Introduction

This review will discuss two negotiating styles. One being strategic and the other being tactical. A strategic approach treats negotiations analytically while the tactical approach considers the personal and practical aspects of negotiating. My goal is to help the reader consider how to best use these approaches in different contexts.

1.1 Summary of texts

Beginning with Voss’s Never Split the Difference, Voss writes in a conversational style, which makes the book both readable and engaging. It also has the narrowest application of the books examined. By that I mean Voss focuses purely on negotiations in formal settings such, as on a sales call or at high-level meetings.

Voss’s negotiation style is tough, one that is unwilling to accept compromises. However, there is also a curiosity to it. Voss discusses at length the right way to guide a probing conversation. The purpose of which is primarily information gathering. In conversation people will elicit “tells” that can be used to judge their state of mind, what they value and where they want the negotiation to go. This knowledge is helpful in manoeuvring someone into a position aligned with yours.

Diamond’s Getting More is written in a more academic style. Diamond presents many practical examples to illustrate his arguments. At times, this makes the book feel like it consists of a series of HBR case studies. However, Diamond’s tone is also often conversational, and at times humorous, making the book mostly a pleasant read. Ultimately though, I did find Getting More to be overly long and occasionally verbose. I think a more tightly edited volume would have been more engaging.

The negotiation techniques Diamond presents are generally psychological, with a strong emphasis on relationship building. He repeatedly tells us to "getting inside the head" of the other side. This means sincerely engaging with your counterpart, trying to understand what they value, and using this as a starting point to nudge them towards your values.

This sounds somewhat like Voss’s approach, but Diamond’s style seems less hardnosed. Indeed, I would suggest that for Diamond “splitting the difference” could be a valid outcome, so long as the deal can be renegotiated further down the line. Diamond's attitude is to "get more", which does not mean winning on every occasion. It means incrementally pushing your agenda forward while maintaining relationships to the maximum extent possible.
Diamond also intends Getting More to have a broad application, beyond formal business negotiations. To that effect, Diamond includes chapters on negotiating with family (spouses, kids) and service personnel (airlines, utility services, etc.).

For me then, both books have a mainly tactical focus, they discuss how to negotiate once a decision has been made to start a dialogue. However, I believe that it is also important to have a framework to consider why embarking on a negotiation is worthwhile in the first place. Hence, I decided to bring in a text that has a more strategic interest. That book is Dixit and Nalebuff’s Thinking Strategically. Their book is not about negotiations per say, it was written as an introduction to the application of Game Theory (mainly in business contexts).

Thinking Strategically has an even more academic style than Getting More but despite that I found it a more readable book. Dixit and Nalebuff present their arguments succinctly and clearly, they rarely use jargon and have a good sense of humour.

Their book presents summaries of classic game theory models (the prisoner's dilemma, the Nash Equilibrium) and discusses major topics in the field (voting, managing externalities and auctioning). The book follows a structure of initially focusing on one of these models or topics, presenting a case study and then discussing how to apply this material in a business or personal context.

Their arguments apply to even more contexts than Diamond's; really they apply to any situation in which human beings (perhaps even other species) have some interaction.

1.2 Why review these books together?

As touched on already, the ideas put forward by Voss and Diamond significantly overlap. They both advocate an approach to negotiation that is particularly concerned with what might be called emotional intelligence and their techniques are rooted in the practice of negotiation. Hence while there are some differences, their arguments mostly support comparable styles.
As a result, the insights that can be gained in comparing them alone are limited. Where does this discussion go from here then? This review will now compare their approaches against a style that both Voss and Diamond are in fact critical of.

Both Voss and Diamond critique techniques that rely on purely “rational” analysis. They share the view that this style has been far too prevalent in the literature and as a result has made negotiators less effective. Their arguments are somewhat reminiscent of the recent developments in behavioural economics, where modern economists have increasingly sought to consider how psychological states (fear for example) can lead actors to make seemingly irrational decisions and ultimately lead to collapses of financial markets.

Like behavioural economists Voss and Diamond encourage us to thoroughly consider emotional factors that influence negotiation. These can include emotions such as pride, the need for respect and idiosyncratic individual preferences. Both authors suggest we should get “inside the heads” of the other party, think about how they see the world and what they value versus what we value.

From my reading, the main reason this is important is because of the chance of miscommunication. Simply presenting a deal in a basic matter of fact way might not work, even if from a purely rational standpoint it’s a good offer. This point comes through strongly in the sections in which Voss writes on sales.

For instance, Voss quite astutely observes that customers will frequently feign agreement in sales calls because they want to avoid conflict. However, sooner or later the sale will fall through because the customer was not really interested. Had the discussion been structured properly this potential customer could have either been won over or quickly eliminated from the pool of leds.

