Professor Wehner’s book “Legislatures and the Budget Process; The Myth of Fiscal Control” was published in 2010 by Palgrave-Macmillan. It has three main aims: (1) to establish a framework for assessing the budgetary role of legislatures; (2) to explore the determinants of cross-national variation in institutional arrangements; and (3) to assess empirically the impact of legislative budget institutions on fiscal policy.
The author rightfully asserts that “the cross-national study of legislative budgeting, despite some progress in recent years, is in a lamentable state”. After examining the institutional foundations of budget control by the legislature, Professor Wehner constructs an index of the “power of the purse”—budget processes that are controlled by parliaments and which allow them to influence budget outcomes. All thirty OECD member countries[1] are included in the study. The index shows an extremely wide variation of the strength of the legislatures in different countries.
Professor Wehner examines why there is such a wide variation. He empirically establishes that colonial heritage is a powerful determinant of legislatures; budgetary powers and organizational arrangements. Other factors, notably presidential government, divided government, and the maturity of democracy, are weaker explanatory factors. Since most countries in the sample are parliamentary régimes (only four OECD countries are deemed to have “presidential” governance systems), one conclusion is that simple macro-constitutional considerations such a regime distinction (e.g., parliamentary versus presidential) may have been overrated in legislative research. Professor Wehner appeals to researchers to take lower level budgetary institutions more seriously.
Indeed, an important contribution of the book is that it examines the relative importance of various “lower level” budgetary institutions, which are defined as formal budgetary rules decided in a political process. The two main groupings of budgetary institutions are: (1) the specific powers of the legislature to change budget outcomes; and (2) the way the legislature is organized (internal committees, research bodies and the time law-makers have to consider annual budgets). One important finding is that the legislature’s effect on the level of public expenditure is driven principally by one budgetary institution, namely the extent of power that law-makers have to amend the executive’s draft budget.
The evidence of the importance of the legislature’s budget amendment rights is well supported. Other variables may have been over-rated in empirical studies. Take, for instance, the “reversionary budget” variable—the formal rule that specifies how the budget is to be implemented at the start of a new fiscal year in the event that the legislature does not vote the budget in a timely manner. Here, the existence of a formal rule is probably less important than the frequency of such an event. In the study, although 10 countries’ legislatures are scored as having the capacity to force the government to shut down (no spending takes place until the annual budget is adopted), such drastic action by the legislature rarely takes place. To my knowledge, only the United States legislature (i.e., only 1 out of the 10 countries) has ever forced a government shut-down by deliberately not providing funds at the beginning of the fiscal year.
The construction and weighting of the components of the budget institutions index is open to discussion. For example, for reversionary budgets, the United States and France are scored at opposite ends of the spectrum (10 and 0 respectively) even though they have similar legal provisions for interim budgets, which are (or would be) typically implemented on the basis of the previous year’s budget.[2] In contrast, the United Kingdom and other Westminster countries’ practice of Vote on Account is scored higher than for France. Vote on Account is a traditional British practice that enables the U.K. government to begin spending 40 percent of its own budget for a new fiscal year without giving parliament any opportunity to object.[3] Procedures for implementing the government’s annual budget by stealth should be scored zero, in my view. For this variable, there are three main cases: (1) parliament has no influence over interim budgets—procedures are in place to ensure that the government’s proposed budget is adopted; (2) parliament authorizes funds on the basis of the previous year’s budget; and (3) parliament refuses to adopt any budget until it ensures that its new policies are included. For scoring purposes, the second option could be refined. For example, a higher score could apply if parliament provides interim budget spending authority for only a limited time period, e.g., less than six months.[4]The non-availability of high-quality data needed for more in-depth empirical work is acknowledged by Professor Wehner. Concerning data limitations, he notes that more finely grained measures of the “budget amendment” variable (e.g., the number of amendments or the quantitative impact on total expenditure of all amendments) would be preferable. However, the required detailed data is not readily available in OECD surveys of budget practices (the main data source used in the empirical work). Similarly, the “flexibility in implementation” variable could be refined. For example, the size of virement of budget appropriations or programs is more important than the formal authority provided to the government by the legislature to swap between spending items or programs.
A similar remark applies to other variables used to measure “flexibility in budget implementation”. For example, concerning a centralized reserve fund for emergency spending, the degree that parliament does not control budget implementation depends on the size of the budgeted reserve, not the mere existence of a budgeted reserve fund (data from OECD surveys are only available for the latter). In Liberia, for example, Parliament is emasculated from all proposals by the government for the use of funds out of the Contingency Fund, whose budgeted amount may be up to 5 percent of domestic revenues[5]—much higher than normal.
Aware of the limitations and availability and quality of data, Professor Wehner points to the need to supplement the empirical research with case studies. I fully agree with this approach. Case studies can highlight how particular budgetary institutions—either formal or informal—are important in some countries, but not in others.
