34-35 - The science of the deal

When one looks for megatrends in the life sciences industry, externally sourced innovation stands out. In contrast to the past, much discovery and early stage development now occurs in very small, highly focused firms who never expect to go to market themselves.

Instead, business development executives travel the world, seeking to buy or sell innovative science at the sweet spot where risk and return are optimised. It is estimated that more than 75% of business development deals fall into this asymmetric R&D category. a figure made more significant because it is concentrated into those areas of advanced science where uncertainty is at its highest. The success of these deals is often mission-critical to both parties and depends heavily on how the deal is structured. The most successful contracts are those that are structured to both mitigate risk and ensure that the synergies between the partners are fully exploited.

But is there such a thing as the best kind of deal? In our research at the University of Hertfordshire, we’ve examined contracts and agreements across many different therapy areas and at every stage of the product life cycle. We expected to see a single dominant model emerge but none did. The variables of these asymmetric R&D deals are so numerous and so diverse that there is no single, optimal deal structure, only a best deal structure for any given situation. So, the question arises: what’s the best way to craft a deal? The management literature contains many theories to answer that question but they have been developed across all kinds of business development sectors. There is little empirical evidence for how well these theories, that are well accepted in academia, explain asymmetric R&D deal construction in the unusual, high-uncertainty environment of the life sciences industry. It was that gap in our knowledge that our research set out to fill.

The importance of deal structure is shown by the number of academics that have tried to explain it. When one searches the literature for theories that address this question, an overwhelming range of ideas emerge. They differ widely in their account of how deals are structured but they all share similar weaknesses; they mostly rely on large scale quantitative studies that fail to get ‘under the skin’ of the deal process and they assume that dealmaking is a static, one-off event. As we read these studies, we joked that such studies looked like scientists trying to understand a marriage by using satellite photographs.

Not all of these many competing academic explanations are of similar standing. Four dominant theories emerge as having the most developed ideas and the best, if limited, empirical support. These are complex ideas, each of which could fill a book, but we have tried to capture their essence in box 1.

Each of these four theories takes a different perspective on what is important in a deal structure. If their logic was applied to identical situations, each would predict and recommend very different outcomes, both in whether the deal should happen, and how. In practical terms, then, it is very important which of these four theories is the most applicable in the real world. As part of a set of research projects into the evolution of the life sciences industry, we set out to find which of these theories mostly closely described what occurs in the real world of high uncertainty, asymmetric, R&D deals in the life sciences.

The four most important academic theories about how to structure a deal take different perspectives on deal structure:

Resource-based theory sees the deal structure as a way of optimising the exchange of assets between very different partners
Real options theory sees the deal structure as a way of deferring and mitigating the risk inherent in any deal
Transaction cost theory sees deal structure as a way to minimise the economic ‘friction’ costs involved in any economic exchange
Property rights theory sees deal structure as a way of allocating control over specific rights, such as development and commercialisation.

The weaknesses in prior research guided our choice of methods. We knew that we had to look inside the heads of the dealmakers, not only the ‘buyer’ or the ‘seller’ but both. We also realised that we had to look at many kinds of life sciences deals to ensure we were seeing the whole picture. Eventually, after years of secondary research into the nature of deals, we executed a series of 30 in-depth interviews using sophisticated sampling frameworks and carefully ‘laddered’ interview techniques. While having a far smaller sample than earlier quantitative approaches, we intended to build on those high-level analyses by diving much deeper into the intricacies of how each deal was wrought. As we interrogated the respondents, we listened carefully to what was on their minds as they built each deal. In particular, we listened for indications that real-world practice aligned to the four theories that are so dominant in the literature.

In qualitative research of this kind, findings emerge gradually, like the scraping away of soil in an archaeological dig. Steadily, indications of many different theories surfaced but, just as the decades of prior research suggested, the four described in box 1 emerged as dominant. But, to our fascination, none of the four appeared to be salient. All four could be seen and heard in the words of the respondents but none of the four was more pronounced than the others. These findings told us something useful – all four theories were correct to some degree, but they didn’t provide much guidance to the real-world practice of dealmaking. So, we did the only thing researchers can do in those circumstances – we dug deeper still.

