Tze Donn Ng, an investor at Brevan Howard Digital, argues that crypto has already solved the token issuance game. In an essay translated and published by TechFlow, he says launch factories in the style of Pump.fun have pushed token creation and distribution into a highly industrialized process, while average holding periods have collapsed to a matter of seconds. In that market, he sees only two routes for a token to survive across cycles: it must become either a belief asset or a cash-flow asset.
His starting point is simple. The industry has become extremely efficient at producing new tokens, but building an asset that investors would want to hold for more than a year has become much harder. To think about durable demand, he reduces the problem to two variables: the strength of belief behind an asset, and whether economic value that reaches token holders is real and dependable.
Belief assets: value sustained by collective conviction
Ng defines belief assets as tokens with no underlying cash flow, where value exists because a community agrees that it exists. He writes that he does not buy BTC out of blind adherence to Bitcoin doctrine. He buys it because he believes enough other people hold an unshakable conviction in it.
That leads to the distinction he tries to draw between short-lived meme tokens such as WIF or BONK and what he calls a lasting “super meme” like BTC. Looking back across market history, he says durable belief assets tend to share five traits.
A sacred figure or leader
In his framework, a belief asset needs a leader who functions as a spiritual symbol. BTC has Satoshi Nakamoto, ETH has Vitalik, SOL has Anatoly, and HYPE has Jeff. When people hear the token’s name, the associated figure should come to mind immediately.
A near-death moment followed by survival
He points to Mt. Gox for BTC, The DAO split and reconstruction for ETH, and the collapse of the FTX empire for SOL. HYPE, he writes, also went through a severe test during the JellyJelly attack, when its price briefly fell into single digits and the market questioned whether it would disappear. Surviving that kind of event filters out tourists and leaves behind the most committed believers.
Life-changing gains for early supporters
Ng writes, “Number go up is the best BD.” In his view, price appreciation is the strongest form of distribution because people who become wealthy through an asset often turn into its most committed evangelists. He cites BTC delivering a 200x return for early participants between 2013 and 2018, ETH rising 40x between 2020 and 2022, and SOL producing its own 100x story. HYPE, he says, rose 30x while maintaining meaningful liquidity and also distributed a large airdrop to early believers.
A community with symbols and inside language
Money alone is not enough. A belief asset also needs a meme system that spreads on its own. BTC has laser eyes, ETH has .eth names, SOL has ◎ and .sol, and HYPE has its active, sometimes confrontational, .hl supporters. Those markers create identity and reinforce belonging.
A short slogan that travels well
Ng says every digital religion needs a compact line that can be repeated. BTC is “digital gold.” ETH is the “world computer.” SOL has been framed as an “Ethereum killer” or a “decentralized Nasdaq.” HYPE, in his description, is the “house of all finance.”
He adds that ZEC has recently moved closer to this standard. He points to its survival through a major shielded pool vulnerability, its move from $40 to $1200 within a year, and narratives such as “privacy Bitcoin” and “the coming post-quantum Bitcoin.”
At the same time, he argues that these qualities cannot be manufactured on an assembly line. A team cannot simply create a community and then declare, awkwardly, that it has built a religion. Among early-stage names, he says he is watching PRL, described as “AI Bitcoin,” as an asymmetric bet to see whether those traits can emerge over time.
Cash-flow assets: real economic surplus that reaches the token
The second route is more direct. A token can survive by becoming a cash-flow asset, meaning the underlying business generates real economic surplus and that value actually accrues to token holders.
Ng says such assets are rare in crypto for two reasons. First, selling tokens and narratives to the market as exit liquidity is much easier than building a business with positive free cash flow. Second, if a team truly owns a cash machine that can grow without outside financing, it may prefer not to issue a token at all and instead keep the profits, avoiding governance and compliance headaches.
He names DexScreener, Axiom and Trojan as examples of profitable teams that have long kept 100% of business profits for themselves. In his telling, Hyperliquid changed the conversation by showing that a project can build a powerful cash-flow business, direct value to the token, and end up with a much larger market-cap and ecosystem outcome than it would by simply pocketing short-term profits.
To evaluate this category, he lays out five questions.
Is there durable economic surplus and business profit?
He calls HYPE and PUMP the clearest examples on this dimension, saying they generate $1 million to $2 million in real profit every day throughout the year. Perpetual DEX revenue, he argues, is usually more durable and less cyclical, especially after the addition of U.S. equities and other real-world asset stock perpetuals. Launchpad revenue is often viewed as less durable. He says the market was right to doubt the staying power of platforms such as PONS, STONK and GP, but that PUMP has continued to challenge that assumption.
