Welcome to Galaxy Research's Weekly Top Stories. Subscribe to get this newsletter delivered to your inbox every Friday morning.

In this week's edition, Will Owens unpacks Polymarket’s decision to start charging trading fees; Lucas Tcheyan explains the significance of Tether’s partnership with fast-growing consumer internet marketplace Whop; and Thad Pinakiewicz looks at the wind-down of Balancer Labs.

Got feedback on this newsletter? Email [email protected]. We’d love to hear from you.

Market Update

The total crypto market cap stands at $2.37tn, down 5.06% from last week (when it stood at $2.50tn). Bitcoin's network value is 4.23% of gold's market cap. Over the last seven days, BTC is down 5.22%, ETH is down 6.75%, and SOL is down 6.43%. Bitcoin dominance is 56.34%, down 16 basis points from last week.

Polymarket’s Free Ride Is Over

On March 30, Polymarket flips the switch. After months of charging fees only on crypto and sports markets, taker fees are expanding to nearly every category on the world’s largest prediction market platform: politics, finance, economics, culture, weather, tech, and more. Only geopolitics and world events markets stay fee-free.

The timing isn’t subtle. Polymarket just printed ~$9.55 billion in 30-day volume. While people love to criticize prediction markets, the platform is no longer a scrappy experiment. Polymarket is a legitimate business backed by Intercontinental Exchange, the parent of the New York Stock Exchange. The free era built the user base by incentivizing traders to add liquidity to markets. The fee era now has to justify the price tag.

Today, only crypto and sports markets carry taker fees. Crypto peaks at 1.56% effective rate (at 50% probability); sports at 0.44%. On March 30, those rates will climb and fees for new categories come online.

Upcoming fee structure on Polymarket.

Fees are not flat; they are dynamic. They’re calculated using a formula that peaks when shares trade near $0.50 (maximum uncertainty) and drops toward zero at the extremes (minimum uncertainty). A 100-share trade on a crypto market priced at 50 cents (fifty-fifty probability) costs $0.90 in fees. The same trade priced at 95 cents costs $0.32. Near-certain outcomes are essentially free to trade for market participants.

Fees will remain taker-only. Limit orders pay nothing and are eligible for daily USDC rebates funded by the fees that takers generate. So, market makers are actually getting paid and are exempt from this new fee structure.

Fees apply only to new markets deployed on or after March 30 (pre-existing markets are untouched), and, as before, there will be no Polymarket fees on deposits or withdrawals.

Our Take

The category-by-category structure isn’t random. Crypto gets hammered hardest because it’s the highest-velocity, highest-manipulation-risk corner of the platform. These short-duration price markets are swarmed with bots. Fees here act as a speed bump against wash trading and high-frequency trading (HFT). Sports and politics stay cheaper because they’re Polymarket’s flagship categories, the ones that drive brand recognition and mainstream adoption. Finance gets the richest maker rebate, a clear play to attract professional liquidity providers where order book depth matters most.

Geopolitics contracts stay at zero. This is the crown jewel, the category where prediction markets produce their purest and most useful information signal. Charging fees on geopolitical outcomes would risk distorting the very thing that makes everyone love the platform as a public good.

On the revenue side, Pine Analytics ran the numbers using onchain Dune data. At current taker volume, Pine estimates ~$1.2M in daily gross fee revenue post-rollout, with a blended effective rate of about 0.76%. After maker rebates and referral payouts, net protocol revenue lands between $570K-$950K/day.

That annualizes to somewhere in the $209 million to $342 million range. For context: Pump.fun (everyone’s favorite, or least favorite, Solana memecoin launchpad) generates ~$1m/day in net revenue and Hyperliquid sits around $2m/day. Even at the low end, Polymarket enters the conversation among crypto’s top revenue-generating applications.

The bull case writes itself: fees fund deeper liquidity via maker rebates, tighter spreads attract more volume, and the flywheel spins. Polymarket becomes a real business without sacrificing the geopolitical signal that made it culturally relevant. The bear case is also straightforward. Takers now pay 0.5% to 1.8% on contested bets. Retail traders and arb bots that thrived on zero-fee markets might compress taker volume or go limit-order-only.

The soft rollout does help, though. Only new markets are affected, so existing positions are grandfathered in. The real test will come as legacy markets expire and fee-enabled ones replace them. Watch crypto volume first. It carries the highest fees and will be the canary in the coal mine. If taker activity holds there, the rest of the categories should be fine.

Polymarket and Kalshi are in scorching competition to dominate the prediction markets space. Let’s see how this decision holds up. - Will Owens

Whop, There It Is: DeFi Meets the Creator Economy

Whop, a fast-growing online marketplace where creators sell digital products, launched a suite of financial tools this week. The first offering, Whop Treasury, is built fully on a crypto stack.

