1999 Fiscal Conference: Sustainable Public Sector Finance in Latin America

Confronting Fiscal Imbalances Via Intertemporal Economics, Politics, and Justice: The Case of Colombia1

Colombian National Planning Department


Fiscal adjustment packages are traditionally designed to tackle excessive public expenditure; the reduction of current deficit and explicit debt levels is achieved by using a flow approach to government financing. When a long-term perspective is incorporated into fiscal policy decisions, the result is the pursuit of a dynamic equilibrium that is related to public sector net worth as opposed to explicit debt and deficit targets. The latter approach has fiscal adjustment implications that differ from the former, even with regard to the definition of fiscal imbalance.

A completely new set of problems emerges when implicit liabilities such as subnational government financing, financial sector bailouts, public sector guarantees, or other country-specific liabilities, such as the imputable cost of a peace process, are included in fiscal strategy. The feasibility of the entire fiscal strategy depends on congressional and high court approval of reforms which, in turn, depends on the political desirability and constitutional coherence of said reforms. Additionally, objectives specific to each public body are pursued, and thus the public sector budget and fiscal adjustment program are directly affected by the specific needs of different areas of government. This paper discusses the current Colombian fiscal adjustment model and how economic, political, and judicial considerations affect public policy making.

The hypothesis is that the feasibility of a particular fiscal package depends on not only a sound intertemporal (that is, stock) economic approach but also the incorporation of a new approach to political and judicial decision making. Whereas the intertemporal approach seeks to define how the effects and the assignment of those effects are distributed over time, this new approach requires that politics, and the concept and enforcement of social justice, have an explicit intertemporal dimension as well. A static view of politics and social justice may result in an unsustainable fiscal program and aggravate the fiscal imbalances that it is designed to eliminate.

If an intertemporal, cross-government approach is to be implemented, open dialogue is necessary between the executive, legislative, and judicial arms of government so that consistency of their respective objectives can be achieved in time. Recent developments in Brazil and Colombia illustrate the importance of such dialogue.

The growing consensus with regard to taking a stock approach to public financing is discussed in the following section. A third section deals with the flow and stock perspective relative to Colombian public finances and briefly discusses current fiscal adjustments. The final section explores both the political and judicial considerations that affect fiscal strategy and the risk of adopting a static approach to public policy making. Finally, the need for an explicit intertemporal approach to decision making, in all arms of public power, is justified as a necessary condition for the achievement of fiscal and economic stability.

Fiscal Policy

Pursuing fiscal stability via deficit reduction may be conceptually inconsistent as a deficit, in itself, is an arbitrary accounting construction that is subject to changes in accounting practices; it therefore depends on the categorization of receipts and payments. Thus, data manipulation is made possible, allowing misleading balances to be created that invalidate current fiscal objectives. Kotlikoff points out that "from a neo-classical perspective the deficit is an arbitrary construct with no necessary relationship to the fundamental stance of fiscal policy (1993)." The real effects of fiscal policy are independent from the size of the public surplus or deficit and depend only on the extent to which surplus and deficit alter economic incentives through the redistribution of income and expenditure.

Fiscal adjustment in the form of flow correction, as related to the reduction of public deficit or debt, may not represent an effective solution. It may also be misleading to use conventional accounting to calculate the extent and impact of an adjustment package in which variations in explicit government liability (that is, debt) are considered while variations in government assets, or implicit liabilities, are ignored.

Anticipated government responsibilities, which are not specified as legal obligations, can be direct (predictable and therefore expected) or contingent (triggered by a discrete event and therefore uncertain). A government commitment to accept these unknown obligations that depend on future events amounts to a hidden subsidy. This could cause immediate distortions in the markets and, therefore, result in an unexpected major drain on government resources in the future. The relevance of implicit liabilities such as natural disasters or legally binding court decisions is related to expectations of government intervention, which can be interpreted as being a moral hazard, the scope of which depends on the magnitude of government-led efforts to minimize market failures.

Bearing this in mind, a more appropriate deficit calculation would reflect variations in the net present value (NPV) of public sector assets, minus all liabilities. This approach to deficit assessment would emphasize changes to the government's net worth, and constitutes a stock perspective of fiscal imbalances—one that allows for an intertemporally optimal fiscal policy path in which objectives are considered in terms of government net worth.

The development of fiscal policy along these lines requires the explicit incorporation of an intertemporal framework that reduces fiscal adjustment to the fall in the government discount rate. This in turn is linked to generational accounting, a system of long-term fiscal analysis and planning aimed at the sustainability of fiscal policy and the measurement of current and future fiscal burdens.

Such intertemporal analysis of fiscal policy is conducted using the fiscal balance rule (FBR), which measures the impact of current policies on the tax burden of future generations relative to that of current generations. Thus, the intergenerational balance guards future generations from having to suffer a higher tax burden compared with that of current generations.

The FBR implies that governments should extract enough from each successive generation so that if it were in a steady state—in which government net worth and consumption grow at the same rate—it would remain there, with no need to impose a larger or smaller tax burden on subsequent generations. It is evident, therefore, that the FBR is based on the efficient extraction of resources from, and allocation to, different generations, as is consistent with an optimal path of government expenditure. An element of long-term analysis is incorporated into the FBR, which guarantees government financing through a dynamically consistent scheme that also seeks to provide equal funding for different generations (Kotlikoff 1993; Auerbach, Kotlikoff, and Leibfritz 1999).

