Chapter

VII Public Sector Finances: Balance Sheets, Financing Needs, and Sustainability

Author(s):
Eliot Kalter, Steven Phillips, Manmohan Singh, Mauricio Villafuerte, Rodolfo Luzio, and Marco Espinosa-Vega
Published Date:
June 2004
Share
  • ShareShare
Show Summary Details
Author(s)
Steven Phillips

This section examines the financial position of the Chilean public sector. The focus is on analysis of balance sheet information, taking account of financial assets as well as liabilities, and of the structure of balance sheets, in terms of foreign exchange position and liquidity. The analysis is facilitated by recent enhancements of the available information, especially by the Public Debt Report issued for the first time in late 2002.1

The analysis begins at the central government level, where account is taken of the government’s structural balance target and is then broadened to consider other parts of the public sector: in turn the central bank and the public enterprises. Significantly, subnational governments in Chile are subject to strict budget constraints (see Section III) and indeed have no financial debt. Accordingly they are not analyzed here.

The central government has been in the past, and will likely continue to be, the key to movements in Chile’s total public debt. Still, attention is due also to the central bank accounts, because the bank’s total debt is significant (as are its assets), and because much of the central government’s debt is owed to the central bank. For some purposes, it is useful to analyze the consolidated central government/central bank position.

The assessment of the public sector finances is favorable. Since risky balance sheet structures have been avoided, exposure to currency and interest rate risks is limited. Taking account of the government’s structural balance target, it is difficult to see debt sustainability problems emerging, as long as this target (or other restrained fiscal policy) is implemented. The central bank’s balance sheet is also examined, noting its considerable strengths in terms of foreign exchange and liquidity positions, but also its tendency to imply an operating deficit. Though the bank’s deficit has been fairly stable and has not interfered with its monetary policy objectives, its situation is not ideal, and some steps, including a capital injection from the government, that could be taken are noted. Less complex is the situation of the public enterprises: their aggregate position appears sound, in light of their overall profitability and limited indebtedness. On the whole, the public enterprise sector is an asset rather than a drain on government finances.

Central Government Finances

Significant aspects of the central government’s balance sheet are the following:

  • Substantial financial assets, about 5 percent of GDP in 2002, mostly claims on the private sector. (In recent years, the drawdown of such assets has played an important role, financing about half of the government’s deficit.)

  • Gross debt is relatively low, about 16 percent of GDP in 2002.

  • Debt to the domestic private sector is zero.

  • Almost two-thirds of the government’s gross debt is owed to the central bank. This old debt was issued in 1983 to compensate for the balance sheet effect of a financial sector bailout (Section II); it is mostly denominated in U.S. dollars.

  • The remainder is external debt, also mainly denominated in U.S. dollars. More than half of this debt consists of medium- or long-term sovereign bonds, all issued within the last several years. Most other external debt is owed to international institutions.

  • The average interest rate is quite low. For example, the ratio of interest payments to government debt was only about 2 percent in 2002 (and about 4 percent over the previous 10 years).

Financing Needs

Gross financing needs can be expected to remain moderate:

  • The government does not face a steep amortization schedule, as the average maturity of its debt is evidently rather long. Thus from an existing debt stock of 16 percent of GDP at end-2002, amortization will be only 1 to 1 ½ percent of GDP annually during 2003–06 (and will rise only moderately thereafter).2

  • The fiscal deficit is modest, not more than 1 percent of GDP for 2003, and is expected to decline in the next few years.

  • A further consideration is that the government has some room to further reduce its financial assets.

Exposure to Currency Depreciation

Currency depreciation does not pose a major risk to the public finances. While most of the government’s financial assets now are denominated in domestic currency, nearly all of its debt is denominated in U.S. dollars. Considering only these two quantities, the government’s net foreign exchange position would appear to have been about -14 percent of GDP at end-2002. Taking a broader view, however, the picture improves:

  • Account needs also to be taken of the stream of foreign currency earnings the government receives, mainly from its ownership of the copper company CODELCO. Though this income varies with world copper prices, in most years, the government’s primary balance on foreign-denominated flows is positive.

  • Most of the government’s U.S. dollar debt—and thus most of its dollar-denominated interest bill also—is owed to the central bank. In turn, the central bank has a substantial positive net foreign exchange position (as discussed in the next section, about +23 percent of GDP at end-2002).

Exposure to Interest Rate Fluctuations

As a consequence of the structure of its debt, the government has little near-term exposure to interest rate shocks. In fact, the government is nearly unaffected by domestic interest rate fluctuations. The avoidance of short-term external debt means there is also little near-term exposure to changes in the risk premium international investors apply to Chilean sovereign debt.

