A recent Fidelity Investments report makes a strong case for why blockchains can be valued on the traditional GDP formula. But…there are problems. Let’s unpack it. 

— Donovan

$40 million GMX hack triggers $135 million TVL exodus:

GMX’s total value locked (TVL) has plunged by at least $135 million after a $40 million exploit on its Arbitrum v1 deployment sent nervous traders rushing for the exits. DeFiLlama data shows protocol-wide TVL — combining v1 and v2 on Arbitrum, Avalanche and Solana — has slumped from over $1.2 billion at its 2023 peak to under $400 million today.

The recent attack exploited a design flaw in GMX v1’s AUM calculation, where the “global short average price” was updated immediately when opening large short positions. Attackers used re-entrancy during order execution to manipulate this price, inflating the redeem value of GLP tokens and draining $40 million in a single atomic transaction. GMX quickly froze v1 trading and capped GLP minting, while confirming v2’s isolated-pool design remains unaffected.

The exploiter bridged stolen USDC to Ethereum and swapped for DAI to avoid an asset freeze.

Even so, the exploit damaged trader confidence at a critical moment. GMX v2’s 2025 launch on Solana had delivered a boost in perps volume, but the brand now faces a trust deficit, with TVL in retreat despite multichain growth ambitions and claims of more robust security in its newer deployments.

— Macauley Peterson

Blockchain GDP

A recent report by Fidelity Investments proposes valuing blockchains on the basis of GDP:

“... it is more appropriate to compare decentralized blockchains to sovereign nations and their economies rather than web2 companies or products because of the embedded currency.”

Here’s the GDP formula: C + I + G + (X-M)

C is consumption, I is business investment, G is government spending, X is exports and M is imports, so X-M is net exports.

Fidelity uses ETH as an example. So, when you transpose the GDP formula onto Ethereum blockchain metrics:

• C = What users are spending as gas, to use Uniswap or mint an NFT.

• I = The quantity of staked assets or capital in liquidity pools.

• G = Ethereum Foundation expenditure, issued ETH to validators.

• X-M = How much stablecoins are minted/burned, bridge flows to/from other chains and DePIN rewards.

You can see the entire table here:

It’s a comprehensive effort by Fidelity, but it provokes some questions.

GDP is a measure of domestic production. Think “the value of everything made here.” When a country exports, that’s domestic production. When it imports, that’s spending. That’s why we “net” imports for GDP.

But if millions of stablecoins are bridged onto (import) or off of (export) Ethereum, that bloats a blockchain’s “GDP” even though nothing productive occurs onchain.

Contrast that to when a stablecoin is minted onchain, or when a Helium miner is paid in tokens for providing a useful mobile cellular service. These are productive “imports” that would rightfully count toward a blockchain’s “GDP.”

So measuring “net exports” by bridge flows is conceptually sound, but it needs to account for CEX cold-wallet sweeps, as Blockworks’ Dan Smith aptly pointed out.

Explicit in Fidelity’s valuation model is also the claim that L1 tokens should be valued on the basis of “money,” or more specifically: a medium of exchange and store of value.

Fidelity argues: “Ether is the dominant trading pair on exchanges and serves as a primary asset to borrow against.”

I think that at best justifies the “medium of exchange” aspect of money, but is silent on the “unit of account” aspect.

Early crypto investors have questioned the ability of L1 tokens to serve as a unit of account. As John Pfeffer wrote back in 2017:

“It is thus overly simplistic to assume that people will hoard that which they use to make payments as opposed to converting their store of value via the payment rail at the time of payment in the exact amount needed and for as little time as possible.”

Account abstraction (ERC-4337) even formalizes this reality, since it enables paying gas fees in any ERC-20 token. That vastly improves the user experience but it removes the need to hoard ETH, thereby undermining the “monetary premium” of the L1 token.

The final aspect of why I think the GDP analogy is somewhat strained looks to accounting for staked ETH under the “Investment” bucket of GDP.

Staking locks up existing assets, but no new productive capacity is created.

In economist jargon, it doesn’t push the “production possibilities frontier” in the same way investment does in the real economy.

So the “I” in blockchain GDP loses its predictive link to future growth.

Even worse: LP deposits can migrate and earn purely extractive airdrops or MEV.

— Donovan Choy

Compound DAO terminated its security contract with OpenZeppelin (OZ) on Wednesday following an onchain vote. Faced with the expiration of its previous provider and no ready alternative, the DAO had voted to renew the deal with its longtime service provider on July 1 only to immediately turn around and exercise a 60-day early termination clause.

At first glance this sounds like a contradiction, but it was actually strategic: The DAO needed to avoid any security coverage gap while buying time to run a competitive RFP process, according to Michael Lewellen, the former head of solutions architecture at OpenZeppelin.

“It'll allow other vendors to bid alongside OZ in a more structured way,” Lewellen told Blockworks.

The termination limits new spending to two months (~$0.67 million) and ensures a smooth handoff to any new provider.

The maneuver was also a result of apparent cost fatigue: OZ has earned over $16 million from Compound since 2021. Critics like FranklinDAO (formerly Penn Blockchain), which voted against the renewal, have accused the company of overcharging and using its last-minute renewal timing to pressure the DAO. FranklinDAO now threatens to oppose any future deal with OZ entirely.

Could OpenZeppelin salvage goodwill? Yes — by discounting or waiving fees for the final 60 days, or offering pro bono transition support. Such a gesture would defuse “vendor lock-in” criticism and improve its standing in the RFP process.

Ultimately, this episode signals Compound’s shift away from entrenched single-provider deals toward a more transparent, competitive procurement, stewarded by a new Compound Foundation  in the process of being set up. This may be the start of a trend that rejiggers the ways DAOs handle major service contracts across DeFi protocols.

— Macauley Peterson