In the spring of 2022, Babel Finance had the kind of week most startups only dream about. On the ninth floor of an office on Hong Kong Island, its founders closed an $80 million funding round that valued the four-year-old crypto lender at $2 billion. Sequoia's China arm was already on the cap table. Tiger Global had written a check. Circle, the company behind the USDC stablecoin, had joined too. Babel was not just riding the crypto boom - it wanted to be the institution the boom grew up into. The comparison its founders liked was JP Morgan.
Thirty-nine days after that round was announced, Babel Finance told its customers they could no longer take their money out.
The distance between those two moments - a headline valuation and a frozen withdrawal button - is the entire story of Babel Finance, and of a particular kind of ambition that ran through crypto in 2022. It is worth walking through slowly, because the parts that worked and the part that failed sat side by side the whole time.
01 / The BusinessA bank for people who own bitcoin
Babel Finance was founded in Hong Kong in 2018 by two men who came out of traditional finance rather than into it from software. Flex Yang had spent nearly four years at PwC before starting the company and serving as chief executive until October 2021. Del Wang, his co-founder and later the company's CEO, had worked at ICBC, one of the largest banks on earth, and had built a Ripple-based payment system before turning to crypto.
Their first customers were Chinese bitcoin miners, and their first product solved a genuinely mundane problem. A miner in 2019 had bitcoin in a wallet and bills due in yuan: electricity, hardware, payroll. Selling the coins to cover costs meant giving up the upside they were mining for in the first place. Babel offered a different deal - post your bitcoin as collateral, borrow stablecoins or cash against it, keep the coins. It was a pawnshop for the crypto age, and it worked.
From that base the company expanded into the full menu of an institutional crypto desk: structured asset management, prime brokerage, and derivatives strategies. It kept a deliberately conservative rule about which assets it would touch - bitcoin, ethereum, and stablecoins, and little else. No long tail of speculative tokens. By early 2021 Babel said it served more than 500 institutional clients and had roughly $2 billion in outstanding crypto loans, a figure that grew to about $3 billion in balances by the end of that year.
02 / The CustomersAsia's institutions, not retail
This was never a consumer app. Babel's clients were miners, exchanges, and investment funds - businesses that were rich in crypto and short on working capital, or that wanted yield on idle coins. That concentration was a strength when the market rose. It also meant that when confidence cracked, the people asking for their money back were sophisticated, well-connected, and capable of moving fast. There was no cushion of sleepy retail depositors. There were 500 professionals watching the same charts Babel was.
The pitch to those clients was that Babel behaved like a real financial institution. It held licenses in Luxembourg, Hong Kong, and the United Kingdom for relevant activities. It restricted itself to blue-chip assets. It hired from banks. All of that made the eventual failure more jarring, because it did not look like a fly-by-night operation. It looked, from the outside, careful.
03 / The Money$122 million and a $2 billion sticker
Babel raised in two big strokes. In May 2021 it closed a $40 million Series A led by an unusually blue-chip group for a crypto startup: Zoo Capital, Sequoia Capital China, Dragonfly Capital, BAI Capital, Bertelsmann's investment arm, and Tiger Global Management. For several of those funds it was a first bet on Asian crypto finance.
A year later, in May 2022, came the $80 million Series B at a $2 billion valuation, with Jeneration Capital, 10T Holdings, returning backers Dragonfly and BAI, and Circle Ventures. What makes that round remarkable in hindsight is its timing. Bitcoin had already lost more than half its value for the year. The crypto lender Celsius was days from its own freeze. Babel priced a two-billion-dollar valuation into the teeth of a bear market.
04 / The BreakOne volatile week
On June 17, 2022, Babel Finance suspended withdrawals and redemptions, citing "unusual liquidity pressures." The phrase was doing an enormous amount of work. In plain terms: the company could not give people their assets back.
What had actually happened came out weeks later. Babel had not only been lending customer assets - it had been trading them. In the week that bitcoin fell from around $30,000 to $20,000, unhedged directional positions in the firm's proprietary trading accounts blew up. Forced liquidations wiped out roughly 8,000 BTC and 56,000 ETH, a loss reported at about $280 million of customer funds. The trades were not hedged. When the market moved against them, there was nothing to cushion the fall.
This is the hinge of the whole story. Babel ran two businesses under one roof. One was a collateralized lending book - conservative, real, and by most accounts sound. The other was a proprietary trading desk running unhedged bets. On paper they were different desks. On the balance sheet they were the same money. When the trading desk lost, the lending customers were the ones who could not withdraw.
05 / The DifferenceWhat set Babel apart, for better and worse
Babel was often grouped with the other institutional lenders that failed in 2022 - Celsius, Voyager, and later Genesis. It shared their fatal structure: promising liquidity it could not guarantee once collateral values dropped. But Babel had a distinct flavor. Where Celsius chased retail deposits with double-digit yields, Babel stayed institutional and Asia-focused, courting miners and funds rather than the crowd. Where some rivals sprawled across dozens of tokens, Babel kept to bitcoin, ethereum, and stablecoins.
That discipline in what it held is exactly what makes the collapse instructive. Babel did not die because it bought the wrong coins. It died because of how it used the right ones. Picking safe assets is not the same thing as running a safe operation. You can lose everything holding only bitcoin if the position is wrong and the hedge is missing.
06 / The AftermathA long restructuring
Babel did not immediately file for bankruptcy. Instead it entered restructuring, sought extended creditor protection into 2023, and began trying to raise fresh debt and equity to recover client assets. It became one of the reference points in every write-up of the 2022-2023 crypto lender collapses, a name filed alongside the larger failures of that period. The recovery process has been slow and, for the clients whose coins were caught in the freeze, deeply unsatisfying.
- 2018Founded in Hong Kong
Ex-PwC and ex-ICBC founders start a crypto lender for miners who need cash without selling coins. - 2019Lending book scales
Stablecoin loans against BTC and ETH collateral spread to exchanges and funds across Asia. - 2021$40M Series A
Sequoia China, Tiger Global, Dragonfly and BAI back the firm. - 2022$2B valuation, then the freeze
An $80M Series B in May; withdrawals suspended in June on "unusual liquidity pressures." - 2022$280M in losses surface
Reports reveal unhedged proprietary trades cost ~8,000 BTC and ~56,000 ETH. - 2023Restructuring
Babel seeks creditor protection and capital to recover client assets.
07 / The TakeawayWhere it fits
Read the name again. Babel. A tower built high on borrowed material, brought down by its own reach. The founders almost certainly did not intend the omen, but critics reached for it quickly once the tower wobbled. What Babel Finance leaves behind is not a story about crypto being fake or its customers being naive. Its lending insight was real, its clients were serious, and its investors were among the best in the world.
The lesson is narrower and more useful than "crypto is dangerous." It is that a company holding other people's money cannot run a hidden, unhedged trading book against it and call the result "liquidity pressure." The wall between the safe business and the risky one has to be real, not just an org chart. Babel had the customer, the license, the assets, and the backing. It did not have the wall.