I recently came across a framework by Oxford Economist John Kay called Robust and Resilient Finance. Drawing from other disciplines such as engineering, Kay’s central premise is that modern finance is prone to accidents because of a high degree of complexity, interconnectedness, and “too big to fail” concentration. He posits that much of the fragility that led up to the Global Financial Crisis may have inadvertently been caused by processes initially designed for safety (e.g., complex derivatives initially designed to help hedge exposures) that ultimately created more problems than before because they interacted with other parts of the financial system in unexpected ways.

Kay suggests learning from the engineering discipline and building financial systems that are simple, modular, and have redundancy if we want them to withstand stress.

• Simplicity means having few moving parts that interact in straightforward ways with each other. Think Steve Jobs’ incessant focus on simplicity in terms of Apple products’ functional design and usability.
• Modularity, as opposed to interconnectedness, means pieces can be removed or replaced without materially affecting the whole. Think Lego blocks.
• Redundancy means key components have alternatives in case of failure, even if that may not be optimal during good times. Think backup power systems in airplanes, or many of the fail-safes in nuclear reactors.

All these features should bring balance to a system and smooth out sharp edges. It’s about preparing, as opposed to predicting; carrying an umbrella instead of forecasting precisely when it will rain.

While Kay’s ideas are aimed at the financial sector at large, I quite liked his framework and see clear parallels with how we manage risk in our portfolios.

People and process

It starts with people and the investment approach. A clear, easy-to-understand investment philosophy brings alignment and focus. Ours is to buy wealth-creating companies, run by excellent management teams, priced at a discount to intrinsic value. Everyone on the team sticks to this simple time-tested “North Star.”

We execute on this investment philosophy in a modular way. We’ve organized the Research team into different asset class pods, and these teams make decisions within their asset classes independently within the philosophical guidelines. Furthermore, we have democratized many aspects of our process. One example is our learning stipend program, which empowers individuals within the team to experiment, innovate, and share ideas to improve our process from the bottom up instead of relying on centrally planned and approved dictates.

Finally, we build redundancy by following a team approach, where team members operate as generalists. The decision-making model is that of “leader decides with input,” which ensures accountability but also that there be ample input, support, and continuity in managing investment portfolios.

Individual companies

Many of the best management teams have built their businesses with these same principles in mind.

Regarding simplicity, we own a few Japanese drug store operators. It’s a straightforward business model: convenience-focused general merchandise is sold in the front of the shop, and pharmaceutical prescriptions are dispensed in the back. Each store operates as its own unit, and growth to the overall enterprise comes by adding more stores to take market share away from smaller competitors, thereby reinforcing purchasing scale and cost advantages. This cookie-cutter growth is simple, repeatable, and carries low execution risk over the long run.

On modularity, we have found that decentralized operating models are often a great way for businesses to both reduce risk by having less interdependence, and increase returns by making it easier to scale given less reliance on coordination. Diploma, a distributor of industrial and healthcare components, has a very lean head office and lets the underlying units operate with a high degree of autonomy. This gives the business unit leaders, who are closer to the customers, the autonomy to be entrepreneurial, and allows the head office to focus on bolt-on acquisitions. Any single business unit facing pressure has limited risk of breaking the whole company.

Finally, in terms of redundancy, we prefer companies with lower customer or product concentrations, so that businesses can readily pivot should a given source of sales come under pressure. Balance sheets with limited leverage shelter the need to seek additional capital during adverse conditions. Of course, during good times, higher degrees of leverage can bolster returns on equity. But as value investor David Dodd once remarked: “You build a bridge that 30,000 pound trucks can go across and then you drive 10,000-pound trucks across it. That is the way I like to go across bridges.”

Portfolio construction and risk management

This discussion of concentration applies to portfolios overall.

One of our most important portfolio construction tools is our Matrix process, a simple framework that involves scoring the relative attractiveness of new and existing holdings across the various elements of our investment philosophy (strength of business model, strength of management team, and ability to purchase at a discount to intrinsic value). The Matrix ultimately helps us determine what weight to ascribe to a given security and is a wonderful tool in focusing debate by providing a common language for comparing the relative merits of investment ideas across the Research team.

With respect to modularity, we’ve recently supplemented our long-standing risk management process with a weekly meeting designed to improve our collective situational awareness. We get to learn and share potential risks and opportunities—at a macro, business model, or internal culture level—decentralizing the process of identifying potentially important signals and allowing us to incorporate ideas from across the platform into each asset class.

Finally, we set broad industry, liquidity, and individual position-size guardrails to “protect us from ourselves.” And to further ensure redundancy, we seek to understand how business models and risks may aggregate at a portfolio-level in more qualitative ways that might be hidden at first blush. For instance, what percentage of companies in the portfolio does our investment thesis rely on management’s ability to continue allocating capital effectively through acquisition? In this way, we reduce the likelihood of highly concentrated and correlated risks.

Preparing, not predicting

Crucially, the foundation to resiliency in managing investments—and everything we’ve discussed so far—isn’t the portfolio at any given point in time; rather, it’s the culture that underpins the ongoing decision-making process.

High levels of curiosity and candour within a team help to identify and debate fragilities. High levels of trust to maintain an even keel and focus on process when a storm hits (as opposed to assigning blame). High levels of humility to learn lessons and update beliefs once the skies are once again clear. These cultural attributes form the foundation of effective risk management and increase the probability of preparedness going forward.

Over the long-term, whether it be in building portfolios, bridges, or nuclear reactors, it pays to be balanced, not bold.