Italy - Background Economic Developments and Issues: Supplementary Information Appendices

This paper reviews the flaws of the Italian tax system during the 1990s and the rationale for reform. The paper highlights that a fundamental reform of the tax system took place in 1971–73, and introduced three main innovations: the concentration of tax authority almost exclusively at the level of the central government; a value-added tax, in line with the rest of the European Union; and the extension of the base of the personal income tax to additional income categories. This paper also reviews the management of public spending in Italy.


This paper reviews the flaws of the Italian tax system during the 1990s and the rationale for reform. The paper highlights that a fundamental reform of the tax system took place in 1971–73, and introduced three main innovations: the concentration of tax authority almost exclusively at the level of the central government; a value-added tax, in line with the rest of the European Union; and the extension of the base of the personal income tax to additional income categories. This paper also reviews the management of public spending in Italy.

APPENDIX II: The Management of Public Spending 1/

This appendix summarizes the existing arrangements for preparing and implementing the annual government expenditure budget in Italy. It also identifies some weaknesses of the present approach, particularly when compared with practices in other industrialized countries in Europe and elsewhere. Finally, the appendix suggests some reforms that could improve public expenditure management in the future. The appendix is wholly concerned with the management of overall government expenditure, not policies on the standards and levels of individual public services.

The appendix is set out in five sections. Section 1 summarizes how the annual budget is formulated, while Section 2 describes budget implementation, that is the exercise of control over public spending; cash planning and management; and the in-year response to overspending, contingency expenditures, etc. Section 3 outlines the arrangements for monitoring and reporting on public expenditure developments. Section 4 briefly discusses the present approach to public sector management at the program and subprogram levels and the related issue of public accountability. Finally, Section 5 indicates possible reforms in each of the above areas viz. budget formulation; budget implementation; fiscal monitoring and reporting; and public sector management and accountability.

1. Annual budget setting

The approach to setting the annual government expenditure budget in Italy is unusually complex. The responsibility for preparing the budget is divided among three central ministries, the Ministry of Finance (mainly for tax revenues), the Ministry of Budget, and the Ministry of the Treasury; but the Audit Court and the Ministry of Public Administration are also involved. Different budgets are prepared, covering varying definitions of government and the wider public sector, and on different accounting and legal bases. Much effort goes into adhering to stipulated legal forms of budget presentation. There is little attention to macroeconomic constraints in the initial stages of the budget process, and there is relatively little input from evaluations of previous years’ budget outturns or value for money studies of individual public services or agencies. The interaction between the proposals which emerge from the government (as executive) and the additional changes which may then be mandated by the legislature, further complicates determination of the budget.

The annual budget emerges from the integration of two budget-setting processes (Table 3).

Table 3.

Italy: Budget Process: Calendar of Events

article image
article image

RGS is the Ragioneria Generale dello Stato, an arm of the Ministry of the Treasury.

  • The first step is the circulation of the Budget Circulars in March by the Ragioneria Generale dello Stato (RGS), an arm of the Ministry of the Treasury. These form the basis for bilateral discussions between the RGS and spending Ministries on projected outlays for the next year on the basis of current legislation. The outlays are expressed on a commitments, rather than on a cash basis, and cover central administration (stato) programs only, not the state sector (settore statale)—whose coverage is similar to that of the central government—nor the wider public sector (Table 4). While preparing the projected outlays, spending Ministries also identify spending needs that cannot be met under current legislation, and hence require new legislative authority. A distinction is thus drawn between a budget under current legislation (bilancio a legislazione vigente) and new legislative measures required to enable spending Ministries to meet assessed spending needs.

  • Slightly later (in May/June) the Document of Economic and Financial Planning (DPEF) is prepared by the RGS and the Ministries of Budget and Finance. The DPEF contains a baseline projection (bilancio tendenziale) and broad fiscal targets under the program projection (bilancio programmatico) for the next three years; it is expressed in cash, not commitment, terms; it allows some limited pre-planned policy changes, for example on wage adjustments and the renewal of legislation; and it mainly covers the state sector, with some material on the wider public sector. The baseline projection is predicated on current policies; the program projection quantifies the cash implications of the overall targets proposed, but does not specify the specific measures that would achieve these. 1/

  • The two processes are integrated in September. First, the specific policy measures required to implement the program projections (bilancio programmatico) are established. Second, the consequent legal changes, relative to the present legal base as specified in the bilancio a legislazione vigente, are identified. Third, these legal changes are then embodied in the Finance Bill and other accompanying legislation. The Finance Bill includes individual limits per spending program, but only on a commitments basis and only for the central administration.

