¹ See, for example, Dye, Thomas R., Politics, Economics and the Public: Policy Outcomes in the American States (Chicago: Rand McNally, 1966)Google Scholar; Fisher, Glenn W., “Interstate Variation in State and Local Government Expenditures,” National Tax Journal, 17 (03, 1964), 57–64Google Scholar; Sachs, Seymour and Harris, Robert, “The Determinants of State and Local Government Expenditures and Intergovernmental Flow of Funds,” National Tax Journal, 17 (03, 1964), 75–85Google Scholar; Morss, Elliott R., Fredland, J. Eric and Hymans, Saul H., “Fluctuations in State Expenditures: An Econometric Analysis,” Southern Economic Journal (April, 1967)CrossRefGoogle Scholar; Dawson, Richard E. and Robinson, James A., “Interparty Competition, Economic Variables, and Welfare Politics in the American States,” Journal of Politics, 25 (May, 1963), 265–289CrossRefGoogle Scholar; and Sharkansky, Ira, “Economic and Political Correlates of State Government Expenditures: General Tendencies and Deviant Cases,” Midwest Journal of Political Science, 11 (May, 1967), 173–192CrossRefGoogle Scholar.

² Most state governments have several administrative units with educational responsibilities, for example, and their titles and program responsibilities vary considerably from one state to another.

³ Davis, Otto A., Dempster, M.A.H., and Wildavsky, Aaron, “A Theory of the Budgetary Process,” this Review, 60 (09, 1968), 529–547Google Scholar.

⁴ Fenno, Richard F. Jr., The Power of the Purse: Appropriations Politics in Congress (Boston: Little Brown, 1966)Google Scholar, Chapters 8 and 11.

⁵ Lindblom, Charles E. “Decision-Making in Taxation and Expenditure,” in Public Finances: Needs, Sources and Utilization (Princeton: National Bureau of Economic Research, 1961), pp. 295–333Google Scholar.

⁶ The states included in this study—and the nature of the findings—may be affected by the requirement of separate indications for the agency's budget request and the governor's recommendation. One tenet of certain administrative reformers prescribes that public agencies not submit their own budget requests to the legislature. Instead, they should submit their requests to the chief executive, who should then review each agency's budget in the light of his whole program, and submit his own recommendations for each agency to the legislature. Presumably the legislature will consider only these budget recommendations of the executive, and thereby force each agency to operate within the budget limits set by the leader of the “administration team.” A number of states have adopted these budget procedures, and their budgets show only the governor's recommendation for each agency. Our sample, in contrast, may be overly representative of states with “unreformed” budget procedures.

The “budget period” employed is the fiscal year or the biennium, depending upon the usage prevailing within each state. Although the use of a single year or a biennium varies from one state to another, this should not have a material effect on the findings insofar as the principal findings grow out of correlation coefficients that are calculated separately for each state. The particular years employed in the analysis of each state are indicated with the name of each state in Table 1. In each case, the most recent budget period was employed for which complete data was available. A small number of additional states publish budgetary information that would qualify them for this type of analysis. They are not included here, however, because the appropriate documents were not available.

⁷ In some cases it is a budget review board, rather than the governor per se, that is directly responsible for the recommendation to the legislature. In each case, however, it is reported by observers native to the state that the governor's role is critical in the recommendations.

⁸ All of the findings reported below for the within-state analysis of agency budgets reflects data about the “general fund” for those states that segregate the “general” fund from the “total” fund. For most states, the designation of “general” fund pertains to those moneys subject to state legislative appropriation. For those states showing both general and total fund accounts, the correlation analyses were performed twice: once with each account. However, the findings from the two analyses were not so different as to warrant separate reporting here. For each state, agencies requesting at least $500,000 were included. These are considered the “major” agencies. Supplementary appropriations are excluded from the legislature's appropriation due to lack of data. In the case of one state for which supplementary appropriations were available, the within-state analysis of agencies was run with and without the supplementals; the differences in findings appear inconsequential.

⁹ Cf. Anton, Thomas J., The Politics of State Expenditure in Illinois (Urbana: University of Illinois Press, 1966)Google Scholar, Chapters 5, 6.

¹⁰ Jewell, Malcolm E. and Patterson, Samuel C., The Legislative Process in the United States (New York: Random House, 1966), p. 251 ff.Google Scholar

¹¹ The data used in this study meet two of the primary assumptions of correlation analysis: the distribution of each variable approximates normality; and all two-variable relationships are linear.

¹² Several other variables were considered, but excluded in this report because of their failure to add anything of importance to the findings reported below. Those excluded are:

a. Schlesinger's index of the Governor's formal budget powers

b. Schlesinger's index of the Governor's formal powers of appointment

c. Schlesinger's aggregate index of the Governor's formal powers

d. Ranney's measure of Democratic party strength

e. biennial compensation paid to members of the state Legislature

f. state expenditures per capita for personal services

g. total population of the state

h. number of state government departments with elected heads.

