Binding fiscal policy rules are likely to influence the level and composition of government expenditure and taxation. In addition, fiscal rules have major macroeconomic consequences for inflation, external indebtedness, and economic growth. Of particular concern is the effect of balanced-budget rules on the short-run variability in output and income. Given the limited experience with fiscal rules, some of these effects must be examined with the help of simulations. By comparison, it is noteworthy that, in the monetary policy area, experience with inflation targeting, applied since the early 1990s, has been even more limited.
Experience
For the most part, economic performance under fiscal rules has been mixed. Besides a number of successes, some rules have been ineffective, suspended, or abandoned. At the national level, in the advanced economies (including the recent convergence in the fiscal position of most EU member countries), attempts to comply with fiscal rules contributed to a decline in inflation and interest rates, mitigated the crowding out of private investment, and alleviated the external imbalance. In developing economies, in the absence of sufficiently deep internal financial markets, restrictions on bank financing or domestic borrowing were partly accommodated by a substantial buildup in foreign indebtedness. To some extent, the latter was facilitated by the credibility gains associated with implementation of the fiscal rule.
Apart from broadly favorable macroeconomic effects, compliance with fiscal rules has led to distortions in the composition of government expenditures or to tax increases. On the expenditure side, often the brunt of the adjustment has been borne by cuts in public investment. In some instances, fiscal rules induced a lack of transparency in the budget process (for example, accumulation of payment arrears), proliferation of creative accounting practices, and recourse to one-off measures (such as financing from privatization receipts). Also, distortions in tax structure and administration (say, by advancing tax payments) may have been compounded, along with an increase in the overall tax burden. Nevertheless, more than in the case of discretionary policies, the future maintenance, and thus the credibility, of a fiscal rule is jeopardized by excessive reliance on low-quality, one-off measures.
Actual experience with fiscal policy rules has been limited mainly to borrowing limits at the national level and balanced-budget rules at the subnational level of government. In general, prohibition on central bank credit to governments has been effective; leakages (mainly, through short-term advances) have been relatively infrequent. In some countries, however, a buildup of expenditure arrears or an acceleration of tax collections has taken place; also, if permitted, indirect central bank financing, mainly through the government securities market, has proved to be an additional channel to bypass the intent of the rule. It has been far more difficult to ensure that ceilings—in contrast to a prohibition—on central bank credit (in proportion to past revenue or expenditure) to governments are, in fact, binding, as calculation of the ceiling is likely to leave room for circumvention. Although a successful reduction in inflation depends mainly on monetary control, rather than on rules limiting central bank financing of budget deficits, in practice it is not possible to maintain an appropriate monetary stance if the government has virtually unlimited access to central bank financing. Thus, prohibition on central bank credit removes an important source of inflationary pressure, and binding limits on such credit contribute to the moderation of inflation.
The experience of the CFA franc zone countries illustrates the difficulty of enforcing a ceiling (as compared with a prohibition) on central bank credit to the government. The statutory limit on central bank lending and advances to the government has been reached or exceeded in most member countries. Excesses have occurred when a decline in government revenue was not necessarily matched by the required cut in central bank financing and when, in some cases, the central bank assumed the external debt obligations of the member government without formally violating the rule. Partly through these outlets and partly through heavy reliance on external financing and payment arrears, most countries incurred sizable budget deficits in the 1980s. The resulting crowding out of private activity and appreciation of the franc contributed to stagnation in output and widening of external current account deficits.48
Indonesia‘s rule disallowing domestic financing of budget deficits has provided a remarkable degree of fiscal discipline for almost three decades. During most of this period, while enjoying sustained growth, Indonesia has managed to avoid the high inflation experienced by many other countries at a comparable stage of development, particularly the oil-producing developing countries. Calculation of the cyclically adjusted budget balance since the early 1970s reveals that the rule has provided scope for authorities to conduct a countercyclical fiscal policy, particularly during the 1980s.49 However, although effectively restraining domestically financed deficit spending, the rule has permitted a significant accumulation of external debt in recent years.
