The end of executive dominance in state appropriations.
Beginning in the 1970s, state legislatures underwent a resurgence as they became more professional (Moe, 1988, 12). In a survey of state legislative and executive budget officials in 1982, Abney and Lauth (1987) found that gubernatorial dominance in the appropriations process was not as pronounced as might be expected. In that study, 52 of the officials surveyed reported the governor to be more influential compared to 38 who cited the legislature as more influential. In 14 states both legislative and executive officials identified the governor as more influential compared to nine states in which both officials said the legislature was more influential. In noting the surprising influence of the legislature, they cited the observation of Rosenthal (1981, 206) that legislatures came to the budgetary process far better prepared than in earlier years.
Based on the results of a survey of state executive and legislative officials conducted in 1994, we find that this trend has continued and that gubernatorial dominance appears to be over. Altogether, in 1994 we received responses from 99 of the 100 chief executive and legislative budget officials; only the chief executive official in Massachusetts declined to participate. Of our 99 respondents, 36 indicated that the governor is more influential in the appropriations process, 32 cited the legislature as more influential, and 31 stated that they are about equal. Surprisingly, executive officials were not particularly more likely to perceive the governor as more influential than were legislative respondents (see Table 1). The similarity of the perceptions in regard to relative gubernatorial influence supports our conclusion that the dominance of the governor in the appropriations process documented at mid-century has ended.
Table 1 Perceptions of Relative Gubernatorial Influence by Type of Budget Official
Executive Legislative All Officers Officers Officers Relative Influence (N = 49) (N = 50) (N = 99) Governor is more influential 19 17 36 Influence is equal 15 16 31 Legislature is more influential 15 17 32
The respondents were asked the following question: In comparing the influence of the governor and the legislature, who has the greater influence? (A) The governor has greater influence; (B) the influence of each is about the same; or (C) the legislature has greater influence.
As seen in Table 2, combining the perceptions of the two respondents from each state gives further evidence of this parity. When we averaged the observations of both officials from the same state, the two officials in 22 states either leaned toward the governor or both cited the governor as more influential. Officials in 20 states either leaned toward the legislature or both cited the legislature as more influential. Of the 21 states where both officials agreed as to relative ranking, in 10 states the respondents said the governor was dominant, and in 9 states both respondents cited the legislature as being dominant.
Table 2 Perceptions of Relative Influence in State Appropriations by State
Perceptions of Both Respondents by State Both cited Lean to Divided to Lean to Both cited governor governor legislature legislature legislature HI AR IA CT AL IN CA MI ID AK KY DE MN ME AZ LA GA NH MT CO MD IL ND NE FL NJ KS OH NM MA PA MO WA NC MS TN NV WV OR OK VT NY SC TX WI RI UT SD WY VA
The perceptions of the two respondents from each state were averaged. Massachusetts had only one respondent. Where a state is designated as leaning to the governor or legislature, one respondent cited the influence of the two as equal, and the other respondent cited the governor or legislature as more influential.
These changes in relative influence have not gone unnoticed. In his Governors and Legislatures, Rosenthal (1990, 201) described the changes in the competition between the two branches as follows:
What is new about the institutional bases of conflict
is the sense of independence and the assertiveness
of legislatures nowadays, and the defensive
posture assumed by governors. Legislatures are
breaching the invisible boundary between the two
branches; governors are doing their utmost to
draw the constitutional line and repel such incursions.
The National Governors Association notes
that its members have been expressing strong concern
about the intrusion of the legislature into
Reasons for the Decline
We believe that the end of gubernatorial dominance in the state appropriations process has resulted from at least four factors. All of the factors relate to a reduction in the advantage given to governors by the executive budget. First, legislative reforms have resulted in governors' losing the ability to control the appropriations agenda--an advantage that the executive budget had given them. Second, the item veto has not generally given governors the ability to protect the executive budget from legislative encroachment. Third, the rise of greater party division between the branches in state government has increased legislative resistance to gubernatorial leadership. Fourth, states have generally not adopted reforms guaranteeing the significance and independence of the executive budget.
