This thesis addresses the issue of the existence of a causal relationship between the change in the food component of the consumer price index and change in the money supply (money base and M1) by using monthly Canadian data from January 1968 to March 1997. A bivariate vector autoregressive (VAR) model is used to describe the relationship between money supply and food price on the basis of identifying the two conditions that the three series are first difference stationary series and the money stock and price series are not cointegrated. Saunders' lag order selection criterion developed from Akaikes final prediction error (FPE), and Hsiao's lag selection procedure are used to identify the order of lags of each variable in VAR model. The models perform well through all diagnostic checks. All hypotheses are tested by using the likelihood ratio statistic and the chi2-statistic. The Granger test of causality is discussed and implemented in this research. By comparing the causality results of the seasonally adjusted series with those of seasonally unadjusted series we find that they are very sensitive to the seasonal adjustment of the series. For the seasonally adjusted series there is no relationship between money base (MB) and the food component of the consumer price index (FCPI) and there is a unidirectional causal relationship between M1 and FCPI (from M1 to FCPI). But for the seasonally unadjusted series there is a feedback relationship between M1 and FCPI and there is a unidirectional flow from MB to FCPI. The difference of the results is consistent with the findings of Sargent and other writers.

This thesis studies fourteen Canadian processed food industries and their pricing behaviour. Pricing models for each industry for the period of 1971-1977 and 1978-1984 have been established. This study also tests wether changes in a pricing behaviour occurred in the middle of 1970s in which shifts in Canada-U.S. exchange rate occured. Food prices change for several reasons. The main reasons for changes in processed food prices are expected to be changes in input costs and demand factors. Input costs consist of material, labour, capital and fuel cost. Changes in demand side - import competition and excess demand - are are important factors. This study attepmts to establish, identify, and analyze pricing models by employing such variables for fourteen Canadian processed food industries at the wholesale level. Karikari (1988) has shown that the Canadian manufacturing industries changed their pricing behaviour as the U.S.-Canada exchange rate shifted in the middle of the 1970s. This study also tests if the changes (shift) in pricing behaviour of the food processing industries took place between two sub-periods: pre-depreciation of U.S.-Canada exchange rate (1971 to 1977), and post-depreciation of U.S.-Canada exchange rate (1977 to 1984). After analyzing the characteristics of the Canadian food processing industries and the distribution channel, three economic theories - which are considered to be appropropriate in reflecting the characteristics and the pricing behsviour - have been discussed. The Mark-up Pricing Theory is employed to explain the food processors' oligopolistic pricing behaviour. From the Mark-up Pricing Theory, relative changes in mark-up, material price, labour price, energy price, capital price, and productivity of each input are derived as independent variables in the pricing model while change in industrial selling price of processed foods is shown as a dependent variable. Excess demand and import competition are the main sources for the fluctuations in the mark-up factor. The Bilateral Monopoly Theory is applied to explain bargaining processes, from which prices of processed foods are determined, between processors and retailers. A shipment variable has been derived from the Bilateral Monopoly Theory as one of the substitutes for the mark-up variable. An International Trade Theory is discussed for the industries that face import competition. From this theory, it is concluded that import price would also influence Canadian food processors' markup. Also discussed is a theory on how the pricing behaviour would change in a situation in which shifts in exchange rates occur. Quarterly data in rate of changes form are used for the estimation of the pricing model. Lags are allowed for independent variables to proferly reflect the characteristics of food processors. First, assuming changes in pricing behaviour, the pricing model is regressed for each industry in each sub-period, respectively. Variables for each industry in each sub-period are selected. It seems that the finalized regression results indicate a possibility of changes in pricing behaviour. A statistical test incorporating dummy variables is used to check if the changes in pricing behaviour which occurred in the middle of 1977 are statistically significant. The results can be summarized as follows. Different variables and different lags fit for each industry in each sub-period. The material prices-in different lag forms - are the main factors that influence changes in the industry selling price. In some industries in a certain period, the material prices are not important at all; only the U.S. prices are shown as important factors. The wage - current or lagged - is an important variable in some industries (at least in one period). The shipment variables are important in most industries with a positive or a negative sign, indicating the food processors' monopolistic pricing behaviour is influenced or interupted by the foods retailers' behaviour. The U.S. price variable(s) is a significant factor in most industries. The statistical test indicates that most of the industries have experienced structural changes and/or model changes between the two periods, except poultry, sugar cane & beet, vegetable oil, brewery, and winery industries. This study, however, does not necessarily conclude that the Canadian processed foods industries' pricing behaviour was changed according to the Karikari's hypothesis. / Land and Food Systems, Faculty of / Graduate

No description available.

No description available.

No description available.

This study is concerned with- the effects of changes in industrial concentration on average costs of production in 17 Canadian 4—digit food manufacturing industries over the period 1961-1982. The model employed is a dual Translog cost function adapted to include a concentration variable (Herfindahl index) and technical change, and is estimated using pooling techniques to allow simultaneous analysis of all 17 industries. The results indicate that there was a significant relationship between concentration and average costs for this sample. In particular, there appears to have been a decrease in average costs for low-concentration industries as concentration increased, ceteris paribus, while in high-concentration industries, increases in concentration led to increases in costs. Concentration changes have also had an effect on the relative shares of factors of production for these industries. An evaluation of employment effects across industries indicates that the benefits in efficiency due to increases in concentration in low-concentration industries must be weighed against apparent decreases in the overall employment (of labour) for these industries. Alternatively, the efficiency losses in high-concentration industries appear to have been offset by increases in overall employment as concentration has increased. Thus, depending on the criterion used, relative concentration effects may have been beneficial or detremental to social welfare; the outcome is not unequivocal. / Land and Food Systems, Faculty of / Graduate