$C | Citi Never Sleeps - Net Interest


Welcome to another issue of Net Interest, my newsletter on financial sector themes. Every Friday I go deep on a topic of interest in the sector and highlight a few other trending themes underneath. If you enjoy Net Interest, please spread the word. It really makes a difference when you recommend this newsletter to others.

Citi Never Sleeps

Every investor has a stock they can’t help but scratch. Maybe it made them a lot of money and they look on it like a golden goose. Maybe they invested so much time getting to understand it, they can’t let it go. Whatever the reason, the stock seduces its investor with whispers of promise.

In my career as a professional investor, that stock was Citigroup.

On Monday next week, Citigroup welcomes a new CEO to its helm. So, like a moth to a flame, it’s time for another look.

Citigroup: A Long Term Value Stock

Jane Fraser takes over Citigroup with its stock pretty much where it’s always been. In the past ten years it has traded in a range between around $30 and $75, usually at a discount to its book value. In the grand sweep of history it hasn’t moved very much either. Near the top of the market almost a hundred years ago, in 1929, it traded hands for $31 a share; a 2.2x return over a century isn’t going to leave any investor with FOMO.1

That doesn’t mean it’s impossible to make money investing in Citigroup. These long-term perspectives mask a lot of underlying volatility.

As Howard Marks says:

Rule number one: most things will prove to be cyclical. Rule number two: some of the greatest opportunities for gain and loss come when other people forget rule number one.

Few stocks reflect such cyclicality as Citigroup.

One of the first investors to exploit it was a merchant called Charles Lawton. He bought a majority of the bank’s shares in 1825 at a discount to par, before selling them on for a profit. Citi had been founded in the previous decade as the City Bank of New York, just before the outbreak of the War of 1812. It spent its first few years helping to fund the war effort but, like other emerging market banks, soon lapsed into lending to insiders and rolling over poor credits. As its stock fell, Lawton spotted an opportunity; he bought after a crisis.

The thing about Citigroup is that it’s quite prone to a crisis. Lawton may have been the first, but he wouldn’t be the last investor to ride the cyclicality of Citigroup.

After Lawton, the next investor to pull it off was none other than John Jacob Astor. In 1837, America suffered a sharp financial panic which led to severe recession. Many banks collapsed. Out of 850 banks in the country, 343 closed entirely. City suffered a sizable run-off of deposits and may not have survived had it not been for the support of one of its biggest customers – John Jacob Astor. Astor bought stock close to the lows at 55% of par and installed his associate, Moses Taylor, to the board.

Together, the two men profited as City Bank not only recovered but grew alongside New York’s emergence as a major financial centre. The stock rallied back to 120% of par, where it stayed for many years.

Moses Taylor had survived the crisis of 1837 by retaining plenty of cash on hand, and he brought that philosophy with him to City Bank. In 1856 he assumed the role of president of the bank. He retained a firm handle on the bank’s balance sheet. Although he let the equity capital ratio drift down from its fortress-like 50% a few years after he joined the board, he never let it drop below 16% (not exactly like-for-like but today’s ratio is in the region of 12%).

Taylor had skin in the game, too, accumulating stock in the bank to the point where he had a controlling interest. When he died in 1882, he left a fortune of $40 million, equivalent to $1 billion in today’s money.

Taylor left the bank in such good shape that it was years before City would be subject to another crisis. While many other banks struggled during the rolling financial panics that occurred over the second half of the nineteenth century, City stood tall. Each panic allowed it to pick up deposits from its competitors. A year after the panic of 1893, City became the largest bank in America. By now it had a national charter and was called National City Bank.

Soon enough, though, it found itself once again tottering on the edge of failure, shortly after the end of the First World War. Faced with regulatory restrictions on branch openings at home, City had begun expanding abroad. In 1914, it opened the first foreign branch of any nationally chartered US bank, in Buenos Aires, Argentina. It went on to open branches in Brazil, Chile, Uruguay, Cuba, Russia and elsewhere. By 1920, City had 81 branches in outposts around the world.

Unfortunately, these new outposts exposed City to all kinds of problems it hadn’t encountered at home. The Bolsheviks nationalised City’s St. Petersburg branch within a year of its opening. And in Cuba, the bank built up a huge exposure to the sugar industry just in time to see the price of sugar collapse as European production came back on stream. (By early 1921, sugar was trading below 2 cents per pound; borrowers had invested on the premise of 20 cent sugar and City had $79 million of exposure – almost as much as its capital base – including $63 million of non-performing loans).