We lack longitudinal data to document the decline in gubernatorial influence or to relate the decline to the above factors. We can show that relative gubernatorial influence is related to the above factors. Also, we know that these factors have either changed during this period or have not developed in a manner so as to protect gubernatorial advantage.
Proposition 1. Governors have lost the ability to control the appropriations agenda.
Beginning in 1916, executive budget reform swept the nation. By the 1950s, most of the states had an executive budget in place. This power to assemble and propose an executive budget gave governors a distinct advantage over the legislature, especially in regard to information and ability to define the legislative agenda. The importance of the executive budget in maintaining gubernatorial influence continues. When our respondents were asked to indicate the factors affecting gubernatorial influence, 84 of the 99 budget officials cited the executive budget as an important factor; half cited it as the most important factor.
The executive budget gave governors an informational advantage. In the last 40 years, however, the increased capabilities of the legislative branch have offset that advantage (Balutis and Butler, 1975; Rosenthal, 1990, 140). Only 11 legislatures had a budgetary staff in 1955 (Council of State Governments, 1956). Today, all of the state legislatures have budgetary staff. This change has undermined gubernatorial domination of information. The roles of the governor as chief executive officer and in many respects as chief budget officer may always place legislatures at a disadvantage in information. Yet, in 29 of the 50 states legislative budget officials indicated that legislatures had independence in information from the governor; 47 of the 99 respondents indicated such independence. Specifically, the respondents agreed with a statement characterizing the appropriation process in the legislature in the following manner: "The legislature is largely independent of the governor in regard to budgetary information."
This legislative capability has undermined the informational advantage of the governor and is probably a factor in the declining influence of governors. Our data indicate that those respondents citing legislative independence were more likely to perceive legislative bodies as more influential. Of the 47 respondents citing legislative independence in information, 19 (41 percent) said the legislature is more influential compared to a similar response by 13 of the 52 respondents (25 percent) not reporting such independence. When given a list of factors affecting gubernatorial influence in the appropriations process, only 30 percent of the respondents cited gubernatorial superiority in information as a factor. Of the respondents citing such superiority as a factor, 63 percent indicated the governor to be more influential compared to 25 percent of the other respondents. Informational superiority matters, but the bias is much less in favor of governors now than at mid-century.
Proposition 2. The failure of the item veto to protect the executive budget from legislative control has exacerbated the governor's difficulty in the appropriations process.
Contemporary proponents of the item veto cite its ability to reduce governmental deficits and curtail pork barrel spending. However, the creators of this reform intended to use it to strengthen the executive's budgetary role in dealing with the legislature (Wells, 1928). Framers of the constitution for the Confederate States of America wanted to institute aspects of the parliamentary system and yet maintain the separation of powers. In particular, they wanted to enhance the role of the executive in the appropriations process by giving the president the power to put together an executive budget and an item veto to defend the budget. The framers believed that the ordinary executive veto had become a useless instrument for appropriations bills because such legislation contained a number of subjects that made it difficult to use the general veto. Allowing the president to delete particular items would guarantee the president's role in the appropriations process.
Although the Confederacy did not last long enough to test the effectiveness of the item veto in promoting executive influence in the appropriations process, the experience of the states raises questions about the underlying assumptions of the framers of its constitution. In many states, the item veto has not protected the executive budget, even though many practitioners of state government have often praised its effectiveness. When we asked our respondents whether the item veto was an important instrument aiding the influence of the governor in the appropriations process, 58 of the 99 agreed that it was. Furthermore, surveys of governors support these observations (Moore 1992; Abney and Lauth, 1994). However, the significance of the item veto in affecting gubernatorial influence appears small. Only 19 of the 58 budget officials who said that the item veto is an important factor in affecting gubernatorial influence in the appropriations process also said that the governor is more influential while 13 of the 58 said that the legislature is more influential. When we compare respondents from the 42 states that have the line-item veto with those from states that lack it, the former do not perceive the governor as more influential as would be predicted.(1) Of the 83 respondents from the item-veto states, 28 (34 percent) ranked the governor as more influential in the appropriations process as compared to 6 of the 14 respondents from the other states (43 percent). The presence of the item veto is not positively related to perception of gubernatorial influence.