Fiscal policy complies with governmental intertemporal budget constraints, which are, in turn, dependent on the lifetime net payments of individuals. This concept is related to positive and negative flows between government and individuals in terms of taxes, loans, and transfers. Thus, generational accounting is based on the government's intertemporal constraint, which determines the sustainability of the policy. This constraint requires that the future net tax payments of current and succeeding generations be sufficient, at today's values, to cover the present value of future government consumption as well as the government's initial net indebtedness. Thus, the zero-sum nature of intergenerational fiscal policy is revealed.

The achievement of fiscal adjustments through reductions in the discount rate—a strategy that results in an increase in the current young generation's lifetime net payment—constitutes a real adjustment with respective implications for saving, investment, and capital accumulation. Such a policy, which can be interpreted as an intergenerational transfer, can work with a balanced budget, a deficit, or a surplus.

However, when adjustments are limited to flow variables, targets such as deficit reduction can be achieved by manipulation of accounting categories, a reduction in asset accumulation, or an increase in hidden liabilities. These adjustment packages were criticized by Easterly (1998), who defined them as mere illusion. The use of fiscal deception can reduce a cash deficit, but it is only a transient reduction and is not a permanent shock that shifts government intertemporal net worth. Thus, a deficit target is achieved through policies that turn out to be short-sighted; they do not consider the intertemporal dimension.

Such is the example of a reduction in public investment, which might be dynamically inefficient since it deprives the economy of future revenue that could have more than compensated for the expenditure. A further example is the reduction of operational and maintenance spending, resulting in a reduction of current asset values and requiring future spending on the restoration of assets. This offsets the current value of the reduction in expenditure, thus making it dynamically inefficient.

Another way to meet current deficit targets is to protect current public consumption by shifting expenditure and revenue across time. This is strictly a flow manipulation and does not have any structural impact on a government's net worth over time.

Thus, a government can shift taxes over time or delay financial crises in the banking sector by altering flow temporality, but it is obliged to face the consequences of dynamically inconsistent decisions later. The outcome is persistent and aggravated structural fiscal imbalances as a result of the dynamic nature of fiscal policy.

The reduction of current deficits through time reallocation of revenue and expenditure flows may overlook or elevate fiscal risks. Sound fiscal packages may favor budget programs that do not immediately require cash and which, therefore, temporarily hide the underlying fiscal cost. The dynamic nature of fiscal policy calls for an up-front analysis of risks and the future implications associated with contingent forms of government support, in order to guarantee the dynamic efficiency and consistency of fiscal policy.

Implicit contingent liabilities carry great risks and are relative to the strength of the macroeconomic framework, the vulnerability of the financial sector, the efficacy of regulatory systems, and the availability of information; these factors that affect contingent liabilities are in turn related to transaction costs. The presence of contingent liabilities requires that the cost of uncertainty be incorporated into the decision-making process.

Prudent fiscal policies need to be developed within a long-term framework, and policymakers need to be explicitly aware of the long-term consequences of their decisions. This requires a recognition of the fact that short-term flow stability does not necessarily mean fiscal stability. The elaboration of prudent fiscal policies entails, therefore, the construction of dynamically efficient strategies. This, in turn, requires the development of a single portfolio that contains a stock of (complete and accurate) contingent liabilities, public sector debt, and other public liabilities, and which permits the evolution of a correlation sensitive to macroeconomic and policy scenarios as well as overall risk exposure through time. This approach results in the determination of optimal risk exposure, taking into account the state's ability to manage risk and absorb contingent losses.

These strategies depend on the existence of an appropriate institutional framework in which the behavior of political arms, such as the legislature, judicial boards, or the Supreme and Constitutional Courts, is crucial. The importance of policy consistency in the different areas of government, in order to achieve fiscal balance targets through the conciliation of their respective objectives, is illustrated by situations such as the one currently faced by Brazil.

In the pursuit of fiscal adjustment, the Brazilian Congress recently approved two reforms aimed at achieving a primary fiscal surplus equivalent to 3.25 percent of gross domestic product (GDP). The surplus was a condition of the austerity program negotiated with the International Monetary Fund in order to receive a US$41.5 billion loan. The reforms consisted of an increase in pension contributions paid by high earners in the public sector and a deduction of contributions from the pensions of those who had already retired. The reforms were also in line with the Brazilian government's efforts to achieve equality among the public and private sector pension schemes ("Nuts in Brazil" 1999).

However, the Brazilian Supreme Court recently judged such reforms to be unconstitutional—a judgment that will result in a US$1.2 billion shortfall in next year's public accounts. This move, which will result in further budget cuts, has created additional uncertainty over exchange and interest rates. Uncertainty has also developed regarding political support for the reforms, ultimately engendering a lack of credibility for future reforms. Doubt has been cast on the ability of the Brazilian government to comply with the required fiscal austerity program, thus hindering a speedy economic recovery from a deep recession.

In sum, a trade-off exists between long-term fiscal stability and budget deficit and debt target levels, and between the quality of fiscal adjustments and the speed of deficit reduction. This may lead to fiscal opportunism and, in turn, a bias toward excessive accumulation of contingent financial risks and unsustainable, dynamically inconsistent policies. Additionally, the extent and effectiveness of fiscal reforms employed in the pursuit of fiscal adjustments requires guarantees from both the judicial and political systems. Such guarantees require that the policies incorporate an intertemporal framework that includes the dynamic implications of decisions for the long-term confrontation of fiscal imbalances.

