Welcome to another issue of Net Interest, where I distil 25+ years of experience analysing and investing in financial stocks into a weekly newsletter. FTX again this week and for paying subscribers, comments on Japanese bank, Norinchukin; investment bank, Jefferies; and the concept of ergodicity. To gain access to the broader suite of content, do sign up as a paying subscriber by clicking here.

Finance revolves around trust and much of the growth in finance over the years has been built on methods to scale it. Grand banking halls were an early innovation, designed to signal to customers that the bank would still be there in the morning and that their money is safe. In some cultures, financial services providers allied themselves to religion to leverage the trust available there – the pawnbroking industry, for example, sprang up in the Buddhist monasteries of China. More recently, central clearinghouses, credit scoring and deposit insurance are all technologies the finance industry has invented to scale trust. Clearinghouses substitute the need to trust any particular counterparty with trust in a single centralised entity. Credit scoring makes the process of underwriting trust in a consumer much more efficient. And deposit insurance eliminates the need for depositors to undertake due diligence on the financial condition of their bank. 

Yet despite the progress, we’re a long way from securing trust completely. The recent collapse of FTX highlights how trust can break down on a number of levels. 

Trust in the Man

On December 18 and 19, 1912, the Bank and Currency Committee of the House of Representatives in Washington DC held an inquiry to investigate “money trust” on Wall Street. Giving testimony was John Pierpont Morgan, in the twilight years of his life (he would die three months later). The attorney for the committee asked him some questions:

Attorney: Is not commercial credit based primarily upon money or property?

JP Morgan: No, sir; the first thing is character

Attorney: Before money or property?

JP Morgan: Before money or anything else. Money cannot buy it.

John Pierpont Morgan was addressing the question of trust in commercial credit. Since then, large corporate borrowers have become more institutionalised, making the question of character more diffuse. These days, if a bank were to underwrite on the basis of character before money or anything else it would be a red flag. Credit Suisse and other creditors are currently facing losses of more than 50p in the pound after misjudging the character of industrialist Sanjeev Gupta. And we’ve discussed here before the folly of Voyager Digital, whose CEO explained his rationale for not taking sufficient collateral on loans: “The people we lend to are some of the biggest names in the industry.” 

In other areas of finance, though, character remains a critical part of the underwriting process, particularly in investment management. Hand someone a piece of your wealth and you want to trust they will manage it well. One of the reasons indexation has been so successful as an investment strategy is that it obviates the need to do this. No wonder – building trust is a difficult process. 

Graham Duncan is co-founder of East Rock Capital, a multi-family investment office that makes investments in actively-managed funds. He has interviewed and assessed more than 5,000 investment managers. “In my role investing in and seeding investment firms, I conduct extensive reference checks on people in order to try to accelerate the process of building trust,” he says. His process is “to gather private information in the form of multiple perspectives about how someone has played a repeat iteration game and more accurately project how they’ll play in the future.”

Few allocators have the patience of Duncan and instead lean on a series of heuristics. One is past performance, even though it is widely known not to work in public markets (and certainly doesn’t scale: strong performance in a small fund has very little bearing on how a fund will perform with more assets). Other heuristics include patronage by a well-respected figure; credentials, for example from academia; and social proof – copying what other investors do.

Social proof shows up in all areas of investment. In his memoir, What It Takes, Steve Schwarzman describes how hard it is to raise a fund without a track record. Two insurance companies committed to his fund early, but only if others would invest alongside them. Eventually, Jack Welch, CEO of General Electric, realised the power he had. “I’ll give you $35 million,” he told Schwarzman and his partner. “Why? Because you are great, both of you. That way you can use the GE name to help get some other people.”

In venture capital, the imprimatur of certain investors has become especially important. Likely, this is because the process of assessing character is both harder, given the lack of a track record, and more important, given so much of the financial return from an early stage investment is tied into it. But once a high profile investor is on a cap table, it can have a big impact on the credibility of the underlying company. One of those high-profile investors is Sequoia. 

From the outside, it’s not clear what process Sequoia itself uses to assess the character of founders. Hundreds of historic investments give its partners a sense of pattern recognition, which can often present as a hunch. In his book, The Power Law, Sebastian Mallaby describes the moment when Sequoia decided to invest in Stripe. He writes that Michael Moritz, a leading partner at the firm, was an astute judge of character, and his questions to Patrick Collison, Stripe’s founder, were designed to detect resilience and ambition. But, “when Patrick said he climbed the Old La Honda in less than twenty minutes [on his Cervélo bike], he felt that he might have passed a test. The fact that he was competitive at a gritty sport said something about his aptitude for entrepreneurship.”

Sequoia’s backing of Sam Bankman-Fried at FTX is equally difficult to assign to a robust process. Before FTX collapsed, Sequoia had a profile of its founder, Sam Bankman-Fried, up on its website. The profile describes a Zoom call Sequoia partners had with Bankman-Fried ahead of their investment in FTX in summer 2021. “I LOVE THIS FOUNDER,” typed one partner into the Zoom chat window towards the end of the call. “I am a 10 out of 10,” pinged another. “YES!!!” exclaimed a third. 

