Hook:
Alpaca Securities, a self-clearing broker-dealer you've likely never heard of, now sits at the center of $1.5 billion in tokenized equity. According to data from RWA.xyz, as of July 2024, Alpaca clears or custodies approximately 94% of all tokenized U.S. stocks and ETFs. That is not a market; that is a hub-and-spoke system wearing a crypto costume. The irony is so sharp it could cut through EVM bytecode. The industry sold tokenized stocks as the end of intermediaries. We got a new one, more fragile than the old.
Context:
Tokenized stocks are a peculiar beast. They work like this: an issuer (Ondo, Dinari, or an exchange like Kraken) mints tokens on Ethereum, Solana, or other chains. Each token is nominally backed by one share of the underlying stock. But that share never touches a blockchain. It sits in a brokerage account held by a licensed broker-dealer. That broker-dealer is responsible for custody, trade execution, and handling corporate actions like dividends and stock splits. The token is merely a ledger entry on a public chain, a receipt for a promise that lives in a traditional clearing system.
The pitch was always about disintermediation. No more waiting for T+2 settlement, no more limited trading hours, no more gatekeepers. The reality, as any forensic systems analyst would tell you, is that the gatekeeper simply moved from the front door to the back office. The question is: who holds the keys? Alpaca does.
Core:
Let us trace the logic gates back to the genesis block. Alpaca is not just any broker; it is a self-clearing broker-dealer, meaning it holds FINRA license to settle trades itself without needing a third-party clearing house. That makes it the only player willing to plug into the tokenization ecosystem. As an Alpaca spokesperson told CryptoSlate: "Alpaca holds the underlying stocks on a one-to-one basis, executes and clears trades, and runs real-time minting and redemption through our instant tokenization network." That single sentence contains the entire attack surface.
Here is what the 94% concentration means in practice:
- Single point of custody risk. If Alpaca goes down—through regulatory action, operational failure, or insolvency—the entire tokenized stock market freezes. There is no fallback. No alternative broker has stepped up because "few established brokers are willing to offer services for this business," as the article notes. The market has already priced in a monopoly, and the premium is negative.
- Legal ownership vacuum. In January 2024, the SEC drew a line: tokens sponsored by the issuing company can carry legal rights to the stock, but third-party tokens (which is what Alpaca supports) give holders only economic exposure plus new intermediary risks. No voting rights, no direct dividend rights. The claim to the underlying asset flows first to the token issuer, then to the user. If Alpaca fails, the token holder has no direct line to the stock. They are creditors to a promise, not owners of an asset.
- Corporate action opacity. Alpaca handles dividends, splits, and other corporate events for every partner. That is a single point of truth. When a dividend is declared, Alpaca moves cash. When a stock splits, Alpaca adjusts the token supply. There is no on-chain verification; it is a centralized database feeding a decentralized ledger. The blockchain becomes an append-only log of Alpaca's internal state.
- The SpaceX IPO case study. In June 2024, a tokenized pre-IPO product for SpaceX was offered on Binance xStocks. The event was cancelled; users got refunds. The tokens never existed. This exposed the model's fragility: the issuer can simply cancel the order, and the user has no recourse. It was a dress rehearsal for the real black swan.
Based on my audit experience with multi-signature contracts and bridging protocols, I have seen this pattern before. The security assumption is not "code is law" but "Alpaca is solvent and honest." That is a trust model, not a trustless one. Read the assembly, not just the documentation. The underlying EVM opcodes reveal nothing about the real asset; they only encode a mapping from token ID to a database row controlled by Alpaca.
Contrarian:
The obvious takeaway is that Alpaca is a single point of failure. But the deeper counter-intuitive insight is that this failure is not an accident—it is an inevitable consequence of the legal framework for real-world assets. Tokenizing a stock under U.S. securities law requires a registered broker-dealer, custody, clearing, and compliance. That cannot be done on a permissionless blockchain without a regulated bridge. The industry tried to have both: the liquidity of crypto and the legal finality of TradFi. The result is a hybrid that inherits the weaknesses of both systems.
Here is the contrarian angle: the 94% concentration is actually the market's rational response to regulatory uncertainty. Alpaca is the only self-clearing broker willing to take this risk. Other brokers stay out because the legal exposure is too high—if the SEC decides these tokens are unregistered securities, the broker becomes liable. Alpaca's monopoly is a signal of how broken the underlying legal structure is, not a failure of competition. The market centralized because the regulatory environment made it too dangerous to compete.
Furthermore, the narrative that "tokenized stocks are a stepping stone to on-chain finance" is backwards. They are a trap. They create an illusion of composability (you can put these tokens into DeFi pools) while hiding the legal time bomb. If the SEC acts, these tokens become un-tradeable. The DeFi composability becomes a vector for contagion, not a feature.
Takeaway:
The next 90 days will be pivotal. The DTCC—the backbone of U.S. equity settlement—plans to launch its own tokenization service in October 2024. If DTCC offers a legally sound, regulated token with actual ownership rights, the Alpaca model becomes obsolete overnight. But if DTCC's solution is similarly centralized or delayed, Alpaca's monopoly will harden, and the risk will concentrate further. Either way, the current crop of tokenized stocks are not investments. They are unsecured promises issued by a single broker. As with any single point of failure, the only question is when it breaks, not if.
So the next time you see a pitch for "24/7 trading, zero slippage, no intermediaries," ask yourself: who clears the trade? If the answer is a single name, run an audit on that name before you run your transaction. The code may be open, but the vulnerability is closed.