To examine the descriptive validity of the economic theory of teams and, more particularly, to study observation, communication, and decision behaviors, sixty business and government executives were asked to make recommendations for the optimal design of a hypothetical sugar plantation information system. Few of the subjects made mathematical calculations; instead they used observation, communication, and decision heuristics which seem representative of behavior occurring in real situations. Ten percent recommended the optimal team theory solution while the remainder exhibited inefficient and sometimes mutually inconsistent heuristics. In some cases too little information was collected and distributed: (1) information expected at particular communication or decision centers was not available at that center, and (2) information observed or communicated was not extensive enough to yield maximum expected net profits. In other cases too much information was collected and distributed: (1) the information, when used, did not yield big enough improvements to cover its cost, (2) the information, when used, resulted in a lower expected gross profit due to interdependencies, and (3) the information observed or communicated was never used. Implications of the results for both theory and practice are discussed.