 50 percent of the recreational homeowners reported annual before-tax household incomes in excess of $50,000, while only 13.5 percent of resident respondents reported incomes in this category. Whereas residents had an average of 12.9 years of formal education, the average for recreational homeowners was 14.8 years. We asked respondents a series of questions about their support for land use controls (see Table 1). Seasonal and permanent residents hold significantly different positions regarding land use and zoning; seasonal residents are much more supportive than permanent residents of land use controls and zoning. Permanent residents are more likely than seasonal residents to believe that use of private land should be based on individual rather than on community preferences. Table 2 reports respondents' (seasonal and permanent residents) rankings of various economic development strategies as a means of

Local Dependency Table 2.

Green et al.

437

Views about economic development among full-time and seasonal residentsFull-Time Seasonal

Growing and harvesting trees Processing trees into wood products Extracting minerals Processing minerals Tourism development Native American gaming Attracting manufacturing firms Development of retail and service industries Supporting entrepreneurial activities Helping existing businesses remain viable

t-Value

- - - - - Mean - - - - 1.51 1.90 -4.89*** 2.26 -8.42*** 1.56 2.97 3.86 -7.69*** -6.77*** 3.16 3.92 -4.10*** 1.98 2.39 3.67 3.99 -2.98*** 1.73 2.75 -10.08*** -7.44*** 1.75 2.39 2.07 2.56 -5.48*** -3.21** 1.67 1.92

** p < .01; *** P < .001. These items were coded on a scale ranging from 1 (very important) to 5 (not important). Thus, the lower the score, the more favorable it was rated by respondents. a

improving the quality of life in Forest County. For many items, the differences are primarily a matter of degree rather than different positions toward growth and development. For every economic development strategy considered, permanent residents are more likely than seasonal residents to favor funding the projects locally. The greatest support is for growing and harvesting trees and helping existing businesses remain viable. Both groups also agree that supporting local businesses is important. Moreover, there are few differences in support for Native American gaming, and for extracting and processing minerals. Exxon has proposed to develop in Forest County what would be the largest underground mine in North America, and almost one-half of the respondents reported that extracting and processing of minerals is not important to improving the quality of life in the county. Less than 10 percent said that extracting and processing minerals was very important. Seasonal and permanent residents differ in their assessment of the various economic development strategies as a means of improving the quality of life in the community. For every single strategy, permanent residents are more likely to consider the economic development strategy as an important contributor to the quality oflife in Forest County. Some of the differences are relatively small, such as for Native American gambling and helping existing businesses remain viable. There is a fair amount of consensus within both groups that helping existing businesses remain viable is desirable, and supporting Native American gaming is not. The differences are strongest for attracting manufacturing firms and processing trees into wood products.

438

Rural Sociology, Vol. 61, No.3, Fall 1996

Property ownership has an effect over and above the effects of socio-demographic characteristics (see Table 3). Differences in land use and zoning attitudes between seasonal and permanent residents exist even when controlling for gender, age, education, and household income. However, household income, education, and age are positively related to support for these regulations. Overall, class (measured by education and income) does not explain away the differences between seasonal and permanent residents with regard to land use and zoning attitudes. Length of residence was not statistically significant; however, the longer residents live in the community, the less likely they are to support land use controls. Permanent residents are more supportive of growth activities than are seasonal residents (see Table 3). Other socio-demographic variables do not achieve statistical significance in the model. The only other variable that is statistically significant is the length of time the resident has lived in the community. The longer the residents have lived in the community, the more supportive they are of local development activities. Next, we examine the support for land use controls and economic development activities among seasonal residents only (see Table 4). We conduct this separate analysis to examine the effects of their social contacts and ties with full-time residents in Forest County on growth and development attitudes. The local dependency hypothesis suggests that these social ties and contacts should produce more similar attitudes to full-time residents. First, we include the number of days the seasonal resident has spent in their recreational home during the past year. One hypothesis is that the more time the seasonal resident spends in their recreational home, the more likely they are to have contact with permanent residents, and thus the more likely they are to develop similar attitudes toward growth and development. Alternatively, the more time that seasonal residents spend at their recreational home, the more likely they are to develop ties with other lake residents who will be less concerned with the exchange-value of their seasonal home. Next, we include four statements about seasonal residents' relationship to the larger community. They were asked to respond to statements that were rated on a six point likert scale (coded: O=strongly agree; 6=strongly disagree). The four statements are: (1) I feel at home with year round residents; (2) I have interest in nonlake related issues in Forest County; (3) The majority of my friends in Forest County are recreational homeowners; and (4) I feel welcome in non-lake related activities in Forest County. The coding was reversed in this analysis so that the higher the score, the more tied the seasonal resident is to Forest County.

