W A S H I N G T O N D C O C T O B E R 2 0 1 0 Risk Modeling: The Brazilian Experience Andre Proite Brazilian National Treasury Investor Relations Office- Manager
Motivation Cost-Risk Analysis – Long term view Model Description Connecting the Model to Reality Next Steps 2
What have we pursuit? Minimize long-term borrowing costs, with maintenance of prudent risk levels; at the same time, seek to contribute to smooth operation of the public bond market. The role of transparency Besides a well established legal and institutional framework, transparency and accountability are also requirements for good governance practices Transparency generates predictability, that reduces market uncertainties about the Debt Managers objectives and actions, contributing for volatility and risk premium reduction 3
Federal Debt (FPD) Composition 4
Motivation Brazil walked on thin ice for a long time poor debt structure Very risky and volatile Then public debt was improved: where do we want to go? Benchmark Optimal composition Instrument for risk management and for strategic planning How to measure debt management performance? Guidelines (World Bank and IMF) According to the their Guidelines, the benchmark could work as a powerful management instrument as it represents the debt structure that the government would like to have, based on its expected risk and cost preferences. Thus, the benchmark could guide the debt administrator in his decisions regarding issuance and risk management 5
Motivation Cost-Risk Analysis – Long term view 1 st and 2nd phases • Model Description Connecting the Model to Reality Next Steps 6
The process of debt planning in Brazil is in a sophisticated stage…. … but, it is a result of a long process of institutional advances and of a simultaneous development of the technical framework 7
1 ST phase – Central Gov’t ALM Short-term analysis The Annual Borrowing Plan 2002 ALM as a tool to map and manage risks of the public debt portfolio ALM: a dynamic analysis of assets and liabilities that takes into consideration the public debt management strategy and the macroeconomic policies enforced by the government the basic premise for outstanding debt insulation is the balancing between the characteristics of assets and liabilities Refinancing and market risks are key elements market risk categories: inflation, exchange rate, floating rate, and fixed rate refinancing risk 8
1 ST phase – Central Gov’t ALM Integrating assets in the analysis helped identifying opportunities… Asset - Liability Mismatch 20% 10% 0% % GDP -10% -20% -30% -40% Inflation FX Linked Floating Rate Fixed Rate Others Linked Central Government Cashflow 400 dez/02 dez/09 350 300 250 Billion R$ 200 150 100 50 0 uo to 1 year from 1 to 2 from 2 to 3 from 3 to 4 from 4 to 5 above 5 years years years years years 9 Assets Liabilities
2 nd phase – Sensitivity Analysis and… Impact of 1% FX devaluation on Net PS Debt/GDP* 0.40% 0.30% 0.20% 0.10% 0.00% -0.09% -0.10% -0.20% Aug/02 Feb/03 Aug/03 Feb/04 Aug/04 Feb/05 Aug/05 Feb/06 Aug/06 Feb/07 Aug/07 Feb/08 Aug/08 Feb/09 Aug/09 Feb/10 Aug/10 Source: National Treasury Note: Stress scenario considered of 3 standard deviations over the medium interest and exchange rate observed at 2002, equivalent of an overshooting of 56.6% on 10 exchange rate and an increase of 7.8 on the Selic rate. *Net Public Sector Debt
2 nd phase – … stochastic analysis Schematic Summary of the CaR Definition Relative Cost-at-Risk of DPF* * Considers the 2007 share for Floating and Exchange Rates plus 10 percentage points. The 2008 projected composition is estimated as the midpoint value of the target ranges of the PAF 2008. Source: National Treasury 11
Simulating strategies INPUTS Projected • Outstanding Debt results: (bonds and contracts) Debt • Macroeconomic • Outstanding payments flow • Maturity Scenario: GDP, forecast • Profile inflation,fiscal balance • Cost • Financing Strategy • Risk • Financial Variables: interest rate, FX-rate • Dynamic Analysis Treasury´s GERIR system
Motivation Cost-Risk Analysis – Long term view Model Description Connecting the Model to Reality Next Steps 13
Building a Benchmark Model A Stochastic Finance Approach Stochastic Scenarios – models * 1 dJ ( J J ) dt J dz Interest rates: Cox, Ingersoll and Ross – CIR t t 1 t t 3 dI I dt I dz Inflation: Geometric Brownian Motion t t t t 3 Exchange Rate: Chan, Karolyi, Longsta and Sanders – CKLS * 2 dC ( C C ) dt C dz t t 2 t t Bond Prices B ( t , T ) J Prices come from CIR P ( t , T ) A ( t , T ) e Nelson-Siegel approach to premium kT P ( T ) e remium 0 1 2 14
Building a Benchmark Model Debt’s Carrying Cost y R Define a Carry cost for each type of bond (y) t D LFT LTN FX NTB R R R R R t LFT LTN FX NTB Federal Public Debt (FPD) Dynamics Where M t , is the monetary base, prim t is the primary balance D FPD ( FPD M prim )( 1 R ) t 1 t t t t Debt Sustainability indicator Because the Treasury control FPD, but the Net Public Sector Debt (NPSD) is the relevant indicator, it is hereby described as the following function NPSD FPD M SelicLiab FXLiab FXreserves GDPAsset SelicAsset others ( ) t t t t t t t t 15
Benchmark model: Searching the optimal composition of the public debt Stochastic Scenarios FPD FPD Carrying Composition Cost Other NPSD NPSD Dynamic Parameters Cost and Risk 16
Results – Efficient Frontier Efficient Frontier Δ Risk 0.00% 0.20% 0.40% 0.60% 0.80% 1.00% 1.20% 1.40% 1.60% 1.80% 0.00% -0.50% -1.00% -1.50% Δ Cost -2.00% -2.50% -3.00% -3.50% -4.00% Efficient Frontier Current Examples 17
Benchmark model: Searching the optimal composition of the public debt Correlation - Single Value Decomposition (SVD) 18
Motivation Cost-Risk Analysis – Long term view Model Description Connecting the Model to Reality Next Steps 19
Constraints may be in the way Development of Debt Management Capability is not necessarily a long process, but it depends on a number of factors Public Bond Cost/Risk Analysis Market Macroeconomic Institutional Environment Framework Public Debt Management Development Demand Constraints Parsimonious usage of the Benchmark Model – 2D analysis 20
Motivation Cost-Risk Analysis – Long term view Model Description Connecting the Model to Reality Next Steps 21
Next Steps Building a more robust macro-structural basis Work on other functional forms of key variables Ex: Interest rates: Nelson-Siegel 1 e 1 e i ( ) ( e t t t 1 2 2 t Having a macro-based description of key variables behind the model will enhance the debt strategies stories Incorporate the transition strategy in the optimization model Long-rung stationary optimal debt story VS today’s problem 22
Contact Investor Relations Office brazildebt@fazenda.gov.br Staff: Andre Proite – Manager Flavia Barbosa – Deputy Manager David Athayde Juliana Diniz Mathias Lenz You can also reach the Risk Management Unit: Luiz Alves-Manager Andre Melo-Deputy http://www.tesouro.fazenda.gov.br/english/index.asp 23
Recommend
More recommend