Another advantage in taking a more emotionally engaged approach is that it enables sellers to tease out exactly what the client needs and what they are willing to pay. There may be scope to work out a better deal than had been initially anticipated but this won’t’ be noticed if the conversation is not structured correctly.

While Voss and Diamond have made some strong points, I felt neither author fully considered why so much literature has focused on more strategic or analytic methods. Even applying their own techniques would require some strategic planning.

Both authors do strongly advocate for planning as an essential part of negotiation, so surely the more thoughtful and developed the plan is the better? Indeed, how could we evaluate if a negotiation is desirable without some strategic analysis in the first place?

Ultimately, I decided to contrast Never Split the Difference and Getting More with Game Theory. The reason I chose Game Theory is that it is a system that is quantifiable and emphasises strategic analysis. Yet it also encompasses an imperative to consider how different parties interact with one another, but I would suggest from a different perspective than Voss and Diamond. Finally, it is such a broad discipline that I believe it can also synthesise the merits of the more tactical style presented by Voss and Diamond with a strategic approach.
I selected Thinking Strategically for this review then as it is an excellent introduction into the concepts and application of Game Theory. The remaining sections of this review will consider three topics:

•   Understanding your opponent
•   How to manage failure
•   Rationally deploying irrationality

2.1 Understanding your opponent

As already noted, Voss and Diamond both recommend planning before beginning negotiations. However neither have discussed in detail how to do this. Planning is an essential component of Game Theory as well and Dixit and Nalebuff discuss it at length.

The most fundamental principal of planning in Game Theory is to think ahead and reason back. This means first consider what possible ends states exist for you, decide which outcome is best for you given the options and then consider how you will reach that point from where you are.
That may seem simple enough but the next step is crucial and it highlights another deficiency in how Voss and Diamond consider planning. Once you have decided your own strategy you must then assume that your counter-part is doing the same thing. Again this may seem a somewhat superficial or obvious point but doing this exercise has significant implications. It may cause you to fundamentally reconsider plan since you must anticipate how your counterpart will react to your actions and behaviour. This gives Game Theory strategies a kind of meta-analysis where you must consider not just your incentives but also how the interaction between yourself and the other side is structured as a whole.

I feel this gives Dixit and Nalebuff’s account of strategy and planning a dynamic quality that is missing from Voss and Diamond’s accounts; this is unfortunate and I believe it exposes two blind spots in Voss and Diamond’s accounts. First despite the fact that Voss and Diamond repeatedly emphasis how important it is to “get inside the head” of the other side, for the most part they don’t treat the other side as a credible agent. The result is that when they describe applications of their techniques, they basically always succeed because the other side lacked the skill to either counter the technique or manipulate you in some other way. If my point is not clear, this rather trite scenario should make it explicit.

Assume an executive has studied the work of Voss and Diamond extensively in preparation for an important upcoming meeting. He gets there and finds that not only has his counterpart read both books, but he’s attended several personal seminars hosted by Voss and happens to have personally consulted with Diamond in preparation for this meeting. What happens then?

There are two points to draw from this scenario. First , there are a range of skill levels and we can’t always assume that the people we negotiate with are necessarily naïve or unskilled. Further, if we assume our opponents are credible, we should also assume that they will use the same techniques to manipulate us, which has implications for how the negotiation is structured.

The second point is Voss and Diamond haven’t explicitly accounted for what happens if you are dealing with a more skilled negotiator. I’m not talking about someone with access to better resources or greater power (Diamond does consider this possibility) I’m referring to someone who just happens to be a more competent negotiator.

In general, that shouldn’t be a problem in Game Theory as, if you can anticipate you won’t benefit from the negotiation you would just avoid it, however realistically there are scenarios in which a negotiation must be undertaken even if the outcome is likely failure. There are ways through Game Theory to respond to this scenario and I will consider them in the next section.

2.2 How to deal with failure

The possibility of failure exists for any endeavour. It is unfortunate then Voss and Diamond haven’t discussed how to manage failure in detail. To his credit, Diamond does touch on it, although he does not consider explicitly what happens when his techniques fail. He just reiterates that the purpose of his book is to show the reader how to “get more”, without fully acknowledging there may be situations in which his techniques could fail. Similarly Voss does not discuss any absolute failures. He does refer to moments in negotiations where he struggled but the outcome always seems to be positive when the right technique is applied.