z-ratio

[-ratio

.013

.775

6.928***

2.632***

-.047

-1.285

Growth activities 3.373 (.339) .369 (.031) 2.141 [E-04] (5.833 [E-04]) -.096 (-.058) -.007 (-.002) .036 (.122) 19.437*** .179 .169

b

a

Note: Standardized beta coefficient in parentheses. Coded: 1 = $250,000. The question asked for 1993 before-tax annual household income.

-1.678

.126

3.370***

2.349*

-.323

-2.807***

Land use controls -1.090 (-.135) -.124 (-.013) .031 (.106) .207 (.151) .327 (.126) -.018 (-.078) 13.167*** .122 .112

b

OLS analysis of support for land use controls and growth activities

*p < .05; ** P < .01; *** P < .001.

Constant R2 Adjusted R2

Length of residence

Property ownership (1=full-time; O=seasonal) Gender (1=male; O=female) Age (number of years) Education (years of schooling) Household income's

Table 3.

\0

~

l"'""

l;l

~

s~

I

~

1~

~ ~ .....

Note: Standardized beta coefficient in parentheses.

* P < .05; ** P < .01; *** P < .001.

Constant R2 Adjusted R2

Welcome in non-lake activities

Majority friends are recreational owners

Interest in non-lake issues

At home with full-time residents

Days at residence

Length of residence

.735

1.126

2.885***

-1.277

3.075***

-.178

1.334

3.057***

.358*

.117

t-ratio

Land use controls .062 (.007) .009 (.028) .244 (.192) .204 (.089) -.004 (-.023) .007 (.195) -.262 (-.086) .507 (.176) .171 (.069) .144 .049 12.070*** .125 .089

b

t-ratio

.

2.176*

-.406

~

-1.587

*

~

~

~VJ

~

~~

r-

~

~

~

~

~

i::l .....

~

-s,

-.580

1.081

.834

.733

-.979

1.123

1.141

Growth activities .842 (.075) .039 (.035) -.109 (-.066) .155 (.212) .028 (.069) .003 (.072) -.162 (-.042) -.391 (-.109) -.085 (-.027) .592 (.156) 16.404*** .063 .022

b

OLS analysis of support for land use controls and growth activities among seasonal residents

Gender (l=male; O=female) Age (Number of years) Education (Years of schooling) Household income

Table 4.

4l..

Local Dependency -

Green et al.

441

Overall, the model works considerably better for land use control attitudes than for support for local economic development activities. The variables most strongly related to support for land use controls are number of days spent in the seasonal home, education, and interest in non-lake issues: The more days seasonal residents spend at their recreational home, the less similar their attitudes are to fulltime residents regarding land use controls. Seasonal residents who spend more time in Forest County are more supportive of land use controls than those seasonal residents who spend a limited amount of time in the county. One possible reason for this finding is that, as they spend more time at their lake home, they develop relationships with other seasonal residents who support their attitudes toward land use controls. Another possible explanation is that, the more time they spend at their lake home, the more likely they are to emphasize the use-value of their home. Seasonal residents who become more interested in non-lake issues are more supportive of land use controls. These findings clearly support the local dependency thesis. The only variable that is statistically significant in the local economic development activity model is whether the seasonal resident feels welcome in non-lake activities in Forest County. Those who strongly agreed with this statement are more supportive of local economic development activities. Again, this provides further support for the local dependency thesis. Conclusions and discussion

Data from the focus group sessions and from the survey of seasonal and full-time residents in Forest County, Wisconsin, suggest the two groups differ significantly in their level of support for growth activities and land use (including zoning) controls. Based on growth machine and local dependency theory, we developed two sets of hypotheses explaining these attitudes-one focusing on class and the other on ties with permanent residents. Class appears to be an important factor in distinguishing support for land use and zoning attitudes within the full-time resident population, but it has a minimal effect among seasonal residents. Wealthier full-time residents appear more like the seasonal homeowners in their support for land use controls and zoning. However, class does not explain away the effects of the type of homeowner. We need additional research to explore the relationship between class and land use attitudes. Is the relationship a consequence of the fact that wealthier residents are more likely to have an opportunity to shape land use and zoning policies at the local level? Our hypotheses about local dependency gain some support from the analysis of seasonal homeowners. Time spent at their recrea-

