Copyright: 1957
Publisher: John Wiley & Sons
ISBN: 0-471-55341-7
Games and Decisions is supposed to be a "non-mathematical" foray introduction to Game Theory. I am not a mathematician and I would hate to read the "mathematical" treatment of the subject! This book is a difficult read for those of us who are not mathematically inclined, however it is still a good introduction to some of the concepts. My goal was to lay the groundwork for further more practical study of the Theory and how it applies to decision making.
Notes
The authors mentioned that they considered the title "Conflict, Collusion and Conciliation" for the book as these are the core concepts of Game Theory. I found their discussions of utility to be very informative considering how much I have heard the term used in the context of economics.
The authors point out that you do not prefer A over B because A has more utility, but rather we say that A has more utility because you prefer it. The notion that utility is not causative but rather representative is important when you get into the social implications of Game Theory.
Another note on utility is that utility comparisons between individuals is impossible. For person A to lose a single dollar might be devastating but to person B it might do little to change their actions. The dollar has the same "value" to each but certainly not the same utility.
Non-strictly competitive games can develop temporal collusion through learned behavior in iterative games. When a decision is played out over and over between two individuals, patterns can emerge in which they understand the move the other is going to make and can plan accordingly. Thus, while we usually consider each iteration to be a separate game, we should be careful to watch for temporal collusion that creates a priori knowledge of the other player's move.
Towards the end of the book, the authors offer a number of practical applications of game theory, particularly n-person games where collusion and arbitration are key. One interesting note, particularly with regard to the current political landscape, is their analysis of the power structure of a bicameral government with an executive veto power. They analyzed our government with regard to the two party system and found that when one party gains a 2/3 majority, the President is the weakest out of all scenarios, regardless of whether he is in the majority or minority party. The recent stimulus plan fight was indicative of this as President Obama, despite his immense popularity (perhaps only immense in the press) seemed unable to corral Speaker Pelosi and Majority Leader Reid.
Conclusion
Do not pick this book up unless you, like me, have a hard core determination to learn the fundamentals of Game Theory! It is not a fun read and unless you are mathematician it is easy to get lost in all the numbers. The book does not contain all the formal proofs and the authors do try to keep it simple, but even so, the book is very intense and difficult to get through.