Chapter

V Treatment of Inflation-Induced Changes in the Budget Balance

Author(s):
Ahsan Mansur, Richard Haas, and Peter Heller
Published Date:
May 1986
Share
  • ShareShare
Show Summary Details

This section provides a conceptual and empirical analysis of the implications of alternative approaches for adjusting the conventional measures of budget balance and fiscal impulse for the impact of inflation on the value of real debt and government interest payments.27

Conceptual Issues

Inflation has several effects on the government budget. Aside from its impact on revenues and non-interest expenditure, an acceleration of inflation is likely ultimately to increase government outlays on interest, as interest rates are pushed up and as investors try to ensure an adequate real rate of return on new lending to the government. However, inflation also allows the government to realize a percentage reduction in its outstanding real liabilities that is greater than the share of government debt in longer-term maturities.

The overall impact of higher inflation on the government’s debt-service expenditure will be a composite of four effects: (1) to the extent that interest rates on existing debt are fixed in nominal terms, real interest payments and, ceteris paribus, real public spending will be reduced through inflation; (2) higher inflation or the expectation of higher inflation in the future will lead to higher nominal interest payments on new and refinanced debt issues; (3) if counterinflationary monetary policies lead to an increase in real interest rates, the interest component of the budget may be raised further in the medium term; and (4) inflation implies an implicit amortization payment equivalent to the reduction in the real value of outstanding liabilities.

Should such effects be considered in appraising the degree of stimulus provided by fiscal policy in an inflationary period? Conceptually, one motivation for adjustment for these effects reflects the argument that there is an absence of money illusion in the private sector, such that the reaction of individuals’ consumption to a real change in their net worth would be the same as their reaction to a wealth tax of an equivalent amount, in terms of the effect on aggregate demand.28 In a similar vein, it is argued that to the extent nominal interest rates include an “amortization component,” private wealth holders view this component of interest payments as simply an accelerated repayment of principal, and replenish their holdings of wealth in real terms without affecting private sector income and expenditure.29 Do people in fact have higher marginal propensities to save out of such interest income? Is their consumption behavior changed by the effects of inflation on the real value of wealth in the same way it would be changed by a tax on wealth? These assumptions regarding private sector behavior are clearly hypotheses that have not been firmly established through empirical tests. They may not accurately reflect the way in which people act. If they do, then significant issues arise as to the appropriateness of the current measure of the budget balance for an appraisal of the direction of fiscal policy.

The conventional measure of the government budget balance reports the current cash-flow position of the government, which is obtained by subtracting the cash value of disbursements from the cash value of receipts. Expenditures include interest payments but not repayments of principal; the latter are placed below the line to obtain a net measure of government borrowing. Inter alia, this definition is limited in its coverage, in the sense that it ignores the changes in the government’s net worth arising from inflation as well as from changes in market interest and exchange rates.30 This is true even when expenditure and revenue are evaluated at constant prices. Gains or losses caused by changes in the real value of outstanding public sector debt are generally not included in the flow of funds.31 A decline in the real value of outstanding public debt induced by inflation is conceptually equivalent to a “windfall” gain to the government. In a wealth-accounting framework, such gains would accrue to revenue, thereby lowering the conventional measure of the deficit or raising the corresponding measure of the surplus.

The present fiscal impulse measure also assumes that any increase in the measured fiscal deficit, at a given level of output and from whatever source, may provide a short-term stimulus to economic activity. Thus, an inflation-induced increase in the nominal interest rate on new public debt may increase interest payments and the budget deficit as well as appearing to stimulate economic activity. The effect of inflation on interest payments may be substantial. If most debt were short term, and if the inflation rate were to rise from zero to 10 percent (raising the effective nominal interest rate on outstanding debt from say, 3 percent to 13 percent), interest payments could more than quadruple. This would be in contrast to other forms of expenditure, which might increase by only 10 percent as a consequence of the inflation. Buiter (1983a) has argued that the component of interest payments that reflects compensation for inflation should be treated as the equivalent of an amortization payment, thus maintaining the real value of the government debt instrument to its holder. Under that interpretation, measuring the deficit inclusive of unadjusted gross interest payments tends to overstate government expenditure and to understate amortization payments.

