The Silvergate Autopsy: Alan Lane Blames Washington, But the Ledger Blinked First

Alan Lane finally broke his silence. The former Silvergate CEO's defense, delivered long after the bank's March 2023 voluntary liquidation, runs like this: Silvergate was solvent, well-capitalized, and properly managed. It wasn't bad loans or a broken balance sheet that killed it. It was the Biden administration, federal regulators, and a coordinated campaign to sever crypto from the banking system. The data doesn't cooperate with that narrative's neatness. In the fourth quarter of 2022, Silvergate lost 70% of its demand deposits — not over a year, not through normal attrition, but within a few weeks. Deposits tied to crypto companies made up roughly 90% of the liability side. When FTX collapsed, that concentrated, hyper-connected depositor base moved as one organism. The charts blinked, and the liquidity didn't blink back. The exit liquidity was already gone before most of the market understood the bank was even in trouble. For a decade, Silvergate occupied a strange, essential niche: the regulated bank that crypto institutions could actually use. Its real product wasn't deposit accounts. It was the Silvergate Exchange Network — SEN — a 24/7 dollar settlement rail that let exchanges, market makers, and institutional desks move fiat between each other in real time. SWIFT sleeps on weekends. Crypto trades at 3 AM on a Sunday. SEN filled that gap and, in doing so, became critical infrastructure for the 2017-2022 boom. The client list read like the industry's top table: Coinbase, Kraken, Gemini, Galaxy Digital. The network effect was real. Switching banks meant re-doing KYC, rebuilding credit lines, surviving weeks of compliance review — a switching cost that made SEN stickier than any loyalty program. This was the period I call the crypto-banking honeymoon, and it felt permanent. It wasn't. Then November 2022 happened. FTX's bankruptcy didn't just take down a trading platform. It detonated a confidence bomb inside a client base that all read the same news, all suspected each other's exposures, and all reached for the exit simultaneously. Silvergate wasn't a major FTX creditor. It didn't carry fatal direct credit exposure to Alameda. The run wasn't about what Silvergate owed. It was about what Silvergate's clients feared — and fear, in a networked industry, compounds faster than any balance sheet can shrink. Three days after Silvergate announced its wind-down, Signature Bank — the other crypto-friendly bank, with crypto deposits around 30% — was seized by New York regulators. The crypto dollar banking corridor didn't narrow. It vanished. Two banks. One week. A structural era closed. The technical reality of SEN is often misunderstood. It was not blockchain. It was a centralized real-time payment network layered on top of traditional banking rails. Its innovation was operational, not cryptographic. No smart contracts. No code-audit risk. Smart contracts don't panic — but bank depositors absolutely do, and they don't need a governance vote to do it. SEN's genuine value was real-time finality: 24/7 dollar movement between Silvergate account holders, settled instantly inside the bank's ledger. For institutions that needed same-day settlement across exchanges and custodians, that was transformative. The same feature that made SEN valuable became a liability mechanism in a crisis. When a client base that large and that correlated decides to withdraw at once, the network doesn't protect you. It accelerates the drain. Lane's technical defense rests on one claim: Silvergate managed its balance sheet for crypto's volatility, holding high-quality liquid assets to cover deposit swings. At one level, that's true. The bank maintained solvency through the worst of the run and eventually repaid depositors in full through orderly liquidation. That's not nothing. In the history of bank failures, full depositor repayment through a voluntary wind-down is closer to a dignified exit than a disorderly collapse. Based on my audit experience across DeFi protocols and lending desks, I can tell you that most balance sheets never get the chance to close that cleanly. But the defense has structural holes. First, the January 2023 10-K delay. Silvergate told the market it might not maintain 'well-capitalized' status — that's the language of a bank whose capital position is deteriorating, not one comfortably defending its ratios. Second, the 'highly liquid assets' weren't liquid at cost. Rising interest rates had pushed the bond portfolio into deep unrealized losses — the same disease that killed Silicon Valley Bank days later. 'Forced to sell highly liquid assets at a loss' is not a phrase from a healthy balance sheet. It's the sound of duration risk meeting a maturity mismatch. The bank chose asset sales over the Federal Reserve's discount window — an unusual move for a solvency-confident institution, and one Lane never adequately explains. Then there's the AML question. Lane's sharpest line: no regulator ever proved Silvergate's anti-money-laundering controls ineffective. That's a lawyer's masterpiece. 'Not proven ineffective' is not the same as 'effective.' The Federal Reserve had a cease-and-desist draft in circulation. The bank's own compliance reporting lagged on suspicious activity filings. When I was mapping Alameda's on-chain outflows during the FTX collapse — $1 billion tracked to offshore entities within hours — the lesson was consistent: the industry's real blind spots weren't on-chain. They were inside institutions whose compliance teams couldn't keep pace with the velocity of crypto flows. Speed eats strategy for breakfast, but only when the strategy is built to survive contact with reality. Here's the part that matters most for anyone still holding crypto assets or running a crypto business: the regulatory pressure Lane describes as the cause of death is better understood as the accelerant, not the fire. The fire was the business model itself. Ninety percent of deposits from a single, highly correlated industry. A client base whose informational density made traditional bank-run dynamics obsolete. An asset side exposed to rate shocks at the exact moment the industry convulsed. Volatility is just velocity without direction — but concentration is a one-way door. Lane's account also conveniently omits the market's own verdict. The people with the most information — the depositors — ran first. That's not politics. That's the market's final audit. Panic is a lagging indicator for the prepared, and the prepared had already left by the time the story became public. Regulators may have applied pressure, but they didn't need to manufacture the vulnerability. The model contained the mechanism of its own failure. What happened next tells you where the industry actually went. In the weeks after Silvergate and Signature closed, exchanges shifted settlement flows to USDC and USDT. Asset managers began using stablecoins for intraday moves that previously ran through SEN. Stablecoin market share rose precisely because the banking rails disappeared. Value didn't leave crypto. It left the banking system. That structural shift — not Lane's testimony, not the political blame game — is the real story. The crypto industry discovered it could survive the loss of its preferred banks, but only by leaning harder on the very stablecoin issuers that regulators were also scrutinizing. The fragility didn't disappear. It migrated. Watch stablecoin dominance and the emergence of new bank substitutes in 2025. The regulatory mood has shifted since the Trump administration took office, and crypto-friendly banks are cautiously re-emerging. But the lesson of Silvergate is not that Washington killed a good bank. It's that bank counterparty risk is an asset-liability problem no amount of regulatory friendship can fix. When Silvergate's successor emerges, demand proof of deposit diversification — not promises, not narratives, not post-hoc defenses from former CEOs. The charts will blink again before the next crisis. Know what your liquidity is actually doing this time.