The Fragility of Scale

In my last blog post I lamented the obstacles to innovation in mining arising from the lack of small-scale mines. TLDR: efficiency trumps flexibility, creating large-scale operations that must be protected from experimentation. I believe that the small-scale gap is harming the industry by shielding it from innovators. The result is fragility. This may be counterintuitive if you think that big = resilient, but consider the case of Rio Tinto’s Juukan Gorge incident in 2020: no-one dared interrupt the large scale plan, so no-one blew the whistle to prevent the calamity (see pp23-26 of the Parliamentary Report). The criticism that the company incurred continues to this day, despite enormous efforts to change. That’s the fragility of scale.

A quick reprise of the case for change. A profound challenge the industry faces today is this: what worked for “complicated” doesn’t work for “complex”. In the face of complexity, we have to be more human and less expert. We have to build relationships and trust. This requires learning (behavioural, not just cognitive), which requires room to manoeuvre. In other words flexibility.

In this post I want to dig deeper into how flexibility has been a casualty of the pursuit of scale. It’s a systemic problem. Here’s a summary:

Summary of the vicious circle of the pursuit of scale

When I encounter patterns that seem “stuck”, I reach for “Thinking in Systems” by Donella Meadows. She described how systems can take the form of self-reinforcing linkages between people and organizations, in which local optimization leads not only to poor outcomes at the system level, but also to great resistance to change.

Over the past dozen years I have looked the lifecycle of a mine through this lens: I have observed who does what, how they are engaged, what the incentives are, and what priorities arise from the business models of their organizations. I have drawn the conclusion that the entire mining ecosystem has lost the ability to balance flexibility with efficiency, for reasons of local optimization. In other words, because everyone is doing what their organization requires of them.

Before describing this systemic problem, I want to destabilize any notion that “bigger = better” is a natural law. If it were, then mainframes would still be the only computers, we would all be connected to district heat schemes, there would be no retailers smaller than Wal-Mart, and perhaps worst of all, there would be no craft breweries.

The reason we don’t live this way is that we have diverse and shifting needs and preferences, so we value flexibility. We want to take advantage of unexpected opportunities. These are defining features of dynamic capitalism – indeed, economy of scale seems philosophically closer to centrally-directed communism than to capitalism, at least to my eyes.

Back to the systemic problem. Over the past half century or so the mining industry has converged on a dangerously uniform and static business strategy, even as the stability of its operating context has deteriorated. The business strategy and the engineering practices are mutually reinforcing, aided and abetted by some human factors relating to status, reward and prestige. Here’s the complete stable loop I see:

Pursuit of scale - the complete vicious circle

Capital Intensity

Because the problem is circular, we could start anywhere, but I find it easiest to start at the top, with the assertion that mining is a capital-intensive business. Of course, it’s not true if you’re an artisanal miner – that’s labour-intensive. So strictly speaking, only capital-intensive mining is capital intensive, but capital-intensive mining is where the power is, because it is capital intensive: the executives who hold that power project their strategy onto a whole industry. No matter, let’s take it as the starting point for our vicious circle.

Shareholder Primacy

Once we hold the mindset that accessing capital is the top priority, we tend to make design choices in whatever way makes most sense to investors. I’m not sure investors (or perhaps more accurately, their analysts) fully comprehend that they set the agenda for how the engineering gets done. But shareholder primacy is alive and well even in an industry that talks about satisfying all stakeholders. The investors are still the customers, and they call the shots.

Design for Investors

In the past, when a broader base of retail investors knew little about the risks of mine design or even geology, some proponents traded ethics for capital, and thus were born the regulations that (to a debatable extent) protect investors from being misled today. It seems not many share my view, so I will write a separate article on this, but regulations are part of the problem. Regulations frame the design task as complicated, not complex, and focus on the technocratic expertise and impartiality needed to address technical risks – this is not surprising, as they are written by technocrats. The relegation of non-technical risks (which are inherently complex) to the auxiliary status of ‘modifying factors’ – and the scant attention these factors typically receive in studies – are evidence that the regulations have not kept up with the times. They are a relic of an era when the problem was merely complicated.

Codified Design Products

Regulations that protect investors sit at the apex of a canon of guidelines and codes of practice whose combined effect is to standardize the design process. The rules also confer exclusive practice privileges upon “Qualified Persons” with defined requirements for independence, experience and qualifications. They even lay out the format of studies by chapter title and number. This is a dream come true for the engineering and geoscience consultants – a highly repeatable “sausage machine” of study production by a privileged elite. It also fits nicely with the misconception that the problem is still merely complicated, because the best strategy for a complicated problem is to outsource to experts, especially if you are focused on long-life assets that you think only need to be designed once.

Outsourced Design

Thus an effect of regulation has been to change the structure of the industry by turning design into something that is bought, not something that is made.  This also makes it very difficult to build trust with stakeholders by bringing them into the design process, but let me deal with that in a future article. For the present purpose, the so-what is that the design task has been commoditized.

Stale Design Task

In a commodity market, differentiation is impossible. Engineering companies might say that they are innovative and exceptional, but the work they are contracted to do rarely takes advantage of those qualities. For the most part, they are ‘bounded gig economies’ of QPs sharing a logo and a project management system, and their Achilles heel is their internal transaction costs in the face of complexity, which create a sub-optimal design process. However, the target audience for studies – investors – has too many prospects on its plate and would rather all the studies looked the same.

There is a conspiracy of mediocrity here, and it’s another vicious circle that goes like this: engineers are not contracted to do anything novel => investors are not exposed to anything novel => proponents experience no investor pressure to study anything novel => engineers do nothing novel. This has directly created the present challenge for senior miners’ M&A desks, in which all prospects have feasibility studies that are not compatible with the acquirer’s commitments, for example to future net zero operations. But again, let me cover that in a future article.

Static Models

For the present topic, the major failing of the conspiracy of mediocrity is that it is satisfied with static economic models – models that assume stability in metal prices, costs, regulations and ore body intelligence. This is convenient for the engineers because static models can be constructed in a spreadsheet, which we all know how to use. And it is enshrined in the canon of rules, which still (egregiously) define a feasibility study as comprising a single scenario. But in a volatile world, static models are becoming increasingly inadequate. Indeed, I would argue that the spider diagram of straight lines showing model sensitivities, which we almost always see in published studies, is borderline professional malpractice today.

Flexibility Neglected

What static models prevent is a calculation of the economic value of flexibility. I will explain why in a future article, but suffice it to say that they see only the cost of flexibility, for example the additional capital cost of a modular plant. They never calculate the value of flexibility. This means that economy of scale always wins, especially where the owner wants to minimize capital expenditure as well as maximizing net present value. The lowest capex solution is always the most fragile – no redundancy, just one big cheap basket for all the eggs.

Economy of Scale

The end result is that mines and process plants are made as large as possible because alternative evaluation strategies never see the light of day, especially at the portfolio level (because regulations are focused at the project level). Our self-reinforcing loop is complete.

The Strategy Lock-In

What sustains this model is the notion that large mines mean large companies, which are resilient (and pay well). This is analogous to the strategy of taking to a hazardous road in an armoured vehicle, rather than a manoeuvrable car.

In a world of increasingly diverse and empowered viewpoints, it would be a profound mistake to believe that we will always have permission to drive that tank down the highway.  

I think we need to get into the collision avoidance business, and that will demand at the very least a new mindset of small, agile and scalable operations.