The Flow and Stock Perspective on Colombian Public Finance

The Flow Perspective. Colombia's applied fiscal accounting practice has traditionally based the sustainability concept on deficit dynamics; an unsustainable situation has been explained in terms of growing and persistent deficits. The result has been a fiscal policy analysis based on the achievement of short-term goals, such as reducing the deficit in the current period, as opposed to a long-term dynamic perspective. Given the conceptual and practical deficiencies that arise when using the government deficit as a measure of fiscal balance, policies aimed at reducing the short-term deficit have not resulted in equilibrium. On the contrary, short-sighted fiscal policy decisions have resulted in larger future fiscal burdens.

Traditionally, unsustainable fiscal flow refers to the explosive tendency of explicit public debt when considered as a proportion of GDP. Public debt increases each year on a scale equal to the primary fiscal deficit, in addition to nominal interest payments, excluding the portion financed through seigniorage. When expressed as a proportion of GDP, debt declines in accordance with inflation and economic growth.


Thus, if the primary deficit exceeds seignoriage income, and the real interest rate exceeds the GDP growth rate, the debt/GDP ratio will grow without limit.

The average real interest rate of domestic and foreign public sector debt in Colombia over the past two decades has been approximately 8 percent while the average GDP growth rate has been 3.7 percent. Such a scenario is clearly unsustainable when combined with increasing primary deficits as a proportion of GDP. According to Trujillo (1999), the sustainable level of debt for Colombia is 29 percent of GDP under reasonable assumptions of economic growth and money creation. This level is three points below the current figure and seven points below the level projected for 2002.

One approximation of the analysis of fiscal imbalances in Colombia relies upon measures of deficit and debt, which are in turn based on the long-term performance of income and expenditure flows. Stylized facts reveal an increasing gap between the government's annual tax income and expense flows between 1980 and 1998 (Chart 1). The unsustainable nature of this relationship implies that future expenditure, excluding contingent liabilities, cannot be sustained within the actual government income flow framework.

When limited to a study of flows, an analysis of fiscal instability underestimates the real dimensions of the fiscal problem and hampers the design of a sound fiscal policy. This implies that the forecasted deficit reduction, as a percentage of GDP, does not necessarily result in the achievement of an intertemporally sustainable fiscal scenario.

Chart 1

Therefore the importance of pursuing a stock-based analysis of fiscal policy in Colombia is stressed. This line of investigation includes various factors that have an impact on the long-term sustainability of fiscal policy (and which are often overlooked by the simple flow approach). The dynamics of a stock-based analysis require the incorporation of contingencies into the public sector balance sheet, thus redefining the perception of fiscal imbalance through the consideration of major risks to fiscal stability. Furthermore, this type of analysis shows that the underlying, long-term nature of current balanced budgets is unsustainable, resulting from the mislabeling of accounts and underestimations. A stock-based analysis also reveals the importance of intertemporal equilibrium between public sector power arms in the design of public policy—an equilibrium that has an impact on the dynamic structure of a long-term fiscal program.

The Stock Approach. According to Easterly (1999), an analysis of fiscal sustainability requires a study of intertemporal government net worth, rather than of current deficit levels. With this in mind, the Colombian National Planning Department amplified the official public sector balance sheet produced by the National Accounts Office (1997). Implicit and contingent assets and liabilities were included, as well as a reestimation of explicit liabilities and assets, in order to present a more accurate picture of the fiscal imbalance (Echeverry and others 1999). This balance sheet amplification shows the relevant contingent liabilities in the elaboration of fiscal policy by displaying the long-term consequences, and thus the dynamic structure, of the policy.

The following contingent liabilities were included: (1) Explicit liabilities: Public body debt guarantees (public enterprises and regional and municipal government); and infrastructure guarantees (linked to traffic volume, income, debt service, excess costs, and delays). (2) Implicit liabilities: Bailing out of financial institutions and financial sector deposits; bailing out of regional bodies; natural disasters (earthquakes and floods); contingencies resulting from legal actions against the nation; constitutional obligations with no contractual basis; and anticipated cost of peace negotiations. (3) Asset recalculation: Included gas, oil, and coal reserves and the electromagnetic spectrum (that is, air waves such as radio and cellular communications).

The official balance sheet, as a percentage of GDP, is presented in Table 1. The balance sheet, including the recalculation of existing assets and liabilities, in addition to the inclusion of contingent liabilities, appears in Table 2.2 A breakdown of the estimated contingent liabilities appears below in Table 3.

A grave underestimation of the value of pension liabilities, due to miscalculations, was the most important variation of the amplified balance sheet when compared with the official results. This is a direct, implicit liability that constitutes the most poignant fiscal sustainability problem, due to both its size and its long-term implications. Unfunded pensions constitute a hidden cost of approximately 159 percent of GDP. Social Security beneficiaries (48.5 percent) and teachers (30.1 percent) are the largest parts of this liability.

TABLE 1: Public Sector Balance Sheet, 1997
(as a percent of 1997 GDP)

Total Assets   140.5
  Current Assets 29.5
  Fixed Assets 110.9
    Natural Resources 16.5

Total Liabilities   78.2
  Current Liabilities 26.1
  Long-run Liabilities 51.8
  Other Liabilities 0.3

Government Net Worth 62.3

Total Liabilities + Government Net Worth 140.5

Source: Colombian National General Accounting Agency (Contaduría General de la Nación). Numbers have been rounded.


