20th May 2024

Better business. Better community

Business Industry and Financial

Social Media and AI Make for a Dangerous Banking Cocktail

Credit Suisse won’t be the last casualty as banks and regulators are clearly on the back foot in an age of flash crashes and misinformation. Now even science is starting to weigh in on what needs to change.

It is not often that a venerable scientific publication usually preoccupied with the newest discoveries in theoretical physics, astronomy, math, and biology feels compelled to comment on the state of the world of banking and finance. But with the advent of social media panics and natural language processing AI chatbots, these are not normal times.

An opinion piece in the «Scientific American» by the head of research at a large sovereign wealth fund and a professor at the Massachusetts of Technology that was published in early August concludes that the entire financial system needs to be overhauled after the dramatic bank runs earlier this year.

It is something that finews.asia has also extensively commented on (see the take by the FED, a comment by Konrad Hummler, and a view of risks in wealth management), with the key difference here being that they seem to take a more dispassionate, bird’s eye view on what has gone wrong.

No Monitoring

According to them, the regulators, including those in the US, have been caught flat-footed as they are not «continuously monitoring» banks for faulty investments that make them unstable.

«The speed at which social media can spur massive financial movements is astounding. Before Twitter (now renamed X) and Facebook, a spooked investor or customer would have to call, personally visit or even e-mail and text colleagues to urge them to withdraw funds from a troubled bank» the authors maintain.

This brings into question the centuries-old banking model of investing in long-term assets backed by short-term deposits. According to them, the model is failing.

Destabilizing Actions

Banks need to be rebuilt from the bottom up so that they are more immune to bank runs while promoting the types of investments where deposits cannot be taken out all that quickly. Regulators also need to continuously monitor banks to catch destabilizing actions at the outset, before they become anything serious.

«Lending is risky by nature, and there will always be issues, but surprisingly, both U.S. bankers and regulators know that today’s complex regulations could not have stopped what happened to SVB. Regulators and banks even expect repeated failures yet only have emergency procedures in place to respond,» they indicate.

According to them, the current situation is like an individual who is fully aware of the approaching flu season but can’t get a vaccine and ends up in the emergency room every single time no matter how sick they are.

No Default Protection

They also believe it is a big issue that current rules cannot protect against default, given insurers expect that the average bank has a 1 to 2 percent chance of failing (hence credit default swaps). Although this is similar when compared with other industries, it is also «problematic» given the systemic importance of banks.

But there are two key obstacles in making banks more resistant to runs. The first is the mix of payments and deposits with lending. The second is that banks are always subject to failure if enough depositors want to take their money out.

Narrow and Fractional

So far, so good but none of this is earth shatteringly new per se. But they go further than that with their solutions. The authors maintain that commercial banks need to be broken up into «narrow» and «fractional» banks that separate daily financial services from the loan business in a similar fashion to the former Glass-Steagall Act of 1933 introduced after the Great Depression and only fully repealed in the 1990s.

«Narrow banks that do only day-to-day operations and make their money from small fees can’t go broke because of runs on another bank. If the narrow bank’s operation and liabilities are legally distinct in segregated accounts, fractional bank clients will not have a claim on the funds of narrow banks’ clients. If the loan bank fails, day-to-day banking operations can continue, protecting the average consumer,» they commented.

The second part is closer adherence to the way that investment firms run their business. They use the example of Blackstone, saying it limits the money that large depositors can withdraw with legally binding agreements, which clients agree to because of the promise of «very good» returns. According to them, this makes them relatively immune to the «panics and fads» of social media.

Digital Regulation

They also call for a wholesale change in regulation, saying flat out that authorities must start digitally monitoring banks «wholly and continuously» rather than the current system of annual audits and stress tests.

«After all, banks settle their books at least once a day, so auditors could easily flag problems in real-time. Contemporaneous audits of bank books will improve banking behavior and discourage some riskier activities these firms might otherwise engage in. Continuous auditing also minimizes the need to seize the bank, destroy much of the value of the bank’s investments and scare everyone unnecessarily,» the authors write.

According to them, all this is becoming even more urgent as countries adopt digital money and artificial intelligence-driven advisers start managing finances, not to mention that in a separate article, the same publication indicates that AI-based models can deteriorate over time, which itself raises significant implications should its use become prevalent in the industry.

High Hysteria

For some in the wealth industry, both authors’ views might come across as naïve. But it is also clear that, unlike past industry inflection points, private wealth management can’t sit and hide behind other sectors and businesses as it is squarely in the middle of the maelstrom. 

The example of Credit Suisse only serves as a poster child of the risks of banking high-net-worth clients, who, if anything, amplified and exacerbated the prevailing social media hysteria.

Also, much of what the authors call for above poses clear and direct challenges to the industry’s standing private banking business model, which mixes narrow and fractional banking with investment services.

Swiss Challenges

For example, what would qualify as collateral for Lombard lending if investment asset withdrawals were gated or otherwise limited? 

Then there is regulation. How would a principles-based regulator such as Finma survive if there was indeed a move towards more continuous forms of digital monitoring? Would there be any appetite for legislators and the country’s citizens to change its fundamental role or not?

In all this, it would be good if Switzerland, as a country, was on the forefoot of where the industry was going, not the back one. Lest we forget, it now has the rare record of having its last two major banks fail – or near fail – in a fifteen-year timespan.