One approach to adjusting the budget deficit for inflation is based on the “purchasing power accrual accounting” (PPAA) concept.32 This concept requires the recording of all assets and liabilities at their “market” values and expressing the balance sheet figures for all time periods in the same units (i.e., using appropriate historical deflators to measure the changes in purchasing power of a currency).33 It is justified by assuming that economic units are rational and free of money illusion. Other partial adjustments for inflation have been proposed in the literature of the United States, Canada, and the United Kingdom. One such concept of budget balance is the inflation-adjusted government financing requirement, according to which an inflation adjustment is applied to interest payments on the outstanding government debt. This measures the government’s financial balance in nominal terms and without considering stock-flow consistency in the framework of appropriate wealth accounting.

Alternative Approaches to Adjusting the Budget Balance for Inflation

Consider an economy with outstanding government debt D in the form of bonds denominated at their nominal values.34 If the government budget is in deficit, the difference between expenditure G and revenue T can be financed by floating new bonds (ΔD) or by creating money. Where the deficit is fully financed by bonds,

However (ΔD/P), which is defined as the real government deficit, and Δ(D/P) are not stock-flow consistent, where P denotes the price index. This is obvious from the relationship

where π = ΔP/P = the inflation rate.

Equation (13) shows that the change in the real value of outstanding liabilities equals the conventional measure of the deficit in real terms less the rate of inflation times the real value of outstanding debt less a small cross-product term. The term [– π (D/P) – π (ΔD/P)] is conceptually equivalent to the amortization of public debt through inflation. In a real-wealth-accounting framework, the effective amortization payments implied by inflation should be put below the line in measuring the net “real” government borrowing requirement.35 Above the line, one would conceptually have some offsetting adjustments in the form of an inflation-induced windfall gain equivalent to a wealth tax on the revenue side. To the extent that the higher inflation has been reflected in increased interest payments on new and refinanced debt, this would also, of course, be reflected in observed government expenditure. The smaller the share of short-term debt in the government’s outstanding debt, the more important will be the “gain” to the government arising from inflation. In summary, the financial balance, stated in real terms and reflecting the inflation adjustment would be

The nature of the overstatement of the budget deficit can be illustrated by noting that the deficit should measure the year-to-year change in the real outstanding government debt when all of the deficit is financed by borrowing. This may be illustrated by using a hypothetical example

Measured in 1980 U.S. dollarsMeasured in 1981 U.S. dollars36
Total net government debt at end of 1981100.0115.0
Total net government debt at end of 198090.0103.5
Real increase in debt10.011.5
Percentage increase in real debt11.111.1

The numbers in the columns are mutually consistent and show the real change in the outstanding government debt between 1980 and 1981. The numbers depend on the monetary units in which these are expressed. However, both numbers are very different from the conventionally measured deficit of 25.0 (= 115.0 – 90.0).

In nominal terms, the financial balance adjusted for inflation would be

An approximation of this measure, the “financial deficit adjusted for inflation,” B′, has been used by the Bank of England37

Comparing B′ with the right-hand side of equation (15), it can be seen that the only difference is the cross-product term π ΔD, which is presumed to be a second-order effect.

A slightly more sophisticated variation of this inflation-adjusted approach has been used by the U.S. Department of Commerce, where the nominal interest rate has been econometrically related to the rate of inflation and other factors (de Leeuw and Holloway (1982)); as interest rates tend to reflect changes in inflation with a lag, the interest rate effect works in the opposite direction from the effects of inflation on the real value of outstanding debt, as noted above. Using the regression estimates, a certain component of the interest rate is assumed to be an inflation-related component and is used to adjust the deficit in a manner analogous to equation (16). This approach is not used in this study.

One consequence of using the inflation-adjustment method in equation (16) is that the magnitude of the adjustment is particularly sensitive to fluctuations in the inflation rate. When a large proportion of the debt is of medium-term to long-term maturity, these fluctuations will have a significant effect on the adjusted budget balance.