Table 4.
Table 4.

Italy: Public Sector

Citation: IMF Staff Country Reports 1995, 037; 10.5089/9781451819779.002.A002

1/ Includes regional and local governments, universities, health care units, chambers of commerce, and public housing administrations.2/ Includes CDDPP, the State Forest Department, the South Agency (whose operations were terminated in 1993), and the State Road Department.3/ In 1993, the State Railroad Company and the Telephone Company were taken out of the coverage of the state sector.

The Parliamentary debate centers on the size of policy measures (manovra) needed to bridge the gap between the baseline projection (bilancio tendenziale) and the program projection (bilancio programmatico), rather than the total amount of spending per program. In recent years, Parliament has respected the aggregate figures in the government proposals, while negotiating a different mix of expenditure programs and sub-programs.

Even after the Finance Act is passed, further steps are necessary before cash allocations to spending programs 2/ can be determined.

Expenditure commitments, which were appropriated in previous years’ budgets but not used, are considered as “carry-over.” These “carry-overs” are reallocated in bilateral discussions with the RGS to individual programs (up to three years for recurrent expenditure and for five years for capital expenditure, where most of such “carry-over” arises). Thus, the annual cash allocations to each program are, in principle, determined by the budget approved by Parliament plus any “carry-over” from the previous year, as expressed in commitment terms. All commitments are then converted into cash figures by the application of standard rules for each economic category, called “realization coefficients.” These cash figures on a program-by-program basis are not subject to Parliamentary discussion and approval as such. Moreover, the first such cash figures are not available until March, i.e., after the first quarter of the budget year. Formal revisions are made in June (halfway through the financial year), when the final bilancio di assestamento is approved by the Parliament.

This approach to annual budget setting reflects a long-established legal tradition in Italy. It is an approach which draws both on French practices, but, particularly in the revised arrangements which date from 1978, also on certain features of the US model. Yet, as the authorities acknowledge, questions arise on whether too much attention is paid to legal form in setting the budget and not enough to economic substance. Some differences between the budget process in Italy and those in other large European countries are striking.

(i) The starting point—costing existing legislation—illustrates the pre-eminence of the legal dimension; it also means that macroeconomic constraints are only brought to bear later. Budgeting everywhere involves striking a difficult balance between “top down” macroeconomic constraints, on which the Treasury or Finance Ministry bases its case, and “bottom up” pressures as presented by individual spending ministries. At the outset, the arrangements in Italy give more weight to costing those “bottom up” pressures, rather than setting an overall “top down” constraint. Indeed, starting by costing current legislation tends to create an expectation among spending ministries, both of the continuation of existing policies and (more implicitly) of existing public service standards and practices. There are incentives to inflate the program’s requirements; there is no stimulus to identify the scope for savings.

(ii) The main budget aggregates—either central administration (stato) or the state sector (settore statale)—cover only central government transfers to local government and spending agencies. 1/ Expenditure financed by their own charges and taxes, which forms part of General Government Expenditure (GGE), is not included. Many other countries focus on a consolidated GGE measure of public sector spending in setting the annual budget. In turn, the tendency to target the borrowing requirement (fabbisogno) for the state sector, rather than the wider definition of public sector borrowing needs, may create temptations to underprovide for transfers to local government and spending agencies. Any resultant additional borrowing by local government and spending agencies is outside the targeted borrowing concept. Under the Maastricht Treaty, it is a broader concept than the state sector fabbisogno that is relevant to measuring both annual government borrowing and outstanding public debt.