TABLE 2: Amplified Public Sector Balance Sheet, 1997
(as a percent of 1997 GDP)

Total Assets   162.3
  Current Assets 29.5
  Fixed Assets 132.8
    Natural Resources 38.4

Total Liabilities   251.7
  Current Liabilities 26.1
  Long-run Liabilities 225.5
    Contingent Liabilities 173.7
    Pension Liabilities 159.2
    Other Contingent Liabilities 14.5
  Other Liabilities 0.3

Government Net Worth –89.6

Total Liabilities + Government Net Worth 162.3

Source: Echeverry and others (1999). Numbers have been rounded.


TABLE 3: Principal Public Sector Contingent Liabilities


Natural Disasters   1,360   1.1
   Earthquakes   1,343   1.1
   Floods      18   0.0

Financial Entities' Bailout   2,700   2.2

Pension Liabilities 193,727 159.2

Infrastructure   7,283   6.0
   Roads     122   0.1
   Airports     863   0.7
   Energy     617   0.5
   Telecommunications     281   0.2
   Mass Transportation   5,400   4.4

Foreign Debt     860   0.7

Judicial Decisions     204   0.2

Territorial Debt     591   0.5

Peace   5,000   4.1

Total 211,725 174.0

Source: Echeverry and others (1999). Numbers have been rounded.

In terms of infrastructure, contingent liabilities generally consist of guarantees associated with private sector concessions; these guarantees originate when the role of the state transforms from that of a direct service provider to that of a guarantor of minimum sales to private sector service providers. The first generation of contracts included such clauses, whereas recent contracts specify a flexible duration. Private investors obtain a good level of return on their investment, and no public sector guarantee is required.3 However, the previous contracts generated uncertainty over whether or not public financing would be required in the future, in response to the private sector guarantees.

Contingent liabilities in road construction consist mainly of traffic volume guarantees and, to a lesser extent, excess construction cost guarantees. The former are invoked if anticipated income falls below an agreed minimal level, related to predicted traffic volume. However, if income rises above an allowed maximum it becomes a contingent asset and the excess is returned to the state. The calculated NPV of this contingent liability, to be incorporated in the balance sheet, is 0.1 percent of GDP until 2009.4

The construction contract for the second runway at El Dorado International Airport (Bogotá), in addition to requiring maintenance of both runways, includes traffic volume guarantees that are in effect minimum income guarantees; the liquidity for such guarantees has been ensured by the creation of a government fund. Other characteristics of the contract include government compensation for tariff modifications. The NPV of this contingency for the duration of the project (twenty years) is estimated to be 0.7 percent of GDP.

Contracts for energy provision have been granted under onerous conditions; the Colombian government has had to assume the difference between the actual market value and the price agreed with investors. The NPV of this contingency is 0.5 percent of GDP until 2009.

Contingent liabilities in the telecommunications sector are related to minimum income guarantees for the contract holders, linked to the number of minutes of conversation, contained in four-year, joint-venture contracts. The NPV of calculated contingent liabilities in this sector is 0.23 percent of GDP until 2002.

Other contingent liabilities may emerge from the construction of the Bogotá Metro. In this case, the contingency is related to the probability of liabilities being larger than predicted and is partly a consequence of regulatory ambiguities. The Metro Law states that the national government may guarantee up to 70 percent of the debt associated with this project. Thus, if excess costs imputable to the public sector are generated, the national government is obligated for 70 percent of them. These contingencies are particularly important given the risks associated with delays, cost of public services (such as piping), and currency risks related to external financing.

The Colombian case is particularly interesting as well because the government has to consider the cost of the peace negotiations as a contingent liability. The approach chosen was to consider the guerrilla as a business and to calculate the sum needed to "purchase" this business. Hence a calculation of the profitability of guerrilla violence was needed. This was achieved by calculating average expected income and expenditure per guerrilla member; total values were calculated from a figure of 20,000 guerrilla members.5

Colombian guerrilla groups present an expected profit of US$430 million per year; the cost of achieving peace is considered to be the amount of investment necessary to eliminate a business with such high levels of profitability. Assuming the guerrilla to be risk-averse, and taking into account a dollar interest rate of 6 percent a year, an investment of approximately US$7 billion dollars is required. However, efficiency gains from army restructuring and the acquisition of superior military hardware should reduce this cost. Not making this investment would be short-sighted, given that the contingent liabilities of a prolonged armed conflict would be much greater. Therefore, attaining peace over the next five years is estimated to have an NPV equivalent to 4.1 percent of GDP.

A dynamic fiscal balance is additionally affected by implicit liabilities resulting from external factors such as natural disasters. Colombia's topography makes it particularly vulnerable to such liabilities, the most common being earthquakes and floods (calculated as having a four- and two-year cycle, respectively). Government obligations after such incidents require the provision of resources for the reconstruction and rehabilitation of affected areas, as well as accommodation for future damage prevention. An estimate of the most recent earthquake in January 1999 indicates that the net current value of earthquake contingency is equal to 1.1 percent of GDP, and 0.04 percent of GDP for flood contingencies.