One simple approach to smoothing out erratic movements in the adjustments due to inflation may be to assume a historically reasonable long-term (ex ante) real rate of interest, ȓ (e.g., 2 or 3 percent) and to calculate the difference between the total measured interest payments and ȓ percent of the outstanding market value of government debt. This approach may be called the “real-interest-adjustment” method. It can be shown that the difference between total interest payments and the product of the assumed real rate of interest and the market value of the debt will equal the product of a measure of the average rate of inflation over the term of the public debt times its market value. This method will reduce the volatility of the adjustment factor. Rather than measuring the decline in the real value of government debt in the current period, it measures the average trend decline resulting from inflation.

Perhaps the most obvious problem with this latter approach is the difficulty of ascertaining what the appropriate real interest rate is. The assumption of a constant real interest rate seems also at variance with the fluctuations in the observed real interest rate in recent years. However, the actual variability may in fact be less than observed for two reasons. First, if one corrects for the effects of taxation, it is likely that the variability in the after-tax real interest rate is considerably less. Second, it could be argued that recent increases in real interest rates simply reflect a widened gap between the actual and expected inflation rates, such that bondholders continue to assume a higher implicit amortization component in actual interest payments than would be implied by the actual inflation rate.

Estimates of inflation-induced changes in the real value of outstanding government debt appear to be very large for most of the major industrial countries owing to the high rate of inflation in the recent past. This may be seen in Table 7, which contrasts interest payments on the outstanding central government debt, the change in the nominal value of outstanding public debt owing to inflation obtained using equation (16), and the real-interest-based adjustment factor (the amount by which interest payments exceed “real” interest payments), all as percentages of GDP. Lacking data on the market value of debt, our estimate of the realinterest-adjustment method perforce uses an estimate of the nominal debt value.

Table 7.Seven Major Industrial Countries: Comparison of Alternative Adjustments for the Interest Impact of Inflation on the Central Government Budget Balance, 1975–85(As percentages of GDP)
19841985
Country197519761977197819791980198119821983Estimated
Canada
Interest payments12.22.42.42.83.13.34.14.74.54.44.3
Inflation-adjustment factor21.51.31.42.22.62.52.72.01.71.8
Interest payments less inflation adjustment factor0.91.11.40.90.71.62.02.52.72.5
Real-interest-based adjustment factor31.71.91.92.12.42.63.33.83.53.33.1
United States
Interest payments11.71.91.82.02.22.63.13.53.74.04.3
Inflation-adjustment factor21.41.61.92.22.42.31.71.41.41.5
Interest payments less inflation adjustment factor0.50.21.00.20.81.82.32.62.8
Real-interest-based adjustment factor31.01.11.01.21.41.82.32.62.73.03.2
Japan
Interest payments10.60.81.11.31.51.92.22.62.93.13.2
Inflation-adjustment factor20.80.91.00.70.90.90.80.30.90.9
Interest payments less inflation adjustment factor0.20.30.81.01.31.82.62.22.3
Real-interest-based adjustment factor30.20.40.60.60.71.01.21.41.61.71.8
France
Interest payments10.90.80.80.91.01.01.51.61.71.71.7
Inflation-adjustment factor21.41.31.31.41.71.81.81.51.21.3
Interest payments less inflation adjustment factor−0.6−0.5−0.4−0.4−0.7−0.3−0.20.20.50.4
Real-interest-based adjustment factor30.40.40.40.40.60.61.11.11.21.21.2
Germany, Fed. Rep. of
Interest payments10.60.70.80.70.80.91.21.41.61.61.6
Inflation-adjustment factor20.40.40.50.50.70.70.90.60.60.6
Interest payments less inflation adjustment factor0.30.40.20.30.20.50.51.01.00.9
Real-interest-based adjustment factor30.30.40.40.40.50.70.81.01.01.0
Italy
Interest payments11.72.54.25.35.35.76.88.28.59.2
Inflation-adjustment factor27.88.57.18.410.49.810.19.99.9
Interest payments less inflation adjustment factor−5.3−4.3−1.8−3.1−4.7−3.0−1.9−1.4−0.7
Real-interest-based adjustment factor30.31.12.83.73.64.05.16.36.46.9
United Kingdom
Interest payments13.33.53.43.43.63.94.34.33.93.93.9
Inflation-adjustment factor24.34.43.54.55.83.82.51.81.91.8
Interest payments less inflation adjustment factor−0.8−1.0−0.1−0.9−1.90.51.82.12.02.1
Real-interest-based adjustment factor32.22.52.42.42.62.93.23.22.82.82.7
Source: Fund staff estimates as of February 29, 1984.

