Reducing alliance risks
Consultants are in business to disagree with each other. Which makes it odd that they all agree on at least one thing, somewhere around a half of all alliances 'fail'. Some pitch the figure slightly lower and some pitch it slightly higher but they tend to agree within a narrow range for the failure rate of alliances. But are they right? And can any analysis that does not factor into to the headline failure rate the inherent risks of a science-based business model be at all insightful? Probably not..
The process of bringing drugs to market is subject to lots of risks and some of those risks are beyond the control of the developing the drug. An alliance can fail because of scientific risks (a promising compound shows unexpected side effects), regulatory risks (a regulator's opinion of a class of drugs unexpectedly toughens because of problems with another drug) or market risks (third world countries decide to ignore patent protection for the drug. Does it make sense to treat all alliances alike when analysing failure rates, distribution agreements, marketing agreements and agreements based on ground-breaking scientific research? Probably not.
Does it make sense to treat all alliances alike when analysing failure rates, distribution agreements, marketing agreements and agreements based on ground-breaking scientific research? Probably not.
The challenge is to develop an analytical framework that takes into account the nature of the alliance and separates failures arising from the inevitable risks involved in scientific research from failures caused by sloppy execution of a business plan or by a company's senior management taking their eye off the ball.
To try to get to the bottom of this issue Silico and IBM conducted a survey among 148 senior executives in the biopharmaceutical industry to assess the extent to which the success or otherwise of partnerships in this sector were due to a failure of one of the partners to execute or due to external pressures outside the control of each party. Respondents (n=148) were asked to complete a number of questions designed to assess whether or not a partnership, from their point of view, had been a success.
The survey included three different subjective independent measures. The first measure was the respondent's response to the question "To what extent have the negotiations or partnership matched your expectations at the outset of the relationship between the parties?" Respondents were asked to rate the relationship on a five-point Likert scale from 1 for 'not at all' to 5 for 'a lot'.
The second measure was the respondent's answer to the question "Would you enter into negotiations or a partnership with your partner again even if these negotiations or partnership were not successful?" The third measure was the respondent's reply to the question "Would you recommend your partner as a partner to other companies in the sector?" For the last two questions respondents were asked to rate the partner from 1 for 'under no circumstances' to 5 for 'without reservation'.
Analysis of the data collected on the respondent's willingness to go through a partnership again with the same partner (n = 91), showed that where a partner's expectations were largely met they were almost always willing (86%) to enter into another partnership with the same partner. Fourteen per cent of respondents whose expectations had been met were nonetheless unwilling to enter into another partnership with the partner. Part of this might reflect very low expectations on the part of the respondent at the outset or that the partnership was successful in spite of the partner's failings. In 55% of partnerships where the partner's expectations had not been met the respondent was not willing to enter into another partnership with the partner. In 30% (n=26) of the partnerships the respondent was not willing to enter into a partnership again with the partner. This was taken as the true 'failure' rate.
The correlation between the respondent's expectations being met and the willingness to enter into another partnership with the partner was 0.45698 (p = 5.84e-06). Converting the data into binomial datasets and using logistic regression techniques on the new dataset Silico Research and IBM discovered that the relationship between the respondent's expectations being met as the binomial dependent variable and the respondent's willingness to enter into another partnership had an odds ratio of 2.15.
There was no significant correlation (0.09, n=502) between the geographical distance between the partners and the success or otherwise of the partnership or the willingness of the responder to enter into an another partnership with the company. Being part of a geographical cluster was not an important factor in making a partnership successful.
From this research it would seem that far fewer of the partnerships which analysts reported as having failed can be regarded as true failures.
Whatever the true figure, any failure is expensive. According to Pfizer's 2006 accounts, the company wrote off an average of $754 million a year between 2004 and 2006 in goodwill impairments caused by a failure of licensing agreements and partnerships to produce the expected revenues. Pfizer is a company which accounts for approximately 10% of the revenues of the top 50 pharmaceutical companies and top 100 biotechnology companies. On this basis the failure of partnerships and licensing agreements to perform as expected costs the biopharmaceutical industry up to $7.65 billion a year.
Similar articles
AstraZeneca's acquisition of Medimmune
The impact of alliances on share prices
The impact of cultural factors on alliance performance
The key drivers of alliance formation
The impact of conditional payment mechanisms on alliance outcomes
Learning from game theory
Procter and Gamble relies on collaboration to drive innovation
Balancing control and trust in alliances
Using balanced scorecards as a tool for partnership analysis
Exubera shows the importance of having an exit strategy
The impact of large alliances on share prices
Alliances drive the hunt for clean energy
Getting to the bottom of alliance failure rates
TAGS: Alliances Biotechnology Partnering Pharmaceuticals || Post to del.icio.us 
|| Mail us about this analysis 
