This is an interesting thread:
I worked in bank liquidity and treasury for the better part of the last decade, and there is one very interesting risk factor from the SVB collapse that I do not think we have ever witnessed before in modern finance,
First, some context. Following the 2008 financial crisis where a massive driver of stress was caused by low quality assets littering the balance sheets of massive institutions, the regulators sought to create a framework to ensure all banks were verifiably sound.
This meant forcing Federally registered banks to conform to capital and liquidity ratios. If they did not comply, they would not be allowed to issue dividends or buy back stocks, hurting the value of their enterprise.
There was the introduction of the concept of RWA, or risk weighted assets. A simple way to think of these ratios is, the riskier your asset base on a weighted basis, the more high quality capital (think cash) you would need to hold to stay in compliance.
Not only were banks required to regularly be in compliance with these ratios as a going concern, but needed to prove they could maintain certain limits when met with a time of stress across a number of incrementally adverse scenarios.
I believe this framework was developed based on the technical limitations of the time period, which is the early 2010s. Recall that companies like Twitter and Reddit and many fintech start-ups were in infancy or did not exist.
Digital customer experience in banking was improving, but many people, certainly more than today, still relied on branches to deposit checks, met with their bankers and everything had a lot more friction. And communication innovations of social media were in their infancy.
So when setting a framework, it was logical to believe that even if 50% of your depositors were to leave the bank, it would take considerable time to a) process transactions and b) have the word spread, giving these institutions a good buffer and time to recapitalize.
But as we just saw yesterday, these assumptions no longer hold, and in some ways, new tech has forced us into a very interesting position. Ask yourself, could the swiftness and awareness of what was happening yesterday happen 10 years ago? We didn't even all have smartphones.
The massive risk factor that has become very clear from all of this is what I am calling for purposes of this thread "social media risk".
I started to dig into SVB's financial situation on Thursday and had some former co-workers who were there months ago when I quit my consulting job, so I had a decent idea on how they were evolving their risk management processes and their balance sheet.
For all intents and purposes, they appeared to be in compliance with capital requirements, their balance sheet was constructed of very strong HQLA (High Quality Liquid Assets) and looked like they could very easily handle a stressed outflow.
Let me caveat here. A stressed outflow for example, would be a day where some large clients pulled down on their credit lines all at once, a few big depositors pulled out and a portion of their loans became non performing. NOT a full on run.
In fact, it can be argued that no going concern bank has the capital available to handle a full blown instantaneous bank run, thats just not how bank business models work. You can argue the merits of this but at the end of the day, this is how the system is set up.
And so this is where I started to get very confused. SVB had announced they were liquidating their available for sale security portfolio for some extra liquidity and do some rebalancing to capture higher yielding instruments to offset what they were paying in interest deposits.
To stay in compliance, they were raising ~$2B to cover the loss they realized from the sale. Clearly things weren't going great, but raising $2B in capital compared to $175B in deposits to be compliant with conservative regulations didn't seem like a bank-ending event to me.
Apparently in my infinite wisdom, I was dead wrong. The equity holders, rightfully, started selling because they were about to be substantially diluted, making each individual share less valuable post new issuance.
This of course caught the eye of many, making it a top 5 headline on Thursday, but certainly not flashing any crazy alarm bells.
Now, enter twitter.
I mentioned in passing to my co-founders "Hey this SVB stuff is crazy eh?" and one of them immediately responded "I heard from a friend they were insolvent". That's when I go worried. First off, who is his friend and second of all, what public evidence was there?
Then the industrial virality complex kicked in. Minutes later, my twitter feed was awash of prominent accounts being retweeted and raised up basically spelling the doom of SVB. The grave dancing party had started, and from there it looked grim.
Low and behold, this is how it all went down. You then hear Peter Thiel told his Founder's Fund companies to pull out and I have to imagine every single venture telegram chat was blowing up. Pull out, pull out was the message.
There was no time for investigation, no time for nuance, it was time to move, and move fast. Pull out your phone, check twitter quick to verify, switch to your SVB app and conduct a wire transfer (I'm assuming its a similar process to my business account at Chase).
We all know where we are now, and it's dripping with uncertainty. I fear we might have shot ourselves in the foot, and with how fast the world works now, there was no real way to stop it. If we had a time machine, I'm pretty sure the past 48 hours were unavoidable.
Back of the envelope, there is a very high probability that depositors of SVB may recovery anywhere from 90% to 100% of their accounts given the quality of assets on their balance sheet and the ability to liquidate the loan portfolio, even assuming 10% losses.
But it will take months now to recover all the funds, months that some companies do not have to wait for cash to hit their accounts. It's painful, but it could have all been avoided, especially if the outcome is indeed max or near max recouping of funds.
So how could this have been avoided? Well for one, SVB should have been doing more prudent portfolio rebalancing and should have been much more communicative way ahead of this incident, both to investors and large depositors.
But the reality is, the banking industry has to start adding this "social media risk" into their frameworks and find a way to get ahead and control the narrative, to avoid this type of situation from happening again.
It's scary to think, but the same types of tactics that can manipulate an election can be used to undermine the strength of a bank. This time around, it was pointing to unrealized losses on high quality assets to pile on and justify a panic. But it can manifest from anything.
The only way I see to combat this is for firms to have a stronger social media presence, one that they take seriously, not just to push certain social messaging and promotions.
Meet your customers where they are, in their preferred medium. Start to do breakdowns of your earnings reports across all channels. Give ad hoc updates during the quarter and make sure everyone is aware of your status. Transparency is the key to this mitigating this risk.
I also strongly believe we, as the consumers, need more opportunities to be better educated on a deeper level on how this system, that we trust with our money, operates. It's been distressing to see some of the commentary on this site the last two days.
TLDR:
- Tech advancement has made finreg very dated
- SVB wasn't on the greatest footing, but was largely taken down by virality and panic
- Social media risk needs to be defined and taken seriously
- Banks need to be more transparent about their business on socials.