On outstanding central government debt.

The inflation rate multiplied by the midyear level of outstanding debt.

Gross interest payments less the product of outstanding government debt times the assumed long-term rate of interest (assumed to be 3 percent).

Source: Fund staff estimates as of February 29, 1984.

On outstanding central government debt.

The inflation rate multiplied by the midyear level of outstanding debt.

Gross interest payments less the product of outstanding government debt times the assumed long-term rate of interest (assumed to be 3 percent).

The effects of the rise and subsequent decline in the inflation rate are mirrored in the inflation-adjustment factor, which peaks for most countries in the high-inflation period of 1979–82; sometimes these adjustments exceed total gross interest payments on outstanding government debt. The increase in effective real interest payments for the period extending from 1982 to 1985 is also quite clear, as is illustrated by the difference between total interest payments and the inflation-adjustment factor. Use of the real-interest-adjustment method leads, as one would expect, to a more stable trend in the adjustment factor in most countries, as it fluctuates in reaction to movements in the outstanding government debt, which is generally very stable, and is less dependent on the rate of inflation. Table 8 provides estimates of budget deficits that are adjusted by the alternative inflation factors, as well as exclusive of interest payments. In some years, the adjusted budget balance is in surplus for some countries, even without cyclical adjustment.

Table 8.Seven Major Industrial Countries: Comparison of Conventional and Inflation-Adjusted Central Government Budget Balances, 1975–851(As percentages of GDP)
19841985
Country197519761977197819791980198119821983Estimated
Canada
Budget deficit excluding interest0.1−0.61.11.80.3−2.01.12.21.5
Budget deficit less inflation-adjustment factor0.32.23.21.20.7−0.43.14.74.22.5
Budget deficit less real-interest-adjustment factor0.6−0.11.62.51.00.7−1.22.03.22.61.2
Actual budget deficit2.31.83.54.63.43.32.15.86.75.94.3
United States
Budget deficit excluding interest3.21.40.9−1.0−0.2−0.60.82.41.40.7
Budget deficit less inflation-adjustment factor1.91.10.1−1.00.22.64.83.93.4
Budget deficit less real-interest-adjustment factor3.92.31.70.8−0.20.60.21.73.42.41.9
Actual budget deficit4.93.32.72.01.22.42.54.36.15.34.9
Japan
Budget deficit excluding interest3.74.24.04.04.74.23.72.92.21.81.0
Budget deficit less inflation-adjustment factor4.24.24.35.55.25.04.74.84.03.3
Budget deficit less real-interest-adjustment factor4.14.64.54.75.55.14.74.13.53.22.4
Actual budget deficit4.35.05.15.36.26.15.95.55.14.94.2
France
Budget deficit excluding interest1.70.40.21.70.50.11.11.21.21.31.2
Budget deficit less inflation-adjustment factor−0.2−0.31.30.1−0.60.81.01.41.81.6
Budget deficit less real-interest-adjustment factor2.20.80.62.20.90.51.51.71.71.81.7
Actual budget deficit2.61.21.02.61.51.12.62.82.93.02.9
Germany, Federal Republic of
Budget deficit excluding interest3.02.11.41.41.00.81.00.50.4−0.1−0.6
Budget deficit less inflation-adjustment factor2.41.81.61.31.01.51.01.40.90.3
Budget deficit less real-interest-adjustment factor2.51.81.71.41.21.51.11.00.5
Actual budget deficit3.62.82.22.11.81.72.21.92.01.51.0
Italy
Budget deficit excluding interest6.35.14.710.55.85.26.16.88.27.9
Budget deficit less inflation-adjustment factor1.30.57.52.70.53.15.06.97.4
Budget deficit less real-interest-adjustment factor8.76.57.212.27.56.97.88.710.310.3
Actual budget deficit10.79.19.014.611.110.912.915.116.817.3
United Kingdom
Budget deficit excluding interest4.62.0−0.31.61.61.0−0.2−1.41.0−0.5−1.0
Budget deficit less inflation-adjustment factor1.2−1.31.50.8−0.90.30.43.11.51.1
Budget deficit less real-interest-adjustment factor5.73.00.72.62.72.00.9−0.32.10.60.2
Actual budget deficit7.95.53.15.05.34.94.12.94.93.42.9
Source: Fund staff estimates as of February 29, 1984.

