The NYSE Closed Every Wednesday in 1968. BitGo's CEO Says Tokenization Finishes What That Fix Started

In 1968 the New York Stock Exchange closed every Wednesday. Not for a holiday. Not for maintenance. It shut down because brokers were drowning in physical stock certificates and needed a day to dig out.

Volume had blown past 15 million shares a day that year, and back offices were still hand-carrying paper between firms. The crisis got bad enough that the exchange ran a shortened week for months. By the time it ended, the industry had committed to a fix. Centralize the paperwork. That fix became the Depository Trust Company in 1973, and it's still the spine of US equity settlement today.

BitGo CEO Mike Belshe keeps returning to that history when he talks about tokenization, and his framing is the most useful thing I've heard on the subject in a while. Tokenization isn't really a trading story, he says. It's an access story.

That's a quieter claim than the usual pitch, and a much better one.

The Fix That Left Most People Outside

The 1973 fix worked. Settlement times went from measured in days of paper shuffling to something automated and reliable. In September 2017 the US moved to T+2. In May 2024 it moved again, to T+1. Faster every decade.

But here's the thing about the system that replaced the paper. It didn't remove the gate. It just changed who holds the keys.

Direct access to the settlement layer sits with a small club of banks, custodians, and broker-dealers. Everyone else rides on top of them, paying a fee at each layer. Your brokerage account isn't a vault. It's an IOU from a member of the club. In simple terms, you own a beneficial interest in a database entry, and that entry is administered by someone with better access to the plumbing than you'll ever have.

Belshe calls part of this ghost stocks, which is blunt but fair. Most of what people think they own in the equity market is a claim on a claim. Fine when everything works. Revealing when it doesn't.

Now compare what a wealthy account holder can do with a normal one. Someone with a $10 million portfolio calls their private bank and gets a securities-backed line of credit at roughly 5% to 7%, often priced off SOFR plus a spread. They borrow against the portfolio, keep the positions, defer the capital gains, and stay exposed to the upside.

Someone with $10,000 in the same stocks gets margin at 8% to 12% if they qualify at all. And if those assets sit in a self-custody wallet, they get nothing. No bank will lend against them. So when cash is needed, the only move is to sell.

Think of it this way: two people hold the same asset, and only one of them is allowed to use it.

That's the K-shaped economy in one sentence, and it has almost nothing to do with who works harder. It's about collateral eligibility.

Why the Plumbing Matters More Than the Speed

This is where tokenized equities and tokenized treasuries stop being a novelty and start being a structural argument. If an asset lives on a public execution layer with clear ownership and programmatic rules, it can be pledged without a custodian's permission. The collateral question changes. The access question changes with it.

Is that guaranteed to happen? No. Regulators have real concerns about consumer protection, and they should. Tokenized equities raise hard questions about shareholder rights, voting, and what happens in a bankruptcy. But the current arrangement raises hard questions too, and we mostly stopped asking them because the paperwork is invisible.

The change comes at a time when tokenized treasury products have already crossed into the billions in assets under management, and BlackRock's entry into the space pulled in serious institutional money in a matter of months. Treasuries came first because they're simple. Equities are next in line, and equities are where the lending markets actually live.

Who wins if this works? Issuers, who get a direct line to holders. Retail, who get collateral rights they've never had. Startups building lending against on-chain positions.

Who loses? Custodians and transfer agents collecting toll revenue on every movement. Brokerages whose profit centers depend on margin spreads. And anyone whose business model assumes ordinary investors have no alternative but to sell when they need cash.

Belshe also floated a governance angle worth watching. Executives locking their own shares on-chain for a set period, verifiable by anyone, instead of filing a plan under Rule 10b5-1 that the public can't audit in real time. Proof of reserves applied to insider conviction. I'd like to see a single Fortune 500 board agree to that. But the mere existence of a verifiable option makes the opaque version harder to defend.

He touched on AI agents too, which sounds like a tangent until you follow it. An agent that manages assets needs to hold value and move it without a human clicking approve on every transaction. That requires programmatic permissions and finality you can verify, not a login screen and a phone call to a custodian. The same rails that fix collateral access happen to be the rails agents need.

On the dollar, Belshe walked back the literal reading of his own past comments. He's not calling for zero. He's describing debasement, a slow erosion of purchasing power that quietly pushes anyone holding cash away from saving and toward borrowing. Which, if you follow the thread, is the same mechanic widening the K.

What to Take From This

For everyday users, nothing changes overnight. Tokenized equities still face a regulatory maze, and the first versions will probably look like wrapped products from familiar names rather than something you self-custody on a Tuesday.

But the fight worth watching isn't about faster trades or lower fees. It's about whether your assets count as collateral, or whether they're just something you're allowed to look at until you need cash.

The 1968 crisis got a fix. It just never got finished. And the unfinished part has been quietly sorting people into two groups for fifty years.