Another important contingent liability is the bailing out of financial institutions. The analysis of this contingency is based on the current financial crisis and the implications that it has had for the government. Poor performance in the financial sector has resulted in government liquidation of insolvent institutions and the recapitalization of those that have presented net worth deterioration. Recapitalization requires substantial resources and, therefore, additional tax income. The liquidation of financial organizations creates depositor liabilities for the government; these are implicit liabilities that are covered in order to prevent a general crisis of trust in the financial system. The contingent liabilities of the recapitalization process result in a difference between the resources provided by the state and those that cannot be recovered from selling financial institutions in the future. The current financial recession presents contingency liabilities of between 1.24 percent and 1.68 percent of GDP.

Additional implicit contingent liabilities are derived from public sector debt guarantees, which constitute the hidden subsidies previously mentioned. These guarantees avoid tainting central government debt but generate moral hazards, such as the incentive to incur unsupportable levels of debt due to the guarantee of central government assistance should the debt become unserviceable. While these liabilities are subject to subnational government borrowing restrictions, current regulations (based on the 1991 Constitution and laws issued until 1997) are insufficient, allowing rapid growth of regional debt. Since 1997, borrowing has been limited to the payment capacity of the respective regional entity and thus associated with the generation of operational savings. However, the cost of bailing out regional governments is calculated to be 0.48 percent of GDP.

Judicial rulings against the state, which have been increasing over the past decade, are another source of contingent liabilities growing from 0.02 percent of GDP in 1990 to 0.3 percent in 1998. Since 1995 the majority of these rulings have been related to the infrastructure sector, though demands resulting from insufficient provision of social services (health and housing) have also risen dramatically recently.

When analyzing Colombian government net worth, proper calculation of public sector assets, such as oil reserves, is important (see Table 4). Verified oil reserves that belong to the state can be considered as income for the government. Considering that they represent a volume three times greater than current reserves, and taking into account the terms of reference of the association contract with the state oil company (ECOPETROL), these reserves have a current value of 14.8 percent of GDP. Equally, coal and gas reserves generate a net present income equal to 12.1 and 9.2 percent, respectively, of GDP. The electromagnetic spectrum is also considered as a source of income, although dependent on licensing; under the current legal framework this resource will provide profits over the next ten years, with a net present value equal to 18.6 percent of GDP.

TABLE 4: Positive Public Sector Contingencies

Contingencies Billions of 1997 Pesos Percent of GDP

Oil Reserves 18,103 14.9

Gas Reserves 11,146 9.2

Coal Reserves 14,785 12.2

Electromagnetic Spectrum 22,657 18.6

TOTAL 66,691 54.8

Source: Echeverry and others (1999)

This revaluation of assets results in a new total value that equals 162.3 percent of GDP, in contrast to the 140.5 percent reflected in the official balance. Government liabilities, on the other hand, increase from 78.2 percent of GDP to 251.7 percent, after the recalculation of pension and contingent liabilities. This translates into a deterioration of government net worth from 62.3 percent to –89.6 percent of GDP (see Table 2). Therefore such net worth should equal the present value of the future flow of fiscal surpluses in order for the transversality condition to be fulfilled.

The amplification of the government balance sheet shows a level of current assets sufficient to cover short-run obligations. Thus, the public sector does not have a medium-term liquidity problem but one of long-term solvency. The implications of this analysis illustrate how misleading the simple flow approach can be in determining fiscal sustainability. In the long-term analysis, insolvency factors are evident, as highlighted by the high value of pension liabilities. The result of these findings is that, in order to achieve a dynamic fiscal balance, the country needs to pursue a greater level of natural resource exploitation, accompanied by both higher economic growth rates and a simultaneous decrease in contingent and known liabilities, including pension liabilities. Of paramount importance are subnational government finance reforms and resolution of the financial sector situation. The results emphasize the importance of using an intertemporal analysis of fiscal policy to achieve a sustainable path of government financing.

Analysis of Colombian Fiscal Reforms. This section presents an analysis of the fiscal adjustment packages implemented by Colombian authorities over the last year. Until the beginning of the nineties, the Colombian macroeconomy was considered stable by Latin American standards. This stability was mainly due to the maintenance of a prudent fiscal stance. Recent instability can be attributed to the decentralization of a number of public sector contracts, among which agreements with labor unions have played a prominent role. Unified budget constraints tend to be disregarded after decentralization.

In this context, the main problems faced by authorities are the size of the necessary adjustment and institutional constraints. The complexity of the situation suggests that the allocation of savings should be made within a three-dimensional space of economic sectors, geographical regions, and time.

The following reforms reflect the main efforts made in the reallocation of savings, along the path described, aimed at confronting the flow and stock fiscal imbalances:

  • Tributary reforms: These reforms consist of various policies related to tax collection; they have both regional and national implications. The value-added tax (VAT) reform implies an expansion of the tax base, with a simultaneous rate reduction from 16 percent to 15 percent, starting in November 1999.
  • Fuel price liberalization: The elimination of fuel price controls has created additional income flows for both central and regional government.
  • Modification of the association contract for oil exploration: The state's participation in the contracts will be reduced from 50 percent to 30 percent after the commercial viability of the well has been established. This gain in competitiveness is designed to result in greater levels of foreign investment and thus in additional income for both central and regional government. A substantial impact on public sector net worth is expected.
  • Reform of subnational government income and expenditure: The main measures center on a reduction in expenditure. In addition to a reduction of regional pressures on the central government via further decentralization, policies are directed toward more efficient expenditure, including such measures as the rationalization of payroll and acquisition plans and of strategies for funding regional and teacher pension funds. Reductions in central government expenditure refer essentially to public investment, with an emphasis on shifting infrastructure investment to the private sector through a concession system.
  • Reform of regional transfers from the central government: Currently, transfers from the national government to the regions constitute the budget item with the greatest relative importance, being equivalent to 39 percent of total national expenditure. Additionally, transfers have experienced a high growth rate, increasing from 2.8 percent of GDP in 1990 to 5.3 percent in 1999.