One of the partners behind the deal elaborates on what they saw in Sam Bankman-Fried. “Not only had he been a top trader at a top firm – and thus, the ideal customer [for an exchange] – but both his parents were lawyers.” 

Sequoia doesn’t have responsibility towards parties outside of its limited partner investor base. And it did have skin-in-the game – its partners lost $213.5 million when Sequoia wrote down the value of its FTX investment. But its brand, and that of peer firms, is increasingly used as a stamp of approval. After the collapse of Greensill Capital, David Cameron, former prime-minister of the UK and advisor to the company, told a parliamentary hearing: “To be frank, one of the things that I took the greatest comfort from was that General Atlantic, which I hugely admire as an investor, had just invested in the business.”

Finance hasn’t yet invented a surefire way to build trust in a founder or an investor, but outsourcing the process to others is not it.

Trust in the Firm

Around any founder there’s a set of structures built into the fabric of their firm to provide safeguards to underpin trust.

Or at least there should be.

I’ve said before that the best insights into how companies work often come after a failure is exposed and an official inquiry is launched. The post-mortem into the Archegos affair sheds more light on internal processes at Credit Suisse than any outsider would have been able to assess ex-ante; the same for the London Whale scandal at JPMorgan and the account fraud scandal at Wells Fargo. 

With proper systems in place, it shouldn’t be easy to lose billions of dollars! Before FTX, MF Global was the last major brokerage firm to have dipped into client funds but even then, it wasn’t an easy thing to do. MF Global maintained $7 billion of client assets in a segregated fund. As the firm reeled towards bankruptcy in October 2011, employees cast around for ways to raise liquidity. The first dip into the client segregated account looks like it may have been an error and was reversed the following day. Then, CEO Jon Corzine put pressure on an assistant treasurer in the firm’s Chicago office to find funds to make a payment, telling her to “resolve the issues”. A mad scramble for cash in the final few days of the firm’s existence led to $1.6 billion of funds being drained from customer accounts. One explanation is that the chaos that descended on the firm in that last week meant that controls normally in place lapsed. In the event, all the money was later recovered by trustees overseeing the liquidation.

That does not appear to have been the case at FTX, where controls were lacking even in a normal operating environment. This week, John J Ray III, the restructuring expert installed to clean up FTX filed his first day pleadings after less than a week looking at the company. “I have over 40 years of legal and restructuring experience,” he writes. Yet “never in my career have I seen such a complete failure of corporate controls and such a complete absence of trustworthy financial information as occurred here. From compromised systems integrity and faulty regulatory oversight abroad, to the concentration of control in the hands of a very small group of inexperienced, unsophisticated and potentially compromised individuals, this situation is unprecedented.”

Ray describes poor record keeping, little documentation of client account balances, extensive company loans to insiders such as Bankman-Fried, company purchases of assets for employees, no independent directors on the board, no board meetings, no permanent records of management decision, no centralised control of cash, weak separation between the various business silos, poor disbursement controls, not even a comprehensive list of staff. The expense disbursement process was particularly unsophisticated: Employees of the FTX Group submitted payment requests through an online ‘chat’ platform where a disparate group of supervisors approved disbursements by responding with personalised emojis. 

Nor was the complete disregard of client assets just a regulatory matter. FTX’s terms of service tell customers that “title to your digital assets shall at all times remain with you and shall not transfer to FTX Trading.” If they’ve disappeared, that looks like theft.

Such weak controls partly stem from FTX’s rapid growth. The company was only founded in 2019 and by the end of 2021 had $15 billion of assets on its platform. In finance, growth is not your friend. It is also notable that FTX did not have to file financial reports. In the US, even non-public financial institutions need to publish financial statements on a regular basis via regulatory bodies; in many jurisdictions outside the US, all companies need to file statements. Clearly, this is no safeguard against fraud – John Ray cautions that he does not have confidence in any of the balance sheets he’s recovered from FTX – but in a business of trust, it gives stakeholders something to monitor.

It’s fitting that John Ray has Enron’s restructuring on his CV. As I wrote last week, Enron was targeted by one short-seller as being “the only financial institution that cannot produce a balance sheet.”

Trust in the System

Ironically, crypto was designed precisely as a solution to the problem of trust. The Bitcoin white paper proposes a “system for electronic transactions without relying on trust.” But a smooth consumer experience requires intermediaries and if trust in those intermediaries disintegrates, it damages the overall system. 

Indeed, the same week that FTX collapsed, Australian stock exchange ASX dropped plans to upgrade its clearing and settlement system to a blockchain-based platform. The project was launched seven years ago with the exchange “determined to deliver the Australian market a post-trade solution that balanced innovation and state-of-the-art technology with safety and reliability” according to its chairman. Yet after a cumulative investment of A$250 million ($168 million) the exchange changed its mind. “After further review, including consideration of the findings in the independent report, we have concluded that the path we were on will not meet ASX’s and the market’s high standards.” It seems that the centralised approach to clearing and settlement still has legs.

There’s a cyclical element to trust at the system-wide level. Traditional finance had its crisis of confidence during in 2008. But like in any evolutionary system, agents adapt, institutions change. Amidst the collapse in crypto, trust in the traditional financial system has only got stronger.