The Art of Distinguishing Real Risks from Pretenders

These days, financial institutions and their risk managers seem to be more focused on geopolitical risk than on data-driven credit risk issues. It’s certainly fair to question, though, whether this is a worthy tradeoff.

Are geopolitical dangers really a big threat to banks or should they be more concerned with credit-fueled asset bubbles, regulatory limitations, and the excessive and foolish risk taking connected to exotic financial instruments?

Tony Hughes

Risk managers' shift in thinking was driven home to me when I attended a recent conference in London. While all the presentations were interesting and informative, they were focused primarily on geopolitical threats and other “big-picture” problems currently facing the global economy.

Perhaps I’m a nostalgic dinosaur, but I remember past industry risk conferences that were dominated by new sources of data and innovative modeling techniques for stress testing and credit scoring. I’m sure this work is still going on, but few seem to be talking about it publicly.

As I attended the sessions, I kept asking myself if the issues being discussed are truly dangerous for banks and, even if they are, whether there is any response other than closing down and refunding the investors’ money.

When risk management is overwhelmingly conjectural in its approach, literally everything becomes a potential threat. This is very distinct from an evidence-based philosophy in which a variety of potential perils may be assessed but many will be dismissed, either because the data indicates a high level of robustness or because an effective strategy to mitigate the risk cannot be identified.

Geopolitical Risk Impact: Follow the Data

The irony is that banks have historically done pretty well during periods of geopolitical turmoil, with communist revolutions being the obvious exceptions.

Bank of England data on bank profitability stretches all the way back to the 17th century. Prominent upward movements can be seen in the data during the Revolutionary and Napoleonic Wars – and especially during the First World War.

Fast forwarding to late 2021, the ECB reported that bank profitability returned to normal very quickly following the COVID pandemic. Obviously, there were high-profile bank failures in 2023, but these were due more to mismanagement of what should have been rapidly improving conditions for banks.

More contemporaneously, reports have emerged from Ukraine suggesting that local banks are performing well, despite the ongoing hostilities with Russia. While the lives of Ukrainians are seemingly in grave danger, their bank deposits are currently safe. (Thank heavens for small mercies.)

The clear conclusion is that while wars and pandemics are serious threats to humanity, they are largely benign events for banks. Indeed, they normally provide an opportunity to profit.

Regulators and Risk Perspectives

The other data point on the impact that conjecture is having on risk management comes from a Financial Policy Summary, published by the Bank of England (BoE) in late June, that strikes me as a bit odd. Investors, that report states, are “… continuing to put less weight on risks to the macroeconomic outlook” in an “uncertain” risk environment. “Valuations and risk premia are therefore vulnerable to a shift in risk appetite that could be triggered by factors, including a weakening of growth prospects, more persistent inflation, or a further deterioration in geopolitical conditions,” the report elaborates. 

However, the UK has recently dragged itself out of a shallow post-Brexit recession, and data tells us that its outlook is now looking slightly more rosy. UK inflation, for example, dipped to two percent in May, in line with the BoE's target range.

The global economy, meanwhile, has been performing better. Most global credit indicators remain fairly quiescent, and the U.S. has been rapidly adding jobs.  

You’d think that the key British policy setters would be pretty happy with the situation, all things considered. Nonetheless, the BoE's report focuses on the idea that “market prices remain vulnerable to a sharp correction.” 

How to Identify Real Threats: Benefits of a Data-Based Approach

It may turn out, of course, that the views the BoE expressed in its recent report ultimately prove correct. Financial markets may be on the brink of a correction. That’s one of the defining features of conjecture: it can sometimes turn out to be prescient. But conjecture is not as valuable as historical data.

A data-driven approach to identifying an overbought stock market would involve a detailed discussion of price-to-earnings ratios and a demonstration that prices have recently risen well above the established long-term trend. It would perhaps point out that earnings growth had fallen relative to nominal GDP, and would include a careful discussion of Sharpe ratios and the VIX index.

Instead, the BoE is placing an emphasis on a vague notion of elevated macro risk. Its recent report does not specify exactly why the central bank views today’s “risk environment” as unusually uncertain – but seems to suggest that investors should sell off their stock portfolios to avoid a looming loss, based on a range of unspecified geopolitical risks that may or may not be impactful. This obviously conflicts with the slight economic data improvement we’ve seen recently.

Risk management executives and regulators who favor the conjectural approach will say that damaging new shocks are never fully reflected in historical data. They will point out that models failed during the pandemic and that a narrative-based approach is somehow superior to one that relies on historical data analysis.

But the problem they fail to acknowledge is that the fashionable, subjective approaches cannot rule anything out under any circumstances. This means that everything is a potential threat. Indeed, under conjectural approaches, financial managers are given no indication of the perils that may be especially hurtful – and that should therefore be their primary focus.

The U.S. Fed Chairman Jerome Powell gave a useful counterpoint when he was recently asked about the inflationary effect of immigration. He ruled out the threat in a straightforward manner, citing evidence to make the case. Mr. Powell’s colleagues at the BoE should learn from his example and improve the clarity and quality of future Financial Policy Summaries.

Parting Thoughts

Historical data informs us that wars, pandemics and high levels of immigration are not major threats to the financial system. It is likely that other geopolitical threats will also have little impact on the performance of banks, which can adapt to a wide range of issues and which remain highly profitable.

On the contrary, our experience from the GFC and the Great Depression suggest that credit-fueled asset bubbles have a high potential to inflict damage on banks.

Looking at past bank failures, fraud and lack of oversight are common features. The lessons of both the GFC and the Savings and Loan Crisis suggest that an erosion of regulatory restrictions on exotic forms of banking can trigger mass failures. Indeed, the failure of institutions like Silicon Valley Bank suggest that exotic approaches to running a bank can be very dangerous.

When we’re musing about a rumored threat, we may not be able to build a specific model, but we can still apply an empirical mindset. If the characteristics of the potential shock mimic past events that had little financial effect, it would be reasonable to shrug it off and concentrate on something else.

Finally, I’d argue that regulators must hold themselves to higher standards while acting as representatives of the public.

If banks want to respond to a non-threat, only their investors will be harmed. Public officials, on the other hand, risk a devastating loss of credibility if their discussion of threats, absent any analysis of data, either distorts market behavior or simply doesn’t pan out.

Chairman Powell provided a useful model for financial regulators to follow.

Tony Hughes is an expert risk modeler. He has more than 20 years of experience as a senior risk professional in North America, Europe and Australia, specializing in model risk management, model build/validation and quantitative climate risk solutions. He writes regularly on climate-related risk management issues at UnpackingClimateRisk.com.