Can the revenue be verified on-chain?
For Ng, the cleanest revenue is revenue that happens entirely on-chain. HYPE and PUMP fit that description and therefore require little trust. By contrast, GRASS generates revenue off-chain through private commercial agreements with leading AI labs. That leaves room for market skepticism, even though he says he personally believes the revenue is real.
Does the value belong to the token or to an equity company?
This question has become a live debate, he writes. Venice recently raised money through equity while distributing tokens to the community for free. GRASS also held a call to clarify that commercial value belongs to token holders.
His clearest negative example is SHFL. He says the underlying gambling business produces strong revenue, but the actual cash goes to a traditional equity company registered in Curacao, which has just paid out $18 million in cash dividends. Token holders in the secondary market, by comparison, receive little more than a lottery ticket.
How does value get captured by the token?
Buybacks and dividends are the most direct routes, but Ng places more weight on automatic buybacks and distributions enforced by smart contracts. He cites HYPE, LIT and SKY as examples. In his view, those mechanisms offer much more transparency and certainty than discretionary buybacks that depend on team decisions, such as PUMP.
If distributions are delayed for growth, can the team be trusted?
This is the quadrant that most resembles venture investing. He places GRASS, ENA, CHIP and CARDS there: projects that say commercial value ultimately belongs to token holders but are currently reinvesting profits into expansion.
The problem, he says, is that this model depends heavily on moral trust. Token holders usually lack legal protection, and often there is no board oversight or audited financial reporting. That is why he says frameworks such as Blockworks’ Token Transparency Framework, or TTF, matter.
Ng does not argue that every early-stage project should immediately distribute all profits. Doing so too early could cap long-term growth. But he also says token holders are justified in preferring present, certain cash returns. If holders cannot tell how retained earnings will eventually flow back to them, then “investing for growth” asks the market to underwrite both the team’s operating ability and its future integrity. He adds that automatic buybacks already reduce a large part of that uncertainty, and says he appreciates recent improvements by CHIP in clarifying how token value accrues.
The names he highlights now and the ones he is watching next
Ng says only four tokens currently satisfy all five conditions in full: HYPE, LIT, PUMP and SKY. New tokens that can check every box remain rare.
He also mentions PAPER as one of the best recent attempts. In his view, it had the potential to generate substantial revenue, but the team abandoned the token because its design created an arbitrage vulnerability that incentivized malicious attacks on exchanges.
He believes the cash-flow token category is approaching a major shift. HYPE established the template for a high-value accrual token, while PAPER showed why the structure is attractive. He offers Polymarket as a thought experiment: if a business generating roughly $20 million in monthly profit were to issue a token and direct all of that profit to token holders, the valuation implications would be significant.
He also says the Shuffle team has recently been discussing internally the launch of a fully on-chain decentralized casino. If such a system goes live and uses 90% of net gaming revenue each day to buy back tokens in the open market, SHFL would move to the top of his ranking list. Buying projects near the bottom of that list, he writes, is effectively a bet that teams will eventually reform and move toward the upper-right corner of the framework.
Looking ahead, he says he is closely watching a new generation of perpetual DEX tokens, including VAR from Variational and XTEN from Extended. He also names ARCUS, ENTROPY and a potential future token from Polymarket as assets worth following.
Crypto’s endgame: two mountains, not many roads
Ng closes by reducing token evolution to two mountains. If a digital asset is going to survive multiple bull and bear cycles, he argues, the path is narrow.
BTC, ZEC and early HYPE climbed what he calls the mountain of belief. HYPE, LIT and PUMP are climbing the mountain of cash flow. Among all digital assets, he says HYPE offers the most comfortable holding experience because it combines high-certainty cash-flow characteristics with the possibility of evolving into a durable belief asset as well.
Most other tokens, in his view, keep trying to climb and then fall back into the valley. That includes newer Layer 1 tokens in an already oversupplied market, as well as older DeFi tokens that lack real protocol revenue and do not return value to holders. Narrative-driven speculative assets, including shallow AI concept coins, may repeatedly try to scale the belief side of the map, but he argues they struggle to stay there over time.
The essay was published by TechFlow as a translation of Ng’s original piece. The byline identifies him as donn, also known as Tze Donn Ng of Brevan Howard Digital.
