Wharton’s latest warning is simple: tokenization can improve market plumbing, but it does not magically create liquidity. A token can trade 24/7 and still be a terrible fit for a slow asset underneath.
- Fast tokens, slow assets
- Tradability is not liquidity
- Oracle risk is still very real
- Treasuries are one thing; private credit is another
Wharton’s Financial Policy and Regulation Initiative says the next phase of real-world asset tokenization will hinge on liquidity design, in plain English, whether the trading speed of a token matches the speed at which the underlying asset can be priced, sold, or redeemed. In a recent Wharton Warns RWA Tokenization May Trigger Liquidity Risks note, that warning was laid out even more bluntly: the tech can move faster than the asset, and that gap is where trouble starts.
The paper, Tokenizing Real-World Assets, was written by Lin William Cong of Nanyang Technological University, Simon Mayer of Carnegie Mellon University, and Daniel Rabetti of the National University of Singapore, who is also a visiting scholar at Harvard Business School. Its point is blunt: tokenization changes access, settlement, and rails. It does not change the economics of the asset itself. A related Tokenizing Real-World Assets: A WIFPR White Paper Webinar went deeper on those same mechanics.
That distinction matters because crypto loves to slap “24/7” on things and pretend time itself has been upgraded. It hasn’t. A fast-moving token sitting on top of a slow-moving asset can create the same old financial fragilities, only now they arrive with blockchain branding and a slicker dashboard.
The core idea in the paper is a speed-matching principle: the token should not trade faster than the asset can realistically support. If the token can be bought and sold instantly, but the underlying loan pool, mortgage book, or private fund can only be valued or unwound slowly, you get a mismatch. The token looks liquid. The asset is not. That gap is where run risk, stale pricing, and broken redemptions tend to show up.
That is not a uniquely crypto problem. It is classic finance trouble wearing new clothes.
According to the WIFPR analysis, public-blockchain RWA data excluding stablecoins stood at roughly $29 billion as of April 2026. Including Figure Technologies’ home-equity and mortgage-related products, that total rises to around $46 billion. The paper also says the market has grown from about $2.9 billion in 2022 to $12 billion by the end of 2024, with industry forecasts projecting as much as $18 trillion by 2033.
Those numbers point to momentum, but they should not be mistaken for destiny. Tokenization is real, but the jump from billions to trillions is still a forecast, not a law of physics. Crypto markets have a long history of turning “could” into “guaranteed” with all the restraint of a sales team on espresso.
What gets tokenized first, and why
The report separates RWAs into three broad buckets:
Highly liquid assets like Treasuries, gold, and large-cap equities.
Money-like claims such as stablecoins and tokenized deposits.
Illiquid private-market assets like private credit, loans, real estate, and private funds.
That order matters. Liquid assets are the easiest fit for blockchain rails because the underlying market already has depth and standard pricing. Tokenized Treasuries are the obvious example: they can improve settlement speed, collateral mobility, and access without pretending that a Treasury bill has suddenly become something exotic. The broader trend has been tracked in pieces like Tokenized Treasuries Hit $11B in 2026 and in broader market research such as 2026 Digital Asset Outlook: Dawn of the Institutional Era.
Wharton says Treasury and money market fund products have overtaken commodities as the largest non-stablecoin RWA segment on public chains. That makes sense. The safer and more standardized the underlying asset, the easier it is to justify tokenization.
Gold is another useful case. The report says tokenized gold tracked spot gold with a correlation near 0.99. That suggests the structure can work when the asset is well understood and the market is already liquid. But correlation is not a magic shield. A product can track fine in calm conditions and still struggle if redemption or custody mechanics go sideways under stress.
The real trouble starts with private credit, mortgages, real estate, and similar assets. These are the kinds of products that get marketed as “accessible” or “fractionalized, ” which sounds great until everyone wants out at once. Then the phrase “sell anytime” starts sounding less like a feature and more like a lie with a blockchain logo on it.
That is why the report pushes function-based regulation. A tokenized Treasury fund and a tokenized private loan pool may both live on-chain, but they do not belong in the same risk bucket. Same rails, very different plumbing.
Why liquidity can fail fast
One of the paper’s main warnings is about par redemption, meaning a token can be redeemed at a fixed value, usually 1:1. That structure works only as long as investors believe the underlying asset pool is sound and liquid enough to honor the promise.
Once that confidence cracks, the exit rush begins. And if the underlying assets are slow to value or slow to sell, the token can’t magically redeem itself into stability. Blockchain may speed up trading, but it also speeds up panic.
Wharton’s point is not that tokenization is bad. It is that the industry keeps confusing tradability with liquidity. Those are not the same thing. A token can change hands easily in tiny size and still be impossible to exit fairly at scale when stress hits. For a plain-English definition, see liquidity risk.
The report also flags concentration risk. It says the top 10 issuers account for about 82% of tracked RWA value, and one Figure Technologies mortgage-related token represents roughly 37% of the dataset cited. If those numbers hold, the market is still highly dependent on a few issuers and structures. Figure’s own Consumer Loan Marketplace and Financial Performance Metrics underscore just how much of this market still runs through a small number of players.
