¹ See, for example, Kavanagh, Dennis, ‘Thatcher's Third Term’, Parliamentary Affairs, 41 (1988), 1–12.Google Scholar

² For a discussion of the ‘new urban left’ see Gyford, John, The Politics of Local Socialism (London: Allen and Unwin, 1985).Google Scholar

³ See Rhodes, R. A. W., Beyond Westminster and Whitehall (London: Hyman Unwin, 1988).Google Scholar

⁴ Rose, Richard, Do Parties Make a Difference? 2nd edn (London: Macmillan, 1984).CrossRefGoogle Scholar

⁵ For studies referring to the post-1974 period see Jackman, R. and Sellars, M., ‘Local Expenditure and Local Discretion’, Centre for Environmental Studies Review (05 1978), 63–72Google Scholar; Storey, D. J., ‘The Economics of Bureaux’, Applied Economics, 12 (1980), 223–34CrossRefGoogle Scholar; Jesson, D., Gray, J., Ranson, S. and Jones, B., ‘Some Determinants of Variations in Expenditure on Secondary Education’, Policy and Politics, 13 (1985), 359–91CrossRefGoogle Scholar; Boyne, George A., ‘Bureaucratic Power and Public Policies: A Test of the Rational Staff Maximisation Hypothesis’, Political Studies, 35 (1987), 79–104.CrossRefGoogle Scholar

⁶ Boaden, Noel T., Urban Policy Making (Cambridge: Cambridge University Press, 1970).Google Scholar For a review and discussion of early British outputs studies see Newton, K. and Sharpe, L. J., ‘Local Outputs Research: Some Reflections and Proposals’, Policy and Politics, 5 (1977), 61–82.CrossRefGoogle Scholar

⁷ See Hofferbert, R., ‘Comment on Morgan and Pelissero’, American Political Science Review, 75 (1981), 722–5CrossRefGoogle Scholar; Marquette, Jesse F. and Hinckley, Catherine, ‘Competition, Control and Spurious Covariation: A Longitudinal Assessment of State Spending’, American Journal of Political Science, 25 (1981), 362–75CrossRefGoogle Scholar; Jones, J. E., ‘Policy Change and Spending Shifts in the American States’, American Politics Quarterly, 2 (1974), 159–78CrossRefGoogle Scholar; Tucker, Harvey J., ‘The Nationalization of State Policy Revisited’, Western Political Quarterly, 37 (1984), 435–42.CrossRefGoogle Scholar The methodology used in this article was initially prompted by Tucker's work.

⁸ This method is used by George A. Boyne in ‘Bureaucratic Power’.

⁹ Van Meter, Donald S., ‘Alternative Methods of Measuring Change: What Difference Does It Make?’, Political Methodology, 1 (1974), 125–40Google Scholar; Hoggart, Keith, ‘Does Politics Matter? Redistributive Policies in English Cities, 1949–1974’, British Journal of Political Science, 17 (1987), 359–84, p. 366.CrossRefGoogle Scholar Whether this statistical relationship between change rates and base levels results from a catching-up process, with low tax and low spending localities catching up with others, or because of the mathematical effect of smaller denominators, is not at issue here.

¹⁰ Sharpe, L. J. and Newton, Kenneth, Does Politics Matter? (Oxford: Clarendon Press, 1984)Google Scholar; Hoggart, , ‘Does Politics Matter?’, p. 366.Google Scholar

¹¹ In tracing the differences between Labour and Conservative councils over time, Sharpe, and Newton, in Does Politics Matter?, pp. 198ffGoogle Scholar, compare changing mean levels of spending at four-year intervals by different types of county boroughs: those that remained Conservative throughout, those that remained Labour throughout, as well as those that experienced changes in party control. Among the problems of this analysis are that the mean spending levels are derived from variable numbers of authorities; sixty in 1961, diminishing gradually to thirty-four in 1973. The reason for this decline in cases is not made clear in the text. However, whatever the reason, this unsystematic treatment of missing data casts doubt on the validity of the findings. The twenty-six cases omitted in 1973 but included in 1961 might well be expected to affect the mean values calculated for the difthirty-four cases remaining throughout. Also, the significance of the differences in means between categories cannot be assessed because standard deviations are not given. Hoggart's reliance on statistical significance tests as a measure of the impact of party can be questioned on the grounds that with many hundreds of cases trivial and powerfully determining impacts alike may show up as significant.

¹² See Hofferbert, , ‘Comment’, p. 725.Google Scholar

¹³ Chartered Institute of Public Finance and Accountancy, Finance and General Statistics (London: CIPFA, annual). Formerly published as Return of Rates.Google Scholar

¹⁴ For a discussion of alternative methods, see Danziger, James N., Making Budgets (Beverly Hills: Sage, 1978).Google Scholar In our data set, changes in total spending expressed per capita and as a proportion of rateable value were strongly correlated (+0.91).

¹⁵ Because some of the higher spending and rating authorities failed to make their returns to CIPFA in time to be included in the publication on which our data is based there was a danger that our measures would be skewed by the sudden exclusion of these authorities in some years and their sudden inclusion following this. In order to avoid these problems, the mean values on which the scores are based were calculated using listwise deletion of missing values – if a particular authority has a missing value in one year, then that authority is excluded from the calculation of the means for all years.

¹⁶ Municipal Year Book (London: Municipal Services, annual).Google Scholar

¹⁷ For a discussion of the use of significance tests see Oakes, M., Statistical Inference: A Commentary for the Social and Behavioural Sciences (New York: John Wiley, 1986).Google Scholar

¹⁸ See Bunce, Valerie, Do Leaders Make a Difference? (Princeton, NJ: Princeton University Press, 1981).Google Scholar

¹⁹ Boaden, , Urban Policy Making, pp. 21–35.Google Scholar

²⁰ Boaden, , Urban Policy MakingGoogle Scholar; Davies, B., Social Needs and Resources in Local Authority Services (London: Michael Joseph, 1968).Google Scholar

²¹ Sources: OPCS, Local Authority Vital Statistics; Census 1971; Census 1981; CIPFA, Financial General and Rating Statistics.

²² See, for example, Clark, T. N. and Ferguson, L. C., City Money (New York: Columbia University Press, 1983), pp. 75ffGoogle Scholar; Boyne, , ‘Bureaucratic Power’, p. 99.Google Scholar

²³ See King, David, Fiscal Tiers: The Economics of Multi-Level Government (London: George Allen and Unwin, 1985).Google Scholar

²⁴ We cannot include changes in spending financed by balances as an independent variable predicting spending change in the same year. Decisions about the use of balances are not necessarily temporally and causally prior to spending and rating decisions; it is more likely that they are taken simultaneously. The inclusion of previous spending decisions allows us to retain a measure of the use of balances which avoids this problem.

²⁵ Stewart, J. D., Local Government: The Conditions of Local Choice (London: Allen and Unwin, 1983), 209–10.Google Scholar

²⁶ Used by, among others, Goldberg, Arthur S., ‘Discovering a Causal Pattern Among Data on Voting Behavior’, American Political Science Review, 60 (1966), 913–22.CrossRefGoogle Scholar

²⁷ As discussed in Ascher, Herbert B., Causal Modelling (Beverly Hills: Sage University Papers on Quantitative Applications in the Social Sciences, 1976).Google Scholar