442

Rural Sociology, Vol. 61, No.3, Fall 1996

tional home and interest in county-wide issues affect attitudes toward land use controls. The more time seasonal residents spend at their home and the more interest they have in county-wide issues, the more likely they are to support land use controls. These findings have several interesting implications for the community sociology literature. As rural communities are restructured and become less oriented toward production and more driven by consumption issues, land use and the consequences of growth become paramount issues (Marsden et al., 1993; Urry 1995). On the surface, these conflicts appear to be based on social class, with working class individuals supporting growth and professionals and managers favoring land use controls and environmental regulations. Cox and Mair's (1988) theory suggests that these positions are often the product of different levels of local dependence. Because seasonal residents are not well integrated into the larger recreational community, they maintain their positions regarding growth and development, and these positions are different from those of full-time residents. These seasonal residents had developed strong social ties with other recreational homeowners, and they frequently belonged to local organizations and associations established on the lake. Their lack of local dependence, and the development of a sense of community among lake residents, prevented them from developing attachments with permanent residents. Thus, these differences appear based primarily in the local dependency rather than in the individual characteristics of residents. These results suggest several possible directions for future research. First, to further examine the interaction between class and local dependency on attitudes toward land use controls and zoning, it may be important to examine other communities where the class position of seasonal residents is different than it is for those in Forest County. Because seasonal residents may have similar backgrounds in terms of occupation and education, we need to examine other communities with a similar level of recreational homes. We are planning to replicate this study in several Northwest Wisconsin counties, and this should provide a broader understanding of the interaction between class and local dependency. The class position and patterns of use of seasonal residents in these Northwest Wisconsin counties should be different from the seasonal homeowners in Forest County. Second, to further test the hypothesis about the interests of seasonal versus permanent residents, it would be interesting to assess the attitudes toward land use controls and local development activities in the home community of seasonal residents. If the local dependency thesis is correct, we should expect seasonal residents to have a different position regarding their permanent residence. Finally, these findings have some interesting policy implications. One issue concerns the local control of most land use decisions in

Local Dependency -

Green et al.

443

rural commumnes (Geisler and Martinson 1976; Huddleston and Krauskopf 1980). Because of the local nature of most land use decisions, the interests of recreational homeowners may not be represented in the policy process. In parts of the Upper Midwest, this exclusion is contributing to a revolt among seasonal residents. Recreational homeowners are organizing to challenge the assessments on their property and their lack of influence in the local political process. In their minds, it is truly a case of taxation without representation. References Allen, L.R., P. Long, R.P. Perdue, and S. Kieselbach. 1988. "The impact of tourism development on residents' perceptions of community life." Journal of Travel Research 27:16-21. Buttel, Frederick H. 1987. "New directions in environmental sociology." Annual Review of Sociology 13:465-88. Capek, Stella M., andJohn I. Gilderbloom. 1992. Community Ver.rus Commodity: Tenants and the American City. Albany, NY: State University of New York Press. Castells, Manuel. 1983. The City and the Grassroots: A Cross-Cultural Theory of Urban Social Movements. Berkeley, CA: University of California Press. Cox, Kevin R., and Andrew Mair. 1988. "Locality and community in the politics of local economic development." Annals of the Association of American Geographers 78:307-25. Davis,John Emmeus. 1991. Contested Ground: CollectiveAction and the Urban Neighborhood. Ithaca, NY: Cornell University Press. Deller, Steven. 1995. "Economic impact of retirement migration." Economic Development Quarterly 9:25-38. Gartner, William. 1987. "Environmental impacts of recreational home developments." Annals of Tourism Research 14:38-57. Gartner, William C. and D.E. Chappelle. 1988. "Segmentation analysis of property ownership types in rural areas." Land Development Studies 5:3-15. Geisler, Charles C. and Oscar B. Martinson. 1976. "Local control ofland use: profile of a problem." Land Economics 52:371--81. Girard, T.C. and William C. Garner. 1993. "Second home second view: host community perceptions." Annals of Tourism Research 20:685-700. Gottdiener, Mark, and Max Neiman. 1981. "Characteristics of support for local growth control." Urban Affairs Quarterly 17:55-73. Green, Gary P. 1996. "Community change in Harmony, Georgia, 1943-93." Paper presented at the annual meeting of the Rural Sociological Society, Des Moines, Iowa. Huddleston, Jack R., and Thomas M. Krauskopf. 1980. "Further evidence concerning local control of land use." Land Economics 56:471-76. Jordan,J.W. 1980. "The summer people and the natives: some effects of tourism in a Vermont vacation village." Annals of Tourism Research 7:34-55. Kasarda,John D. and MorrisJanowitz. 1974. "Community attachment in mass society." American Sociological Review 39:328-39. Krueger, Richard A 1994. Focus Groups: A Practical Guide for Applied Research. Thousand Oaks, CA: Sage Publications. Lew, A 1989. "Authenticity and sense of place in the tourism development experiences of older retail districts." Journal of Travel Research 27:15-22. Logan, John R. and Harvey L. Molotch, 1987. Urban Fortunes: The Political Economy ofPlace. Berkeley, CA: University of California Press.