(iii) The emphasis in recent annual budget discussions has been on the manovra, the size of policy measures (in the case of expenditures, the size of spending cutbacks). These are presented as the difference between the estimated cost of a program in the baseline budget (bilancio tendenziale) and the post-cutback cost in the program budget (bilancio programmatico). But this comparison gives an exaggerated impression of cutbacks. First, in the past, there has occasionally been a tendency to inflate the size of the baseline projection. Second, the frequent recourse to one-off measures in recent annual budgets has increased the size of the manovra required simply to compensate for the expiration of their effects on the baseline projection. In these circumstances, focussing on changes relative to the trend deficit exaggerates the economic impact of the adjustment, which depends on the change in the actual fiscal stance from year to year. For example, the Parliamentary debate on the 1995 Budget focussed on measures designed to increase the primary surplus by 2½ percent of GDP relative to the baseline projection (bilancio tendenziale). But, on a year-on-year basis, the budgeted increase in the surplus was only 1 percent of GDP.

(iv) Upward pressure on government spending can arise when alternative measures proposed by Parliament are included in the budget. Under the copertura requirement, a Member of Parliament can only propose additional spending in one area, if an offsetting expenditure reduction or additional revenue measure is also identified. However, this requirement applies only within the state sector: this safeguard can therefore be undermined where the Parliament proposes additional spending to be implemented by a decentralized spending entity, such as the social security agency. The copertura would require compensating measures only to the extent that budget transfers to that agency are increased; these transfers do not always take fully into account the impact of the higher spending proposals put forward by Parliament. Thus, extra general government spending can arise even when the copertura requirements are formally met. Also, as happens in other countries, copertura measures included in the budget are sometimes overestimated, or in later years their yield is lower than the cost of the measure that they are supposed to cover.

(v) There is no clear, consistent and comprehensive cash budget for spending programs within Ministries. Thus program and sub-program managers do not face hard budget constraints. On the contrary, they must cope with considerable uncertainty about cash availability—in extreme cases until the bilancio di assestamento is set in the middle of the financial year. This makes both management of public services and resource allocation amongst sub-programs more difficult.

(vi) The lack of a single clear definition of the cash resources available by program and sub-program also hinders comparisons between budget plans and outturns. The Audit Court has noted that present arrangements unduly complicate the audit task. Parliamentary scrutiny of budget plans vis-à-vis outturns is also under-developed. The profusion of budget formats make scrutiny more difficult. In short, it obscures public accountability.

(vii) Finally, the emphasis in budget presentation remains on the functional and economic categories, with relatively little attention to classification by administrative unit. Such a more managerial approach, perhaps developed furthest in both New Zealand and Australia, is now increasingly used in Europe to help improve the delivery of program objectives. Dividing programs and sub-programs into identifiable and manageable units can both set the basis for improving management, e.g., through running costs or activity cost concepts discussed later; and for better accountability through the specification of program objectives and then assessments of outcomes through measurements of outputs and inputs.

2. Budget implementation

The Treasury has overall responsibility for budget implementation and for the management of the government’s cash resources, chiefly through the unified Treasury account. For the central administration (stato) budget, the accounting system follows conventional control and accounting features—through the registration of appropriations and the normal processes of commitment, verification, payment authorization, actual payment and accounting records. Within the Treasury, it is the RGS which exercises this control and authorizes all payments through the unified Treasury account (Tesoreria unica).

In addition to the accounting and payment roles for central administration, the unified Treasury account also covers two related functions. First, the Bank of Italy acts as government banker, with all receipts and payment transactions going through a single Treasury bank account. Cash forecasts are prepared on a monthly basis and revised throughout the year; daily reports on the cash position in this single bank account are passed to the Treasury. This, and the associated cash management function, are well-advanced with good coordination between the Treasury, the RGS, and the Bank of Italy. Second, and rather more unusually, the Treasury account also functions as a kind of deposit-taking institution for local government and certain state agencies. These place on deposit both transfers received from central government and the bulk of their own resources collected from local taxes, charges or borrowing. However, like a deposit-taking institution, this arm of the Treasury account may make advances to agencies over and above their budget allocations, while duly recording a liability in the account books of the treasury account. If the state sector borrowing requirement (fabbisogno) requirement is not to be increased, any such advance to one agency or spending program requires lower payments to another.