Positive figures are deficits, and negative figures are surpluses.

Source: Fund staff estimates as of February 29, 1984.

Positive figures are deficits, and negative figures are surpluses.

Net or Gross Adjustment

In making the above adjustments for inflation, one operational issue that arises is the appropriate measure of outstanding debt or interest payments to use. Should one use gross or net outstanding debt? Interest payments net or gross of interest receipts?38 The above analysis has used gross measures of interest and debt. The argument for using a net measure is that the government should gain or lose symmetrically in comparison with the private sector with respect to the effects of inflation on its own holdings of financial assets. While it is true that the central government’s interest receipts are not negligible in some of the countries, a distinction needs to be drawn between interest receipts derived from loans for resource allocation (net lending) as opposed to receipts derived from the holding of financial assets for portfolio (investment) purposes. Presumably a government is sensitive to capital losses caused by inflation in the latter case and is less so in the former.

Even if one accepts the argument that the government treats the inflation-related component of interest receipts (or the effects of inflation on the real value of its assets) as if it were a form of amortization payment, one would have to distinguish between the amortizations on the two types of asset. It can be argued that in its net lending for resource allocation, the government attaches less significance to the value of the corresponding financial assets than it would if it were holding them as investments.

This suggests that to make these adjustments, one should net out only interest receipts on financial assets that are related to portfolio holdings. While these holdings are likely to be negligible at the central government level,39 in some of the industrial countries, lower levels of government may hold significant financial assets (either liabilities of the central government or of the private sector) related to pension schemes for their employees. This suggests that if the inflation adjustments are to be made at the level of the general government, one would want to use a net approach, excluding that component of general government interest receipts derived by the lower levels of government. If the data on general government interest payments and receipts, or on outstanding debt, were consolidated for intra-general government transactions, there would remain only the problem of ascertaining the amount of interest receipts derived from holdings of private sector assets for portfolio purposes.

Incorporating Inflation-Adjustment Procedures into the Fiscal Impulse Methodology

If the budget balance were adjusted for the effects of inflation, how would one incorporate such an adjustment into the methodology for estimating the fiscal impulse?

The decision to adjust the fiscal impulse measure ultimately reflects some assumptions on the responsiveness of the private sector to a reduction in the real value of its net worth held as claims on the government. To the extent that the current fiscal impulse measure eschews any attempt at assigning weights for the relative impact on aggregate demand of different components of government revenue and expenditure, it is unclear whether the implicit multiplier effects associated with these inflation adjustments should be taken account of. Similarly, should one attach equal significance to the effects of an inflation tax and those of an explicit tax on wealth?

Also, should one attribute to the fiscal impulse measure the effects of an increase or a decrease in inflation, independent of data on the source of such inflationary pressures? Should one make the same adjustment for inflation regardless of whether the inflation arises from an independently expansionary monetary policy, exogenous price shocks, or an excessively expansionary fiscal policy? If one asserts that the adjusted balance is a better measure of the government’s budgetary position in relation to the rest of the economy, abstracting from changes in net wealth position, then shifts in that adjusted budgetary position are what should be examined to measure the initial thrust (or lack thereof) imparted by fiscal policy.

Empirically, the issues that are posed can be seen by reviewing the basic formula used to estimate the fiscal stance in equation (2)40

Incorporation of an inflation adjustment into this formula implies the necessity of adjusting both the observed budget balance B and the cyclically neutral balance Bn. Assume that ΔV represents the absolute adjustment for the effects of inflation, using one of the approaches described above. The adjusted actual budget balance B” would equal

The question remains as to what the adjustment for Bn should be.