There is no record of any investor stepping up to the plate this time, except for the US Government. The Federal Reserve system had been created a few years earlier via an Act of Congress which City and other bank executives had a hand in crafting. The Act allowed regional Reserve Banks to lend to member banks in times of difficulty. City drew down $28 million of borrowings from the New York Fed in 1919, increasing that to $95 million in 1920 and then to $144 million in 1921.

Aided by a modest rebound in sugar prices, City was able to pay it all back.

But then the Big One came.

To clean up the Cuba problem, City brought in a new president, a bond salesman called Charles Mitchell. In his first few years he pursued a cautious path, writing off bad loans and slimming down the bank’s balance sheet. Pretty soon, however, he started taking more risk. He doubled down on Cuba, moving into the sugar business directly via the General Sugar Corporation. He continued the bank’s expansion abroad, growing the number of foreign branches to 97. And, more than anything, he wanted to bring Wall Street to Main Street. Back in 1911, the bank had established a securities business called National City Company, and Mitchell invested heavily in it, opening brokerage offices across the US. By the late 1920s its profits matched those of the bank. And by 1929, City was the largest bank in the world.

At its peak, just before the Crash, City stock was trading at 9x book value. Over the next few years, it would fall precipitously.

Howard Marks says about value investing:

Value investors score their biggest gains when they buy an underpriced asset, average down unfailingly and have their analysis proved out. Thus, there are two essential ingredients for profit in a declining market: you have to have a view on intrinsic value, and you have to hold that view strongly enough to be able to hang in and buy even as price declines suggest that you’re wrong. Oh yes, there’s a third: you have to be right.

Charles Mitchell averaged down unfailingly. (He described his stock purchases as “throwing my fortune into the breach” for “the benefit of this institution.”) Unfortunately his analysis lacked Marks’ third essential ingredient – he wasn’t right. Over three years beginning in September 1929, National City Company’s stockholders’ equity declined by more than 80%; in the bank, losses emerged in Germany and Chile; and in Cuba, sugar exposures came back to bite.

Mitchell’s share dealings drove him deep into debt and led to charges of tax evasion. He was acquitted, but forced to resign from City Bank in 1933. He was also held personally responsible for the Crash by lawmakers. Senator Carter Glass, the man behind the famous Glass-Steagall Act said:

The recent trouble is due largely to Charles E. Mitchell’s activities. That man more than 40 others is more responsible for the present situation. Had the Federal reserve acted and dismissed him, the trouble might be less. The crash has shown that stock gambling has reached its limit.

Shortly after his resignation, the Government injected $50 million of preferred stock into City Bank to save it. City used the capital to clean itself up. After write-downs, it showed an equity to assets ratio of 8.5%. And not one year through the rest of the 1930s did charge-offs exceed recoveries, even as the economy continued to slide.

As it did after the financial panic of 1837, the brush with failure cast a long shadow. It was many years before City – renamed Citibank in 1976 – tottered back into a crisis. In 1987, it took $3 billion of provisions against Latin American debt, wiping out the preceding four years of earnings. And then, in 1990, it was floored by exposure to commercial property developers at home.

Once again, a value investor stepped up.

At the end of 1990, Prince Alwaleed bin Talal bin Abdulaziz al Saud, a member of the Saudi royal family, quietly bought 4.9% of Citi’s common shares on the open market for $200 million. In February the next year he was approached by Citi to support a capital raise and agreed to buy $590 million of preferred stock, which, if converted to common, would give him a stake of 15%. Alwaleed’s timing was superb. He bought his first tranche at an average price of $12.5 a share and the second at $16. Notwithstanding a few wobbles along the way, Citi’s stock price would head up towards $500 by the middle of 2000.

That kind of return left others salivating. So when Citi skidded into trouble yet again, in 2007, the sovereign wealth fund of Abu Dhabi was quick to jump in. Weeks after announcing billions of dollars of subprime mortgage related losses, the fund put $7.5 billion into the bank in a similar preferred structure.

Howard Marks says, “Being too far ahead of your time is indistinguishable from being wrong.”

Abu Dhabi was wrong. The subprime related losses Citi suffered in the third quarter of 2007 were just the beginning. Over the next twelve months waves of losses would keep crashing in. An internal Citi analysis projected that “the firm will be insolvent by Wednesday, November 26 [2008].” Like in 1934, it fell to the government to steady the ship, this time with over $500 billion of support.