The skill with which the Treasury account has been developed and operated reflects an attempt to cope with burdensome legal and administrative constraints on budget implementation, while maximizing the efficient use of the government’s cash resources. However, the very flexibility offered in optimizing use of cash resources has given rise to other problems. Underlying most such problems are underprovision in the budget for certain expenditures and a lack of coherent contingency arrangements. The resultant pressures have frequently been accommodated through the Treasury account more by accounting rather than real contingency measures. Although there are a number of formal reserves, the Treasury account acts as a main source of funds for overruns and other contingency spending. In addition, the emphasis on the narrow measure of borrowing (fabbisogno), which matches the coverage of the Treasury account, has also made it tempting to divert transactions outside the targeted borrowing aggregate. The Treasury account has thus become a kind of supplementary resource allocation mechanism after the budget. Issues not faced up to in budget formulation are resolved at budget management stage—a stage much too late for any fundamental fiscal adjustment.

There are several instances of serious practical consequences of this approach.

a. Underprovision of transfers to the state pension agency (INPS)

Mandatory spending on pensions has to be met as an entitlement program. When faced with consistent underprovision, rather than increasing the allocations (or requiring higher social security contributions), the Treasury account has made advances to the state pension agency (Table 6). This in turn has been financed partly by additional borrowing and partly by delaying access to funds deposited by/or for other institutions. Also, the Treasury account advances to the state pension agency are in part matched by reduced commitments for local authority transfers (typically by stretching into the future the planned provision for local authority or spending agency capital programs). At end-1993, the accumulated stock of these advances had reached an estimated 10 percent of GDP. In reality, the pension agency is unlikely to repay the advances made.

Table 5.

Italy: Consolidated General Government, 1992

article image
Source: Relazione Generale sulla Situazione Economica del Paese (1992) Vol II.

Local health units (Units Sanitaria Locali (USLs)).

Table 6.

Italy: State Transfers to INPS 1/

(In billions of lire)

article image
Source: INPS.

Cash transfers from the central government (settore statale).

Transfer from the commitment budget (bilancio di competenza).


b. “Hidden debt”

Local health units, faced with inadequate budget resources, have from time to time been encouraged to borrow from commercial banks, because such borrowing is outside the targeted borrowing figures. These debts are then regularly refinanced “below-the-line” by the Treasury through bond issues (replacing the commercial borrowing) without affecting the state sector deficit in the current year, but nonetheless increasing its outstanding debt. As in the case of pensions, the Treasury account is used to resolve a problem that lies in underprovision at budget formulation stage, while helping facilitate compliance with a targeted borrowing figure.

c. The use of cash balances (giacenze)

This is the practice of using fungible cash resources deposited by one agency to meet the cash needs of another. In principle, it is supposed to contribute to an efficient use of cash resources; in practice, such advances are often not repaid within a year but instead lead to deferral of payments or other services and to additional government borrowing. This illustrates how the Treasury account, rather than just the budget, can in practice determine spending priorities.

d. The diversion of borrowing

First, delay has been used to help meet the target borrowing requirement. Second, however, local government has been encouraged in the past to divert some borrowing from the Cassa Depositi e Prestiti—a spending agency which lends to them and whose resources are within the Treasury account—directly to the commercial banks or indirectly to suppliers (through arrears). Again the aim is to help meet the targeted borrowing figures by pushing more borrowing outside the target concept.

Whatever the accounting advantages, there are real economic costs to such operations within the Treasury account. Uncertainty about the actual availability of cash resources for program managers undermines normal budget management discipline. Delaying provision for local capital projects can increase their costs, when it results in sub-optimal phasing of work or the expensive renegotiation of contracts. Diverting borrowing to commercial banks adds to borrowing costs and may allow uneconomic capital projects to proceed if commercial banks believe there is an implicit government guarantee on the loan. Allowing local health units to go into arrears not only adds to borrowing costs but encourages suppliers to add a premium to their prices, in the expectation of delayed payment.

