¹ This literature is now quite extensive. I have found the following to be especially useful: Fisher, Gordon R., Effects of Monetary Policy on the United States Economy: A Survey of Econometric Evidence (Paris: Organization for Economic Co-operation and Development, 1972)Google Scholar; and Hansen, Bent, Fiscal Policy in Seven Countries (Paris: Organization for Economic Cooperation and Development, 1969)Google Scholar; and Cooper, J. Phillip, Development of the Monetary Sector, Prediction and Policy Analysis in the FRB-MIT-Penn Model (Lexington, Mass.: D. C. Heath, 1974).Google Scholar

² For useful discussions on the basics of these issues, see Johnson, Harry G., ‘Major Issues in Monetary and Fiscal Policies’Google Scholar, McCracken, Paul W., ‘Monetary versus Fiscal Policy’Google Scholar, and Wallich, Henry C., ‘Monetary and Fiscal Policy – Comparisons and Alternatives’, all printed in Entine, Alan C., ed., Monetary Economics: Readings (Belmont, Calif.: Wadsworth, 1968)Google Scholar; and Pierce, Lawrence C., The Politics of Fiscal Policy Formation (Pacific Palisades, Calif.: Goodyear, 1971).Google Scholar

³ See Boughton, James M. and Wicker, Elmus R., The Principles of Monetary Economics (Homewood, III.: Richard D. Irwin, 1975), Chap. 13Google Scholar; Laird, William E., ‘The Changing View on Debt Management’Google Scholar, in Entine, , Monetary EconomicsGoogle Scholar

⁴ Helpful discussions of incomes policy are found in Ulman, Lloyd and Flanagan, Robert J., Wage Restraint: A Study of Incomes Policies in Western Europe (Berkeley: University of California Press, 1971)Google Scholar; and Galenson, Walter, ed., Incomes Policy: What Can We Learn from Europe? (Ithaca, N.Y.: New York State School of Industrial and Labor Relations, Cornell University, 1973).Google Scholar

⁵ Boughton, and Wicker, , Principles of Monetary EconomicsGoogle Scholar; Barger, Harold, Money, Banking and Public Policy (Chicago: Rand McNally, 1968)Google Scholar; Simpson, Thomas D., Money, Banking, and Economic Analysis (Englewood Cliffs, N.J.: Prentice-Hall, 1976)Google Scholar; Kirschen, E. S. and associates, Economic Policy in Our Time (Amsterdam: North-Holland, 1964).Google Scholar

⁶ At least one study has indicated that, for the American case, approximately 85 per cent of open market sales and purchases are defensive. See Boughton, and Wicker, , Principles of Monetary Economics, p. 370.Google Scholar

⁷ Constraints on the setting of reserve requirements vary from country to country. Some are set by statute, as in the Netherlands, while others are set by custom, as in Britain. See Kirschen, and associates, Economic Policy in Our Time, Chap. IIIGoogle Scholar; individual country chapters in Bank for International Settlements, Eight European Central Banks (New York: Praeger, 1963)Google Scholar; and Sayers, R. S., ed., Banking in Western Europe (Oxford: Oxford University Press, 1962).Google Scholar

⁸ See individual country chapters in BIS, Eight European Central Banks.Google Scholar

⁹ BIS, Eight European Central Banks.Google Scholar

¹⁰ Quarterly economic and policy data were recorded from the International Monetary Fund's International Financial Statistics, 1950–1975, Washington, D.C.Google Scholar and the Statistical Office of the United States' Monthly Bulletin of Statistics, Volumes IXGoogle Scholar through XXIX. Data availability differs by country and thus dictated the following time periods for observation: (1) the Federal Republic of Germany: 1951–75, 1st quarter, ninety-seven observations: (2) the United Kingdom: 1951–74, 2nd quarter, ninety-five observations; (3) France: 1950–74. one hundred observations: (4)Sweden: 1959–74, sixty-four observations; (5) the Netherlands: 1959–75, 1st quarter, sixty-five observations; (6) Austria: 1959–75, 2nd quarter, sixty-six observations: and (7) Italy: 1953–75. 1st quarter, eighty-nine observations.

