@article{Implications:4242, recid = {4242}, author = {Scotland, Francis}, title = {Investment: A Survey of Models with Some Implications for the Effects of Monetary Policy}, publisher = {Bank of Canada}, address = {1981}, pages = {1 online resource (ix, 54 pages)}, abstract = {Investment is one of several linkages through which monetary policy may influence the economy. The nature of the presumed linkage has been defined in various theories of investment, each apparently emphasizing a different channel of influence. It is useful, therefore, to understand the differences among the theories by assessing their strengths and weaknesses from both theoretical and empirical points of view, in order to arrive at a clearer understanding of the issues involved in determining the impact of monetary policy on aggregate demand. The four models of investment surveyed in this report, chosen on the basis of popularity and vintage, are the accelerator, the cash flow, the securities valuation or Tobin's q, and the standard neoclassical models. Briefly, the accelerator model attributes variations in net investment to movements in the demand for output and disregards direct sensitivity to monetary or fiscal parameters. The cash flow model points to corporate profits and depreciation allowances as the major driving force behind investment demand. The securities valuation or Tobin's q model indicates that the driving force of investment is the market value of the real capital stock (as determined in the securities market) relative to its replacement value. In this model, monetary factors affect investment via their effect on the valuation of capital in the stock market. The standard neoclassical model suggests two possible investment specifications, (a) as a function of output and relative factor prices, and (b) as a function of output and the relative prices of output and capital. Under specific conditions the two specifications are theoretically equivalent. In this model the implicit rental price of capital is defined to be a function of a number of policy levers, including the discount rate, thus providing a direct linkage to monetary policy. These investment models are surveyed in a fashion similar to U.S. studies by Charles Bischoff and Peter Clark; that is, they are compared on the basis of motivating theory, estimation properties, intra- and extra-sample simulation performance and their implications for the effects of monetary policy. The empirical findings of the survey indicate that there is not much difference among the models when they are assessed on their ability to track investment. When their theoretical properties are compared, however, the standard neoclassical model is found to be the most appropriate framework for determining the effects of monetary policy on investment as well as for future research. The survey does not explore at least two important issues: the effect on investment of inflation and uncertainty and the impact of the rise in energy prices. These issues represent, in themselves, major research projects.}, url = {http://www.oar-rao.bank-banque-canada.ca/record/4242}, doi = {https://doi.org/10.34989/tr-29}, }