One approach would be to define the adjusted cyclically neutral balance Bn′′ as

In effect, one defines a base-year expenditure parameter, g0ʺ in terms of the share in GDP of total expenditure less the adjustment for inflation that would have been made in the base year—namely, ΔV0.

A problem with this approach may be that interest payments adjusted for inflation-induced changes are negative or very small in the base year. This can affect the final outcome by reducing the value of the cyclically neutral budget deficit. The magnitude of this bias may be significant if the base year happens to be a year of high inflation, so that real interest payments are negative, in the sense that adjustments owing to inflation are larger than gross interest payments, as was the case with France, Italy, and the United Kingdom in 1978 (Table 7).

In Table 9, measures of fiscal stance and impulse for the central government have been calculated according to the two alternative inflation-adjustment procedures, and the results are compared with the unadjusted results of the Fund’s World Economic Outlook exercise. The results suggest that the adjustment for inflation will clearly affect the observed fiscal impulse, and the type of adjustment procedure used will determine the effect of the adjustment. Specifically, with an increase in the inflation rate, the effect of the full inflation-adjustment procedure will be to increase the adjusted fiscal balance, ceteris paribus, by a greater amount than the previous year’s adjustment, imparting a contractionary bias to the observed fiscal impulse measure, relative to that obtained without the inflation adjustment. In a period when inflation is receding, the opposite result will be obtained: in effect, the government derives lesser “gains” from the effects of inflation relative to the previous year, which implies a lower relative fiscal balance and a more expansionary position. This phenomenon is observed for the United States and Canada; in the United States, a more contractionary impulse is observed during 1978–80, and a more expansionary impulse in 1982 and 1983. A similar result obtains for the United Kingdom.