As we painstakingly analysed the interview transcripts, using specialised content-analysis software, the data began to challenge our thinking. Consciously or otherwise, we had assumed that dealmaking was a more or less continuous and linear process. But the data belied our assumptions.

First, it was evident that the deal process had several discrete stages, between first contact to signature, during each of which the focus and concerns of both parties shifted. In short, what is important when the deal is being considered differs from what is important when it is being designed and different again as the deal is finalised.

Second, it became clear that such deals were never made in a single, straightforward journey. Instead, they are dynamic and iterative processes in which each partial agreement was revisited and redesigned so that the deal progressed in a ‘two steps forward, one step back’ manner.

It became clear to us that the crafting of these deals was a more complex and less linear process that earlier researchers had assumed. This insight gave us a new way to look at the data. Instead of simply asking, ‘which theories, if any, appear to describe the behaviour of dealmakers best?’, we extended our research question to include ‘….at each stage of the deal construction process?’.

As Claude Levi-Strauss famously said, a scientist is not the one who gives the best answers, but the one who asks the best questions. In our case, asking a slightly better question took our science forward in a big leap.

Our better question refined our analysis. The same four theories remained dominant but a time-related dimension emerged. At preliminary stages, resource-based considerations dominated and were then balanced by transaction cost issues. These guided the decision to make a deal but contributed little to its design.

At the deal-design stage, real options and property rights theories provided a good explanation of what each side cared about and how they built the deal. Then, as the design was settled and the process edged towards the contract, transaction cost issues again became important.

Interestingly, throughout the process a fifth theory could be seen in the background. Relational contract theory, which describes how deals are partly formal and partly trust-based, could be seen underpinning dealmakers’ behaviour. And while other researchers had seen dealmaking as a single step, our findings revealed it as a shimmering, iterative dance of stages through which the partners moved backwards and forwards.

From our rich and complex results, which we can only summarise here, we concluded that earlier researchers saw only part of the picture. Each gave a partial answer to how a deal of this kind is structured but not the whole answer. We were reminded of the famous Indian parable in which a group of blind men, stumbling across an elephant, variously conclude it is a snake, a fan, a wall, a tree or a rope depending on whether they laid hands on the trunk, ears, flank, leg or tail of the animal. To their findings, we were able to add that it is essential to understand the ontogeny of the deal, that is its origination and development. Only then could it be seen that factors from all four theories, underpinned by the fifth of relational contracts, were important but that their relative importance changed as the deal developed. What our research showed that others hadn’t is that the concerns and focus of the dealmakers shift over time.

We still have much work to do in analysing our data but we think that three high-level lessons have emerged that will be of practical relevance to both sides in asymmetric, R&D life sciences deals.
The first is that antecedents of the deal – the nature of the assets and the firms – do not dictate the deal’s structure. They may predispose towards a certain kind of deal but they do not dictate its destiny.
The second is that such deals are always constructed in a series of interdependent stages so that rushing through the deal process is counterproductive. Indeed, taking backward steps in order to move forward seems quite common.

The third is that good practice in deal construction involves considering all of the factors implied by theory, from asset exchange to risk deferment and from property rights to transaction costs, but that expert deal-craftsmen adjust their focus as the deal moves from stage to stage. And underlying all of this is the idea that deals depend on trust as much as contracts.

As we begin to conclude our research, we are increasingly comparing dealmaking in life sciences with the process of raising a child. It is important not to prejudge or assume how they will end up based on where they’ve come from. They must go through each developmental stage, which cannot be hurried or omitted. At each stage, their parents need to worry about many different things but the focus of these worries vary as the child grows. And underneath all of it lies trust. Like child-raising, deal-raising is an imperfect art but it can be improved by understanding the science of the deal.

The science of the deal

Following the science

Diving deep into deals

A better question

A better answer

An Indian elephant

Raising the deal