The experience of U.S. states with the balanced-budget rule probably yields the richest empirical evidence in this area. Because of the varying degrees of stringency of the rule among different states, it has been possible to test the effect of institutional features on compliance and the effect of the rule on output variance and interest costs. The U.S. states that are subject to a strict constitutional ex post rule have been successful in achieving balance between current expenditure and revenue, as compared with states that follow merely an ex ante rule under the authority of a state supreme court elected by the legislature or appointed by the governor.50 In other words, the effectiveness of the rule in constraining budget deficits is enhanced if applied to end-of-year performance, with no allowance for expenditure carryover, under the surveillance of a politically more independent authority and on the basis of a constitutional obligation. As regards the composition of the adjustment, deficits tend to be reduced, or surpluses generated, through spending cuts instead of tax increases.51 This result is not surprising in view of the competitive pressures among the states to reduce local sales tax, income tax, and property tax rates.
In contrast to the subnational balanced-budget rules in the United States, in Canada most of these rules have been in effect for only one or two fiscal years. During this short period, preliminary evidence indicates that those provinces or territories with anti-deficit constraints have achieved better budgetary outcomes than those without constraints. This result may be attributed to the relatively rapid response of financial markets in encouraging fiscal discipline through increased borrowing costs for provincial governments that failed to correct a high and increasing level of indebtedness, as well as to the greater public tolerance for fiscal austerity at the subnational level of government.52
Evidence on the macroeconomic effects of the U.S. states’ balanced-budget rules is mixed. Between the early 1970s and mid-1980s and again at the beginning of the 1990s, fiscal performance seems to have been responsive to cyclical fluctuations in output—broadly reflected in a countercyclical stance—through the effect of either automatic stabilizers or of discretionary action. By comparison, in the 1960s and late 1980s, budget surpluses and output seemed to have been either unrelated or inversely related. The more recent trend may be explained, in part, as the consequence of devolution of fiscal responsibility to the lower levels of government and the increased ability of the states to draw on their stabilization funds set up from accumulated surpluses.53 Cyclical responsiveness depends, however, on the coverage of the state budget and on the choice of the cyclical indicator.54 More recent tests have confirmed that the fiscal rule limits budget flexibility, but this in turn does not appear to influence output variance.55 Although encouraging for proponents of balanced-budget rules, these findings are of limited relevance for national fiscal rules given the small weight of subnational governments relative to that of a federal government. A potentially useful lesson, however, is that binding balanced-budget rules can work in limiting fiscal deficits, though they sometimes result in an inefficient expenditure structure; moreover, with the accumulation of sufficient contingency reserves, they are not inconsistent with macroeconomic stabilization.
The track record of U.S. states also provides evidence of the sensitivity of interest costs to the level of indebtedness and to the stringency of the fiscal rule.56 Presumably, the latter can strengthen the market discipline that would be felt (for example, in the context of certain Canadian provinces) in the absence of rules. Again, however, it should be noted that the experience of subnational governments is limited by the much lower debt ratios than those found for sovereign borrowers at the national level. Ultimately, the market response to fiscal stance and indebtedness depends on the perception of whether the higher-level government or central bank will be ready to bail out the lower-level government in the event of default. Sufficiently binding fiscal rules would, of course, obviate such occurrences.57
The fiscal policy stance adopted under currency board arrangements in Argentina and Estonia in the early 1990s helped restore the credibility of financial policies. This in turn contributed to a pickup in investment and growth and to price stability in Argentina—a significant turnaround from the previous period of high inflation and stagnation—albeit at the cost of some external indebtedness.58 In Estonia, the policy rule led to the containment of output contraction and the deceleration of inflation, in contrast to the deep contraction and inflation spiral experienced by Russia and the other countries of the former Soviet Union.59 These episodes corroborate the significant gains that result from adherence to a strict monetary rule supported by an implicit balanced-budget rule.
