and then rewritten as
.
Issuance of domestic and foreign currency debt requires inclusion of a revaluation effect because movements in the exchange rate affect debt valuation (a depreciation will increase the governments effective indebtedness). The change in the debt-to-GDP ratio is then decomposed as follows:
where b–Dt = BDt / Ptytand BDtrepresents domestic currency debt, b–Ft = BFt / Ptytand BFt represents foreign currency debt, Stis the end-of-period exchange rate (soles/foreign currency), and S¯tis the average exchange rate.
Fiscal Sustainability
In this chapter, due to data availability, our definition of “government” is the nonfi-nancial public sector; therefore, it excludes the central bank and other public finan-cial entities. Since in Peru direct BCRP financing of the fiscal deficit is limited, and we assume independence of the BCRP is credible, we assume in our projections that the likelihood that the government would resort to inflationary financing of the deficit is negligible.
Assuming that all debt has a maturity of one period, that debt is real, and that it pays a constant real rate of interest, then the budget constraint identity is rewritten as
where bt= Bt/Ptis the end of period tstock of real debt and xt= Xt/Ptand ris the real rate of interest. By rearranging then
and updating to period t, this implies
. bt = +(1 r)−1bt+1+ +(1 r)−1xt+1
bt−1= +(1 r)−1bt + +(1 r)−1xt bt = +(1 r b)t=1−xt
b b i x b g
z b
S S
t t
t
D t
t
−− = − −− − − − t
+ −
+ + −
t-1 t-1
π π
1 . 1 . tt
t t
F t t
t t
t t
t F
S b S S
S z b
+ −
+ − +
−
− 1 − 1
1
1
1 1
π π
b b i x b
g z b
t t
t t
−− = − −− − − − −
+ −
+
t-1 t-1 t-1
π π
1 . 1 .
b b i x z
z b
t t
−− = − −− − − −
+
t-1 1 .t-1
Then, substituting bton the right hand side of the previous equation, .
Forward iteration of the above equation and combining with the condition that
gives the government’s lifetime budget constraint that states that the government finances its initial debt by running future primary surpluses with a present value that equals its debt obligations. That is, in order to avoid an unsustainable debt, the pubic sector must generate a sequence of primary surpluses such that the net present value of the sequence is worth at least as much as the debt at time t– 1:
If the government also finances its initial debt by raising seigniorage revenue in the future, then the lifetime budget constraint would be defined as
To express the government lifetime budget constraint as a ratio of GDP, defining yt as real GDP, and then defining ¯bt= bt/yt, ¯xt= xt/yt, and ¯σt+i= _t/yt, the budget con-straint is rewritten as
which can also be rewritten as
In a steady state with constant real GDP growth rate g, so that yt/yt_1 = 1+g, the pri-mary balance is a constant ¯x, and seigniorage a constant ¯σ, then the government life-time budget constraint is reduced to
Assuming r> g, then
b
g
r x
i
i
t-1 t
−
=
∞ +
− −
= +
+
+
∑
110
1
( σ)
b r x
y
i y
i
t i t
t-1 t i t i
−
=
∞ − +
− − +
−
=
∑
(1+ )( )( ++ +)0
1
σ 1
b y r x y
i
i
t-1 t-1 t i t i t i
−
=
∞ − +
− −
=
∑
(1+ )( )( ++ +) +0
1
σ
bt r x
i
i
t i t i
− =
∞ − +
+ +
=
∑
+ +1 0
1
1
( ) ( )
( )
σ
bt r x
i
i
− t i
=
∞ − +
=
∑
+ + 10
1
1
( )
( )
Limj r j b
→ ∞ − + t j
+ + =
(1 ) ( 1) 0 bt−1= +(1 r)−2bt+1+ +(1 r)−1xt + +(1 r)−2xt+1
or alternatively, it could be written as
where ¯r is the real interest defined as (r– g)/1 + g). And, given estimates of ¯r, ¯σ, and data on the public debt stock at t– 1, the required size of the primary balance to ensure fiscal sustainability can be determined.
