Lead paragraph
Ethereum’s development roadmap has entered a cadence where incremental protocol upgrades are explicit, named and sequenced — with ‘Glamsterdam’ and ‘Hegota’ the most recent signposts cited by Decrypt (Mar 29, 2026). The network’s engineering narrative since the Merge (15 Sept 2022) has shifted away from consensus-level throughput gains toward rollup economics, calldata pricing and state-management primitives. That strategic pivot continues to shape on-chain metrics — most notably issuance dynamics and validator economics — that institutional desks must monitor alongside application-layer adoption. Regulatory, macro and competitive factors (layer-1 rivals and rollup architectures) will interact with protocol-level changes to determine where value accrues. This note synthesizes documented milestones, gives measured data context, and highlights implications for market participants and risk managers.
Context
Ethereum’s roadmap is the product of iterative design choices taken since major milestones in 2022–2024. The Merge (15 Sept 2022) transitioned consensus from Proof-of-Work to Proof-of-Stake, a change the Ethereum Foundation estimated reduced annual ETH issuance by roughly 90% relative to pre-Merge levels (Ethereum Foundation, Sept 2022). That structural change was followed by Shapella (also known as Shapella/Shanghai), which enabled validator withdrawals on 12 Apr 2023 (Ethereum Foundation, Apr 12, 2023), altering staking liquidity dynamics and introducing a periodic source of net supply movement.
Subsequent upgrades focused on execution-layer optimizations and rollup-friendly primitives: EIP-4844 (proto-danksharding / calldata gas pricing) was targeted to lower rollup costs and was integrated into the protocol roadmap in 2023–24 (Ethereum EIPs repo). Decrypt’s March 29, 2026 coverage names the next stages — Glamsterdam and Hegota — as upgrades intended to refine calldata pricing, state expiry mechanisms and L2 interoperability (Decrypt, Mar 29, 2026). These changes represent a continued move to position Ethereum as a settlement layer for rollups rather than trying to scale all demand on L1.
This context matters because protocol design choices constrain economic outcomes. A 90% cut in issuance (post-Merge) is not equivalent to a supply cap — it is a recurring issuance regime that interacts with burn dynamics under EIP-1559 and with withdrawal patterns following Shapella. As a result, short-term price sensitivity to supply shocks (e.g., sudden large-scale withdrawals) coexists with a medium-term strategic trend toward rollup-led throughput expansion.
Data Deep Dive
Three dated, verifiable anchors help quantify where Ethereum stands: the Merge (15 Sept 2022), Shapella withdrawals unlocking staking liquidity (12 Apr 2023), and Decrypt’s roadmap exposition (29 Mar 2026). The Merge’s issuance reduction (~90%) was measured against pre-Merge block rewards and transaction-fee driven issuance; the Ethereum Foundation reported these figures in its post-Merge documentation (Ethereum Foundation, Sept 2022). Shapella’s activation on 12 Apr 2023 permitted staked ETH withdrawals that had been locked since validators began staking ahead of the Merge; the immediate withdrawal throughput was throttled by protocol limits but created a new, observable flow of ETH into spot markets (Ethereum Foundation, Apr 2023).
On activity and economic metrics, EIP-4844 and related calldata changes materially change L2 economics by lowering per-byte costs for rollups. Benchmarks published by protocol developers and independent researchers (2023–2025) show calldata cost reductions of between 40%–70% for optimistic rollups and significant per-transaction cost declines for zk-rollups when proto-danksharding primitives are active (EIP analysis, 2024). Those reductions are meaningful when compared with 2021–22 gas spikes: for example, median mainnet gas fees spiked above 100 gwei during high demand in 2021 vs more stable sub-30 gwei ranges in quieter post-upgrade periods (on-chain analytics platforms; historical gas data).
Validator and staking metrics are central to macro liquidity modeling. The enabling of withdrawals in April 2023 changed the liquidity profile of staked ETH: previously locked balances could not be redeployed; post-Shapella they could. Public telemetry (Beacon-chain explorers and governance metrics) shows that validator counts and effective staking percentages continued to rise after 2023 as institutional staking products and liquid staking tokens proliferated, but the balance between staked supply and liquid supply remains a key variable for exchange inventory and market-making desks.
Sector Implications
The technical emphasis on calldata pricing and rollup optimization shifts where economic value accrues across the crypto stack. Layer-2 protocols and their operators stand to capture fees and product differentiation as calldata becomes cheaper and rollup throughput improves, while base-layer fee revenues may compress relative to prior eras. That structural shift implies a re-allocation of fee capture from L1 block producers to L2 sequencers and rollup operators — a dynamic analogous to the specialization seen in other technology stacks when a shared transport layer commoditizes throughput.
For institutional counterparties, two consequences are material. First, settlement risk and custody frameworks will need to incorporate more granular monitoring of rollup health and surcharge mechanics; smart contract risk on L2s becomes a first-order consideration. Second, the evolution of staking economics post-Shapella — combined with liquid staking derivatives — changes the financing returns on ETH exposure relative to other yield assets. Comparing yields: staking returns historically fell into a multi-percent range post-Merge (variable with validator participation), which is structurally different from fixed-income asset classes with deterministic coupons.
