A mid-year reset on the dossiers that have moved meaningfully since our January outlook. Three positions we hold more firmly than we did six months ago, and two that warrant fresh humility.
The first half of 2026 has obliged us to revisit several positions stated in our January outlook. Some have hardened with the passage of time; others have weakened, and we owe clients an honest account of both.
The first is that liquid staking has crossed a threshold and is no longer a discount play. Through the spring, validator queue dynamics tightened, MEV revenue stabilised, and the spread between staked ETH and held ETH compressed below 90 basis points on a six-month average. We continue to recommend material exposure to stETH and rETH, but we now treat them as a baseline rather than an opportunistic allocation.
The second is the durability of stablecoin yield. Twelve months ago we cautioned that yields above 5% on regulated stablecoins were unlikely to last. They have. Sustained demand for on-chain dollars from institutional balance sheets — particularly in Asia — has anchored DeFi rates well above the wider rates environment. We have stopped treating those positions as carry trades and now treat them as cash management.
The third is that cold-storage allocations should grow with portfolio size, not shrink. The instinct of clients who have prospered is to deploy more, not less. We disagree. The marginal yield earned on the last 5% of a portfolio rarely justifies the marginal complexity it introduces. Above CHF 100 million in client capital, our advised cold-storage floor is now 18%.
We were too sanguine in January about restaking. The layered architecture of EigenLayer-style protocols — multiple AVS counterparties absorbing operational risk in exchange for thin compounding yield — has not proven robust under stress. We have reduced restaking exposure in client mandates by roughly a third since February. We expect to be at zero in some accounts by autumn.
We also overestimated how much regulatory clarity in Europe would translate into regulatory comfort for our clientele. MiCA has shifted compliance complexity from operators to issuers, but it has not made cross-jurisdictional planning materially simpler. Swiss-resident clients with EU-resident heirs face the same set of questions in 2026 that they did in 2024.
The greatest variable in the second half will not be price. It will be the rate at which protocol risk continues to migrate — from smart contracts into operators, from operators into custody, and from custody into governance. We will write at length on this in late summer.