Farmland conservation policies typically use zoning and differentiated taxes to prevent urban development of farmland, but little is known about the effectiveness of these policies. This study adds to current knowledge by examining the impact of British Columbia's Agricultural Land Reserve (ALR), established in 1973, which severely restricts subdivision and nonagricultural uses for more than 4.7 million hectares of farmland. To determine the extent to which the ALR preserves farmland by reducing or removing the development option, a multilevel hedonic pricing model is used to estimate the impact of land use, geographic, and zoning characteristics on farmland value near the capital city of Victoria on Vancouver Island. Using sales data from 1974 through 2008, the model demonstrates a changing ALR impact over time that varies considerably by improved and unimproved land types. In 2008, landowners paid 19% less for the typical improved farmland parcel within the ALR versus that outside it. This suggests that would‐be developers expect permanency in the zoning law, and prefer non‐ALR zoned land. However, ALR land that is unimproved has a premium of 55%, suggesting that this land is more valuable for agriculture than for development. Farmland located closer to the city or the commuting highway commands a premium if it has a residence on it, with a residence also explaining why smaller agricultural properties sell at higher prices. However, it appears that zoning by itself is insufficient to protect farmland; other policies likely need to be implemented in conjunction with zoning to protect agricultural land.

The Agricultural Land Reserve (ALR) in British Columbia (BC), Canada, is a provincial zoning scheme designed to protect agricultural land from development. Since 1973, landowners have not been permitted to use ALR land for nonagricultural purposes, prompting some to seek recovery of development option value by applying for exclusion from the ALR. Using Geographic Information System (GIS) technology and a binary choice (logit) model, this study examines factors that impact the acceptance of ALR exclusion applications. With data from two regions in southwestern BC, we find that applications are more likely to be approved when the land is closer to the major highway, has a smaller parcel size, consists of a smaller portion of the total parcel area, or has poorer quality soils. Therefore, as intended by public policy, agricultural capability is a key determinant in exclusion decisions, retaining properties of greater agricultural value in the ALR. Even though public opinion has suggested otherwise, the political party in power at the time of the decision was only a weak factor, mostly moderated by the number of applications in a given year. A spatial‐effects model found no evidence of spatial autocorrelation, supporting the conclusions drawn from the nonspatial model estimations. La réserve de terres agricoles de la Colombie‐Britannique (ALR), au Canada, est un plan de zonage provincial destinéà protéger les terres agricoles contre le développement urbain. Depuis 1973, les propriétaires fonciers ne peuvent utiliser les terres de la réserve à des fins non agricoles, ce qui incite certains à tenter de récupérer une certaine valeur d’option en faisant une demande d’exclusion. À l’aide de la technologie des systèmes d’information géographique (SIG) et d’un modèle de choix binaire (type logit), nous avons examiné les facteurs qui influencent l’acceptation des demandes d’exclusion. À l’aide de données sur deux régions du sud‐ouest de la Colombie‐Britannique, nous avons observé que les demandes d’exclusion sont plus susceptibles d’être acceptées lorsque les terres sont situées en bordure d’une route importante, sont de petite taille, ne représentent qu’une portion d’une superficie plus grande ou présentent des sols de mauvaise qualité. En conséquence, comme le prévoit la politique du gouvernement, la capacité agricole est un facteur clé dans les décisions d’exclusion qui permet de conserver les propriétés de grande valeur agricole dans la réserve. Bien que l’opinion publique indique le contraire, le parti politique au pouvoir au moment de la décision ne constituait qu’un faible facteur, principalement réduit par le nombre de demandes au cours d’une année donnée. Un modèle d’effets spatiaux n’a pas réussi à montrer l’existence d’une autocorrélation spatiale appuyant les conclusions tirées des estimations du modèle non spatial.

Specification uncertainty arises in spatial hedonic pricing models because economic theory provides no guide in choosing the spatial weighting matrix and explanatory variables. Our objective in this paper is to investigate whether we can resolve uncertainty in the application of a spatial hedonic pricing model. We employ Bayesian Model Averaging in combination with Markov Chain, Monte Carlo Model Composition. The proposed methodology provides inclusion probabilities for explanatory variables and weighting matrices. These probabilities provide a clear indication of which explanatory variables and weighting matrices are most relevant, but they are case specific.

A survey of consumers at three farmers' markets (FMs) was done near Vancouver, British Columbia. The markets span urban and suburb locales, and the survey's 234 respondents were asked questions about shopping behavior, attitudes toward FMs, and demographic information. The focus of the analysis is on the differences between regulars and non-regulars to the market, where a regular is considered a shopper who shops weekly or bi-weekly. The results show that regulars spend more ($46.36 vs. 33.19 for non-regulars), are much more likely to expect higher prices compared to grocery stores than non-regulars, and buy more products (4.15 vs. 3.1). Regulars also value attributes of FMs differently: they value variety, organic products, and being locally-grown more highly. Organic purchasing behavior is also significantly different with regulars much more likely to say they “always” or “usually” buy organic products. As this is the first study to explicitly analyze regulars at FMs, suggested research directions and methods are offered to help guide future research.

No abstract