In anticipation of the advent of EMU, most EU member countries launched, beginning in 1992, convergence plans to meet the deficit reference value by 1997 and to approach the debt reference value over the medium term. Although some EU members have made progress in reducing the public debt-GDP ratio (Belgium, Denmark, and Ireland), and most members have reduced the budget deficit relative to GDP, success in meeting these criteria is not always assured; further convergence appears difficult in a few cases, partly because of the slower-than-anticipated cyclical recovery in output. In a number of countries, the durability of the convergence is open to question insofar as the fiscal adjustment was accomplished through revenue enhancement rather than expenditure cuts.60 This is remarkable given the already high tax rates across much of Europe, which clearly cannot prevail indefinitely, as lower rates in some EU countries in the context of the single market are likely to result in competitive pressures to reduce rates in high-tax countries. Accordingly, the high rates of payroll contributions and weak financial position of social security institutions—also included in the definition of fiscal criteria—are probably not sustainable over the medium to long run.61 This problem is exacerbated by attempts to meet the deficit limit with reliance on one-off measures.62
Although it would be premature to determine the full macroeconomic consequences of the convergence, it appears that the fiscal adjustment may have been partly responsible for the recent slowdown in certain European countries. On the other hand, real interest rates have eased, especially in high-debt countries, such as Belgium and Italy—in anticipation of the final stage of EMU, as achievement of the fiscal criteria by the majority of members is deemed to be within reach—thus mitigating the slowdown.63 However, the immediate effect of the fiscal adjustment on output can only be inferred on the basis of studies of past adjustment episodes.64 An implication of these episodes is that the nature or composition of the adjustment under a fiscal rule, as in the case of a discretionary regime, can be critical for the macroeconomic repercussions of the rule. Indeed, its salutary effects on investment and growth can be enhanced through permanent cuts in transfers, government employment, and subsidies, rather than through public investment cuts, wage freezes, or tax increases, which tend to be transitory and contractionary. Thus, if the convergence had been supported by more efficient and durable structural reform measures, it could have contributed to a faster recovery in output (or to a milder recession) in some European economies.65
The experience of New Zealand in the aftermath of the recent introduction of fiscal rules has been, on the whole, favorable. Since 1994, the government has followed strictly and transparently the fiscal policy goals it had set for itself. The budget has been in surplus, net debt has declined, and net worth has turned positive. The high degree of fiscal discipline, accomplished through these rules, was accompanied by high growth and low inflation, though also by a widening in the external current account deficit in recent years.
The U.S. social security trust funds have accumulated sufficient contingency reserves for the Old-Age, Survivors’ and Disability Insurance (OASDI) programs, exceeding by a wide margin the minimum reserve ratio to annual benefit payments, whereas the Medicare funds are forecast to be depleted by the end of the decade. However, the reserve buildup has been effectively used to offset part of the federal budget deficits. Thus, social security surpluses have helped reduce the unified budget deficit, so that successive medium-term fiscal adjustment plans undertaken since the mid-1980s look more successful than they could have been without those surpluses.66
In Chile, the Copper Compensation Fund has been managed scrupulously over the last decade. Notwithstanding its symmetrical stabilization function, the fund has accumulated significant reserves and thus—along with the prohibition on central bank financing of government deficits—contributed to government surpluses and to macroeconomic stability.
Simulations
Given the paucity of real world experience with balanced-budget rules—particularly at the national level—and rules limiting government deficits and debt, it is necessary to complement an assessment of their likely macroeconomic effects with two types of fiscal adjustment experiments. The first draws on the existing literature to ascertain the likely effects of the fiscal adjustment, associated with implementation of binding fiscal rules, on interest rates and growth. The other consists of a set of stochastic simulations of the effect of various fiscal rules on the short-term variability of output.