In contrast, Dixit and Nalebuff are upfront that failure is always a possibility, even if you have a great strategy. Failure can occur because you are unlucky or perhaps you didn’t fully understand the stakes before starting the negotiation. But the most intriguing examples Dixit and Nalebuff consider are situation such as the classic prisoner’s dilemma. While I won’t describe this game in detail, it is basically a scenario in which both sides simultaneously make the correct and rational choice but end up creating the worst outcome given their options. This kind of inscrutable situation is a more accurate reflection of the complexities of any business activity including negotiation.

So what do Dixit and Nalebuff have to say on managing failure? It is a complicated topic, there are many possible ways to fail in a complex endeavour so Dixit and Nalebuff simplify the problem to begin with by looking at success and failure within sports. We can always add the complexity of the real-world back in later once we grasp the basic logic.

The game of tennis provides a pertinent analogy. The point to remember, and this is where a more analytic approach has the edge on Voss and Diamond’s argument, is that the game is dynamic i.e. it is not enough to make a strategy for yourself, you have to consider what your opponent is thinking and adjust your plan accordingly.

For instance, how should you respond to a serve? Obviously you could choose a backhand or forehand volley, but which is best? One answer might be to always play to your strengths so always respond with your stronger hand. But if you do that, your opponent will know how you respond to every shot and retaliate accordingly, making your strong hand less effective. Game theory shows that simply playing your strength all the time will be less effective than mixing things up and sometimes playing a weaker hand.

Obviously always playing your weaker hand won’t work either, this will likely be even less successful than always playing your strength. So what’s the ideal mix? Determining this involves considering not just your own strengths and weaknesses but also those of your opponent. Does she play better to forehand shots or backhand shots? Ultimately if you consider your strength and those of your opponent an ideal mix based on your respective skill sets will emerge and this is your strategy.

It’s worth pushing this analogy a bit further as there are two curious points that can be derived from this scenario. First if your opponent also does this exercise, she will arrive at the same mix of forehand and backhand shots, just from her perspective. This fact was discovered by the famous mathematician John Nash and this scenario is known as the Nash equilibrium.

The second point indicates one way we could think about failure. It may be the case that after you’ve done this analysis you find that you are in fact the weaker side. The most likely outcome is that you will fail. However even if this is the case, you should still play to this strategy. If you deviate from this mix the most likely outcome is a poorer result. Your best hope is either that you get lucky or that your opponent deviates from the equilibrium.

In principal this may sound very intriguing, but I do believe this analogy starts to breakdown when we consider how this would work in practice. For instance knowing there is an ideal mix of forehand and backhand shots is all well and good but how can the mix of shots be sufficiently randomised so that the pattern is not obvious. If a player simply played ever third shot as a backhand then the strategy is not going to be effective since the pattern is not so complex as to prevent his opponent from recognising it. To make the Nash equilibrium there must be some degree of randomisation but not so much that the ideal mix of shots is threaten.

In reality, generating such semi-randomness becomes very problematic. How is a player, in the middle of a challenging match, suppose to compute a complicate randomisation calculation. While I do not know the mind of a professional tennis player, I nevertheless have a hard time believing that even a world class player would have the capacity to consciously run such complex calculations in the middle of play.

Having said that I suspect that peak athletes may in some way intuitively work through these calculations as they play. By that I mean they’ve develop a strong enough impression of their own skill and they should develop an impression their opponent’s strengths and weakness either as a match goes or through repeated play if they have played against each other before. Indeed it would be an intriguing research project to review top class tennis matches to see if players do eventually arrive at a Nash equilibrium as the match progresses.

However, the main point here is that although game theory is an important and useful intellectual exercise for planning, implementing it in practice can be extremely challenging. It is here that I believe Voss and Diamond’s techniques become relevant. I would suggest that their more tactical approach is better suited to the practice of a negotiation once it’s begun. At that point the variability of factors becomes so great that calculating a nuanced game theory strategy would be impossible.

Hence Game Theory is extremely useful in planning and analysing the merits of a negotiation, but a more tactical approach is necessary once the meeting has been scheduled.
There remains though an alternative way to respond to the prospect of failure through Game Theory, and it’s another approach the Voss and Diamond have for the most part neglected. This strategy is to attempt to alter the game itself by deliberately behaving irrationally. This is a risky move and in the final section I will discuss a way this can be undertaken in a controlled manner.

2.3 Rationally deploying irrationality

This review has associated Getting More and Never Split the Difference with a style of negotiation that emphasises the human and emotional aspects of negotiations. It has also shown that both authors are sceptical on the merits of approaching negotiations from a purely rational and analytical perspective. It is somewhat curious then that neither Voss nor Diamond have considered how irrational actions can at times lead to a positive outcome from negotiations. They mostly caution against acting irrationally during negotiations. Instead saying that we should remain cool and rational throughout negotiations. While I think this is largely good advice, I’m sure many of us have experienced someone appearing to win a negotiation through an emotional outburst or apparently irrational action. Without encouraging anyone to act in such a manner, I do think it’s unfortunate that neither Voss or Diamond discuss this possibility sufficiently.