The government’s interest rate exposure is largely confined to fluctuations in world interest rates. This is because interest on the government’s old debt to the central bank is tied to the London interbank offered rate.

Sustainability of Central Government Debt

Debt sustainability analysis for Chile must take into account the government’s commitment to an on-going target for its structural balance (see Section II). As long as this practice is maintained, it is difficult to see problems of debt sustainability emerging, in light of two basic considerations:

  • As long as the structural balance rule is followed, debt dynamics are unaffected by a rise in interest rates on government debt: since the target refers to the entire government balance, any increase in the interest bill would have to be offset by a policy reaction to tighten the government’s primary balance.

  • Most important to debt sustainability is the level chosen for the targeted structural balance: a surplus of 1 percent of GDP. It is true that this figure does not take account of the central bank’s deficit (discussed below), and that if a broader view of the public sector were taken, the effective level of the target would not be quite as strong; however, it still would not represent a deficit. Thus the government debt/GDP ratio could well be expected to decline over time even if economic growth were zero.

There is one significant complication to this picture: the fiscal target refers to the government’s structural, not actual, balance. Whereas the target refers to a cyclically adjusted balance, the evolution of government debt of course depends on the actual balance. Therefore, the structural balance rule does not alone determine the exact path of public debt, and further analysis is needed.

Under the rule holding the structural balance constant, the path of public debt depends not only on the chosen level of the fiscal target, but also on the estimated cyclical adjustments. More precisely, given Chile’s fiscal rule, projecting government debt requires projecting not only actual output and copper export prices, but also (the estimates of) potential GDP and the reference copper export price, which will together determine how far the actual government balance deviates from the level of the structural balance target.

Table 7.1 illustrates how the central government gross debt ratio would evolve under various combinations of actual and (estimated) potential output paths. The starting point is the baseline scenario in which the output gap, currently estimated about 5 percent, would close gradually over the medium term. In that scenario, the debt/GDP ratio would decline from 16 percent to 8 percent of GDP by 2008. Several alternative scenarios demonstrate that some rather unlikely assumptions would be needed to generate a rising debt/GDP ratio:

Table 7.1.Sensitivity of Government Debt Projections Under the Structural Balance Rule
2002200320042005200620072008
Structural balance (percent of GDP)0.81.01.01.01.01.0
Baseline scenario
GDP growth, actual (percent)3.34.55.25.55.04.8
GDP growth, potential (percent)3.73.73.84.04.24.5
Output gap (deviation from potential, percent)‒4.8‒4.0‒2.7‒1.2‒0.4‒0.1
Actual balance (percent of GDP)‒0.8‒0.10.40.70.80.9
Government gross debt (percent of GDP)15.915.614.513.011.39.68.0
Alternative scenarios
1. Baseline, except: higher estimate of potential output
GDP growth, actual (percent)3.34.55.25.55.04.8
GDP growth, potential (percent)3.76.47.05.55.04.8
Output gap (deviation from potential, percent)‒4.8‒6.5‒8.0‒8.0‒8.0‒8.0
Actual balance (percent of GDP)‒0.6‒0.6‒0.7‒0.7‒0.7‒0.6
Government gross debt (percent of GDP)15.915.414.814.313.913.513.2
2. Baseline, except low actual growth, with some downward revision to estimate of potential growth
GDP growth, actual (percent)3.32.02.02.02.02.0
GDP growth, potential (percent)3.73.73.82.22.02.0
Output gap (deviation from potential, percent)‒4.8‒6.3‒7.9‒8.0‒8.0‒8.0
Actual balance (percent of GDP)‒0.6‒0.6‒0.7‒0.7‒0.7‒0.7
Government gross debt (percent of GDP)15.915.415.115.015.015.014.9
3. Baseline, except: low actual growth, with no downward revision to potential output estimate
GDP growth, actual (percent)3.32.02.02.02.02.0
GDP growth, potential (percent)3.73.73.84.04.24.5
Output gap (deviation from potential, percent)‒4.8‒6.3‒7.9‒9.6‒11.5‒13.5
Actual balance (percent of GDP)‒0.6‒0.6‒0.7‒1.0‒1.4‒1.7
Government gross debt (percent of GDP)15.915.415.115.015.315.917.0
Source: IMF staff estimates and projections.
Source: IMF staff estimates and projections.
  • In the first scenario, actual growth follows the baseline scenario, but a burst of optimism about potential output takes the estimated output gap up to 8 percent. The actual deficit is therefore larger than in the baseline scenario, but the government debt ratio still declines somewhat.