That concentration creates a clearinghouse-like problem: a handful of entities end up handling major operational and governance functions without the same level of oversight that traditional financial market infrastructure usually has. So yes, the rhetoric may be decentralized. The spreadsheet, less so.
Chain concentration shows a similar pattern. According to the report, Ethereum hosts about 54% of public-blockchain RWA value, followed by BNB Chain at 13%, Solana at 7%, Stellar at 6%, and Liquid Network at 5%. That is not surprising, but it is a useful reminder that “decentralized finance” often runs through a fairly small set of rails. As one recent breakdown on Ethereum Dominates RWA Tokenization as Hong Kong Pioneers pointed out, the ecosystem is still very much clustered around a few dominant chains and jurisdictions.
Oracle risk is not theoretical
The paper’s caution around RWAs also connects to a well-known DeFi weakness: oracle manipulation. An oracle is the bridge that brings offchain data, such as a price feed or valuation, onto a blockchain. If that bridge is weak, the whole system can be gamed.
This is not a simple coding error. It is a structural design risk.
Chainalysis has documented how oracle manipulation attacks can drain protocols by exploiting low-liquidity markets and unreliable price feeds. One of the clearest examples is Mango Markets. In October 2022, Mango Markets suffered about $117 million in losses after manipulated prices and leverage mechanics were used against the protocol.
Chainalysis also reported that oracle manipulation attacks caused $403.2 million in losses across 41 separate attacks in 2022. That is not an edge case. That is a flashing warning light.
Why does this matter for RWAs? Because tokenized assets often rely on offchain valuation, redemption, or collateral data. If those inputs are stale, gamed, or simply bad, the on-chain wrapper does not save you. A beautiful smart contract built on rotten data still ends in the same place: a wreck.
What tokenization does well
There is a real bullish case here, and it should not be buried under the risk section.
Tokenization can improve settlement speed, broaden access, and make assets easier to use as collateral. For liquid assets, that is meaningful. A tokenized Treasury product can move faster than legacy market plumbing and make capital more mobile without pretending the asset behind it is something it is not. Major institutions are already moving this direction; Wall Street Titans BlackRock and Apollo Dive into DeFi is not exactly subtle about where the money is flowing.
That is why the report recommends starting with liquid assets as a testbed before pushing deeper into harder territory. It is the sane path. Build on something that already has depth. Then, if the industry insists on tokenizing slower assets, it should use proper gates, notice periods, redemption limits, and structures designed to match the asset’s actual behavior.
Those tools are not glamorous, but they exist for a reason. Gates restrict redemptions during stress. Notice periods delay withdrawals. Closed-end and interval fund structures are built to handle illiquid assets without lying about how fast investors can get out. Boring? Sure. Also necessary.
The same goes for regulation. Wharton argues that lawmakers should focus on the economic function of the product, not just the fact that it sits on a blockchain. That is a basic but often ignored point. A tokenized Treasury, a tokenized deposit, and a tokenized private loan pool may all use the same tech stack, but they carry very different risks.
Why the dark side matters
The downside is that tokenization can make risky products look cleaner than they are. That is where the industry can get sloppy, or outright dishonest, by marketing a slow asset as if it offers instant liquidity.
That is how run risk gets built. Investors believe they can redeem quickly at a fair price. Then stress exposes stale NAVs, thin secondary markets, or operational bottlenecks. Once redemption confidence breaks, things can go downhill fast.
The paper also points to the old securitization problem known as originate-to-distribute. In that model, originators may weaken underwriting if they can quickly pass risk on to someone else. Tokenization can improve distribution. It can also make it easier to package and move questionable assets if oversight is weak. Faster pipes do not fix bad incentives; they can amplify them.
That is the part the move-fast crowd tends to hand-wave away. The e/acc instinct to accelerate is understandable. Better rails, more access, less friction, all good things. But accelerating straight into a liquidity trap because someone wanted to tokenize a bad asset and call it progress would be peak financial cosplay.
Key takeaways
-
Why is 24/7 trading not the same as liquidity?
Because a token can trade around the clock even when the underlying asset can only be priced, sold, or redeemed slowly. Easy trading does not guarantee a fair exit under stress. -
Which RWAs are the best fit for tokenization right now?
Liquid assets like Treasuries and, in some cases, gold and large-cap equities are the cleanest fit. They already have deep markets, which makes tokenization more useful and less deceptive. -
What is the biggest risk in tokenized private credit or real estate?
Liquidity mismatch. If investors think they can redeem instantly but the underlying assets are slow-moving or stale-valued, stress can trigger run dynamics and broken redemptions. -
Why does oracle risk matter so much?
Smart contracts depend on external price data. If that data is manipulated or unreliable, the protocol can be exploited even if the code itself is working as intended. -
Does tokenization remove traditional finance problems?
No. It often repackages them. Run risk, stale pricing, concentration, and bad incentives can all survive the trip on-chain, just with better branding.
Wharton’s message is not anti-tokenization. It is anti-fantasy. That is exactly the right place to stand.
Tokenization can be a genuine improvement when the asset is already liquid and the rails are clunky. But if the industry keeps pretending that blockchain can turn a slow, opaque asset into instant liquidity, reality is going to collect its invoice sooner or later. And it will not be polite about it.