As mentioned before, this chapter follows Burnside’s (2005) methodology to determine the required primary balance to achieve a target level and to analyze inter-est rate and foreign currency risk. Given a target level, the primary balance required to reach that target is given by
where all the variables are expressed as ratios of GDP, and it is assumed that the inter-est rate remains constant. To account for the possibility of increasing interinter-est rates, even though only about 40 percent of Peru’s debt is variable interest rate debt, and mainly with multilateral institutions and the Paris Club, the debt-to-GDP ratio is analyzed by the following:
and for the exchange rate risk analysis,
where θt–1 is the share of the foreign currency-denominated debt in the previous period, δtis the rate of depreciation of nuevo sols and U.S. dollars in the present period, and Rt–1is the nominal interest rate in the previous period.
b R
z b x z
zm
t t t t
= + + t
+
+
− − +
(1 ) 1 − −
1 1
1 1
δ θ 1
b R
z b x z
zm
t t
= + t
+
+
− − + 1
1 1 1
x r r b b
t r
j t
j t
= +
+ −
−
(( ) * − )
(( ) )
1 *
1 1 σ
x rb−=− − −−
t-1 σ
bt-1 b x t r
− = ≡ +− (− σ− ) /−
Bibliography
Budina, Nina, and Norbert Fiess. 2004. “Public Debt and Its Determinants in Mar-ket Access Countries.” World Bank, Washington, DC.
Burnside, Craig., ed. 2005. Fiscal Sustainability in Theory and Practice: A Handbook.
Washington, DC: World Bank.
Central Reserve Bank of Peru (BCRP). 2005. “Inflation Report: Recent Develop-ments and Prospects.” Lima, Peru.
Garcia, Valeriano, and Jose Valderrama. 2006. “Politica Fiscal y el Systema Tributario Peruano: Perspectives y Retos.” In Peru: Policy Notes. Washington, DC: World Bank.
International Monetary Fund (IMF). 2002. Assessing Sustainability. Washington, DC: IMF.
Loayza, Norman, and Rossana Polastri. 2006. “Macroeconomic Framework for Sus-tainable Growth.” In Peru: Policy Notes. Washington, DC: World Bank.
López-Calix, Jose R., and Alberto Melo. 2003. Restoring Fiscal Discipline for Poverty Reduction in Peru: A Public Expenditure Review. Washington, DC: World Bank.
Ministerio de Economía y Finanzas (MEF), Perú. 2005. “Annual Program for Pub-lic Debt and Debt Management 2006–2008.” MEF, Lima, Peru. August.
———. 2006a. “Informe Pre-electoral, Administración 2001-06.” MEF, Lima, Perú.
———. 2006b. Marco Macroeconómico Anual Multianual 2006-2008 Revisado.
MEF, Lima, Peru.
Patillo, Catherine, Helene Poirson, and Luca Ricci. 2004. “What Are the Channels through which External Debt Affects Growth?” Working Paper WP/04/15, IMF, Washington, DC.
Reinhart, Carmen, Kenneth Rogoff, and Miguel Savastano. 2003. “Debt Intoler-ance.” NBER Working Paper 9908, August, National Bureau of Economic Research, Washington, DC.
Schick, Allen. 1999. “Budgeting for Fiscal Risks.” Unpublished. World Bank, Wash-ington, DC, September.
Endnotes
1. There are several types of public sector liabilities other than “debt.” The following ter-minology is adopted throughout this paper: (i) direct-explicit liabilities are liabilities estab-lished by law or contract, and include “full faith and credit” debt, expenditures prescribed by budget law, and claims for services rendered (the timing and amount of these liabilities may nevertheless be affected by contingencies); (ii) direct-implicit liabilities are liabilities on which it is presumed, with good probability, that the government will make good, but without a legal obligation to do so; (iii) contingent-explicit liabilities are recognized in legally binding docu-ments, but the extent and timing of payment hinges on uncertain future occurrences; and (iv)
contingent-implicit liabilities refer to an expectation that government will accept a liability without having a legal obligation to do so. For more details on the classification of public sec-tor liabilities, see Schick (1999).
2. Net debt is total gross debt minus domestic public debt held by all public sector entities.
The evolution of debt aggregates in the last decade is discussed with reference to gross aggre-gates because estimates of net debt in Peru are not available.