The product lets businesses and individuals earn up to 6% APY on their cash balances with real-time accrual, no lockups, and full liquidity. User funds are held in USDT0, a stablecoin issued by Tether on Plasma, a blockchain purpose-built for stables. Wallet infrastructure is handled by Tether's Wallet Development Kit, onramps are powered by MoonPay, and yield is generated through the Aave protocol using vaults developed by Veda. In addition to yield, users can diversify their treasuries beginning with access to Tether's tokenized gold offering XAUT. The announcement also highlighted plans to introduce additional assets such as bitcoin and ether in the future.

The announcement comes on the back of Whop's newly launched Payments Network, which provides platform users with access to 100+ payment methods across 187+ countries and 241 territories, including crypto. For businesses on the Payments Network, revenue can now flow automatically into Treasury and begin earning yield the moment it lands.

The entire buildout follows Tether's $200 million strategic investment in Whop last month, which valued the company at $1.6 billion and kicked off the integration of Tether's stablecoin infrastructure into the platform. Tether has been on an aggressive investment spree, deploying capital from its profits and excess reserves across AI, financial services, energy, and digital media through its investment arm based in El Salvador.

Our Take

Consumer-facing crypto integrations at this scale are rare. Most of the stablecoin infrastructure buildout over the past two years has targeted institutional use cases like cross-border settlement, treasury management, and tokenized assets. Whop is doing something different, embedding stablecoins, DeFi yield, and self-custodial wallets into a product with 18.4 million users and $3 billion in annual creator payouts, many of whom may not have directly used crypto previously. Notably, Whop Treasury isn't limited to businesses. Individual Whop users can also hold USDT and earn yield directly, giving anyone on the platform access to the same financial infrastructure as the merchants they buy from.

What’s especially interesting is how many layers of the crypto stack are working together under the hood. Stablecoins, a purpose-built payments L1, DeFi lending, vaults, and self-custodial wallets are all in the mix, and end users don’t have to understand any of it. They just see revenue flowing in, a balance earning 6%, and a withdraw button. This setup supports the thesis that blockchains are fundamentally a financial technology that makes it dramatically easier for any platform to integrate financial products — payments, banking, yield, trading — into its core offerings.

That thesis is playing out beyond Whop. As we covered last week, Stripe built an entire blockchain (Tempo) around stablecoin settlement and agentic payments. And in a parallel move in the creator economy, Beast Industries, YouTube star MrBeast's holding company, acquired the Gen Z banking app Step and received a $200 million investment from Bitmine Immersion Technologies (BMNR), the Ethereum treasury company, with explicit plans to integrate DeFi into its upcoming financial services platform. A common thread is emerging. Consumer-facing companies are discovering that crypto infrastructure lets them ship financial products faster, cheaper, and with fewer intermediaries than building on traditional rails.

The Tether relationship also highlights a growth dimension that goes beyond the technology. Tether's ecosystem reaches more than 530 million users globally, with over $180 billion in issued stablecoins. The partnership is designed to connect Whop's platform with that distribution, particularly in emerging markets across Latin America, Europe, and the Asia-Pacific region where traditional banking infrastructure introduces unnecessary friction and cost. For Whop, the stablecoin integration isn't just a better backend, it's a user acquisition strategy for going global.

The gap between "crypto product" and "product that uses crypto" is closing fast. Stablecoin supply is expanding not because retail users are rushing into crypto, but because stablecoins are becoming invisible infrastructure inside platforms that already have the users. The protocols that win from here will be those that become the default yield and settlement backend for fintechs that already have distribution. - Lucas Tcheyan

Balancer, Unbalanced

Balancer Labs, the for-profit entity behind the Balancer protocol, is throwing in the towel: stepping back from the protocol and proposing drastic tokenomics changes on the way out.

Balancer is not pivoting; it is restructuring in public. After a November exploit that drained over $100 million and left behind legal and balance sheet damage, the team is winding down Balancer Labs entirely—explicitly calling the corporate entity a liability rather than an asset. That alone tells you where things stand.

The protocol still works. It still facilitates trades and generates fees. But the structure wrapped around it—legal, operational, and economic—stopped making sense. Total value locked (TVL) collapsed, costs stayed high, and the tokenomics that once bootstrapped growth became a drag on survival.

Our Take

The diagnosis in the governance posts is unusually direct: BAL governance token emissions and the veBAL (vote-escrow) vesting system were designed for a growth phase, and they worked in that context. But over time, they devolved into a circular subsidy loop, paying increasingly more to attract liquidity that wasn’t valuable enough to justify the cost. Meanwhile meta-protocols like Aura stepped in to optimize extraction for their own token holders rather than for Balancer itself. This is the core problem with ve-tokenomics (vote escrow) in mature systems: what begins as a coordination mechanism becomes a rent-extraction layer. The protocol pays more to sustain the system than the system generates in return.

