Wednesday 14 December 2016

Corporate Strategy in Construction



Cost Curves and Competencies


Every so often something reminds me of the fact that building and construction is different from other industries, and Walter Kiechel’s book The Lords of Strategy: The Secret History of the new Corporate World was one of those experiences.

The strategists of the title were the men who founded the firms and created the modern management consultant industry. These men (and with few exception the characters in the book are all men, many of them with engineering degrees) and their ideas have fundamentally changed the world we live and work in through their effect on the corporate world, and in particular their effect on the largest global corporations that dominate the modern industrial and post-industrial economic landscape.

So this book is interesting for several reasons. The first is the story it tells about how management consulting developed from the 1960s and the manner in which it became a central player in business development. The second is that it is really a history of ideas, or from another perspective the intellectual history of an idea. The third is the way the interaction over time between the consultants with their ideas and business facing challenges changed both.

On the face of it the story of how management consulting developed and how it became a central player in business development is not a promising topic. But this turns out to be wrong, mainly because Kiechel knows so much about his subject after a career spent observing managers of major US corporations for many years, as Managing Editor at Fortune magazine and as Editorial Director of Harvard Business Publishing. He gives an inside account of the birth and evolution of strategy as the dominant business paradigm of the second half of the twentieth century.

The founder of the Boston Consulting Group (BCG), Bruce Henderson, was the originator of business strategy as we know it today. His first insight was the importance of the experience curve, or the decline of unit costs as production volumes increase (also called the learning curve effect). No other idea has had such a large impact on corporate consciousness, despite its weak empirical support: "As the 1960s unfolded, fattish, complacent American companies found themselves confronted with competition from unexpected quarters foreign manufacturers, smaller upstart enterprises in their own backyard. What was going on? What to do? The BCG had the answer to both questions in the form of the experience curve”.

The large industrial companies that turned to BCG were of course doomed to decline from their peak in the early 1970s. However, BCG gave them the tools to fight back, mainly in the form of detailed data on their costs, capital structure, customers, competitors and market shares, data these firms had never had or needed before. This data was then integrated by BCG into a corporate strategy based on gaining market share, to drive down costs by increasing production, not maximising short-term profits. Radical stuff.

The best-known concept to come out of BCG is the growth-share matrix. This ubiquitous diagram of cash cows, dogs, stars and question marks pulled together all the elements of strategy BCG thought essential. As a single, conceptual device it was a thing of beauty that captured with brutal honesty a corporations situation and the decisions to be made. It also made BCG a load of money because analysing a companys product portfolio was a lot more lucrative than drawing experience curves.

In 1973 BCGs best salesman left to form his own company. Bill Bain wondered what happened for BCG clients after the consultants report was handed in, with all the insights and data it contained. Did the clients make more profits? Did anyone know? Bain and Company did not take on projects for clients, like BCG, but had an ongoing relationship, paid monthly, with only one client from an industry, or more exactly one client from a competitive set. With this approach Bain stole a march on its competitors by taking on implementation, the nuts and bolts of managing a strategy.

Bains slogan (its value proposition) could have been We dont sell advice by the hour; we sell profits at a discount.” The keys to profits were costs and processes, and Bain developed the ideas of best practice and benchmarking, and measuring the results by a company’s share price growth. If you have wondered where the modern obsession with measurable results originated, Bain and Companys work in the 1970s and 80s would be the place to look.

The third global consulting firm is McKinsey and Company. The oldest of the three it was founded by an accountant in 1926, and in the 1950s and 60s focused on helping large companies shift from the traditional functional form to the new divisional model of organisation. However, in the 1970s it was in the doldrums and falling behind its competitors. The book tells the story of how Fred Gluck went from being a rocket scientist for Bell Labs to the founder of McKinseys strategy practice, and in the process turned it into a firm of strategy buffs’. Indeed, by the mid-1980s McKinsey had arguably become the leading strategy firm, and had put the full weight of its prestige and reach behind the strategy revolution started by Bruce Henderson.

