Cybernetic Contracts

A while back I wrote a post about cybernetic ethics and for some time I have been thinking about cybernetic lawyering. By this I mean lawyering embedded in, responsive to, the social systems it works in, in ways that meaningfully use evidence to improve that social system. And by evidence I mean more than chats in the bar at conferences; ‘experience’ or judgement (important as all three – es even those chats in the bar – might be); but I don’t just mean quants or AI.

Part of this thinking has been inspired by the work I have seen when legal innovators step into the classrooom with my students. One such is Alex Hamilton, of Radiant Law.  Alex has just written a blog which echoes many of the things he has taught my students (and me) in recent years. He posted it on Linked In (what the hell is wrong with him, I wonder, there must be something I am missing!?) and has kindly agreed that I might post it here. My inbox precludes me offering any analysis or critique, I encourage you simply to enjoy and work it out for yourselves. Cybernetic contracting, that works. It begins with a tune and ends up with Alex being reasonable. Go figure.

Hey, baby, how about this dance?

Why?

‘Cause it takes two to do this dance

The beginning of Louis Armstrong’s version of Takes Two to Tango, 1952.

fascinating recent paper discusses what “lay folk” (that’s pretty much all of us, most of the time) think about “sound judgment”.

Drawing on experiments, the authors found that we distinguish between rational and reasonable behaviour as two separate aspects of sound judgment. Rational behaviour is about maximising outcomes to meet our personal preferences. Reasonable behaviour is about balancing outcomes to meet both our preferences and societal norms, aka satisficing or good enough. This is what behavioural economists have been banging on about for decades – humans aren’t just rational.

The paper’s killer point though is that when it comes to negotiating: we prefer to act rationally while wanting the other side to act reasonably. 

Doh!


I’m going to be discussing commercial contracting in this post. These contracts have some characteristics that make a desire to maximise one’s own outcomes, while expecting the other side to play nice, just not make sense:

First, it takes two to tango. We’ve got to get both sides to sign the contract and an approach that starts from a grabby first draft is not a great way to get the other side to agree.

Second, we are trying to create a long-term relationship. Even in transactional situations, there is a huge benefit in being able to do repeat business – trust is valuable and the cost of contracting is high. For a long-term relationship to work, both sides need to feel good about it.

And third, because the relationship is long-term, this isn’t a one-round game of prisoner’s dilemma. Even if you are the overwhelmingly dominant partner, you will need the other side to cooperate. If you think there can’t be retaliation, think again.

Yet what happens too often in practice?

We create a reaction of unreasonableness by being ourselves unreasonable. As described in Give and Take, takers will be punished and yet we start negotiations with unreasonable terms that dare the other side to behave accordingly. The result is not only difficult negotiations, but we also lose more deals than we realise.

We also fail in other ways to engender the reasonableness that we want on the other side. The Reasonable Person Model, from the world of environmental psychology, posits that there are conditions needed for people to behave reasonably: from understanding the situation and being able to participate and being respected, to being able to act with competence and a clear head. We may not be able to solve the latter, but dense incomprehensible agreements and aggressive negotiation styles hardly create the reasonableness we are looking for.

At the same time, we are underestimating the brand impact that our contracts have and hence who wants to work with us. Legal is part of the marketing function (as well as the sales and procurement functions) and our contract terms are part of our brand promise. Our contracts say a lot about how we are as an organisation to deal with, and conflicting signals with our other promises will create a breakdown in trust.

And to top it all, we are fighting the wrong fights. The IACCM has been running a survey for a decade that consistently shows that we spend our time negotiating indemnity and other liability provisions when what goes wrong is the parties not being clear about what they are buying or selling and what it costs. We are literally fighting third order points: not what the core deal is, not even what to do when it goes wrong, rather what happens when damages are being considered (in that tiny proportion of cases that ever make it to actual liability being determined).

If none of this lands, try an experiment. Find a five page NDA and a one page NDA that you have been sent. Observe your physiological reaction as you open them. If the five-pager doesn’t cause you to bang your head against a table (or at least raise your pulse), I can’t help. And if you can’t find a one-pager, weep.