Government support paved the way for private investors to swoop. David Tepper of Appaloosa Management made around $7 billion in his fund in 2009 buying up bank stocks including Citi preferreds when the stock was below $10. He exited his position in the last quarter of 2014, with the stock around $50.2

Howard Marks says, “Large amounts of money aren’t made by buying what everybody likes. They’re made by buying what everybody underestimates.”

Jane Fraser’s To-Do List

Jane Fraser takes over Citi towards the end of a period scarred by the pandemic. The bank took $17.5 billion of provisions against loan losses last year, but that turned out not to be as bad as it might have been. Its equity capital ratio, at 11.8%, puts it on sounder footing than in either of its 1990 or 2007 crises. Yet the stock trades at a 10% discount to book value, and – but for a brief window a few years ago – hasn’t traded above book value since Tepper bought his first shares.

Jane Fraser has given herself three targets: position Citi to win, improve returns significantly and address issues raised by regulators.

Let’s take the regulatory issues first.

Citi’s internal operating systems came under the spotlight this month when they lost a court case – and a lot of money – arising from a mistaken wire transfer. The details of the case are quite amusing. One recipient messaged another: “DFREY5: How was work today honey? It was ok, except I accidentally sent $900mm out to people who weren’t supposed to have it.” But the cause was more serious – clunky and outdated infrastructure.

Citi had already been warned about its systems as long ago as 2009, when one of its regulators, the FDIC, wrote in a memo: “The deficiencies in Board and management oversight are magnified by the institution’s inadequate information technology infrastructure.” With those inadequacies clearly not having been fixed, the bank was slapped with a consent order in October last year together with a $400 million fine.

It’s a track record that goes back a long way. In 1900, an examiner complained about the bookkeeping in City’s foreign exchange department. In 1909, an examiner criticized a repurchase agreement. The whole sugar saga caused anguish among regulators.

Since 2000, Citi has paid $25 billion in penalties against various infractions. Jane Fraser is unlikely to be held in as high regulatory esteem as Moses Taylor, about whom regulators wrote, “[His] very name is the embodiment of integrity, energy, mercantile sagacity and patriotism and strongly reminds us of the Merchant Princes of Venice in the 12th Century who made that Commercial Emporium the seat of wealth, art, genius, luxury and power that has commanded the admiration of the civilized world.”

But there’s clearly upside.

Next: position Citi to win.

This one’s a bit vague, but it touches on Citi’s strategic positioning. On the one hand, “We are the world’s most global bank with a network no one can match.” On the other, “I believe there is value to unlock by simplifying the firm.”

Citi’s global ambitions were sketched out as long ago as 1903, in a speech its vice president (and former assistant Treasury secretary) made at the New England Club in Boston:

Let us suppose that we have an organization of great financial strength…energetically and intelligently representing broadly our exporting interests in the world’s markets… Suppose such an organization should stand behind purchaser and producer, guaranteeing…the delivery of goods…[as well as] the credit of the purchaser. If such an organization were equipped with men of trained intelligence, keen observers of commercial conditions, who would be quick to see an opportunity and devise means of grasping it, and if that organization had behind it the cooperation of great manufacturing interests here, I believe wonders could be accomplished.

It was an ambitious plan. The first foreign branch openings in 1914 laid the seeds, but it wasn’t until after the Second World War that it gained traction. Under the Marshall Plan, the bank arranged commercial letters of credit for shipments to countries receiving US government aid. More broadly, City was able to take advantage of America’s new position as world superpower to expand its international commercial banking operations.

In 1967, the bank recruited its new group president from its Overseas Division. Walter Wriston was a firm believer in the global strategy:

The plan in the Overseas Division was first to put a Citibank branch in every commercially important country in the world. The second phase was to begin to tap the local deposit market by putting satellite branches or mini-branches in a country. The third phase was to export retail services and know-how from New York.

As president, it was on his watch that foreign deposits exceeded domestic deposits at the bank, a situation that persists to this day (Citi has $650 billion of international deposits and $630 billion of domestic deposits).

The challenge with maintaining a global network is the complexity it injects into the group structure. Citi already exited a number of consumer markets under former CEO Mike Corbat and Jane Fraser could follow suit.

Jane Fraser’s final target is to improve returns significantly.

The bank earns a return of around 7% on its capital, which is four to six percentage points lower than peers like JP Morgan and Bank of America. Part of the drag comes from the group’s global sprawl, which adds cost. But it also comes from an inferior deposit franchise in the US, where the bank has a smaller base of cheap retail deposits than peers. This is a difficult issue to correct. One option might have been to acquire a bank with a healthy deposits franchise but Citi is precluded from doing that under the terms of its consent order. The other option is to mount a digital campaign to win deposits, but that’s a slow process.