    The proposed reform will reduce the pressure that these transfers place on the central government by almost freezing them in real terms once they reach their maximum value as a percentage of national current income. The regional implication of this policy, in terms of decentralized government, is a reallocation of flows, favoring central government finance. However, specific regional finance reforms have also been presented to Congress to promote fiscal sustainability in subnational governments.

  • Creation of the regional pension fund: Regional pension liabilities have reached a level equivalent to 39 percent of GDP. This level represents a threat to long-term fiscal stability, and the lack of regional government savings has resulted in serious payment delays. The objective of the policy is, therefore, to direct resources towards the creation of the reserves needed to cover pension liabilities over a maximum of thirty years. The necessary resources will be obtained from the joint participation of regional and central governments. Regional funding is derived from an increasing share of transfers to municipalities.
  • Gambling: The Colombian Constitution provides for a state gambling monopoly and exploitation of the resource is regulated by one, specific administration. Income derived from gambling is destined for the health sector, which is in an extremely critical situation, adding importance to an evaluation of the administration of the gambling monopoly.
  • Rationalization of the social security system: One of the most important factors in the current worsening of the fiscal situation is transfers to the social security fund. These transfers represent 30.4 percent of all central government transfers, of which 17 percent are related to severance payments. Not only do these transfers represent a major burden for the central government, but the design of the social security system implies rapid growth of these liabilities. Under the current system, severance payments suffer geometric growth, which resulted in the doubling of this liability between 1996 and 1999. This geometric growth is the consequence of the drawing of severance payments discounted by the nominal value at the moment in which the contribution took place, as opposed to the moment of the last contribution. The government will not have sufficient resources to guarantee this liability if this remains uncorrected.

    The elimination of the retroactivity of severance payments has been proposed as a way to achieve long-term fiscal balance from a stock perspective. Additionally, the social security reform contemplates the inclusion of teachers' contributions, exempted until now, and a two-year extension of both the contribution period and the age of eligibility for receiving pensions. These policies should offset what is both a regional and national contingency.

A careful evaluation of whether this fiscal package is sufficient to tackle stock-level fiscal imbalances is beyond the scope of this paper. The purpose is to illustrate how the main problems that were identified in the stock analysis are emphasized in the reforms mentioned, with pension reform being the most urgent.

In the event that these reforms are insufficient to achieve fiscal balance, the result would be a cascade of defaults on public contracts. Default would start with groups with the lowest bargaining power—for instance, pension and wage recipients in remote areas—as has recently been the case.6 The ineffectiveness of the adjustment could imply further defaults, in which case the last contracts to be affected would most likely be those with international commercial banks and multinational institutions. This could be the reason that Colombian bond spreads do not reflect the current stock problem.

Once the fiscal package has been designed, the government has to confront two independent bodies in seeking approval of the reforms: Congress and the Constitutional Court, both of which can either hinder or promote the government's efforts to impose long-term fiscal adjustments. Therefore, intertemporal budgeting and financing are a necessary, but not sufficient, condition for long-term sustainability. The commitment of these two arms of public power to harmonize their policies with fiscal adjustment strategies, by assimilating the intertemporal consequences of their decisions and taking into consideration the dynamic structure of budget constraints, has become as crucial as the executive strategy itself.

Intertemporal Economics, Politics, and Justice

Economic planning has traditionally been the responsibility of the executive branch of public power and is related to the annual public sector budget. The 1991 Colombian Constitution specifies that each successive new government must produce a national development plan consisting of (1) an analysis of the main problems affecting the country, that is, of the state of the nation at the start of the new government's term of office; and (2) the priorities of the new government for distribution of the investment budget, specifying spending targets for the four-year mandate.

The importance of intertemporal equilibrium in government is evident, especially when considering the extent to which decisions of one public body directly affect the dynamic structure of public policy decisions taken by a different body. The need for dynamic institutional cooperation in Colombia becomes manifest through analysis of the existing relationship between the policy decisions of Congress and the Constitutional Court and the government's intertemporal fiscal adjustment objectives.

What has become evident over recent years is that sound economic planning is a necessary, but not sufficient, condition for the effective planning of public sector expenditure. Indeed, other branches of public power have proven as effective as the executive in affecting obligations, namely, Congress and the Judiciary.

It is crucial to extend an intertemporal approach to public policy decision making to the other branches of public power for at least two reasons: first, in order to give a true meaning to economic planning; and second, to give effective economic content and feasibility to laws passed by Congress and to judicial rulings.

Some laws passed by Congress have increased state obligations, and the remainder of the fiscal reforms are under serious scrutiny. However, the negotiation process necessary to smooth the passage of the government's initiatives through Congress requires that agreements be made with Congressmen regarding investment in their constituencies.7

In terms of the Judiciary, the 1991 Constitution created a Constitutional Court to rule on the constitutional status of governmental initiatives, and a procedure was implemented to allow individuals to demand the fulfilment of their "fundamental rights" (known as a tutela). Various examples can be cited that illustrate the manner in which Constitutional Court rulings have imposed obligations on the state without contemplating intertemporal budget constraints. This is the result of the judicial purity of these decisions, which fail to consider resource generation and expenditure requirements.