On the other hand, Dixit and Nalebuff discuss at length how apparently irrational actions can shift a negation or deadlock through the idea of brinkmanship. In a certain sense implementing brinkmanship is not itself irrational in that the ultimate goal is to shift the negotiation in your favour. But for the action to be credible it be genuinely irrational i.e. it harms your position and this must be obvious to all sides.

The idea is to introduce an element of risk into the game that threatens not only your opponent’s interests but also, and crucially, yours. It must threaten your own interests because otherwise your opponent won’t see the threat as credible. Therefore the action must have an obvious and potential immediate negative impact on yourself, otherwise he would just treat it as an empty threat. Since the risk must be genuine Dixit and Nalebuff treat brinkmanship as something of a last resort, something to be used only when the game your playing is likely to have a sufficiently bad outcome that you have no other choice.

Of course brinkmanship can only be considered successful if the eventual outcome is positive for you. Hence a second, and just as important factor, as having a genuine element of risk is that there is a clear and obvious way for your opponent to turn the situation back around again so that steady rational negotiations can be resumed.
In that sense then, when someone behaves irrationally in a negotiation and wins, it is not so much that a simple irrational action leads to victory, rather it’s that a specific and targeted element of risk has been introduced into the negotiation, which forced the other side to capitulate.

As noted the risk element must be genuine, this is where the danger arises. If your opponent does not capitulate and instead also decides to behave irrationally then a “slippery slope” of reciprocal bad behaviour can develop leading to a complete breakdown of the negotiation, which is in no one’s interest.

The classic example cited by Dixit and Nalebuff of brinkmanship is the Cuban Missile crisis. Here President Kennedy upped the stakes and forced the Soviets to backdown. I’m sure there are many conflicting accounts and interpretations of what happened during this period, I am no historian, so I can’t judge if Dixit and Nalebuff are factually correct in this analysis. Nevertheless, their account of this incident provides a useful example of how we might think about applying brinkmanship in negotiations.

Essentially their argument is that Kennedy did not simply threaten retaliation if the Russians proceeded any further in building up Cuba’s nuclear capabilities. This would not work since the Russians would assume that, as a rational agent, it would be against Kennedy’s interests to initiate a nuclear war. Instead what Kennedy did was hand over responsibility for the decision to go to war to his Navy. This heightened the risk of retaliation firstly because, as a branch of the military, the Navy would more likely take a hard-nosed attitude to any further incursions than a civilian politician answerable to the electorate. Secondly with so many ships moving back and forth between the USSR and Cuba the chance of a Soviet ship, even by accident, making a wrong move and the Navy misinterpreting such a move as an incitement to war rises significantly. As a result then the Soviets choose to backdown.

Of course, Dixit and Nalebuff acknowledge that there are risks in using this approach. In introducing a slippery slope, it’s entirely possible you might slide down it yourself but as with the topic of failure they should be credited for at least broaching the subject.

3. Conclusion

This review considered two approaches to negotiations: strategic and tactical. Both have uses and in fact one another. Strategic analysis is useful in determining whether to negotiate or not in the first place, devising a plan to stick to during the negotiation and in considering methods to change a losing stance into a winning one.

However deploying a strategy in reality is challenging, this is where it is useful to adopt a more tactical approach. This approach helps us form relationships, understand the stakes in the negotiation and discover options we’d not previously considered.
Therefore I would suggest utilising both the strategic elements of Dixit and Nalebuff’s arguments and the more tactical techniques proposed by Voss and Diamond as the context dictates.

To end on then, I’ll just briefly highlight a couple of topics I feel the authors overlooked. First there is a question of the role that luck plays in success. To some extent Dixit and Nalebuff acknowledge this as they do include odds for failure in the models they describe but there is no explicit discussion on the role that chance can play in producing success.
The second issue relates more to the impact of complexity in modelling a game. By that I mean there may be scenarios that are so complex that doing the mathematics to evaluate the outcomes appears impossible (or in the very least calculations would take longer than the lifespan of the universe to work out).

How should we approach such issues? Beginning with the later point, this is an important topic to be aware of when we attempt to model reality. At best a model is an extremely useful approximation of reality that somehow is close enough to let us control some aspect of the world (think about the triumphs of modern engineering). However they have limitations and we should be mindful of these as we form plans. With luck, it seems to me that we cannot fundamentally quantify entirely unforeseen events (if we could they would not be unforeseen). In that case we should be open to chance when it can benefit us but also take some consolation from its role when we fail.