The argument

The argument of the book can be put in three sentences.

First, flexibility has value. In a highly uncertain market, entering in stages and keeping the right to leave can have a much higher expected value than committing everything at once. This is a result of real-options theory, and Chapter 2 computes it. But it holds only under conditions: when volatility is low, committing at once is better, and the best exit rule is not as simple as “quit after three months if it isn’t working”.

Second, whether the value of flexibility is realised depends on how the other side reads your changes. A partner cannot see your internal reasoning, only that the plan changed. If the conditions for change were not agreed beforehand, every change is evidence, to the partner, about what kind of collaborator you are, and often unfavourable evidence. Chapters 1 and 2 show this with Bayesian updating and a signalling game.

Third, the remedy is not to give up flexibility but to write it into the contract: state on day one the observable conditions under which you will change course or leave, so that a change becomes the keeping of a promise rather than the breaking of one. Chapters 3 to 6 develop this idea from four angles (incomplete and relational contracts, feedback delay and transaction costs, institutional distance, and principal-agent incentives) and also show where it fails.

Chapter 7 reads these modern theories beside the Art of War’s “an army has no constant formation, water no constant shape”, the Book of Changes’ “when a course is exhausted, it changes”, the discussions of “keeping one’s word” in the Analects and Mencius, the Western covenant tradition and the doctrine of changed circumstances. Chapter 8 offers a four-step contract alignment checklist ready for use and a pre-registered randomised vignette experiment.

What this book is and is not

This is a book of models and reviews. Each chapter introduces a published theory or body of research and then runs it in a short Python program, so readers can change the numbers and see how the conclusion moves. A model can show whether a mechanism is logically possible and what it needs; it cannot prove how a particular project will turn out.

Several of the ideas behind this book needed correction or conditions once they were checked. “Staged trial plus a strict stop-loss is the rational, NPV-maximising strategy” holds only when volatility is high enough, and the optimal rule is more patient early on than a three-month stop-loss line. An “explicit state-machine hybrid contract” is not better than other forms of governance in every environment: when the environment changes very fast, purely relational governance does better in the model. National culture scores are widely used and also widely criticised on method. Each chapter says so where it arises, and the Editor’s Notes published with the book files collect these points.

How to read the boxes

As in other 5D Library books, different kinds of statement are marked separately.

Plan of the book

1. Two business operating systems. Rule-based commitment and outcome-based adaptation; high and low context, Hofstede’s dimensions and their critics; trust in joint ventures.
2. Real options and optimal stopping. The abandonment option, staged entry, the Bellman equation, and the stop-loss as a signal.
3. Incomplete and relational contracts. Residual control rights, self-enforcing promises, and when formal contracts complement or substitute for relational governance.
4. The control theory of iteration speed. Feedback delay, correction gain and oscillation; brand assets and the cost of pivoting; transaction costs.
5. Institutional distance and Southeast Asian markets. Cultural distance indices, religion and regulation, Chinese business networks, and the limits of distance measures.
6. Principal and agent: incentives for local teams. Multitask incentives, risk aversion, and contract menus that let partners sort themselves.
7. Classical sources. The Art of War, the Book of Changes, the Analects and Mencius, with covenant and changed circumstances.
8. Practice and a pre-registered trial. A four-step contract alignment checklist and a randomised vignette experiment.

About this edition

This is version 1.0.0, the first complete edition: a preface and eight chapters, in Chinese and English. Corrections and missing references are welcome through the 5D Library website.