Whatever Jane Fraser does, it’s unlikely to create an opportunity on the scale that Charles Lawton, Prince Alwaleed or David Tepper enjoyed. Citi isn’t in distress, it’s just not earning very much money. At 0.90x book, the play is that book value grows rather than that it gets rerated materially.

But then again, as Howard Marks says, “Buying something for less than its value. In my opinion, this is what it’s all about—the most dependable way to make money.” If there comes a time when people tire of paying over 10 times sales for companies that don’t make any money, then Jane Fraser’s Citigroup may provide an alternative.

For more on the history of Citigroup, I would recommend Borrowed Time by James Freeman and Vern McKinley. Genealogy of American Finance by Robert Wright and Richard Sylla provides potted histories of the 50 largest US financial companies as at 2015. The definitive pre-1970 Citigroup history is Citibank, 1812-1970.

More Net Interest


Coinbase released its S-1 prospectus this week, days after shares traded hands in the secondary market valuing the company at $100 billion (I wonder if those investors had seen the disclosures?)

The S-1 contains an unusually long glossary (“Hodl: A term used in the crypto community for holding a crypto asset through ups and downs, rather than selling it.”)

It also contains an unusually long list of risk factors. Highlights include:

  • A caution that “a significant amount of the Trading Volume on our platform is derived from a relatively small number of customers.”
  • An observation that “many of our employees and service providers are accustomed to working at tech companies which generally do not maintain the same compliance customs and rules as financial services firms.”
  • A warning that “the Enforcement Division of the CFTC commenced an investigation…”
  • A resignation that “we have become increasingly obligated to comply with the laws, rules, regulations, policies, and legal interpretations both of the jurisdictions in which we operate and those into which we offer services on a cross-border basis.”

Ultimately, though, it’s a fairly simple business model. In 2020 Coinbase generated $1.3 billion in revenue, of which $1.1 billion came from transaction activity. Coinbase takes 0.57% on the value of every transaction it processes. Given that it sees 11% of the total market value of crypto assets pass through its platform, that’s a pretty valuable toll. But it is a sensitive toll - to the value of crypto assets - and on top of its operational and regulatory challenges, Coinbase’s financial challenge will be managing its revenue volatility.


Klarna filed its 2000 financial results recently. Our discussion of the company in Buy Now Pay Later a few months ago included a detailed financial overview. The company derives revenue from three sources, but the mix is changing.

Historically, net interest income on loans contributed a greater share of revenue, but that is slipping. In the fourth quarter of 2020, net interest income contributed only 23% to total operating income. That’s partly because the yield on loans has come down – it’s now below 10% on average – even while the funding mix has improved, with more of the loan book funded by deposits.

In contrast, commission income is growing as the key driver of revenues, with the weight within total commission income shifting towards the merchant and away from the consumer. In the fourth quarter of 2020, commissions contributed 73% to total operating income, with merchant fees making up nearly 80% of that (versus 65% in 2018). The shift reflects a strategy to move away from late fees which have come under a lot of regulatory scrutiny in some markets and brings the business model closer to Affirm’s.

Meanwhile, the company continues to invest heavily. Having consistently generated a quarterly operating profit up until mid 2019, it has lost a cumulative SEK 2.6 billion since. All part of the land grab as it readies itself for IPO.


We covered insurance company Root in The End of Insurance a week or so before they IPO’d in October. Yesterday they reported earnings and the picture was not pretty. The key in insurance is to keep loss rates low. Root’s loss rate was over 80% in the quarter, a lot higher than the low 70’s target flagged during the IPO. At the time, the company argued that loyal customers would generate a lower loss rate. But that’s not happening – 70% of the quarter’s premium was from renewing customers, yet the loss rate remains stubborn.

It’s inevitable that among all the fintechs coming to the market right now, some will struggle - as the marketplace lenders that came to the market in the first wave have done. Root is guiding for $500 million of losses in 2021, versus cash on hand of $1 billion, so it will need to return to the market at some stage. The question is how receptive the market will be when it comes back.

I’ve adjusted for stock splits, but not for dividends. At the peak in 1929, National City Bank traded at $455 a share, equivalent to $31 in today’s terms, adjusted for splits.

The reason I find Citi so seductive is that I, too, profited from its performance in 2009. In January we were short, covering when the company announced a massively dilutive conversion of preferred stock to common. Shortly after, we went long and held for several years.