The next character introduced is Michael Porter who would eventually become the most famous business-school professor of all time. To get there, though, he would have to fight off academic elders who wanted to deny him a job, and then thoroughly disrupt both the curriculum and the pedagogy of the Harvard Business School.”

When Porter’s book Competitive Strategy came out in 1980 it may have built on the consultants previous work, but it also laid out the possible choices of strategy (three) more clearly than anything done before, and put strategy at the centre of both business management and business school teaching. The book is now in its sixtieth printing and business education has never been the same since. The story of Porters experience at Harvard is an engrossing example of overcoming entrenched conservatism in academia.

One of the criticisms of Porter’s ideas is the lack of a human element, his corporations do not appear to have people working in them. In 1982 Tom Peters and Robert Watermans book In Search of Excellence put that right, and went to the top of the best-seller lists. Peters had left McKinsey a few months earlier and Waterman left after the books success. They emphasised the centrality of people to a companys success, although this turned out to something of a short-term victory against the prevailing management view.

Excellence also turned out to be difficult, within five years half of the 43 companies on Peters and Watermans list were in trouble. The wave of similar books that followed, with their stories of success, all found the same problem. Success is transitory, performance is impermanent, but corporations, especially large corporations, go on. More precisely, the human factors like norms and behaviours that make up what we call corporate culture persist. This turned the focus of attention onto the implementation of strategy, which is neither as sexy nor as stimulating as solving problems for a new client. What it did do was build client relationships and make Porter’s value chain the closest thing we have to a universal concept in business strategy.

Over the final chapters Keichel surveys a number of the key features of the contemporary corporate landscape. These include financial engineering, leverage and buyouts, the arrival and departure of core competencies and capabilities, and the impact of technological development and the internet. This is all interesting stuff, although these chapters lack the focus of the earlier part of the book, and unlike the original ideas in the strategy revolution none of these ideas are secrets because they have been extensively promoted by their authors. What they do is highlight just how influential in forming the contemporary corporate landscape the ideas have been.

That is one of the genuinely important insights this book gives us. Many of the conventional wisdoms heard from chief executives these days are recycled and re- treaded ideas from the early days of the strategy revolution. For better or worse these ideas have been instrumental in developing modern management methods, and in doing so have directly affected the lives of millions.

It is striking how this book shows up differences between the building and construction industry and the manufacturing, banking and finance industries where the strategy consultants were most influential. The most significant difference is the limits to corporate strategy in an industry based on tendering and low-price bidding. For many, probably the great majority, of projects, price is the only strategy. While there are contractors that specialise in a particular type of work, Westfield in shopping malls for example, and the largest have become geographically diversified through takeovers, these are the exceptions that prove the rule. Strategy, in the way it is used in this book and in other industries, is not an obviously important factor for most contractors.

The second key difference is the lack of a meaningful experience curve effect. Costs are not closely related to volume or market share, so Bruce Hendersons idea would not have much impact on a typical construction firm. Closely allied to this is the lack of opportunity and/or unwillingness of most clients to invest in developing long-term relationships with contractors and other industry suppliers. Where this has been done there have been clear gains in efficiency and reliability.

That said, construction is a more diverse industry than most, in terms of types of projects and characteristics of firms. It may be that the technological or organisational conditions for a strategy revolution in construction have not yet been put in place, although there are signs that this is changing.

The type of frontier firms discussed here clearly have strategic intent behind their move from the process-based contracting and project management business model to a more product focused approach. The increasing degree of industrialization and automation across the industry, and the rollout in late 2016 of systems like Trimble Connect, may signal a tipping point is being reached. McKinsey has become deeply involved in the industry since 2010, focused on clients and infrastructure as discussed here and here, and the McKinsey Global Institute has produced some interesting reports over the last couple of years. Other consulting firms like Deloittes are also starting to pay more attention to the industry.

Kiechel, W. 2010.The Lords of Strategy: The Secret History of the new Corporate World. Boston, Harvard Business Press.