The growing scale of such operations within the Treasury account is worrying. Already a significant part of government spending takes place through this deposit-taking arm of the Treasury account, rather than on the central administration programs (Table 7). Moreover, the credibility of the Treasury account assets must be a concern; realistically many of the advances cannot be repaid, except through further borrowing. In short, the arcane and complex nature of the Treasury account, its wider economic cost and the increasing size of the debts have damaged the transparency, credibility, and perhaps ultimately the sustainability of this present approach to public expenditure management.

3. Monitoring and reporting on fiscal developments

Italy has an highly advanced facility for monitoring budget implementation. The RGS Information Center processes all budget and accounting data for the central administration, and also monitors developments in the wider public sector. An integrated network across the country enables all payments stages to be recorded for the central government budget by sub-program and by economic and functional category. The reconciliation of RGS data on payment authorization with cash payments made is facilitated by the speedy receipt of cash data on payments from the Bank of Italy. Data are available daily on cash inflows and outflows on the central administration side of the Treasury account.

For the central administration, the monitoring system appears to work very well. Indeed, for those programs, the spending outturn has typically been less than the final cash budgets in the bilancio di assestamento, though not necessarily less than the original budget before revisions of “carry-over.” But for the state sector, i.e., including the deposit-taking arm of the Treasury account, the picture is much less satisfactory. Typically the overall Treasury account balance has turned out worse than estimated at the beginning of the year. Indeed, RGS officials indicate that the data on flows in and out of the deposit-taking arm of the Treasury account are particularly difficult to project. This both complicates cash management and encourages bad payment practices, particularly delay and deferral of payments.

The monitoring of the wider public sector also tends to be undermined by weaknesses in local government fiscal reporting. Priority is given to securing borrowing data, rather than information on revenues and expenditures. Moreover, the expenditure data routinely arrive only with delay, and are not satisfactorily broken down by economic and functional category. There is also relatively little monitoring of spending commitments within local government and spending agencies. Thus the government is deprived of information on month-to-month developments in government spending for around one-half of total general government expenditure, and cannot identify the real needs nor the appropriate targets for remedial fiscal action in-year.

4. Accountability

The accounting systems for the central administration programs are well-established; conventional financial and proprietary audits are soundly in place. Management accounting, however, is less well-advanced. Neither running costs (that is outlays on personnel and office costs) nor activity costs (that is running costs plus a capital charge which provides the basis for user charges) seem to be widely used. Nor are there deeper assessments of the government balance-sheet and financial position. Many countries, notably Australia, New Zealand, and now also the United Kingdom, are beginning to develop more comprehensive measurement of assets and liabilities, including medium-term perspectives on liabilities, contingent liabilities, and valuations of real and other government assets.

Auditing also has considerable scope for further development toward value-for-money (vfm) work—no doubt, in part, a reflection of the long tradition of focusing on legal compliance. However, proposals to pursue initiatives in this area, drawing on U.K. practices, have been agreed. Indeed, the Audit Court has begun some vfm work in recent years, not only for central administration programs but also for local government. As noted earlier, the accompanying development of more Parliamentary scrutiny of such vfm has yet to be fully taken up.

Accordingly, the concept of establishing and testing public accountability in Italy has some way to go. At heart, such an approach would require fundamental changes from detailed legal control to broader management control. This would help strengthen the accountability of the executive to the legislature, through the setting of program objectives and the testing of outcomes. In turn, this would require greater transparency to ensure that budget outturns can be compared with budget appropriations. Such an approach would also need more accountability of program manager to spending Minister; but that, in turn, would require at a minimum the costing of inputs and measurement of outputs. It would also require greater accountability of individual managers, giving them responsibility for separate cost or activity centers.

5. The way forward

The need for reform in the management of public spending, and the required direction of changes toward greater transparency and public accountability, are well-understood and widely supported within Italy—not least by senior officials and academics, for example in the work of the Technical Commission on Public Spending. More weight needs to be attached to macroeconomic constraints and also to objectives, standards, and value for money in the delivery of public services. Above all, greater attention needs to be paid, both in budget preparation and budget execution, to economic substance and less to legal form.

