226. National Intelligence Estimate1

NIE 93–66


The Problem

To estimate the situation in Brazil and the prospects for the next year or two.


Castello Branco has managed for the most part to preserve constitutional forms without endangering the objectives of the revolution and has retained solid military backing. His economic corrective measures are showing favorable results, but the results have come slowly and the measures have provoked widespread dissatisfaction.
The administration is determined to see that acceptable candidates are chosen in the series of elections scheduled for this fall. It is taking steps to ensure that no opponents will become governors in the [Page 505] indirect elections on 3 September in 12 states. But the touchiest election will be the direct popular one to be held on 15 November for federal congressmen and state legislators; Castello Branco may deem it necessary to interfere directly and obviously so as to retain a working majority in Congress.
Costa e Silva, who has been War Minister, is almost certain to be elected president by the present Congress on 3 October. He will probably not exert much influence in the “lame duck” period before his four-year term begins on 15 March. Castello Branco’s policies will not change much in those months, though there will be some loss of momentum.
General dissatisfactions will persist, but the new government will probably succeed in keeping the opposition off balance and fragmented. At least to begin with, Costa e Silva’s control over the military establishment will be firm, and we do not believe that a military coup against him is likely during the period of this estimate.
Costa e Silva’s administration is likely to be a marked departure from Castello Branco’s, not in its broad goals, but in style of governing, in choice of key advisors, and in certain lines of policy. In some ways he will probably perform better; for example, he will give higher priority to public relations and may reduce popular opposition to some extent. He is likely to try for better relations with students and labor organizations, but will take whatever measures seem necessary to prevent a resurgence of the extreme left.
In other matters, however, Costa e Silva will probably not do as well. In his efforts to “humanize” the economic program, he may weaken present checks on inflation. Because he is less judicious and more a man of action than Castello Branco, we see more chance that he might resort to harsh, authoritarian methods. Finally, we think that he will put more emphasis on Brazilian nationalism and that in time this could cause friction in US-Brazilian relations.

[Omitted here is the Discussion section of the estimate.]

  1. Source: Central Intelligence Agency, Job 79R–01012A, O/DDI Registry. Secret; Controlled Dissem. According to a note on the cover sheet this estimate was prepared in the Central Intelligence Agency with the participation of the intelligence organizations of the Departments of State and Defense and the National Security Agency. The United States Intelligence Board concurred in this estimate on August 18. The estimate superseded SNIE 93–65 (Document 218).