3. Budina and Fiess (2004). See their debt decomposition table results in Annex I, Table A.I.2.
4. Reinhart, Rogoff, and Savastano (2003). Their table is included in Annex I of this chap-ter as Table A.I.3.
5. See Valeriano and Valderrama (2006) and Loayza and Polastri (2006) for details and rec-ommendations on the transitory components in the fiscal accounts.
6. Information as reflected in MEF (2006a). There exist accounting differences between the BCRP´s data and that of the MEF.
7. A small percentage of the domestic debt is indexed to inflation.
8. See Monetary and Exchange Rate Issues in Macro Policy Notes.
9. Such as swap transactions from floating to fixed interest rates on four IBRD loans and Brady Bonds.
10. The debt dynamics decomposition is carried out following Craig Burnside’s methodol-ogy as described in Burnside (2005). Methodolmethodol-ogy is described in Annex I of this chapter.
11. Contingent liabilities are not included in these scenarios. Including the present value of contingent liabilities, estimated by the Ministry of Finance to be about 10 percent of GDP in 2005, would require primary surpluses of 3.1 percent to achieve a 27 percent debt-to-GDP ratio and 3.5 percent to achieve a 25 percent debt-to-GDP ratio by 2010.
12. The share of treasury sovereign bonds accounts now for 41 percent of domestic issued debt with instruments that have maturities of 15 years.
13. This section also follows the Burnside (2005) model to determine the debt paths.
14. Primary balances are defined here with respect to the nonfinancial public sector.
Although in the assumptions we do account for seigniorage, which in the case of Peru has remained very low; in 2005 our estimates have it at around 0.2 percent of GDP and it is expected that in the medium term it will remain at around that level.
15. Financial Accounting Standards Board Statement No. 133, “Accounting for Derivative Instruments and Hedging Activities” (http://www.fasb.org/pdf/fas133.pdf ).
16. http://www.cavali.com.pe/.
17. Datatec is a joint venture of the Lima Stock Market (Bolsa de Valores de Lima-BVL and Mexican SIF Garban. All market participants praised Datatec’s positive contribution to mar-ket development and its excellent operational performance and service. Datatec supported more than 80 percent of government securities trading in 2004. It also supports most trading in BCRP CDs as well as foreign exchange spot and clean interbank.
18. http://www.grupocimd.com/cimd.
19. Government securities are dematerialized and registered with CAVALI, the central depository.
20. General debt law specifies that DGCP caters only to debt longer than one year.
21. BCRP has made an effort to reduce the number of CD series by attempting to have only two maturity dates per month.
22. Indeed, MEF is only indirectly present via CAVALI as a regulator/supervisor.
23. These changes may include reviews of investment limit in government instruments, for-eign investments limit, impact analysis of development of multifunds scheme, projected asset growth, and so forth.
Abstract
Peru’s investment in infrastructure has historically been low, volatile, and generally insuf-ficient for creating the public infrastructure needed to support a dynamically growing economy. Investment has been particularly low in transportation and water infrastruc-ture. These gaps translate into economic costs to the private sector and inadequate public service provision, especially to the poor. To generate adequate infrastructure, Peru should aim to invest 3 or 4 percent of GDP annually, up from the current level of about 1.5 percent of GDP. To do so, in light of budgetary constraints, the government should move forward with public-private partnerships (PPPs), through redesigned contracts that offer financial security to both parties, that address social concerns, and that can also serve to deepen local financial markets.
I. Main Infrastructure Sector Issues
Low and Volatile Historic Investment in InfrastructureInvestments in infrastructure in Peru have been cyclical, ranging from 0.5 to 2.5 per-cent of GDP. From 1981 to 1986, Peru invested around 2 perper-cent of GDP in infra-structure sectors. This flow fell substantially to less than 1 percent of GDP during 1988–93, reflecting the major economic recession of this period. Thereafter, from 1996 to 2001, infrastructure investment once again recovered to levels above the 2 percent of GDP threshold. Until 1995, however, Peru’s infrastructure investments were among the lowest in their peer group (Figure 1).
Infrastructure Concessions: Moving Forward
José Luis Guasch
167