Liquidity preferences of commercial banks : the Canadian case

In conjunction with the recent interest in the liquidity preferences of commercial banks, which is itself part of a new supply theory of money, this thesis investigates the reserve behaviour of Canadian commercial banks from 1920-1939. Several models of bank reserve behaviour are presented including the one to be tested in this thesis. This model differs from the others in that it will be tested with monthly data on individual banks (and hence can explain differences among banks as to the holding of reserves) whereas the others were tested with annual data for a group of banks or a banking system. Since there was no required reserve in Canada prior to March, 1935, there was also no definition of what constituted reserves. This problem had to be investigated before any reserve ratios could be calculated. After calculating some reserve ratios, several interesting observations can be made. The hypothesis that Canadian, commercial banks adhered to a ten per cent required reserve ratio through a gentlemen's agreement within the Canadian Bankers Association was clearly refuted. Also the effect of the establishment of the .Bank of Canada on reserve holdings was noticed. Furthermore, the evidence cast some doubt on the conclusions of George Morrison in his book, Liquidity Preferences of Commercial Banks. The model presented previously was then tested both with monthly data for individual banks and with monthly data for the banking system as a whole. The tests using monthly data for the individual banks indicated the need for further refinement of the model although considering the number of observations and the diversity among banks perhaps the R² was not that bad.. The R² was much improved when monthly data for the banking system was used. This is to be expected as aggregations over banks hides much detail which, thus, does not have to be explained.Arts, Faculty ofVancouver School of EconomicsGraduat