¹³ Schlesinger, Joseph A., “The Politics of the Executive,” in Jacob, Herbert and Vines, Kenneth N., Politics in the American States (Boston: Little, Brown and Company, 1965), p. 229Google Scholar.

¹⁴ Loc. cit.

¹⁵ Austin Ranney, “Parties in State Politics,” in Jacob and Vines, op. cit., p. 65. The measure of competition used here equals the difference between each state's score on Ranney's index of Democratic Party strength and .5000, with the result inverted so that high scores indicate high interparty competition.

¹⁶ The data pertain to the 1964 election for U.S. Representative.

¹⁷ The figure used is total general state government expenditures per capita in 1966.

¹⁸ See the works of Dye and Dawson and Robinson, cited in note 1.

¹⁹ The concept of “slack” is explained in Cyert, Richard M. and March, James G., A Behavioral Theory of the Firm (Englewood Cliffs: Prentice-Hall, 1963), pp. 36 ff.Google Scholar

²⁰ A test for significance is not, strictly speaking, applicable because the units of analysis are not chosen at random to represent a larger population. Nevertheless, the tests for significance provide a convenient device to denote relationships that are “sizeable.”

²¹ The techniques are explained in Blalock, Hubert M. Jr., Causal Inferences in Nonexperimental Research (Chapel Hill: University of North Carolina Press, 1964), pp. 64 ff.Google Scholar; and demonstrated in McCrone, Donald J. and Cnudde, Charles F., “Toward a Communications Theory of Democratic Political Development: A Causal Model,” this Review, 61 (03, 1967), 72–79Google Scholar. In this causal analysis, the author is following the convention of inferring a linkage to be absent if the difference between expected and actual findings are less than the difference for the alternative model, and if the difference between expected and actual findings is less than .10.

²² The correlations indicate that high scores on veto power and expenditures correspond with high negative relationships between agency acquisitiveness and the governor's short-term support.

²³ The correlations indicate that low scores on debt, but high scores on party competition, voter turnout, and per capita personal income correspond with high positive relationships between agency acquisitiveness and the governor's support for budget expansion.

²⁴ The correlations indicate that high state expenditures and a low number of elected officials correspond with high negative relationships between agency acquisitiveness and short-term success in the legislature.

²⁵ The correlations indicate that low expenditures and debt correspond with high positive relationships between agency acquisitiveness and budget-expansion success in the legislature.

²⁶ The correlations indicate that high tenure potential and expenditures, but a low number of elected officials correspond with high positive relationships between the governor's short-term recommendation and the legislature's short-term appropriation.

²⁷ The correlations indicate that high tenure potential and low debt correspond with high positive relationships between the governor's recommendation and the legislature's appropriation for budget expansion.

²⁸ An examination of simple correlation coefficients among variables S₁—S₈ suggests that the relationships which appear most strongly in Table 5 are independent. Except for the expected relationships among turnout, competition and personal income, few of the variables show strong relationships with each other. Thus, the relationships in Table 5 involving the governor's veto, and tenure potential, and total per capita state expenditures and debt are not the product of any of the other independent variables considered in this study. The coefficients of simple correlation among the independent variables are reported here:

* Significant at the .05 level (see note #20).

A test with regression analysis lends support to the importance of many elected officials for the legislature's short-term support of acquisitive agencies and the importance of few elected officials for governor-legislative rapport on shortterm appropriations; this independent variable (S₄) shows a regression coefficient of at least 1.5 times the standard error with the appropriate dependent variables, in the presence of all other independent variables.

²⁹ See, for example, MacMahon, Arthur, “Congressional Oversight of Administration: The Power of the Purse,” Political Science Quarterly, 58 (June and September, 1943), 161–190CrossRefGoogle Scholar; and 380–414; Schilling, Warner, “The Politics of National Defense: Fiscal 1950,” in Schilling, et al, Strategy, Politics and Defense Budgets (New York: Columbia University Press, 1962), pp. 1–266Google Scholar; Huzar, Elias, The Purse and the Sword (Ithaca; Cornell University Press, 1950)Google Scholar; Wildavsky, Aaron, The Politics of the Budgetary Process (Boston: Little, Brown and Company, 1964)Google Scholar; Rufus Browning, “Innovative and Non-Innovative Decision Processes in Government Budgeting,” a paper delivered at the Annual Meeting of the American Political Science Association, September, 1963; Richard Fenno, The Power of the Purse op. cit.; John P. Crecine, “A Computer Simulation Model of Municipal Resource Allocation,” a paper presented at the 1966 Midwest Political Science Association Meeting.

³⁰ See Malcolm E. Jewell and Samuel C. Patterson, op. cit., p. 251 ff.

³¹ See the works by Dye, Dawson and Robinson and Sharkansky cited in note 1; plus Hofferbert, Richard I., “The Relation between Public Policy and Some Structural and Environmental Variables in the American States,” this Review, 60 (03, 1966), 73–82Google Scholar.