In short, we are shooting ourselves in the foot on a daily basis, encouraging drawn out negotiations and unreasonable behaviours as we try to grab the advantage on the wrong issues while undermining the very relationships we are meant to be creating.


If this is sounding a little abstract, let me show you some data.

By May 2018 all European companies had to update their supplier terms to cover new requirements for processing personal data. Many of you will have shivers from the experience.

Radiant Law helped a number of these companies. And a curious thing happened – we noticed that the projects to make the updates were taking wildly different times to complete. Some projects were being closed quickly and others were dragging. Really dragging.

So we plotted some of the projects as measured from the point we issued the contract amendment to the time the amendment was closed, as a burndown chart (this was just the client’s paper). The lines go from 100%, all amendments were open, to 0%, all closed (well, nearly 0% but some were abandoned). And we got this picture:

No alt text provided for this image

The lines represent each company’s project, the numbers are the average time to close and yes, the amendments were taking completely different times to close with the same objective (i.e., meet the requirements of the GDPR). The slowest project took almost seven times longer to close on average than the fastest project.

So we had inadvertently created a split test (aka A/B test) and the next step was figuring out what was different between the projects. The team identified two areas: how together the client was on getting approvals, signatures etc (surprise, mega-enterprises struggle) and how reasonable the first draft was (which was specified by the client). We RAG’d the two dimensions (as shown on the left of the picture) and then did maths to show there was a statistically-significant correlation. In other words, there are reasons to believe that both process challenges and the starting terms had an impact on time to close.

No alt text provided for this image

But if you look closer, it’s so obvious! Project B took over four times longer to close on average than Project D and the only difference was the terms (we were pretty consistent in how we worked) – see the arrow. More precisely, the terms were pretty much identical except for a single provision that significantly bumped the cap.

One provision (not required by the GDPR) caused the average time to close to be four times longer.

On the back of this learning, we started working with clients to radically shorten their standard terms and make them more reasonable. One project with a client required entering into over a hundred new agreements with third parties. We reduced their standard terms from 17 pages to 3 and made them a lot more reasonable. The result was an average time to close of 16 days. That’s a third less than the best example we had with the GDPR.

To understand why changing the terms has such an impact, consider this chart which shows the average amount of time a matter spends with Radiant (blue) v the other side (green) in rounds of negotiations on one of our accounts:

No alt text provided for this image

We typically work to two day turn-arounds and are well on our way to bringing that down to half a day. But us going even faster won’t materially change the time to close. What matters is how many times the matter bounces back and forth, because the other side tends to be so slow (although punchier and nicer terms get faster responses).

How clever is it really to be sending out drafts with unreasonable points in them so that you can concede them later… and hence guarantee multiple rounds of negotiations?


So how do we fix this?

The starting point is to recognise that a strategy to maximise outcomes while expecting the other side to be reasonable is counter-productive. We are creating unreasonableness with our first drafts and suffering from the consequences. Standard terms need to become short, clear, reasonable and relevant.

And to help companies move to reasonable terms, we need to recognise that the individual in-house lawyers can feel very exposed. What’s in it for them to make their terms reasonable and risk later finger-pointing? We have found that there are ways to frame these discussions with data and insights that help build a case for reasonableness. But GCs have to show bravery and provide support and the broader business needs to get behind this change.

And while we are at it, it’s time that we stop bowing to subject matter experts in other departments with ridiculous demands for what should be included in contracts. Skin in the game, not back covering, is needed to make real progress.

Fixing this problem will have a huge impact on your contracting. If it sounds a little too hippyish to care about your commercial relationships, then just think how much time and money you’re wasting with pointless negotiations.

It’s rational to be reasonable.

Alex adds…

If you liked this post, you may also like the one about why we need to be more unreasonable… but in a different context. And we have a bunch of guides on improving your standard terms, automating your contracts and navigating legal tech on the resources page of our website.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s