  • In the second scenario, the output gap is again 8 percent, but this time with actual growth held to just 2 percent and combined with a somewhat less optimistic estimate of potential output. Again, the debt ratio fails to increase.

  • Finally, the third scenario combines a low actual growth rate with an assumption that estimated potential output is no less than in the baseline scenario. The estimated output gap rises to an implausible 14 percent, and this time the debt ratio does begin to rise, after a few years.

Again, these examples are not intended to represent plausible scenarios, only to illustrate the degree of sensitivity of debt dynamics under the structural balance rule.

More practically, under this fiscal rule, the estimate of the output gap is updated regularly. This updating is done with the help of an expert panel, providing the opportunity to lower the estimate of potential output in the event that actual growth turns out less than expected. Moreover, this estimation is performed within a methodological framework (the Hodrick-Prescott filter) that penalizes large gaps between actual and potential output estimates. In this light, the above examples are seen to be artificial, as it is unlikely that an output gap estimate as high as 8 percent would be sustained year after year. Broadly similar considerations apply to the other adjustment used in the structural balance, which relates to the copper export price gap.3

Thus while uncertainty about the size and duration of (estimated) cyclical adjustments used in the structural balance rule is relevant for debt projections, plausible degrees of error (or “optimism”) on potential output along the way would not lead to unstable debt dynamics.4

Central Bank Finances

The balance sheet of the Central Bank of Chile has these essential features:

  • Large size. Assets and liabilities together total about 70 percent of GDP. Assets are dominated by international reserves, along with some (old) central government debt. Liabilities are dominated by domestic debt issues.5

  • Strong liquidity position. While assets are dominated by liquid foreign assets, liabilities are mostly medium-term paper. Importantly, the bank has avoided any issuance of short-term debt indexed to the exchange rate; it also has avoided interest rate indexation.

  • A significant currency mismatch, the bank being long in dollars. Assets that are exchange rate-linked far exceed exchange rate-linked liabilities (Table 7.2). At end-2002, the net foreign exchange position was +23 percent of GDP.

Table 7.2.Components of Central Bank Net Debt, 2002(In percent of GDP)
AssetsLiabilities
Foreign-denominatedForeign-denominated (or indexed)
International reserves23.9Foreign liabilities0.0
Government debt9.1Government deposits0.7
U.S. dollar-indexed paper9.3
Peso-denominatedPeso-denominated
Government debt0.7Government deposits0.3
Peso paper21.5
Of which: inflation-indexed9.3
Source: Central Bank of Chile.
Source: Central Bank of Chile.

Assessment of Risks

The structure of the bank’s balance sheet is not problematic. As noted, the bank’s balance sheet is strong in terms of liquidity. As for exchange rate exposure, strictly speaking, the bank’s currency mismatch is a source of risk: for example, a 10 percent appreciation (depreciation) of the peso rate produces a valuation loss (gain) in excess of 2 percent of GDP. However, this exposure is in the right direction to be considered a form of insurance.6 Thus if “bad times” for the Chilean economy cause the currency to depreciate, there will be associated gains for the central bank, lowering net debt of the public sector as a whole and so helping to support confidence. On the other hand, losses from currency appreciation would tend to be associated with favorable circumstances (e.g., terms of trade improvement), times in which confidence would be improving.

The central bank’s currency mismatch has declined in recent years, and it may be reduced further. Since 1997, the significant issuance of U.S. dollar-indexed debt has reduced the exposure of the central bank, and the public sector more broadly, to appreciation of the domestic currency. Looking forward, it is possible that the denomination of the government debt owed to the central bank will be switched from dollars to pesos.7 This would distribute exposure to peso appreciation more evenly within the public sector.

The Central Bank Deficit: Implications and Outlook

The central bank’s balance on operations has tended to be negative. Rather than from current central bank policies, these losses arise from: (1) the government exercising its option to capitalize interest on its old debt to the central bank; and (2) the inherited structure of the balance sheet, related to the support to banks in the 1980s and to sterilized exchange market intervention during part of the 1990s.8 In recent years, the IMF staff’s measure of the central bank deficit has been close to 1 percent of GDP on a cash basis, and about 0.5 percent of GDP on an accrual basis.9

Though the central bank’s deficit is moderate in size and seems to be stable, it is not to be ignored. The Chilean government explicitly recognized the significance of the central bank’s deficit when it chose to set the target for the central government balance at a small surplus. A further consideration is the potential link between a central bank’s financial strength and its effectiveness.10 There has been no sign that the Chilean central bank’s deficit has interfered with its effective independence, but it is possible that public perceptions of the bank’s independence could be diminished by the existence of a deficit (especially if the deficit were to begin to grow, though this is not expected). At the least, the bank’s deficit complicates understanding of the public sector’s financial position.

