Andres Oppenheimer

Despite the latest polls showing that Brazil's leftist candidate Luiz Inacio ''Lula'' da Silva is widening his lead for the October presidential elections, the majority view in U.S. business and diplomatic circles is that he will not win and that predictions of a dramatic shift to the left in Latin America's biggest country are premature.

  Is it wishful thinking on the part of Wall Street economists who fear a massive economic downturn for Brazil if da Silva's Workers Party wins the election? Or is da Silva, who has already failed in three previous runs for the presidency, incapable of surpassing the threshold of the estimated 35 percent of Brazilians who traditionally vote for the left?

  POLL LEADER

  A poll released this week by Brazil's GPP Institute shows da Silva is leading with 38.7 percent of the vote, followed by centrist government-backed former Health Minister José Serra with 14.5 percent of the vote, leftist-populist former Rio de Janeiro Gov. Anthony Garotinho with 13.9 percent of the vote, and left-of-center former Finance Minister Ciro Gomes with 13.8 percent.

  The latest surveys show that da Silva has widened his lead by about four percentage points from last month's polls. Unless he wins by more than 50 percent of the votes in the Oct. 6 first-round election, he will have to go to a runoff vote three weeks later.

  With the latest polls in hand, I asked some of the leading U.S. Brazil watchers this week whether they foresee a da Silva victory. To avoid wishy-washy answers, I asked them to rate the candidate's chances in percentage points.

  Their answers:

  • Paulo Leme, managing director of emerging markets of Goldman Sachs in New York, gives da Silva a 30 percent chance of becoming president. ''Lula's chances have risen in the last six weeks from 15 percent to 30 percent,'' Leme said. ``There is no question that he has strengthened in the polls, but opinion polls in Brazil this early in the game have in the past been very poor predictors. Fifty percent of the people have not yet made up their mind. And the Serra campaign has not yet taken off.''

  • Michael Gavin, head of economic research with UBS Warburg in New York, gives da Silva a 30 percent chance. ''Brazilians are too cautions to make a leap in the dark,'' Gavin says. ``And there is no period of severe economic and social stress in Brazil that would generate an anti-government vote strong enough to elect Lula.''

  • Paulo Vieira Dacunha, chief Latin American economist with Lehman Brothers in New York, gives da Silva a 30 percent chance. ''The distribution of political resources heavily favors government candidates,'' he says. ``And the second round election in Brazil is inherently conservative.''

  • Robert Berges, senior Latin American stocks strategist with Merrill Lynch in New York, gives da Silva a 30 percent chance, despite his own firm's downgrading of Brazilian bonds earlier this week because of concerns over his climb in the polls. ''You will see the polls changing in July and August, when the campaign really starts,'' Berges says. ``Serra has not been visible at all yet. But he will have the government behind him, and he will have the media establishment behind him. Brazil is still our favorite market in Latin America.''

  • Albert Fishlow, head of Columbia University's Brazilian Studies Institute in New York and until recently a Wall Street business consultant, gives da Silva a 25 percent chance. ''Historically, Lula has always started out with a considerable margin in the polls,'' he says. ``But Garotinho is likely to drop out of the race to run for reelection in Rio, and his votes will not go to Lula. They are protestant, lower middle class and middle class votes, that will ultimately go to Serra.''

  DIPLOMATIC CIRCLES

  In U.S. diplomatic circles, da Silva is not given much better chances. While U.S. officials say they do not rule out any outcome in the Brazilian elections, most are
  privately skeptical of a da Silva victory.

  Melvyn Levitsky, the former U.S. ambassador to Brazil during the Clinton administration who now teaches at Syracuse University, told me that he gives da Silva a ''35 to 40 percent chance'' of winning.

  ''I would give Lula a somewhat better chance this time around, because the other candidates are somewhat lackluster,'' Levitsky said. ``They have not shown the kind of national appeal that [President] Fernando Henrique Cardoso had.''

  My own hunch: I would give da Silva a 40 to 45 percent chance of winning. Serra, the government-backed candidate, does not seem to be a great campaigner, and the opposition will benefit from a growing anti-corruption sentiment, which could lead to a surprise first-round victory by da Silva. If there is a second round, he will most likely be history.