The post-Megaproject Era: an end to Magical Thinking

In his major tome on Industrial Megaprojects[1] in 2011, Ed Merrow poured decades worth of empirical assessments and statistical analysis into this largely unresearched field.

Merrow’s work on megaprojects is worth revisiting, as the oil industry perches over a new landscape, 2016 onwards, that is currently barren of large extraction plays.

His analysis can conceivably be summarized as follows:

  • There is no substitute, ever, for good up front project planning: it does not guarantee success, but neglecting it does guarantee failure
  • Megaprojects are always complex far beyond the technical: owners, project teams, partners and suppliers often intend to work together, but competing priorities, cultures and events generally get in the way of delivery
  • Owners and contractors “live in different worlds: they fundamentally do not understand each other” – they need to, and manage project risks accordingly.

Importantly, Merrow’s first adage regarding up front planning relates to all aspects of project strategy and choice: to concept selection, to alignment with corporate capability, and to ensuring partner positioning, not just technical front-end engineering.

There is much more covered in the book, but these fundamentals need to be thought about in some detail by any firm coming out the other end of the recent era of megaproject investment, where performance has been systemically poor (see here and elsewhere).

Some industry players have in fact called an end to megaprojects themselves, at least for the medium term. And even those who remain more positive agree that without a return to something close to a sustainable $100/bbl oil price ($80/bbl theoretical break-even plus historic over-run), megaprojects as a future option look unlikely.

It’s also worth considering the view that if a system or process is systematically delivering poor outcomes, its time to discard or redesign the system.

In this context, if oil majors and national champions wish to retain large projects as a key device in their tool-box, some basic rethinking is required.

Fixing megaproject delivery will be a long-term endeavor. But it has to start somewhere, and begin quickly.

Look upon it in footballing terms as a somber post-match review after losing a series of major fixtures. Something has to change: doing the same things with more diligence will almost certainly not be enough.

An obvious entry point is that sober assessment of current performance. But this is not as simple or accessible as it sounds.

Merrow uses a neat phrase in his work to describe the inability of major project teams to confront difficult truths: he labels it magical thinking. He then cites three areas where normally “hardheaded” and “practical” megaproject directors are prone to it:

  • Overcoming organizational complexity – assuming teams just come together
  • Preferential contracting strategies –key project risks can be offloaded via negotiations with contractors
  • The existence of a contractor “A” team – deep in the business of every contractor lurks this mythical group

And he points out that the impact of being disposed to it can cause havoc with project outcomes: for example, by assuming teams will manage project complexity by just learning to work together, or that certain contract forms such as lump sum will automatically reduce risks. The dull, hard evidence does not support this.

A new pragmatic realism and skepticism therefore needs to be introduced, one that avoids magical thinking in its various forms, such as optimism bias, and generous screening criteria.

Painful though it will be to examine, there is now a deep literature of poor project delivery to review, and use to improve future implementation.

A new agenda should probably therefore take two forms:

Understand the Nature of Megaprojects

The inherent nature of megaprojects is in fact straightforward: the bigger they are, the more prone to over-run they become. There are profound structural reasons why this is the case, even for relatively straightforward designs, never mind the novel designs required in oil and gas.

This simple insight should not be dismissed either as obvious or only anecdotal. In a study of several hundred infrastructure projects, Flyvbjerg[2] and others showed over half the projects over-ran by more than 40% on cost. But this average out-turn masked the far higher failure rate of projects outside the US, with a significant proportion of runaway projects costing over 2-3 times the original estimates, especially as projects grew large in scope. The chart from Flyvbjerg’s work, below, illustrates the megaproject out-turn by geography, and it should be used as an objective reminder to any portfolio screening.