Yet, in practical terms, such a transformation will not be easy to achieve. Replacing long-established traditions of careful legal compliance as the basis for public sector management will require a thorough and comprehensive new approach. Reforms would be necessary on all aspects of government expenditure management: (i) annual budget setting, (ii) budget implementation, (iii) financial monitoring and reporting, and (iv) the exercise of greater accountability.

(i) In preparing the annual budget several changes are required; most important are setting a global cash target (and then agreeing program figures in cash), and introducing a formal contingency fund within the annual budget.

First, the sequence of budget-setting would need to be changed. Putting in place the macroeconomic constraint of an affordable global total of government expenditure (GGE) should be the first step. Once such a target is agreed (typically with guideline figures for later years) the subsequent bilateral discussions between financial ministries and spending ministries are essentially about program and sub-program shares. Starting from guideline shares consistent with the GGE target, spending Ministers can be asked to specify how they would implement changes to reduce cash spending by 5 or 10 percent, as well as making whatever case they might wish for additions above their guidelines. Ministers would continue to need a baseline in framing specific proposals for policy changes so that priorities can be established within an appropriate ministerial forum. That baseline should be predicated on genuine current policies, rather than either the current legal requirements (bilancio a legislazione vigente) or the present baseline budget (bilancio tendenziale). Although construction of a baseline budget projection can be useful—indeed an essential step when there are growing entitlement programs or one-off revenue measures—excessive emphasis on it during budget preparation can, as noted above, lead to an exaggerated picture of savings. Whatever the variations in actual practices amongst other European countries, this discipline of setting an affordable global total, and then negotiating shares amongst spending programs, lies at the heart of efficient budget preparation.

Second, under the present approach in Italy, the proposed GGE total and the consistent program and sub-program details would then go to the legislature. In discussion of the manovra (policy measures), the copertura requirement would have to be sustained; transparency requires an objective costing of any counterproposals. One option might be to establish an independent agency to cost all such proposals, perhaps modeled on the Congressional Budget Office in the USA. But whether a new agency is created, or the task given to the RGS, two changes are necessary. First, the copertura perspective should be broadened to encompass general government; that should avoid additions to GGE arising from extra spending by local government or state agencies, when only transfers are offset by other proposed measures. Second the full costs, over a period of years, of any copertura measures should be calculated and taken into account in assessing whether a counterproposal would be fully offset by other proposed measures.

Third, a contingency fund within the annual budget and hence the cash GGE target should be set at between one and three percent of total GGE. This should be the only source of finance for overruns or genuine contingencies. All use of contingency fund resources would require approval of the Ministry of the Treasury and be reported to Parliament. But access to the contingency fund should only reluctantly be conceded, as a last resort. Faced with a prospective overspend or contingency, spending Ministries should be required, wherever possible, to undertake a rigorous search for in-year savings in other areas, even including other demand-determined but non-entitlement current spending through rationing of supply, queuing and actions to curb demand. Only if that fails should access to the contingency fund be considered.

Fourth, since the program totals and the contingency fund would be established in cash terms within a GGE target, the whole complex nexus of estimating cash allocations from a combination of commitments, carry-over and realization coefficients could then be discontinued. All monies not spent each year would be surrendered to the Treasury, a standard practice in many countries. 1/ Greater certainty on program and sub-program cash allocations would pave the way for more effective direct management for programs and sub-programs, and improve expenditure control.

However, significant parts of the public sector within GGE are not under direct central government control (i.e., local government and other state agencies). The size of grants and transfers to such bodies are an important restraint on their spending plans. But, increasingly, they can finance additional spending from local revenues or charges. One option for budget planning purposes is to ensure that a realistic forecast is made of aggregate local authority and state agency spending (including from their own resources) within the GGE cash target. Any gap between forecast and estimated spending (evident once local government and agencies set their own budgets) can be “withdrawn” from the contingency fund. Another option, particularly where spending agencies and local government are prevented by law from borrowing for current expenditure purposes, is to accept that any such gap will be financed from local revenues, and has no implications for the borrowing requirement. 2/

(ii) To improve budget implementation within this new framework, spending programs should be cash-limited whenever possible. All stages of expenditure processing, particularly commitments, should be monitored against commitment and cash limits. In turn, all potential overspending against a cash limit would be handled through the proposed contingency fund: any successful bid leads to a transfer of an appropriation from the contingency fund to the budget line item in question. Accordingly no transaction could be financed unless within a cash limit (or explicitly exempted, as in the case of entitlement programs). Transactions would thus no longer be financed within the deposit-taking aim of the Treasury account. Any payment which would breach a cash limit would generate an automatic “stop” feature in the budget system. This should also bring to an end the whole secondary budget allocation role of the Treasury account, through advances and delays in authorized payments.