¹¹ Downs, Anthony, An Economic Theory of Democracy (New York: Harper and Row. 1957), Chap. 3.Google Scholar

¹² I have relied upon the following for specific dates of government change: Mackie, Thomas T. and Rose, Richard, The International Almanac of Electoral History (New York: Macmillan, 1974)Google Scholar; and The Statesman's Yearbook: Statistical and Historical Annual of the States of the World, 1945–1975 (London: St Martin's Press)Google Scholar. When a government changed during the middle of a quarterly period, that period was coded on the basis of which party held power for the majority of days during that quarter.

¹³ In fact, parties of the left may abandon this view altogether as they pursue the goal of national economic growth. This may be especially likely in Scandinavia where social democratic parties and in America where the Democratic party have been particularly concerned with stimulating economic growth with low interest rates. The most complete general discussions of these and other differences in party economic policies in Western nations are found in Kirschen, and associates, Economic Policy in Our Time, Part IIGoogle Scholar; and Martin, Andrew, The Politics of Economic Policy in the United States: A Tentative View from a Comparative Perspective, Sage Professional Paper in Comparative Politics (Beverly Hills, Calif.: Sage Publications, 1973).Google Scholar

¹⁴ For useful discussions and applications in economics of these types of linear models with dummy variables, see Pindyck, Robert S. and Rubinfeld, Daniel L., Econometric Models and Economic Forecasts (New York: McGraw-Hill, 1976)Google Scholar, Chap. 3; and Kmenta, Jan, Elements of Econometrics (New York: Macmillan, 1971).Google Scholar

¹⁵ Pindyck, and Rubinfeld, , Econometric Models and Economic ForecastsGoogle Scholar, and Kmenta, , Elements of EconometricsGoogle Scholar

¹⁶ Cochrane, D. and Orcutt, G. H., ‘Application of Least Squares Regressions to Relationships Containing Autocorrelated Error Terms’, Journal of the American Statistical Association, XLIV (1949), 32–61Google Scholar; Hildreth, G. and Lu, J. Y., ‘Demand Relations with Autocorrelated Disturbances’ (Michigan State University, Agricultural Experiment Station, Technical Bulletin 276, 1960).Google Scholar

¹⁷ Cooper, J. Phillip, ‘Asymptotic Covariance Matrix of Procedures for Linear Regression in the Presence of First Order Autoregressive Disturbances’, Econometrica, XL (1972), 305–10.CrossRefGoogle Scholar

¹⁸ Morrison, Denton E. and Henkel, Ramon E., The Significance Test Controversy (Chicago: Aldine, 1970).Google Scholar

¹⁹ This is a particularly difficult problem for which a precise solution is not apparent. Imagine the case in which discount rates were solely a function of the preferences of two alternative governments. Assume that a conservative government in power during a specific period held the rate at a constant value, m. Upon replacement of that government by a socialist government, the rate may be raised immediately to some value, m + n, and held at that constant value over the lifetime of that socialist government. A regression which conceived of the rate solely as a function of the government-type variable would correctly monitor that difference. The intercept would provide an estimate of m, the mean value of the discount rate under the conservative government; and the slope and intercept summed would provide the comparable estimate for the socialist government. The entry of the autoregressive term would destroy that estimate – even though it represented the ‘true’ effect in the real world, since the rate would be held at its previous value, once a new government took power. Yet, we may still be hesitant to drop the autoregressive term because it represents realistically a natural constraint on change. Note that the correction for autocorrelation will have the same destructive effect. Thus, I have provided two alternative estimates of that linear additive effect with the assumption that the true value must lie somewhere between.

²⁰ There are several indications of the poor performance of these models in the French case. In each case the estimate of ρ is at its maximum value of 1·0. In that extreme case, the estimation process is equivalent to first-differencing. Since ρ is estimated by transforming each variable in the model (e.g., the dependent variable is DR _t − ρDR _t−1), ρ = 1·0 is equivalent to first-differencing. Also, the intercept in such cases must be re-calculated by solving in the original equation measuring each variable about its mean – if the intercept is needed for some particular substantive reason. The extraordinarily large original intercepts for the French case, as they stand, are simply indications of time-trend effects. Furthermore, the trivial coefficients for unemployment and price instability and the large coefficient for the uncorrected effect of the auto-regressive term suggest that the process is simply an auto-regressive one with little or no sensitivity to these external effects. See Pindyck, and Rubinfeld, , Econometric Models and Economic Forecasts, Chap. 4.Google Scholar