Binding fiscal rules would be expected to have, in the first place, similar macroeconomic effects as those of any fiscal adjustment, with the quantity and quality of the effects on interest rates and growth reflected in fiscal multipliers (Appendix II). Furthermore, as the fiscal adjustment under such a rule is perceived as permanent, expected increases in future output, as well as expected declines in future interest rates and tax rates, tend to encourage present investment and consumption, while mitigating the negative withdrawal of demand. For the same fiscal shock, the multiplier can exhibit substantial variation (between negative and positive values), depending on expectations regarding the future course of policy. In general, multipliers tend to be smaller if the fiscal policy adjustment is gradual but credible. In this case, the positive credibility effects (such as lower long-term interest rates and higher future income growth) of a continuous fiscal reduction are anticipated and thus realized early in the adjustment.
Concerning the nexus between fiscal adjustment and interest rates, empirical estimates for industrial countries confirm that interest rates are sensitive to variations in the public debt-GDP ratio (Appendix II). Most researchers have found a positive relationship between interest rates and government debt, with estimates over a fairly wide range: on average, an increase in government debt equivalent to 25 percent of GDP would result in an increase of up to 500 basis points (but at least 125 points) in long-run interest rates. In any event, these results should be regarded as a lower bound of the beneficial long-run effects that may result from debt reduction arising from a permanent rule limiting government debt-GDP or deficit-GDP ratios.
A set of stochastic simulations were performed for the Group of Seven economies, on the basis of shocks derived from historical data, to ascertain the likely effect of various balanced-budget rules on the variability of macroeconomic performance (Appendix III). A basic premise underlying these simulations is that if a major source of the variability is irresponsible fiscal policy, then fiscal rules will tend to dampen macroeconomic fluctuations. If, on the other hand, output variability is the result of disturbances due to shocks other than fiscal policy, then fiscal rules will be procyclical and will tend to exacerbate these fluctuations.67
In general, the source of overall macroeconomic variability is a result of shocks to the behavioral equations of the model, as well as of fiscal policy, with allowance for automatic stabilizers to operate. The effect of fiscal rules on output variability is determined by the relative size and persistence of fiscal policy shocks compared with the size of the other underlying shocks, and by their interaction with automatic stabilizers. On the whole, for most countries, the variability in the baseline simulation is fairly close to the actual variability over the period 1974–95.
The simulations were conducted for several major policy rules: a strict balanced-budget rule, enforced alternatively through adjustment in government consumption, adjustment in transfers, and tax adjustment; a no-deficit target and a 3 percent of GDP budget deficit ceiling, with allowance for automatic stabilizers when in surplus or below the ceiling, respectively; and a debt target, with allowance for short-run debt accumulation and drawdown. All rules are stylized in that they are assumed to be immediately binding.
The results suggest that, except for the most rigid (symmetric) definition of the balanced-budget rule, the simulated rules add very little variability to output as compared with variability under the baseline simulation (Table 2). An important finding is that both the no-deficit target (similar to the proposed U.S. constitutional amendment) and the 3 percent of GDP deficit ceiling (consistent with the EMU reference value) provide a sufficiently comfortable margin to allow for the operation of automatic stabilizers. In addition to providing such flexibility, the debt target has the added advantage of preventing a possible accumulation of shortfalls over time, which may occur under a deficit ceiling.68 Therefore, none of the fiscal rules under consideration or being adopted in major industrial countries would appear to be significantly pro-cyclical and to exacerbate output fluctuations.
Output Variability Under Alternative Fiscal Rules
Source: Appendix III.
All rules are assumed to be binding; except for (2) and (3), the rules are assumed to be met through changes in government consumption.
