David E. Bell


Policy, Probability and Preference – PDF

Author: David E. Bell

Harvard Business School

Howard Raiffa’s influential book “Decision Analysis” introduced a generation to problem structuring in the form of decision trees. Decision trees have three major components; alternatives, uncertainties and values.

In the nearly 50 years that has passed since the book was published, the increasing availability of data and computing power has allowed science, systems analysis, to make ever more powerful forecasts of what will happen under different policies.

Models can be convincing in answering the if …. then part  of a problem – assessing the probabilities associated with alternatives.  If we reduce CO2 emissions by half what does that do to sea levels in 2080? That’s very useful if the policy maker sees sea level as the key issue.  But decisions almost always involve tradeoffs among multiple criteria. We argue that understanding the values of affected parties, and the concerns of key decision makers, is just as important to influencing policy as is forecasting.  The talk will suggest a way forward.


David E. Bell is the George M. Moffett Professor of Agriculture and Business at Harvard Business School. He teaches the MBA course in Agribusiness and also runs the annual Agribusiness Seminar that attracts 200 leading food executives each January to the HBS campus, and each year to one other location around the globe. The next such off-site seminar is in Cape Town in September 2015. He studies all aspects of the food chain, from farming to distribution to trends in consumer eating habits.

During 35 years on the HBS faculty David has taught a variety of courses to both MBAs and executives, including marketing, retailing, risk management and economics. Most recently he has taught the MBA Leadership and Corporate Accountability course. He has held a number of administrative positions at HBS including a previous term as chairman of the school’s marketing faculty (2002-8) and Senior Associate Dean for Planning and Recruiting (2008-12).


Related Sessions

View full schedule