This outgrowing of the ve-tokenomics model, and the negative externalities involved, is exactly what drove Pendle to abandon its ve-tokenomics earlier in the year. The proposal for Balancer goes even further: it is not only sunsetting its ve- model, but also completely severing the economic link between BAL and the protocol’s performance. Emissions go to zero. veBAL is wound down over a year. All fees are redirected entirely to the DAO treasury. Any future value accrual is deferred into discretionary buybacks, contingent on the protocol becoming profitable after costs.

Balancer isn’t leaving token holders out to dry, though. The protocol is offering a buyback at net asset value (NAV), up to 35% of the treasury, after a one-year phaseout of veBAL, and distributing $500,000 in stablecoins to veBAL holders to compensate them for foregone yield, roughly a 20% dividend. The buyback is framed as fair exit liquidity, and in one sense it is—an honest acknowledgment that not everyone signed up for this new contract. But the partial buyout creates its own dynamics. If you believe the protocol can become sustainably profitable, you hold. If you don’t, you line up for the exit. Because only a portion of supply can be redeemed, belief becomes reflexive: if too many people doubt the turnaround, you get a rush to the door. If the market is sanguine enough, Balancer may be forced to sweeten the terms—spending more of its treasury just to get tokenholders to take the exit.

Operationally, the rest of the plan is as stark as the token changes. Balancer Labs is gone, the team is cut down from 25 to 12, emissions are eliminated, and the goal is simple: run the protocol profitably or stop pretending. The foundation-as-agent structure remains, but the experiment with a parallel corporate entity is over, partly for cost reasons, partly in an attempt to isolate the legal exposure from the exploit. Whether that separation actually holds is an open question; winding down an entity is not the same as making liabilities disappear.

This is what happens when a protocol outgrows its own tokenomics. Ve models can be powerful early on: they bootstrap liquidity, distribute control, and create a sense of participation. But as meta-governance layers mature, they introduce externalities that are hard to unwind: governance capture, ineffective incentives, and a steady drain on the underlying business. At some point, the token stops representing the protocol and starts parasitizing it. Balancer is choosing to sever that relationship rather than optimize it.

The uncomfortable implication is that BAL holders are being told, more or less explicitly, that the DAO cannot continue in its current form. Either it becomes a lean, revenue-generating system with optional buybacks, or it winds itself down over time. There is no middle ground where emissions paper over the gap; the protocol has four years of runway left. That is a hard reset, but also a rare one. DeFi is full of marketing pivots; it is not full of protocols willing to zero out their own incentive machine and refocus on keeping the ship afloat. Across Protocol faced similar issues with its structure and is taking its protocol private.

If this plan works, it will set a precedent: mature protocols may need not more clever maneuvers in tokenomics, but fewer of them. If it fails, it will confirm the opposite fear: that once you remove the subsidies, there is not enough organic demand to sustain the system. If DeFi wants to prove it can be more than a subsidy machine, this is the test: when the incentives disappear, does anything real remain? – Thad Pinakiewicz

Chart of the Week

Gold posted its worst stretch in more than a century this month, falling for 10 straight sessions in what appears to be its longest losing streak since February 1920. After sliding as much as 27% from its January peak, the metal found support near $4,400, just above its 200-day moving average, suggesting the selloff may be easing. Even so, gold remains down roughly 17% since the Middle East conflict escalated in late February.

Meanwhile, spot bitcoin ETFs have taken in nearly $2.5 billion over the past month, wiping out almost all year-to-date outflows even as BTC endured a roughly 30% drawdown since its Jan. 14 year-to-date high. March flows have been especially strong, with nine sessions above $150 million, including one near $458 million and two straight days above $200 million. More broadly, ETFs of all stripes now account for 37% of total U.S. equity trading volume, a record share that underscores how central they have become to market activity.

At the same time, bitcoin has held above $70,000 and continued to outperform gold. That has kept the bitcoin-to-gold ratio just below 16 ounces, up from roughly 12 ounces before the conflict began. Put differently, one bitcoin now buys about 30% more gold than it did at those earlier lows, reinforcing bitcoin’s resilience during this tumultuous event.

In Other News

• 🧊Circle freezes 16 wallets’ USDC in sealed court case, unfreezes some

• 🔍Tether says a Big Four firm will audit it, gives no name or timetable

• 🖥️Google moves post-quantum cryptography timeline up to 2029

• 🔷ETH devs won’t prioritize Vitalik-backed quantum-resistance plan in 2H upgrade

• 🇺🇸Senators, White House strike deal to resolve CLARITY Act dispute

• 🏡Fannie Mae to accept crypto-backed mortgages

• ₿ Nonprofit launches to build “conservative” Bitcoin client

• 🕛 Franklin Templeton debuts ETFs that trade in crypto wallets 24/7

• 🇧🇹Bhutan moves $37m in BTC to exchanges; holdings down 66% from peak