Output Variability Under Alternative Fiscal Rules
| (Root-mean-square error, in percentage terms) | |||||||
|---|---|---|---|---|---|---|---|
| Policy Simulations1 | |||||||
| Balanced-budget target | |||||||
| Baseline simulation | Government consumption adjustment (1) | Transfer adjustment (2) | Tax adjustment (3) | No deficit (4) | Deficit ceiling (3 percent of GDP) (5) | Debt target (6) | |
| United States | |||||||
| GDP | 1.78 | 2.60 | 1.89 | 1.92 | 1.97 | 1.83 | 1.82 |
| Fiscal balance | 1.57 | — | — | — | 1.61 | 1.26 | 1.01 |
| Japan | |||||||
| GDP | 3.29 | 5.22 | 3.58 | 3.61 | 3.85 | 3.58 | 3.21 |
| Fiscal balance | 2.11 | — | — | — | 2.02 | 1.70 | 1.81 |
| Germany | |||||||
| GDP | 3.06 | 3.93 | 3.16 | 3.13 | 3.30 | 3.13 | 3.04 |
| Fiscal balance | 2.37 | — | — | — | 2.04 | 1.95 | 1.75 |
| Canada | |||||||
| GDP | 2.34 | 3.02 | 2.44 | 2.45 | 2.46 | 2.35 | 2.35 |
| Fiscal balance | 1.70 | — | — | — | 1.55 | 1.42 | 1.19 |
| France | |||||||
| GDP | 1.97 | 3.33 | 2.15 | 2.19 | 2.41 | 2.09 | 2.08 |
| Fiscal balance | 1.64 | — | — | — | 1.37 | 1.49 | 1.33 |
| Italy | |||||||
| GDP | 2.76 | 4.51 | 2.99 | 3.00 | 3.35 | 3.13 | 3.30 |
| Fiscal balance | 4.00 | — | — | — | 4.64 | 3.89 | 2.03 |
| United Kingdom | |||||||
| GDP | 3.33 | 5.94 | 3.78 | 3.81 | 4.44 | 3.93 | 3.69 |
| Fiscal balance | 2.44 | — | — | — | 1.87 | 1.91 | 2.04 |
Source: Appendix III.
All rules are assumed to be binding; except for (2) and (3), the rules are assumed to be met through changes in government consumption.
Output Variability Under Alternative Fiscal Rules
| (Root-mean-square error, in percentage terms) | |||||||
|---|---|---|---|---|---|---|---|
| Policy Simulations1 | |||||||
| Balanced-budget target | |||||||
| Baseline simulation | Government consumption adjustment (1) | Transfer adjustment (2) | Tax adjustment (3) | No deficit (4) | Deficit ceiling (3 percent of GDP) (5) | Debt target (6) | |
| United States | |||||||
| GDP | 1.78 | 2.60 | 1.89 | 1.92 | 1.97 | 1.83 | 1.82 |
| Fiscal balance | 1.57 | — | — | — | 1.61 | 1.26 | 1.01 |
| Japan | |||||||
| GDP | 3.29 | 5.22 | 3.58 | 3.61 | 3.85 | 3.58 | 3.21 |
| Fiscal balance | 2.11 | — | — | — | 2.02 | 1.70 | 1.81 |
| Germany | |||||||
| GDP | 3.06 | 3.93 | 3.16 | 3.13 | 3.30 | 3.13 | 3.04 |
| Fiscal balance | 2.37 | — | — | — | 2.04 | 1.95 | 1.75 |
| Canada | |||||||
| GDP | 2.34 | 3.02 | 2.44 | 2.45 | 2.46 | 2.35 | 2.35 |
| Fiscal balance | 1.70 | — | — | — | 1.55 | 1.42 | 1.19 |
| France | |||||||
| GDP | 1.97 | 3.33 | 2.15 | 2.19 | 2.41 | 2.09 | 2.08 |
| Fiscal balance | 1.64 | — | — | — | 1.37 | 1.49 | 1.33 |
| Italy | |||||||
| GDP | 2.76 | 4.51 | 2.99 | 3.00 | 3.35 | 3.13 | 3.30 |
| Fiscal balance | 4.00 | — | — | — | 4.64 | 3.89 | 2.03 |
| United Kingdom | |||||||
| GDP | 3.33 | 5.94 | 3.78 | 3.81 | 4.44 | 3.93 | 3.69 |
| Fiscal balance | 2.44 | — | — | — | 1.87 | 1.91 | 2.04 |
Source: Appendix III.
All rules are assumed to be binding; except for (2) and (3), the rules are assumed to be met through changes in government consumption.