Zoning in on Warren Buffett

Over the past seven years I have conducted a study of the American investor, businessman and philanthropist Warren Edward Buffett. During that time my research has gone through three distinct phases: a finance phase, an integral phase and a developmental phase. The meta-narrative embedded within this history is a movement from outside-in, technical explanations of Buffett’s unique success toward an inside-out, developmental, characterological explanation with a stop along the way to consider the multitude of factors, inside and outside, that might be included in any integrated understanding of the Buffett phenomenon. In this note I want to highlight these three phases in my research on Buffett with a particular focus on the integral and developmental phases and a discussion on some of the difficulties I experienced using an integral research approach.

Looking back, my research can be divided into three distinct phases: the finance phase (2003-2005), the integral phase (2006-2008) and the developmental phase (2007- ). In a sense the developmental phase is also part of the integral phase, but the concentration on developmental theory and methods suggests a different label than integral. The finance phase is not that important in this discussion, but it was important in my research journey in that I spent two years looking at Buffett from an outside-in perspective; a traditional third-person objective scientific approach. Ultimately frustrated with the partial nature of that approach, I went to the opposite extreme and adopted the mother of all multi-perspective approaches in Integral Theory. It is hard to describe just how much Integral Theory both exploded my personal consciousness (see footnote 1) and my study of Buffett but also how in the end it was very difficult to bring back the study into a form that would be accepted within mainstream academia (see footnote 2).

At the beginning of the integral phase, I came up with a fairly simple AQAL framework of the issues that might be considered in any approach to the Buffett phenomenon (See figure 1). This was followed by a more detailed exploration from an AQAL/IMP perspective which allowed me to ask eight different questions of the Buffett phenomenon: For instance, from a phenomenological perspective, how does Warren Buffett himself experience the Buffett phenomenon? From a structuralist perspective: what are the structures of Buffett's personality that influence the Buffett phenomenon? From a hermeneutics perspective: how does the Buffett culture (e.g., the Berkshire shareholders, the community of investors who surround him) make meaning of the Buffett phenomenon? From an ethnomethodological perspective: what does the Buffett culture do in relation to the Buffett phenomenon? From an autopoietical perspective: what are the cognitive processes that guide Buffett in enacting the Buffett phenomenon? From an empirical perspective: what are Buffett's investment behaviours that create the Buffett phenomenon? From a social autopoietical perspective: how do the systems operating in the Buffett phenomenon communicate? And finally from a systems theory perspective: what are the systems dynamics that contribute to the Buffett phenomenon? These questions arose gradually and became crystallized in conversations with Sean Esbjorn-Hargens while participating in the Integral Theory program at John F. Kennedy University.

As I outlined in Kelly (2008a) An Integral Approach to the Buffett Phenomenon; A proposed mixed methods study published in the Integral Journal of Theory and Practice, I felt I had found a way of completing an integral IMP case study of Buffett within a mixed methods research framework using a method I called “IMP Coding”, a general coding scheme based on the eight primordial perspectives in IMP. In an attempt to empirically support this hypothesis, I prepared a paper, Kelly (2008b) IMP Coding applied to the Buffett phenomenon: A multi-perspective approach, in which I applied IMP coding to a sample of the Buffett data, i.e., Buffett’s 2007 letter to shareholders. While incredibly time consuming, it was nonetheless possible to match evidence in the data to the eight different perspectives in the theory. I concluded the paper as follows:

It seemed that no sooner had I concluded that one perspective offered the most insight, than I moved onto the next, which was equally insightful. Surely an understanding of Buffett’s personality was all one needed to explain the Buffett phenomenon? But then what about Buffett’s investment approach that he learned from Ben Graham? Or how could the Buffett phenomenon have developed without the Buffett culture? Or where would Buffett’s personality, Buffett’s culture or Buffett’s investment approach be if Buffett had not being born and lived in the United States at this time in history? (p. 40).

I thought I had found what I was looking for; a way of theorising about the multiple perspectives that ought to be considered in any understanding of Buffett phenomenon. In other words, I had answered the question I had set out to answer in my integral phase. But there was something static or ‘flat like’ about the answer that did not feel right. It was as if integral had laid out all the perspectives on the floor and said “these are all the things going on at the moment (in any phenomenon) which you need to consider”. But while it captured what ‘was’ going on, it did not capture what ‘is’ going on in a strange sense. I likened it to looking at financial statements (profit & loss and balance sheet) as a snapshot of a company’s performance, which are revealing but they don’t capture the dynamics of the organisation as a living system, the “is” of the business.

Similarly my ‘meta-integral theoretical approach’ was not capturing the dynamic aspects of Buffett as a living system. Searching for alternative approaches (within integral) I found that a constructive-developmental theory approach was a better match to what I was trying to do. Choosing between Kegan’s Orders of Consciousness, Cook-Greuter’s Ego Developmental Theory or Torbert’s Leadership Developmental Framework (LDF), I opted for the LDF as the most appropriate model of individual leadership development within the business environment.

What I found was that the patterns in Buffett’s life could be matched to the propositions in the theory (LDF), with Buffett’s teenage years correlating with the Opportunist stage, Buffett’s early years in college with the Diplomat stage, Buffett’s introduction and adoption of Benjamin Graham’s value investment approach to the Expert stage, the success of the Buffett partnership (1957-69) with the Achiever stage, Buffett’s ‘go it alone’ period in his early forties with the Individualist stage, Buffett excelling in a wide range of investments and businesses with the Strategist stage, and Buffett’s gradual distribution of his fortune in Berkshire and playing a ‘sage’ like role in the community with the Alchemist stage.

What is really interesting is how at the later stages (post-conventional, particularly the Strategist and Alchemist stages) Buffett begins to integrate a type of integrity, mutuality and sustainability in his behaviour, i.e. a first-person integrity and awareness of himself, with a second-person mutuality in his dealings with others (managers and shareholders) and a third-person sustainability in the businesses he invests in. His leadership approach is not just based on ‘mutuality’ in his dealings with others, but rather a mixture of these three different factors. What you get is a type of ‘distributed’ leadership in which Buffett mirrors how he ‘gets out of his own way’ in investing, in how he ‘gets out of everyone else’s way’ in leadership. It is a great story and unfortunately too long for this short blog, but hopefully I will get the chance to tell it to you in full again soon.

Many thanks and the best of luck

Edward Kelly
 

Figure 1. The Buffett Phenomenon as Quadrivia

Table 1. IMP Coding Applied to the Buffett Phenomenon.

Kelly, Edward (2008a ). An Integral Approach to the Buffett Phenomenon. Journal of Integral Theory and Practice. Summer 2008, Vol. 3, No. 2.

Kelly, Edward (2008b). MP Coding applied to the Buffett phenomenon: A multi-perspective approach. Unpublished paper.

-----------
Footnote 1:  I spent 15 months near full time on the JFKU Integral Theory course

Footnote 2: I don’t make too much of this point here, but it is a real problem doing an integral study within mainstream academia. Firstly, there is the problem of trying to ‘operationalise’ a ‘meta-theoretical approach’(such as integral theory) at the empirical level. This is perhaps not well understood within integral circles, but it is well understood outside. Secondly, it sounds so obvious to say this, but if you want to do an integral developmental study, better to do it at an integral institution, otherwise it might break your heart, which in my case has been a good thing, but it may not suit everyone.