The limitations of the item veto are evident. By using lump-sum appropriations in spending bills rather than detailed itemizations, legislative bodies tend to nullify the effectiveness of the item veto.(2) Only 31 of the 83 respondents from states with the item veto indicated that detailed itemization is used in appropriations bills. In most states with the item veto, governors are unable to reduce appropriations; they must veto the whole item. If the item includes money for matters of concern to the governor, the value of the veto is diminished.
In interpreting the meaning of item-veto power, courts have constricted it (Briffault, 1993). Despite efforts by a number of governors to use the item veto to reduce appropriations, in only 12 states with the item veto have courts approved the power of reduction. When we asked our respondents from executive budget agencies if the governors needed the ability to reduce items of appropriations, 32 of the 41 from states with an item veto responded positively. Given the presence of lump-sum appropriations and the absence of the reduction power, the item veto is not a corrective to the failure of the ordinary executive veto envisioned by reformers.
Respondents from states where governors have the reduction power are clearly more likely to perceive the governor as more influential than the legislature. Fourteen of the 23 respondents (61 percent) from states having governors with the reduction power regarded the governor as more influential compared to 14 of the 60 respondents (23 percent) from other item veto states. Arguably, without the reduction power in 12 states, our results would reflect greater legislative influence than gubernatorial.
A second factor that has restricted the effectiveness of the item veto is the use of narrative in appropriations bills for the purpose of placing conditions on how funds should be spent. In many of the states with the item veto, the governor may not separately veto the narrative while sustaining the monetary item.(3) In these states, the courts have viewed a veto of narrative without a veto of the item of appropriation as an affirmative act of legislation rather than the negative act of a veto. That is, in deleting such narrative, the governor is changing the content of legislation rather than nullifying legislation.
In those states where governors may delete conditions, provisos, or substantive legislation, gubernatorial influence benefits. For respondents from states that have the item veto but lack the reduction power, gubernatorial authority to delete any narrative is related positively to perceptions of gubernatorial influence; 15 percent of the respondents with governors lacking the narrative deletion authority cited the governor as more influential (N = 34), while 35 percent with governors having the deletion authority reported the governor as more influential (N = 26).
Having the authority to delete narrative may not solve the governor's problem. We asked our respondents how the legislature communicated its intent about spending. While narrative in appropriations bills was the most common method, 47 percent cited committee or conference reports, and 42 percent cited personal communications from legislative leaders and staff. Both of these methods are "veto proof."
In conclusion, the item veto may have a significant impact on the relative influence of governors, but not in the manner normally expected. The item veto has largely failed to protect the governor's executive budget. However, where the reduction and deletion power exists, the item veto correlates positively with gubernatorial influence.
Proposition 3. Increasing partisanship in state government has undermined gubernatorial influence in the appropriations process.
In the last four decades party competition in the states has increased. In 1953, the governor and both houses of the legislature were of the same party in 38 of the 46 states with partisan legislatures. By 1993, the number of such states had declined to 19 of 49 states, excluding Nebraska. In 1953, partisan control of the legislature was divided in three states. By 1993, the number of states with divided party control had increased to 18. Rosenthal (1990, 55) explains the reasons:
This [increase] is attributable in part to candidate-centered
campaigns for state legislatures and the
increasing incidence of split-ticket voting. It is
also attributable to the growth of the Republican
Party in the South and the victories of its gubernatorial
candidates in a number of southern states
and to gains made by Democrats in formerly
Republican states, like Vermont and Indiana, in
Rosenthal (1990, 56) also notes that the governor is more affected than the legislature by partisan division. Specifically, he says, "Legislatures are more likely to make life difficult for opposition-party governors than the reverse." This seems particularly true in regard to the executive budget as partisan division can lead to increased attacks upon it. Forty-eight percent of the respondents from states where the governor is of one party and both houses of the legislature are of another reported that "executive budget recommendations for expenditure increases above the base budget are significantly changed by the legislature" (N = 25). Forty-three percent of the respondents from states with divided control in the legislature reported these types of changes (N = 21). On the other hand, only 27 percent of the respondents from one-party states indicated such changes in the executive budget (N = 34).