Future practice should demand that economic assessments of megaprojects requires an objective analysis of the reference class – in simple terms, other megaprojects. This means that the generally poor performance of these types of projects can not be dismissed by an enthusiastic belief that our project will turn out OK. Some practical steps would be:

  • Independent database assessment of oil and gas and other industry megaproject outcomes in the last 10 years – categorization of the risks / issues contributing to outcomes
  • Robust stress scenarios – bias toward probabilistic / fat-tail assessments with appropriate contingencies, rather than linear analyses.
  • Alignment of stakeholders – simplicity in operating model and organization, avoid adding to the difficulty the project team will have in managing the inherent, multiple interfaces in any megaproject

Realism of Project Strategy

Turning to project team beliefs that need to be challenged:

  • Contractor Takes the Risk – Merrow is blunt on this – “significant risk transfer from sponsors to contractors is structurally impossible”. This is due to basic finance, not beliefs – contractors are generally thinly-capitalised service firms, with balance sheets and market capitalisations an order of magnitude less than most oil firms. Even if they could price in vast premiums to accept such risk, most now won’t because of specific policy restrictions.
  • Contracting Strategy The Solution – many commentators including Merrow look at contracting strategy as a second-order phenomenon. By this they mean it is situational, wherein various approaches may succeed (or fail) in driving project success. Preferences for Lump Sum, or Reimbursable contracts and so on need to be placed in this context. There is no silver bullet that will overcome the large array of issues that will beset large-scale projects.
  • Maximise Innovation and Standardisation – which is it to be? Project teams can be guilty of taking on everything to suit everyone, and assuring everyone that it is all achievable within targets. Pressure to drive corporate programs and processes introduces system novelty into megaproject complexity, as does the requirement to enforce global procurement agreements over and above contractor scope and capability. Realism in these two areas is critical, rather than a hopeful belief in accomplishing it all.

This is an incomplete list of where to begin, but the key point is to use the preceding era of poor performance, factually, to drive improvements.

Both parent corporations and their project delivery teams then need to take a more straightforward and clear-headed assessment of how to select and manage megaprojects.

It can be viewed as two-levels of autonomy that today often get confused:

Corporate level – Autonomy of Selection: management of the performance database, setting of insightful stress scenarios, communicating a clear bias for simple project organisations, either as direct operator or partner, and maintaining a credible range of innovation and standardisation initiatives.

Project level – Autonomy of Delivery: appropriate organization design, situational contracting strategies, and a credible implementation plan for innovation and/or standardization requirements.

Too often, as Merrow describes, and recent performance has shown, these two categories can be switched around, with corporate centres setting intricate standardization agendas, while project teams provide their own independent assurance of overall project targets.

Such a new framework calls for more severe screening of projects, especially megaprojects. This will be a negotiation, but it needs to be based on facts of recent performance.

To be sure, such a selection process is likely to result in an unsympathetic and widespread cull of preferred projects, causing concern about how production decline and reserves replacements can be accommodated when so many ventures are frozen or deleted.

At the extreme, it could result in the cancellation of whole portfolios.

That may in fact be necessary.

The system in place has not delivered performance, and the focus on large-scale extraction for a commodity where the biggest investors are all price takers, retains substantial risks. Big, bespoke, complex, immobile assets with huge up-front costs need alternative models. Their long time horizons also provide a late coming to supply pressing energy needs.

The answers to this are not yet clear. Achieving the benefits of big extraction via smaller scale or more scaleable designs is one potential solution (shale extraction via fracking and well tie-back projects are two current straightforward examples), as is a greater focus on developing existing assets. What is evident is that new concepts need to be urgently devised, or alternative energy technology methods will fill the void left by unsanctioned megaprojects.

In sum, solving some of these fundamental problems with the large-scale hydrocarbon extraction model will be an exercise requiring a major change in oil industry leadership, management thinking and technical design. The inherent nature of megaproject complexity needs to be confronted in a clear-eyed way, and more efficient extraction models devised.

It cannot be resolved working with incremental changes to the existing frameworks, or working harder on current concepts, and then relying on magical thinking to do the rest.

The alternative is to convince all stakeholders that upwards of $100/bbl is the price we have to pay to develop oil and gas



[1] E W Merrow, Industrial Megaprojects: Concepts, Strategies and Practices for Success, 2011.

[2] B Flyvbjerg et al, Big is Fragile, arxiv preprint