Competition from other L1s must be assessed quantitatively: throughput, finality times and fee curves on competing chains (e.g., Solana, Avalanche, new modular architectures) are relevant. Yet Ethereum’s unique asset base, developer tooling, and network effects mean comparisons are not apples-to-apples. Market-share metrics such as DeFi TVL, NFT activity, and developer counts remain useful benchmarks; even with compression of some L1 revenue streams, Ethereum’s aggregate ecosystem depth (measured by installed dApps and institutional-grade tooling) continues to differentiate it from most peers.
Risk Assessment
Upgrades of this nature introduce schedule and implementation risk. Even well-tested EIPs can have subtle interactions with execution-layer clients, validator software, and L2 sequencers; hardening testing environments and public testnets provide mitigation but do not eliminate residual risk. For example, proto-danksharding required coordinated changes across client implementations and rollup operator logic; any delayed or uneven rollout increases operational risk for services relying on predictable calldata pricing.
Economic risks are also non-trivial. Shapella’s withdrawal mechanics created episodic selling pressure possibilities; while protocol-level throttles limit instantaneous flow, cumulative withdrawals can be significant if correlated with market stress. Similarly, changes that benefit rollups could transiently reduce L1 fee revenue, affecting validator commission economics and, by extension, staking yields. Correlations between macro market volatility and withdrawal behavior should be included in stress-testing models for desks and custody providers.
Regulatory and custody risk remain active. As the protocol makes staking liquidity more fungible (via liquid staking tokens and withdrawals), custody models and regulatory classifications of staking services will be scrutinized. Operational compliance demands — KYC/AML, asset segregation, and insurance frameworks — need to be re-evaluated as protocol design alters liquidity profiles.
Outlook
In the near-to-medium term (six to 18 months following the Decrypt roadmap update dated 29 Mar 2026), the probability-weighted outcomes center on clearer rollup economics, lower per-transaction costs for L2s, and continued developer focus on application-level scaling. If Glamsterdam and Hegota implement state expiry or calldata-accounting changes successfully, we should expect L2 throughput to become materially cheaper versus 2021–22 baselines, accelerating migration of activity to rollups.
Over a longer horizon, the network’s role as a settlement and coordination layer becomes the dominant valuation narrative: Ethereum’s value capture mechanism may increasingly decouple from raw transaction counts and instead reflect protocol-level fees allocated across a broader stack that includes sequencers, prover services and settlement anchors. Market participants should model fee accrual and capture at the stack level rather than treating L1 fee revenue as the sole indicator of network monetization.
Operationally, protocol upgrades will continue to require multi-client coordination and active observability. Counterparties should maintain rapid deployment and rollback procedures for node software and keep exposure to single points of failure low. Institutional readiness will be judged by process — automated client testing, diversified validator infrastructure, and integrated rollup monitoring.
Fazen Capital Perspective
Fazen Capital assesses the roadmap pragmatically: Glamsterdam and Hegota are not mere feature releases but strategic steps in a longer migration toward modularity. The contrarian view is that the market underprices the systemic value of predictable, low-cost rollup settlement even while it overemphasizes headline L1 fee compression. In other words, a modest decline in L1 fee revenue can coexist with a larger structural expansion in ecosystem value if rollups unlock new classes of application throughput and revenue that were previously unattainable on-chain.
From a risk-adjusted perspective, the non-obvious implication is that custody and trading infrastructures that can natively interact with rollups and manage cross-layer settlement will capture disproportionate operational rents. This suggests institutional priorities should tilt toward architectural adaptability (multi-rollup, multi-client) rather than static exposure to L1 fee accrual. For further reading on how infrastructure adaptation matters for allocators and service providers, see our broader research on platform strategies and custody [here](https://fazencapital.com/insights/en) and our notes on layered scaling economics [here](https://fazencapital.com/insights/en).
Bottom Line
Glamsterdam and Hegota are iterative but material steps in Ethereum’s shift toward being a rollup-centric settlement layer; the changes recalibrate fee economics, staking liquidity and where value accrues across the stack. Institutional actors should prioritize operational resilience across L1 and L2, and reassess fee-capture models in a modular future.
Disclaimer: This article is for informational purposes only and does not constitute investment advice.
FAQ
Q: Will Glamsterdam and Hegota immediately reduce ETH issuance?
A: No. Issuance policy is governed by consensus changes like the Merge (15 Sept 2022), which produced an estimated ~90% cut in issuance (Ethereum Foundation, Sept 2022). Glamsterdam and Hegota are focused on calldata and state mechanics; their primary effect is on fee allocation and transaction economics rather than direct issuance changes.
Q: How should firms model withdrawal risk after Shapella (12 Apr 2023)?
A: Firms should incorporate scenario-based withdrawal flows into liquidity stress tests, using historical post-Shapella withdrawal throughput as a baseline and applying correlation shocks with market volatility. Protocol-level rate limits moderate instantaneous outflows but cumulative withdrawals can be material over multi-week horizons; custody models should reflect this.
Q: Do rollup cost reductions imply L1 revenue destruction versus peers?
A: Cost reductions for rollups can lower L1 fee revenue, but this is not equivalent to destroying ecosystem value. Consider the comparison: a lower-cost settlement fabric can enable substantially higher transaction volume and new applications, shifting fee capture downstream to sequencers, provers and other modular actors. Evaluating peers requires stack-level analysis of where fees and value are captured, not L1 fee totals alone.
