Most of the agencies and departments of the federal government have "planning," "research," or "evaluation" staffs that help their administrators to deal with many of the tactical problems facing the federal government. But the President traditionally has had no comparable staff to do research into the larger or strategic social alternatives that the nation faces. There appears to be a need for such a staff, which might gain visibility and influence through the issuance of annual social reports. One difficulty is that there is no fully satisfactory intellectual framework or theory for the analysis of society-wide social problems. The main acceptable intellectual frameworks are the problem-solving approach, inherent in economic theory and operations research, and structural-functional analysis, which is common in several social sciences. This paper endeavors to show that structural-functional analysis is an incomplete version of the problem-solving approach, and that its main shortcomings can be traced to this incompleteness. The problem-solving approach, in turn, should be used with more concern about "suboptimization," and with more interest in the "softer" variables characteristically considered in structural-functional analysis. Such a broadened use of the problem-solving approach is called "complex systems analysis."