Partisan division is related to decreased gubernatorial influence in the appropriations process. As indicated in Table 3, gubernatorial influence is much greater in one-party states than in states where one party controls both houses of the legislature and the other party controls the governorship. However, gubernatorial influence is not particularly stronger in states where the legislature is divided than in one-party states. Increasing partisanship in state legislatures has probably undermined gubernatorial leadership where it has led to a legislature with both houses dominated by the party in opposition to the governor's party.
Table 3 Gubernatorial Influence by Partisan Division with Government
Partisan Division Same political party controls Divided both leg. control Relative Influence and gov.'s office of legislature Governor is more influential 17 12 Equal influence 10 10 Legislature is more influential 10 5 Partisan Division Different political parties control Relative Influence each branch Governor is more influential 5 Equal influence 8 Legislature is more influential 15
This table does not include respondents indicating a nonpartisan legislature or those who indicated that the partisan arrangement changed within the last two years. The chi square probability is less than .05.
Proposition 4. Since the executive budget, states have generally not adopted reforms to enhance gubernatorial influence; where adopted, reforms have enhanced gubernatorial influence.
In defending the executive budget, governors have two basic problems. The first is preventing the legislature from expanding the revenue estimate and thereby risking deficits. The second problem is defending recommendations for individual programs or items. In both areas, legislative bodies make changes. Forty-seven percent of the respondents reported that the legislative appropriations are usually higher than those proposed by the governor while 10 percent said they are usually lower. Thirty-three percent of the respondents reported that the legislature usually makes "significant changes" in the executive budget.
One reform designed to give governors the ability to constrain the legislature's appetite for spending is to assign to the governor primary authority for setting the revenue estimate. The revenue estimate is the primary authority of the governor in 25 states, of the legislature in five states, of a consensus mechanism in 19 states, and of an independent source in one state (Council of State Governments, 1994, 322). As seen in Table 4, of our respondents from states where the primary responsibility is that of the governor, 25 of 53 (47 percent) reported that the governor is more influential. Only one of the 10 respondents from states where the primary responsibility of the revenue estimate belongs to the legislature indicated that the governor is more influential. Of most importance to governors is the exclusive authority to define the revenue estimate.
Table 4 Gubernatorial Influence by Power to Define Revenue Estimate
Revenue Estimate Consensus between Governor governor & Relative Influence sets legislature Governor is more influential 25 10 Influence is about equal 14 12 Legislature is more influential 14 14 Revenue Estimate Legislature Relative Influence sets Governor is more influential 1 Influence is about equal 5 Legislature is more influential 4
Classifications are based on data from the Council of State Governments (1994). Respondents from Texas are not included here because an independent source sets the estimate.
In one state, Maryland, the legislature cannot increase gubernatorial recommendations within the executive budget. This power significantly limits the legislature's capability to change the governor's budget. Though the legislature can pass individual bills enhancing particular items after the budget bill is passed, these single-subject bills are subject to gubernatorial veto. Both legislative and executive respondents from Maryland cited this reform as the most important factor explaining gubernatorial influence in the appropriations process. Both respondents also ranked the governor as more influential than the legislature. This reform is not new, and it remains confined to one state.
Gubernatorial authority to set the revenue estimate restricts the legislature's ability to increase the items of the executive budget. Another reform, impoundment, allows the governor to negate the legislature's changes after it has acted and after the governor has signed the bill. Any increases beyond those recommended in the executive budget presumably become strong candidates for gubernatorial impoundment. Such impoundment by President Nixon led Congress to prohibit impoundments without its approval. The legality and use of impoundment varies. Twenty-six of the 99 respondents said that it was constitutional in their states or had been given to the governor by statute. Impoundment powers are usually confined to such situations where revenue shortfalls develop during a fiscal year.
Altogether, 51 of our 99 respondents indicated that governors had in fact used impoundment to deal with revenue shortfalls. From the perspective of gubernatorial influence, a far more important form of the power of impoundment is to allow the governor to use it for policy reasons. Only four of the respondents indicated that the governor had used impoundment to promote a policy agenda. Of these, three indicated that the governor was more influential than the legislature and the other respondent said that the two branches were equal in influence. The constitution and statutes of the state of Hawaii grant the governor considerable flexibility in regard to the use of impoundment. Apparently, this power significantly affects gubernatorial influence in that state.
