Back to Blog

Seven Bad Habits of Risk Management

I’ve spent the last few weeks reading through the report of an investigation into the default of Archegos Capital Management. The report was prepared by and for the Board of Directors of Credit Suisse, the Swiss investment firm that lost $5.5 billion when Archegos defaulted. I really wish I could report that the Archegos default was completely out of the blue, that Credit Suisse was in innocent victim in the situation, and that there are brand new, groundbreaking lessons that executive teams could pull from this tragedy. Unfortunately, the precarity of Archegos’ finances was clear to see months in advance, Credit Suisse was hapless in their response, and reasons for the default are all well known.

Steven R. Covey made his mark in the world by identifying the Seven Habits of Highly Effective People, and the title of this post plays off of his excellent work. Credit Suisse lost $5.5 billion because of known issues and the unwillingness of those running and managing the Archegos account to make hard choices in the face of clear evidence. I’ve pulled out seven critical mistakes that contributed to the loss:

1. Mistaking friends for enemies. Every investment bank has two “desks,” or groups of experts involved in doing business: the trading desk and the risk desk. The trading desk brings in money for the firm, and the risk desk makes sure that the cost of the money (in terms of risk) leads to profits and not losses. In well-functioning operations, the two desks have a healthy and productive partnership, and the bank has high quality earnings. At Credit Suisse, the traders viewed the risk desk as policing parents and fought against or worked around limits imposed by the risk desk.

Bottom line: Risk and trading have different incentives that act to check and balance the worst impulses of the other group and protect the interests of the firm as a whole. Make sure that groups with competing incentives see and treat each other as partners working for the same overall goal.

2. Gains today sewing lies about losses tomorrow. The report estimates that Credit Suisse (CS) brought in about $60 million in revenue from the Archegos account. That’s not chump change to the traders involved, and they saw a steady stream of commission in these trades. Those real commissions helped them overlook potential losses in the future. It’s a common refrain: why should we worry about paper losses tomorrow; we’ve earned real revenue today!” This thinking led to losing $91 for every dollar they made.

Bottom Line: CS traders threw their own company under the bus to preserve their own income. CS had a set of systems in place that made it easy, convenient, and lucrative for traders to adopt a myopic perspective. Great compensation systems tie current individual fortunes to the long term health of the firm.

3. Preferring dependence to interdependence with trading partners. Each time the trade desk would push back on trading with Archegos, they’d get the same response: “we can’t push the client that hard; they’ll take their business to others on the Street.” In discounting the risks of increased business with Archegos, CS notes that the client was a “significant relationship” for CS, and that holding the Archegos to company standards “may result in irreversible damage to the client relationship.” This mindset of dependence made traders dependent on bad business when they should have walked away.

Bottom Line: Archegos was a classic bully of a client, threatening to pull their business if CS failed to do what they wanted. Make sure you have a healthy, interdependent relationship of value creation with clients, not a toxic relationship of dependence driven by a fear of loss.

4. Thinking in marginal, “one-off,” terms. The CS risk desk had a clear set of potential exposure and scenario loss limits established for each client based on the risk appetite of the CS trading desk. Between April of 2020 and March of 2021 (when the firm defaulted), Archegos’ Potential Exposure limits exceeded the limits by a factor of between 3 and 30. That’s right, CS accepted 30 times as much risk with Archegos than its internal policies would tolerate. Traders and the risk desk allowed these consistent overages because they reported each one as a unique or “one-off” problem.

Bottom line: treat “one-off” events as “one off” events. The first time might be an accident, and the second, in the best case, is an incident that should be watched. The third time marks a precedent and a new normal. Act decisively the second time a problem occurs, not the 52nd time.

5. Focusing on the trees, not the forest. Just before Archegos went into default, CS teams began to see a very ugly truth: they had focused solely on credit risk in their analysis and planning; they somehow forgot to remember that CS was the counterparty to all the hedging transactions that Archegos completed with CS. I’ve noted that Archegos exceeded internal CS credit risk targets by whopping amounts (and yes, whopping is a technical finance term). Just before Archegos defaulted, CS realized they were counterparty to $25 billion worth of transactions, and the credit risk would morph into market risk as CS would have to unwind all those trades.

Bottom Line: Specialists focus on specialties, that’s what we pay them for. Someone in the shop must have responsibility to look beyond those specialty (trees) and view the entire portfolio of risks (the forest). Threats to the forest as a whole can be another point of leverage to getting those focused on their trees to change course; the health of the forest relies on the health of individual trees.

6. Changing metrics creates confusion. As the relationship with Archegos matured, and as the risk level went up, The CS risk desk was also responding to internal challenges and changes. Credit Suisse rolled out a new formula for calculating Potential Exposure (PE), and the change in the formula caused a significant bump in Archegos’ PE numbers. Rather than trust the numbers as reflective of a more accurate measure of risk, the risk desk discounted the PE numbers as inaccurate and foreign. They confused a siren for noise.

Bottom Line: Measures matter, and changing how we measure risk (and success) is as much a cultural and institutional practice as a mathematical and technical one. If measures change, but not the underlying culture, then people won’t trust the numbers and they’ll make poor decisions.

7. Failing to share bad news with “mom and dad.” The most shocking finding of the report was the narrative that the top managers of CS Prime Services (the part of the organization that dealt with Archegos), and the senior executives of the firm had no knowledge of the risk posed by Archegos until after the firm defaulted. A $20 billion problem seemed to be nothing more than a “local problem” and no one in either trading or risk saw any need to escalate reports of the risks or perils Archegos presented. My sense is that the failure to move bad news up the chain is both a reflection and consequence of the other six habits noted above.

Bottom Line: Create processes and cultural norms that encourage people to move bad news, or big risks, up the food chain to senior management. Senior leaders can call for reinforcements, and they may be more effective in helping manage high risk, touchy clients.

Credit Suisse lost $5.5 billion because of a series of unforced errors. None of the 7 habits noted above are new or shocking, and “off the rack” solutions to each of these problems exist. In fact, CS had many policies on their books that could have averted the crisis, but in the pursuit of todays fees, those rules became a hinderance and a brake. If you and your organization want to manage risks more effectively, do exactly the opposite of what the CS team did: be a highly effective (risk) manager instead.

Get Updates into your inbox