Read time: 5 min

Brazil: Politics vs Policies


By Rita Lavin, Standard & Poor’s MMS


The Brazilian financial markets corrected sharply down since mid-April amid speculation related to the presidential campaign. For almost 12 months now, the opinion polls consistently hint that the left wing candidate Luis Inazio da Silva – “Lula” – of the left-wing Labor Party, is the most popular. In that framework, a handful of investment houses advised clients to lighten their exposure to Brazil ahead of the election. Shortly after that, Brazil’s interest rate spreads to US Treasuries widened substantially and the real sold off. In four weeks, the stripped yield of the Brazilian C bond went from 11.9% (April 12th) to 13.9% (May 20th). In that same period, the real fell 4%. Are the Brazilian bonds relatively cheap now? Has the Brazilian real seen its pre-election lows already? The quick answer is “no”.

Financial markets in Brazil can remain highly volatile in the next few months as there will be national conventions in June, the campaign kicks off in July, the election is in October and the new administration takes over at the beginning of next year. Financial markets are too sensitive to the democratic process in Brazil not because democracy is destabilising but because of Brazil’s track record of traumatic transitions. Recent political history in Latin America doesn’t help either, and market participants fear that Brazil’s democracy could be tested in the same way it was tested in Ecuador (1999), Peru (2000), Argentina (2001-02) and Venezuela (2002).

The Brazilian economy is large and resilient and Brazilian leaders prove to be creative and focused when they have to deal with negative shocks to their economy, but there are vulnerabilities exposed in this election year.

Brazil’s external debt was estimated by the Central Bank at $210 billion at the end of last year, almost 50% larger than the Argentine debt at that time. The Brazilian debt would be equivalent to 39% of gross domestic product, equivalent to 3.6 years of exports. The Liquid International Reserves at the Central Bank were at $36 billion at the end of last year and $37 billion at the end of March this year. CB reserves would cover 17% of the foreign debt.

Unlike Argentina, Brazil also has a sizable domestic debt that was estimated at 530 billion reais at the end of last year and 551 billion reais at the end of March. Overall, Brazil’s total debt at the end of last year would be equivalent to 79% of GDP. IMF’s managing director Ann Krueger tersely warned recently about Brazil’s indebtedness and risk.

Recently, the Brazilian bonds sold off on speculation that the pre-presidential candidate for the Labor Party or PT was the most popular while the government’s candidate – a Social Democrat – was losing popularity. The Brazilian bonds were relatively cheap when the most recent opinion polls were published and the bonds attempted a bounce when the rumour proved true. But why would the bond market be so sensitive to opinion polls that are fundamentally preliminary because the presidential campaign hasn’t even started? The answer can be found in the size of Brazil’s debt and the left-wing parties’ propensity to talk about need for rescheduling the debt service.

The Cardoso administration managed the debt well. They went to Washington to negotiate a stand-by loan from the IMF that would give Brazil an important liquidity cushion during the presidential campaign but would expire a few months before the new administration takes office. Meanwhile, the opposition made, and continues making, their mark with voters with an anti-market approach and a nationalistic and populist rhetoric that challenges the status-quo and rattles the markets.

Vulnerability

Looking ahead, the Brazilian bonds will remain vulnerable if the election leaders present platforms that are perceived as fiscally loose. Chances are high that the PT party presents a platform vague enough that it will be difficult to make a definite judgment about the fiscal approach. The imprecision of the economic platform for Lula won’t prevent, and may enhance, conditions to speculate and profit with the PT’s fame as a populist and nationalistic party that would hurt lenders and potential lenders to Brazil.

Brazil’s currency is also vulnerable. The real is relatively young as it was created in July 1994, supported by programmes drafted 12 months earlier under the coordination of the then-Finance Minister Cardoso, during the presidency of Itamar Franco. The successful launching of a new currency in 1994 gave Cardoso a platform to run for president that year. He won the election in October 1994, reformed the Constitution and got reelected in October 1998, still using the platform of the so-called Real Plan that by then had established a record and had been modified and attuned to the evolving economic situation.

The most visible aspect of the Real Plan was that the real was a managed currency used as a nominal anchor to contain inflation. By late 1998 the plan was a success, inflation was low and predictable, but the real was overvalued. Supporting the real was a high interest rate that was causing difficulties on the fiscal front, to the extent that the Brazilian Treasury had to finance the public deficit at a high and rising cost. At the same time, the overvaluation of the real was making Brazil unattractive to foreigners who made Argentina their South American darling.

While Argentina borrowed easily and relatively cheap, Brazil limited foreign borrowing, introduced the Fiscal Responsibility Law, devalued the real, asked and obtained financial assistance from the IMF, and started running primary surpluses. When Brazil devalued the real, president Cardoso said that the exchange rate would no longer be the nominal anchor for inflation but from then on, the inflation rate itself would be the target. Cardoso did not promise no inflation, he just promised that economic policies would be conducted towards achieving pre-specified inflation rates. Cardoso didn’t even promise that the inflation targets would be reached; it was enough to let the market participants understand that there was an objective for economic policy. That alone provided stability in the last couple of years. It was very good as it lasted, but the Cardoso promise will expire when he steps down by year-end.

The reason is that the policy scheme of the last couple of years is specific to the Cardoso administration and the officers that serve with the president. Like Cardoso did in 1994, helping design the Real Plan and later on making huge changes to the plan to attune it to the circumstances, the new administration will have the option to do the same. The existence of that option and the possibility that the new administration can have an entirely different approach with exchange rate and interest rate policy causes uncertainty and affects markets negatively.

The failure of the outgoing administration is that it will leave Brazil with unconsolidated monetary institutions that can and may be modified by the new government. In its own Brazilian way, exchange rate and interest rate policies are perfectly married to the political cycle. When the president changes in Brazil, the policies can change so market participants ask for a premium for the so-called political risk. It’s a risk inherent to all the Latin American countries. The South American country leader in this decade is undoubtedly Brazil and it is failing to introduce the changes that would be required to consolidate institutions that would be independent of the political cycle.

Rita Lavin is senior economist, Standard & Poor’s MMS

Profit & Loss

Share This

Share on facebook
Facebook
Share on google
Google+
Share on twitter
Twitter
Share on linkedin
LinkedIn
Share on reddit
Reddit

Related Posts in