A Regional Analysis of Employment Changes in Canadian Agriculture

Abstract

<p>The study investigates changes in the size and composition of the farm labour force in the regions of Canada over the period 1946 to 1973. The analysis of changes in farm employment is based upon an econometric model of the market for agriculture labour. The study departs from previous econometric analyses of the farm labour market in several respects. The model is disaggregated into three submodels, each pertaining to a distinct component of the farm workforce. Unlike other disaggregated models, the fundamental processes that determine levels of employment are not assumed to be the same for all types of farm labour. Instead, the submodels for farm operator employment, unpaid family employment, and hired farm labour have different conceptual bases and different structural forms, thus incorporating important elements of the heterogeneity of the farm labour force within the model. The derivation procedures also result in particular interdependence being specified among the components, a further extension of previous models.</p> <p>Empirical estimation of the farm employment model is undertaken at both the national and regional levels, and regional differences in the parameter estimates are analyzed using covariance procedures. This investigation of regional variation in the estimated employment relationships represents a notable development in the analysis of the market for agricultural labour.</p> <p>The results attest to the efficacy of using econometric models to analyze changes in farm employment, and demonstrate that such models also lend themselves to statistical examination of regional variation in the determinants of employment changes. The regional analysis shows that the effects of changes in economic variables on farm employment differ significantly among the regions of Canada. However, these differences are most frequently differences in the magnitude and speed of employment responses to changing economic stimuli, and several of the empirical relationships are relatively consistent among groups of regions.</p> <p>Changes in farm operator numbers are shown to be related to changes in agriculture technology and nonfarm income opportunities, though changes in the relative price of farm products seem to have little effect on operator employment in agriculture. In all regions, the number of unpaid family workers in agriculture is related to the number of farmers, but in Quebec, Ontario and the Prairies, unpaid family employment also varies with changes in the cost of hired labour. In these regions, the results imply that increases in farm wage rates encourage farm operators to employ available members of their families on the farm rather than hire additional paid labour. The results also indicate that in Quebec, Ontario, and the Prairies, the supply of hired farm workers would increase if returns of hired labour in agriculture were raised, as long as operators are willing and able to offer such increases. In these regions also, employment in all components of the farm workforce are shown to be related to wage levels and the availability of jobs in the nonfarm sector. In periods when the nonfarm economy is depressed, agriculture tends to retain surplus labour.</p> <p>It is suggested that the regional differences in the parameter estimates might reflect differences in the structure of agriculture, especially the relative importance of commercial family farms, or they may be related to differences in the degree of urban-industrial development among the regions. In any event, the farm employment model generally fits the data well in Ontario, has minor limitations in the Prairie regions and Quebec, but performs poorly in the Atlantic region and British Columbia.</p>