Other factors besides those cited above surely affect gubernatorial influence. However, the point of this article is not to explain gubernatorial influence in the appropriations process, but to identify reforms and changes in state government that may have contributed to the decline in gubernatorial dominance.
Consequences of the Decline
If we are correct that gubernatorial influence has declined, what results might we expect in terms of gubernatorial-legislative interaction and in terms of the direction of policy? An underlying assumption of proponents of the need for a strong executive in the appropriations process is that legislative bodies tend to be spendthrifts. Our respondents tend to confirm this assumption. Forty-seven percent of the respondents characterized legislative budgets as higher than executive budgets. Only 10 percent characterized legislative appropriations as lower. Based on this finding and the assumption cited above, it is reasonable to presume that state appropriations are being spent less carefully than at any time since the beginning of the century.
It is not correct, however, to conclude that legislatures are leading states into a new era of waste and corruption. The legislatures most likely to engage in spendthrift policies are found in states where the governor dominates the appropriations process. Of the respondents perceiving the legislature as more influential, 18 of 32 (56 percent) reported the legislature to "be much concerned about efficiency in government." In comparison, only 11 of 36 respondents (31 percent) perceiving the governor as more influential reported that the legislature was concerned about efficiency. The legislatures in states with gubernatorial domination are also more likely to pass budgets in excess of the executive budget. Fifty-eight percent of the respondents who said that the governor is more influential also reported that legislative appropriations are higher than gubernatorial recommendations. Thirty-eight percent of the respondents who ranked the legislature as more influential made that characterization. As might be expected, legislative respondents were much less likely than executive officials to perceive legislative bodies as spendthrift. However, executive respondents reflected the tendency of legislative influence to discourage spendthrift behavior.
The effects of the decline in gubernatorial influence may vary with the cause of the decline. If the decline results from partisan considerations, then the governor's budget may receive little deference. Respondents from states with divided party control reported a slightly greater tendency to make significant changes in the governor's budget than did respondents from one-party states.
If the decline results because the legislature becomes more informed, then the result will likely be more fiscal responsibility on the part of the legislature. Our data indicate that respondents reporting superior gubernatorial information were less likely than other respondents to say that the legislature is concerned about efficiency in government. Ten of the 30 respondents claiming gubernatorial superiority in information reported that "the legislature is much concerned about efficiency in government compared to a similar report by 37 of the other 69 respondents (54 percent). Furthermore, respondents reporting gubernatorial superiority in information were more likely to report that the legislature usually increases the governor's budgetary recommendation.
The increase in legislative influence has probably also changed gubernatorial-legislative interactions. Legislative influence puts governors on the defensive. Forty-seven percent of the respondents who claimed that the legislature is more influential said that the legislature makes significant changes in executive recommendations (N = 32); only 22 percent of the 36 respondents citing the governor as more influential made this characterization.
In the states dominated by legislatures, governors must rely on personal skills to defend their budgets because they lack the formal powers or the partisan allies to succeed. Twenty-three of the 32 respondents (72 percent) citing legislative dominance said that the governor's personal skills were important to gubernatorial influence in the appropriations process while only 14 of the 36 respondents (39 percent) citing gubernatorial influence said personal skills of the governors were important. Personal skills are needed most when all else fails. If we are correct that governors are losing their domination of the appropriations process, then governors will need to be more personally involved in legislative deliberations.
Although there are negative consequences associated with the decline of gubernatorial dominance of the appropriations process, there are benefits associated with increased legislative influence. In times past, executive reform was needed to end legislative corruption. Today, based on the respondents' perceptions of legislative interest in efficiency, the most responsible legislative bodies seem to be those with the most influence in the appropriations process. Often, legislative influence leads to responsibility and not to corruption. Influential legislatures tend to be the most independent in terms of information, and they are also the most likely to stress efficiency as a factor in resource allocation.
The present condition of relative parity between governors and legislatures may not continue. Tenure limits on legislators may undermine legislative expertise and shift influence in the appropriations process. On the other hand, increasing professionalism and/or partisanship in state legislatures may further undermine the influence of governors.