Constitutional Court judgments T-296/98 and T-153/98 illustrate the problems of purely judicial decision making (Sotelo 1999). Following the status change of some civil offenses to criminal offenses punishable by incarceration, the prison population increased. The Constitutional Court subsequently ruled in favor of a prisoner who had asserted that overpopulation in Colombian prisons creates living conditions that violate not only prisoners' dignity but also their basic human rights to life, personal integrity, and health. The court found that Colombian jails were overcrowded, characterized by violence, corruption, and serious public service deficiencies, and lacked rehabilitation facilities. The state, therefore, was not complying with the constitutional obligation to provide prisoners with a reasonable quality of life.

The court ruled that prisoners' rights are guaranteed by the constitution and have "an absolute value, not susceptible of being limited under any circumstance" (Ruling T-296/98) and that the necessary resources for the transformation of prisons be set aside. In doing this, the court arbitrarily decided that this expenditure is more important than social expenditure and therefore must be made at the expense of investment in other areas.

The court ordered the Colombian government to formulate a plan for the construction and refurbishment of penal institutions within three months. Furthermore, the court stated that all work must be completed within four years. This call for improvements in the public infrastructure is outside the judicial mandate and makes the judge a policymaker (that is, an expenditure executor); the judge thus has assumed a role constitutionally assigned to other public bodies. Overcrowding is still at critical levels, and an across-the-board reduction in sentences is currently being contemplated by Congress in an effort to find a solution to the problem.

The difficulties arising from these rulings are not limited to the additional fiscal burden but also include the temporal conditions of the resolutions, which represent an obstacle to the pursuit of an optimal intertemporal fiscal program that will ultimately lead to dynamic fiscal sustainability.

Additionally, decisions by the judiciary affect the fiscal adjustment process directly, as the fiscal package is subject to a determination of its constitutional status. Given that the fiscal package is yet to be approved by the Constitutional Court, fiscal adjustment in Colombia faces a risk similar to that faced by Brazil. In order to increase the saleability of fiscal reforms, marketing devices have to be employed; this paper is intended to be such a device.

Thus, when considering the relationship between the Constitutional Court and government financing, it can be seen that some court decisions ultimately become contingent liabilities on the government balance sheet. Judgments may, therefore, constitute a major risk to fiscal stability and the intertemporal sustainability of fiscal policies. The direct impact of Constitutional Court rulings on the government's net worth illustrates the importance of intertemporal and interdepartmental consistency in public policy elaboration.

However, the autonomous nature of these three areas of government makes it difficult to establish a dialogue that will result in more comprehensive levels of planning for the provision of new public services as well as the maintenance of those already provided by the state. It is therefore evident that sound planning requires a comprehensive strategy, the precondition of which is a reasonable level of communication and cooperation between branches of the government without reducing their autonomy. Intragovernment consistency needs to be guaranteed for the fiscal strategy to be successful.

The rulings of Congress and the Constitutional Court should, at the very least, be compatible with incentives. They should be dedicated to reducing problems and promoting solutions; the implications of the ruling should not be more damaging than those of the problem to be resolved. In summary, Congress and the Constitutional Court should explicitly assume an intertemporal stance when making decisions, considering the effects of their rulings on society as a whole. It should be recognized that a better provision of public services depends strongly on the evolution of the economy and not exclusively on the issuing of laws and rulings that cannot have real economic content when they disregard the state of the economy and the intertemporal budget constraint of the government.

Governability and sound planning depend on Congress and the Constitutional Court adopting an intertemporal approach to decision making. A static approach can only lead to a more erratic path of public expenditure, one full of huge contingencies and an instability of norms. The effect of the existing decision-making structure can be devastating for a country, as is clearly illustrated by the recent experiences of Colombia and Brazil.

The experience of Peru illustrates the risk of a narrow interpretation of the independence of different government bodies and the nature of their rulings. Fujimori deemed it necessary to suspend both Congress and the Supreme Court—a decision that initially received both national and international criticism. The policy was, however, later praised as the only way to bring order to Peru's erratic institutional and economic decision-making process. Commenting on this episode of Peru's history, Barro (1997) concluded that Fujimori did the right thing, as an excess of democracy can be damaging for democracy.

The economic lessons learned over the last twenty years as a result of the exploration of an intertemporal approach must now be learned by the political and judicial systems. No country, in particular no poor country, can afford to be short-sighted when considering its economy and the role of incentives, the burden of contingent liabilities, and the damage resulting from policy decisions that do not promote long-term wealth creation.