It might be argued that adding cash limits to an already legalistic system will enhance the complexity of government expenditure management in Italy. Yet, without this essential discipline, and especially without reforms to the Treasury account, there would continue to be scope for avoiding difficult but necessary choices about priorities within limited resources. While pure entitlement spending cannot be contained within cash limits, the implication of additional spending there (as for example in the past on pensions) need to be reflected in genuine reductions elsewhere (including the drawdown of the contingency fund), not accounting arrangements within the Treasury account.

Cash limits would lend themselves to improved managerial control by allowing the establishment of running cost and activity cost centers. A further advantage of this is the ability to track performance by cost center from the initial setting of the appropriation and cash limits, any modification (such as reductions or additions from the contingency fund), and the budget outturn. It would enable the auditor, Parliament, and the public to be better informed about overall budget performance.

This approach would require a strengthened but rather different Treasury account. For central administration programs, monthly commitment and cash limits would be recorded in the accounts, both to assist in cash planning and to enforce control within the appropriation limit. However, both local government and state agencies might also be encouraged to process payments through sub-accounts in the Treasury account (in accounting terms) and within the single bank account (in practice through a local bank’s acting as agent for the Bank of Italy). To make this attractive, the government might offer such a payment service free of change. Most important, there should be no facilities within the revised Treasury account for advances, for switches of cash allocations between one program and another, and for stretching-out commitments into the future.

(iii)To accompany the introduction of commitment and cash limits, and the modified role of the Treasury account, some improvements in fiscal monitoring and reporting would be required. Integrating local government and state agencies within the Treasury account is highly desirable, preferably through on-line links; that would need to be accompanied by strengthened monitoring. Whatever arrangements are made, if the needs and targets for within-year action on public spending are to be identified, the expanded range of indicators should include (by economic, functional and, for central administration programs, cost center classification):

  • monthly limits on cash expenditure;

  • monthly outturn data on cash expenditure, generated within a few days of the end of each month;

  • monthly commitment limits and actuals, to inform central ministries about prospective spending, not only within central administration but also local government and spending agencies;

  • monthly estimates of revenues both in accruals and cash terms; and

  • monthly projection of borrowing needs and sources of financing.

(iv) Finally, there is a need for reforms to enhance the accountability of the executive to Parliament and the public. The introduction of cash limits and cash outturn would in itself improve accountability. So would the creation of cost centers and the application of cash limits and running cost limits to them. But accountability needs to go much further: a new managerial culture must evolve. Thus cost centers might be provided with a written statement of objectives to be achieved and stipulated standards of services, to replace the detailed, cumbersome and often out-dated laws and regulations which at present typically provide the basis for public sector management. The manager of that cost center could be required to report annually on outputs achieved and on the cost of inputs. Internal evaluation within spending Ministries should be enhanced to match the more managerial approach. In turn, Parliament should be able to conduct evaluation through scrutiny of reports prepared by the external auditors. Such an approach could also be extended to local government and state agencies.


Prepared by Barry Potter, on the basis of a paper by Vito Tanzi, Arigapudi Premchand, Barry Potter, Sergio Lugaresi, and Dimitri G. Demekas.


The latest DPEF (July 1994) specified for the first time norms for total revenue and total current primary expenditure that would support the program targets. Still, the measures needed to meet these norms were not indicated.


The DPEF and the Forecasting and Planning Report do set overall targets in cash terms for central government spending, but not for the individual spending programs.


Local government and spending agencies—notably the social security fund—account for more than 50 percent of total general government expenditure (Table 5).


Some minor flexibility is often allowed on capital programs.


The planned tax reform, with the move toward enhanced fiscal federalism, would make this more possible in the future.