Steps could be taken to clarify the financial relationship between the government and central bank. One step would be for the government to pay full accrued interest on its debt to the central bank (i.e., declining to utilize its option to capitalize some of this interest). Going further, the government could consider a more comprehensive solution, ending the bank’s deficit entirely by an appropriately sized recapitalization. Such a move could for example take the form of a transfer of government bonds to the bank; alternatively, it might be possible to transfer to the government responsibility for a part of the domestic debt now on the bank’s balance sheet.

The Chilean authorities are currently engaged in technical work to better understand the central bank deficit. Despite its apparent stability in recent years, the bank’s balance is subject to change from a number of factors, though these sometimes offset one another. The Chilean authorities aim to estimate the bank’s underlying deficit—a difficult exercise, taking into account, inter alia, cycles in interest rates (including variations in term structure) and exchange rates. Although this section cannot anticipate the results of that exercise, two qualitative points on the outlook for the bank’s deficit can be noted. One is that the bank’s net flows are likely to improve in the period ahead, especially as debt issued some years ago at much higher interest rates comes due and is replaced with cheaper debt. Another is simply to observe, in light of the bank’s sizable currency mismatch, that projections of the central bank balance will be sensitive to exchange rate assumptions.

Finances of the Public Enterprise Sector

Significant aspects of the public enterprise sector’s balance sheet are the following:

  • Net debt of about 6 percent of GDP: gross debt being about 6 ½ percent of GDP, and financial assets about ½ percent of GDP, at end-2002.

  • Gross debt is mainly—about three-fourths—external. Domestic debt remains small, still less than 2 percent of GDP, though in recent years a few enterprises, including CODELCO and the Metro, have issued bonds locally.

  • The average interest rate is low: for example, the ratio of interest payments to gross debt was only 3.5 percent in 2002. (For CODELCO, the company that accounts for the largest share of public enterprise debt, interest payments were only 2 percent of current income in 2002, a year not only of low interest rates but also relatively weak copper prices.)

  • Maturity of debt: about three-quarters of the public enterprise sector’s debt to the private sector is long term.11

In terms of flows, the public enterprise sector has been performing favorably in recent years. In particular, its balance on current account—after sending taxes and profits to the central government—has been fairly steady, at around 1 percent of GDP. Its overall balance has been in modest deficit, reflecting capital expenditure usually between 1 and 2 percent of GDP. (The majority of such capital expenditure has been by CODELCO, which is now well into a multiyear plan to expand its capacity.)

Interpretation

Under current circumstances, the public enterprise sector is not critical to the analysis of possible debt sustainability problems of the Chilean public sector. Partly this is because this sector represents a minor share (less than 15 percent) of total public debt. More fundamentally, the public enterprise sector seems to be run mainly on commercial principles and in any case is on the whole profitable. At the same time, it is advisable to continue to monitor the condition of the public enterprise sector, as it is large enough to be of macroeconomic significance, and to be alert to any change in its financial position.

Transparency and analysis of the public enterprise sector are set to take a step forward. The authorities are now developing a new set of statistics to follow the new standard set by the IMF’s 2001 Government Finance Statistics, with its accrual basis and emphasis on capturing changes in net worth. In particular, the new set of statistics will include information on the value of the fixed assets of this sector, and on capital consumption flows (depreciation).

References

Most of the amortization due relates to the government’s old debt to the central bank. There is no possibility of rolling over this debt, or refinancing it with fresh funds from the central bank, since Chile’s constitution (see Section III) prohibits the central bank from financing the government.

For an analysis of sustainability issues in light of uncertainty over copper export prices, also with generally favorable conclusions, A key III.

A key factor limiting the sensitivity of debt projections in this context is the relatively small adjustment allowed under the structural balance rule for a given output gap (see Section II).

Credit to banks and monetary liabilities are relatively small components of the balance sheet.

Whether this degree of insurance is optimal to Chile's circumstances is a technically challenging question beyond the scope of this section.

Such a move has already been approved by the congress; what remains is for the government and central bank to agree on technical details.

See Appendix III of the staff report for the 2002 Article IV consultation for a fuller discussion of the bank's flow losses and their origin (IMF, 2002b).

This measure does not include capital gains/losses arising from currency fluctuations. In recent years, the bank has experienced large capital gains as a consequence of the peso's depreciation against the U.S. dollar.

See, for example, Stella (2002).

An annual amortization schedule for public enterprise sector debt is not available. However, the Public Debt Report gives the breakdown between short-term and all other debt, as noted here.

    Other Resources Citing This Publication