Case studies can also identify missing variables in cross-national studies. For example, the frequency and size of Supplementary Budgets—barely examined in the book—are extremely important when assessing Japan’s fiscal performance. Similarly, the importance of a two stage budgeting procedure, whereby the legislature first approves the annual budget aggregates (revenues, expenditures, budget balance), possibly during a pre-budget review by parliament several months before presentation of the government’s draft annual budget, are important in some countries. For example, it is well-known that this procedure has been instrumental in improving Brazil’s overall fiscal discipline.
Under the chapter entitled “The Promise of Top-Down Budgeting”, the key legislative budgetary arrangements for Sweden and South Africa are examined. These are useful complements to the book. The case study of Sweden highlights how the legislature’s formal adoption of multiyear appropriations was most helpful for improving Sweden’s overall fiscal performance in the 1990s. The absence of serious examination of the role that medium term budget frameworks and macro-fiscal rules have played in improving overall fiscal performance is perhaps the book’s most serious shortcoming. The incorporation of these budgetary institutions in future studies is perhaps the highest item on the research agenda.
Future studies of the role that “medium-term budget frameworks” play in determining outcomes should not be confined to the expenditure side of the budget, as is the case in this book. In discussing the role of budgetary institutions in determining fiscal outcomes, the taxation side of the budget should not be excluded. In this context, in the 80 country cross-sectional study explaining budgetary institutions’ contribution to different aggregate expenditure/GDP ratios, is the statistically significant result for “accept-or-reject” amendment powers a proxy for revenue/GDP ratios[6]? When many countries are being examined simultaneously, are expenditure/GDP ratios primarily determined by the availability of resources, i.e., revenues/GDP?
To examine more fully whether fiscal control by the legislature is a myth Professor Wehner identifies several areas for future research. One such area is the identification of potentially important budgetary institutions that are hidden away in secondary legislation or informal practices. I fully agree with this, as well as the need for more systematic study of the interaction between macro-constitutional and low level budgetary institutions in the quest for understanding how institutional design affects fiscal policy outcomes. Also—and this is highly pertinent for researchers—is the need for more careful theoretical modeling that would clarify the effects of specific institutional arrangements for budgeting. This would help to avoid unfounded or exaggerated claims made by some researchers. Finally, I fully agree that it is especially important to identify how and why budgetary institutions change over time.
In summary, this new book is a must-read for all theoreticians and practitioners who wish to avail themselves to the most comprehensive internationally comparative study on legislative budgetary institutions that has been published to date. By reading this nicely-presented and thought-provoking volume, readers would broaden their understanding of the relative importance of specific budgetary institutions under the control of the legislature.
[1] In 2010, the OECD has expanded the numbers of its member countries beyond 30.
[2] In the USA, “ongoing resolutions” are adopted when Congress does not adopt Appropriation bills before the beginning of the fiscal year (October 1). Ongoing resolutions are usually based on the previous year’s appropriations. Adoption by Congress of ongoing resolutions is very frequent (e.g., 5 of the last 7 election budgets, including this year’s one for 2010/11). In contrast, shutdowns of government are rare. In France’s 1958 Constitution, if the National Assembly does not adopt the budget before the beginning of the fiscal year, funds are made available by government decree to meet the commitments already voted for. The 2001 Organic Budget Law elaborates: “voted services” are the minimum amounts needed to provide government services consistent with the budget approved by Parliament in the previous year. Unlike the USA’s ongoing resolutions, this provision is rarely applied as the National Assembly nearly always votes on the annual budget law before the start of the fiscal year (January 1). Although Congress takes the initiative in the USA and the government in France (since, in the Constitution, Parliament delegated permanent authority to the government on this issue), in both countries, previous-year budgets are used as the basis for keeping the government running.
[3] The Vote on Account is usually slipped through Parliament several months before the fiscal year begins—and well before the Chancellor of the Exchequer presents his annual budget to the House of Commons, which is often very close to the beginning of the new fiscal years. Appropriation Acts are typically adopted a few months after the beginning of the fiscal year.
[4] In some countries, the legal provisions for interim budget authority are applied for the entire 12 months of a new fiscal year. This was the case in the Philippines in 2001, 2004, and 2006 when Congress failed to adopt any Appropriation Acts for those three years (in each year, government services were provided on the basis of previous year’s General Appropriations Act).
[5] Liberia’s new Public Finance Management Act, adopted in 2009, requires the legislature to approve, in each annual National Budget, a contingency fund of up to 5 per cent of total annual domestic revenues. In using the Contingency Fund for “urgent and unforeseen expenditures arising from emergency situations for which payments cannot be postponed until the passage of a supplementary budget or the next annual National Budget without seriously affecting the public interest, the Minister of Finance shall submit a proposal to the President for approval” (see section 13 of the Act).
[6]The “accept-or-reject” amendment variable is reported to be operative in Bangladesh, Gambia, Ireland, Malawi, Nepal and South Africa. On average, these six countries’ revenue/GDP ratios would be much lower than the average for the other 72 countries in the sample.
Note: The posts on the IMF PFM Blog should not be reported as representing the views of the IMF. The views expressed are those of the authors and do not necessarily represent those of the IMF or IMF policy.