Table 9.Seven Major Industrial Countries: Comparison of Central Government Fiscal Stance and Impulse, With and Without Allowance for Inflation Adjustments, 1978–85(As percentages of GDP)
19841985
Country197819791980198119821983Estimated
Canada
Fiscal stance
WEO1−0.65−1.03−2.12−0.76−0.05−0.13−1.32
Inflation-adjustment factor used−1.54−2.19−3.24−2.01−0.55−0.41−1.69
Real-interest-based adjustment used−0.90−1.47−3.29−2.30−1.28−1.21−2.21
Impulse
WEO10.98−0.65−0.37−1.091.360.71−0.08−1.19
Inflation-adjustment factor used0.92−1.54−0.66−1.051.241.460.14−1.28
Real-interest-based adjustment used0.74−0.90−0.58−1.810.991.020.06−1.00
Inflation rate (in percent)6.7010.3011.1010.6010.106.204.904.70
United States
Fiscal stance
WEO1−0.79−0.45−0.290.232.262.192.04
Inflation-adjustment factor used−1.04−0.80−0.630.593.002.872.54
Real-interest-based adjustment used−1.03−0.98−1.37−1.110.870.590.26
Impulse
WEO1−0.790.350.160.522.03−0.07−0.15
Inflation-adjustment factor used−0.38−1.040.240.181.222.41−0.13−0.33
Real-interest-based adjustment used−0.20−1.030.05−0.390.261.97−0.28−0.33
Inflation rate (in percent)7.408.609.309.206.104.203.904.10
Japan
Fiscal stance
WEO11.041.040.730.09−0.36−0.59−1.27
Inflation-adjustment factor used1.351.150.830.320.37−0.44−1.18
Real-interest-based adjustment used0.960.710.19−0.66−1.35−1.68−2.41
Impulse
WEO10.211.04−0.01−0.31−0.64−0.45−0.23−0.68
Inflation-adjustment factor used0.121.35−0.21−0.31−0.52−0.05−0.81−0.74
Real-interest-based adjustment used0.130.96−0.25−0.52−0.85−0.70−0.33−0.72
Inflation rate (in percent)4.602.602.902.602.000.702.002.00
France
Fiscal stance
WEO1−0.85−1.53−0.57−0.55−0.87−1.21−1.44
Inflation-adjustment factor used−0.98−1.88−0.99−0.98−0.97−1.01−1.27
Real-interest-based adjustment used−1.00−1.66−1.19−1.18−1.58−1.94−2.11
Impulse
WEO11.87−0.85−0.680.960.02−0.32−0.34−0.23
Inflation-adjustment factor used1.78−0.98−0.900.890.010.02−0.04−0.26
Real-interest-based adjustment used1.84−1.00−0.660.470.01−0.40−0.36−0.17
Inflation rate (in percent)9.5010.3011.8012.3012.409.507.207.10
Germany, Federal Republic of
Fiscal stance
WEO1−0.38−1.07−2.82−2.93−3.14−3.21
Inflation-adjustment factor used−0.02−0.51−1.20−3.13−2.99−3.19−3.33
Real-interest-based adjustment used−0.04−0.53−1.39−3.29−3.58−3.78−3.85
Impulse
WEO10.12−0.38−0.69−1.75−0.11−0.21−0.07
Inflation-adjustment factor used0.02−0.02−0.49−0.69−1.920.13−0.20−0.13
Real-interest-based adjustment used0.16−0.04−0.48−0.86−1.90−0.28−0.20−0.07
Inflation rate (in percent)4.204.004.504.204.803.203.003.20
Italy
Fiscal stance
WEO1−2.98−3.04−2.45−1.80−1.87−1.97
Inflation-adjustment factor used−4.37−6.44−5.03−4.43−3.94−3.92
Real-interest-based adjustment used−3.11−3.48−3.82−4.274.23−4.82
Impulse
WEO15.30−2.98−0.070.590.65−0.07−0.09
Inflation-adjustment factor used6.80−4.37−2.111.460.600.490.02
Real-interest-based adjustment used4.43−3.01−0.47−0.34−0.450.04−0.59
Inflation rate (in percent)13.9015.9020.7018.4017.5015.1013.7013.90
United Kingdom
Fiscal stance
WEO10.53−1.83−4.29−5.53−3.11−4.32−4.87
Inflation-adjustment factor used−0.52−3.98−4.28−4.26−1.18−2.43−2.25
Real-interest-based adjustment used0.31−2.25−4.95−6.17−3.39−4.54−5.04
Impulse
WEO12.740.53−2.36−2.46−1.252.43−1.22−0.54
Inflation-adjustment factor used3.53−0.52−3.47−0.300.023.08−1.26−0.51
Real-interest-based adjustment used2.740.31−2.56−2.71−1.212.78−1.15−0.50
Inflation rate (in percent)10.8014.5019.8011.607.205.105.004.90
Source: Fund staff estimates as of February 29, 1984.

Based on the results of the Fund’s April 1984 World Economic Outlook (WEO) exercise.

Source: Fund staff estimates as of February 29, 1984.

Based on the results of the Fund’s April 1984 World Economic Outlook (WEO) exercise.

Similarly, the results obtained using the real-interest based adjustment method are sensitive to the implied expected inflation rate, as measured by the difference between the average interest rate on outstanding debt (which is itself affected by the mix of short-term and long-term debt) and the observed historical interest rate. If the implied expected inflation rate is increasing, the adjustments will increase similarly, thus increasing the fiscal balance and imparting a more contractionary (or a less expansionary) impulse; conversely, decreases in the implied expected inflation rate between two periods will, ceteris paribus, lead to a more expansionary impulse than would otherwise be observed using the unadjusted fiscal balances. If the implied expected inflation rate closely parallels the actual inflation rate, the two inflation-adjustment measures will impart a similar bias to the observed impulse estimates. If, however, the implied expected inflation rate is lower than the actual inflation rate, perhaps owing to a large share of long-term debt in total outstanding debt or to a fall in the “true” real interest rate, the adjusted fiscal balance obtained using the real-interest-based approach will be lower than that obtained using the inflation-adjusted approach; movements in this differential across time will thus lead to different biases in the adjusted impulse measure, depending on the choice of adjustment procedure. The differences can be quite marked, leading one adjusted impulse measure to suggest a more expansionary position, and another a more contractionary one, relative to the unadjusted impulse (as was the case in both the Federal Republic of Germany in 1983 and the United Kingdom in 1981).

    Other Resources Citing This Publication