The achievement of parity does not mean that reformers should abandon the search for institutional devices to enable the chief executive to prepare and defend a budget. For example, when governors possess the special form of the item veto permitting them to reduce as well as delete appropriation items, then they have an effective device to defend their budgets without undermining the legislature's role in making changes in the executive budget.
Despite our emphasis in this article on relative influence, we do not believe executive-legislative influence to be a zero-sum game. Although legislatures may gain influence from greater information, the growth of professionalism in the legislature may aid the governor in achieving such budgetary goals as balance and efficiency in state fiscal matters. Indeed, rational decision making in the state appropriations process is most likely where both branches are institutionally strong (Abney and Lauth, 1987).
(1.) The number of states that have the item veto is usually listed as 43. We excluded Maryland from this list, but we did not include it among the states that do not have the item veto. The Maryland constitution allows the use of the item veto against supplementary bills passed after the budget bill. In effect, the governor does not need the item veto because the legislature cannot raise items beyond gubernatorial recommendations in the budget bill. The governor may use the regular veto against supplementary bills, which are single subject in nature.
(2.) Detailed itemization increases a governor's opportunity to use the item veto, but it reduces the discretionary powers of administrators. Not surprisingly, the legislative use of detailed itemization is positively related to the relative influence of legislatures. For example, of those respondents reporting detailed itemization, 50 percent reported the legislature as more influential compared to a similar ranking by 34 percent of the other respondents.
(3.) See, for example, State ex rel Sego v. Kirkpatrick 86 N.M. 359 (1974); Welden v. Ray 229 N.W. 2d 706 (Iowa 1975); and Henry v. Edwards 346 So. 2d 153 (LA 1977).
Abney, Glenn, and Thomas P. Lauth (1987). "Perceptions of the Impact of Governors and Legislatures in the Appropriations Process." Western Political Quarterly 40: 335-342.
--(1985). The Politics of State and City Administration. Albany: State University of New York Press.
--(1993). "Determinants of State Agency Success." Public Budgeting and Financial Management 5: 37-65.
--(1994). "Governors and the Item Veto." Paper delivered at the annual meeting of the Southern Political Science Association.
Anton, Thomas J. (1966). The Politics of State Expenditures in Illinois. Urbana: University of Illinois Press.
Balutis, Alan, and Daron K. Butler (1975). The Political Purse-Strings. New York: Sage.
Briffault, Richard C. (1993). "The Item Veto in the States." Temple Law Review 66: 1171-1206.
Council of State Governments (1956). Book of the States, 1956-57. Chicago: Council of State Governments.
--(1994). Book of the States, 1994-1995. Lexington, KY: Council of State Governments.
Howard, S. Kenneth (1973). Changing State Budgeting. Lexington, KY: Council of State Governments.
Moe, Ronald C. (1988). "Prospects for the Item Veto at the Federal Level: Lessons from the States." Washington DC: National Academy of Public Administration.
Moore, Stephen (1992). "How Governors Think Congress Should Reform the Budget: Results of a Survey of U.S. Governors and Former Governors." Policy Analysis. Washington DC: Cato Institute.
Rosenthal, Alan. (1981). Legislative Life. New York: Harper and Row.
--(1990). Governors and Legislatures. Washington DC: Congressional Quarterly.
Schick, Allen (1971). Budget Innovation in the States. Washington DC: Brookings Institution.
Sharkansky, Ira (1968). "Agency Requests, Gubernatorial Support and Budget Success in State Legislatures." American Political Science Review 62: 1220-1231.
Wells, Roger (1928). "The Line Item Veto and State Budget Reform." American Political Science Review 18: 782-791.
Willoughby, Will Franklin (1918). The Movement of Budgetary Reform in the States. New York: D. Appleton.
Glenn Abney is chair and professor of Political Science at Georgia State University.
Thomas P. Lauth is chair and professor of Political Science at the University of Georgia.
|Printer friendly Cite/link Email Feedback|
|Publication:||Public Administration Review|
|Date:||Sep 1, 1998|
|Previous Article:||Federal unfunded mandate reform: a first-year retrospective.|
|Next Article:||Verifying compliance: social regulation and welfare reform.|