1   The opinions expressed are the responsibility of the authors and do not represent those of the Colombian National Planning Department. This paper has benefited from conversations with Ana María Arjona, Alberto Carrasquilla, Roberto Chang, Juan Manuel Charry, Miguel Gandour, Gabriel Piraquive, Arturo Porzencanski, Natalia Salazar, and Luis Carlos Sotelo. We are especially grateful to Nicholas Perkins for his editorial assistance.
2   For a detailed analysis of the estimation of contingent liabilities and the asset and liability recalculation, see Echeverry and others (1999).
3   Previous contracts involved fixed terms within which a level of profitability was guaranteed by the government. Flexible contract length allows for low profitability to be compensated over time.
4   Balance sheet figures reference values relative to Colombia's GDP in 1997. See Tables 1–4.
5   The average expected income per guerrilla was calculated by considering it as a random variable related to (1) the probability of death of a guerrilla based on the fact that 400 out of 20,000 guerrilla members die each year; and (2) the value of the income variable associated to previously calculated levels of guerrilla income from 1991 to 1994 and their dependence on economic activity. Such data exhibits a high-level guerrilla income of approximately $680 million a year while low-level income fluctuates around $370 million. The principle sources of this income are drug trafficking (41 percent), robbery and extortion (28 percent), and kidnapping (19 percent). Considering that Colombia has been affected by a two-year recession over a six-year economic cycle, a probability of two-thirds was assigned to high-productivity periods. Expected guerrilla income in U.S. dollars = [Y](probability of death) + [(HY)(PHY) + (LY)(1–PHY)](probability of survival) = 576 million, where Y = income; HY = high-income level; LY = low-income level; PHY = probability of high productivity based upon length of Colombian economic cycles (2/3). Average income per guerrilla member = US$28,800. Estimated guerrilla costs (US$ million): yearly endowment = 6; salary costs = 72; other costs = 72; total costs = 150.
6   Presently some of the subnational governments are in default. Such debt is held by domestic commercial banks, which have bargaining power and are therefore putting pressure on the national government to implement new mechanisms to solve the problem. This situation reveals how this cascade of defaults is reaching a point at which the financial system's stability is threatened, and it is therefore consequential at a macroeconomic level.
7   Prior to the 1991 Constitution, the national annual budget gave members of Congress an endowment to be used for projects beneficial to their constituencies. This practice was deemed a source of corruption and was abolished. Congress subsequently began pressing for allocations from the national annual budget as a whole in order to obtain the necessary resources to effect their own initiatives. Thus, ironically, the abolition of the endowment was self-defeating.


AUERBACH, ALAN, AND LAURENCE KOTLIKOFF. 1999. "The Methodology of Generational Accounting." In Generational Accounting around the World, edited by Alan Auerbach, Laurence Kotlikoff, and Willi Liebfritz. Chicago: University of Chicago Press.

BARRO, ROBERT. 1997. Getting it Right: Markets and Choices in a Free Society. Cambridge: MIT Press.

COLOMBIAN CONSTITUTIONAL COURT. 1998. Rulings T-153/98 and T-296/98. Bogotá: Government of Colombia.

COLOMBIAN CONTADURÍA GENERAL DE LA NACIÓN. 1996/1997. Balance General de la Nación e Información Financiera del Sector Público Colombiano 1996 y 1997. Bogotá: Government of Colombia.

COLOMBIAN MINISTRY OF FINANCE AND PUBLIC CREDIT. 1999. "Reformas estructurales para garantizar la sostenibilidad de las finanzas territoriales y nacionales." Mimeo.

EASTERLY, WILLIAM. 1999. "When is Fiscal Adjustment an Illusion?" World Bank Working Paper 2109.

ECHEVERRY, JUAN CARLOS, MARIA VICTORIA ANGULO, GUSTAVO HERNANDEZ, ISRAEL FAINBOIM, CIELO NUMPAQUE, GABRIEL PIRAQUIVE, CARLOS RODRIGUEZ, AND NATALIA SALAZAR. 1999. "El Balance del Sector Público y la Sostenibilidad Fiscal de Colombia." Archivos de Macroeconomía #115 (June 30), Colombian National Planning Department.

EMPRESA COLOMBIANA DE PETRÓLEOS (ECOPETROL). 1999. "La nueva política petrolera de Colombia." Mimeo.

FAINBOIM, ISRAEL, AND CARLOS J. RODRIGUEZ. 1999. "Sostenibilidad de la Política Fiscal Colombiana." Mimeo, FEDESARROLLO.

FEHR, HANS, AND LAURENCE KOTLIKOFF. 1999. "Generational Accounting in General Equilibrium." In Generational Accounting around the World, edited by Alan Auerbach, Laurence Kotlikoff, and Willi Liebfritz. Chicago: University of Chicago Press.

KOTLIKOFF, LAURENCE. 1993. Generational Accounting: Knowing Who Pays and When for What We Spend. Cambridge: MIT Press.

KOTLIKOFF, LAURENCE. 1999. "From Deficit Delusion to the Fiscal Balance Rule: Looking from an Economically Meaningful Way to Assess Fiscal Policy." In Generational Accounting around the World, edited by Alan Auerbach, Laurence Kotlikoff, and Willi Liebfritz. Chicago: University of Chicago Press.

POLACKOVA, HANA. 1998. "Contingent Government Liabilities: A Hidden Risk for Fiscal Stability." Policy Research Working Paper 1989. Washington, D.C.: The World Bank.

POTERBA, JAMES. 1997. "Do Budget Rules Work?" In Fiscal Policy Lessons from Economic Research, edited by Alan Auerbach. Cambridge: MIT Press.

SOTELO, LUIS CARLOS. 1999. "Los derechos constitucionales de prestación y sus implicaciones economico-políticas." Mimeo, National Planning Department (November).

"NUTS IN BRAZIL." 1999. The Economist. October 9: 19.

TRUJILLO, EDGAR. 1998. "Cual es la deuda pública sostenible?" Mimeo, Colombian National Planning Department.

Comments by Arturo C. Porzecanski

Previous  |  Contents  |  Next