444

Rural Sociology, Vol. 61, No.3, Fall 1996

Marsden, Terry, Jonathan Murdoch, Philip Lowe, Richard Munton, and Andrew Flynn. 1993. Constructing the Countryside. Boulder, CO: Westview Press. Perin, Constance. 1977. Everything in Its Place. Princeton, NJ: Princeton University Press. Peterson, Paul E. 1981. City Limits. Chicago: University of Chicago Press. Pfeffer, Max j., and Mark B. Lapping. 1994. "Farmland preservation, development rights and the theory of the growth machine: the views of planners." Journal of Rural Studies 10:233-48. Stack, Carol. 1974. AU Our Kin: Strategiesfor Survival in a Black Community. New York: Harper and Row. Stone, Clarence N. 1989. Regime Politics: Governing Atlanta, 1946-1988. Lawrence, KS: University Press of Kansas. Swanstrom, Todd. 1985. The Crisis of Growth Politics: Cleveland, Kucinich, and the Challenge of Urban Populism, Philadelphia, PA: Temple University Press. Urry,John. 1995. Consuming Places. New York: Routledge. Van Liere, Kent D., and Riley E. Dunlap. 1980. "The social bases of environmental concern: a review of hypotheses, explanations, and empirical evidence." Public Opinion Quarterly 44:181-97. Wilkinson, Kenneth P. 1991. The Community in Rural America. New York: Greenwood Press.

-

-

3 4

.034

.028

.212** -.245**

.068

.023

-.121" -.049

.008

-.073

-.025 .045

.153**

-.074

.005

-.019

.163** -.099*

.172** - .233**

-.008

.030

.092

-

8

.260*

-

10

-.052

.443*

.250** -.075

-.108

.060

-

9

* Correlation is significant at the 0.05 level

.103**

.031

-.176** -.074

-.254**

.199** -.127*

-

7

-

.432**

-

6

.337** - .177** - .308**

.128** -.256**

-

-.052

5

.351** -.014

-.423**

-.106** -.142** -.392**

.162** -.017

-.111**

.075

.184**

.053

.275**

-.149**

.234** -.214

-1.55** .404** .002 .010 -.047 -.027 .060 .022 -.067 .251** -.180** -.311** .080*

-.163**

-

2

** Correlation is significant at the 0.01 level (2-taiIed).

1. Support for Land Use Controls 2. Support for Growth Activities 3. Property Ownership 4. Gender 5. Age 6. Education 7. Household Income 8. Length of Residence 9. Days at Residence 10. At Home with Full-time Residents 11. Interest in Non-Lake Issues 12. Majority of Friends are Recreational Owners 13. Welcome in Non-Lake Activities (2-taiIed).

.291**

.087*

-

11

.106**

-

12

-

13

2.089

1.941

1.727

1.820

122.438

22.578

1.211

1.421

1.230

1.198

102.626

17.355

1.566

.500 .439 13.953 3.024

.496 .741 57.519 13.849 3.042

5.044

4.187

SO

20.832

8.621

Mean

APPENDIX Descriptive Statistics and Correlation Coefficients for Independent and Dependent Variables

4::..

~

l""'"

~

~

s~

I

~

s

s

~

~ ..... b

t--<