Invest2 publishers3 min readPublished Updated
Compound puts $52M behind an institutional pivot, and $24M of it is not code
The DAO approved a record budget with $14M unlocked immediately and the rest on milestones. TVL is down about 90% from 2021, and Aave holds roughly twelve times as much.
The Investor · Invest desk

What happened
- The Compound DAO endorsed a $52 million two-year development program, the largest funding commitment in the protocol's history, to develop infrastructure supporting on-chain credit services for traditional financial entities.
- Since its launch in 2018, Compound has processed roughly $480 billion in deposits and borrowing activity.
- Compound has maintained a record of zero bad debt.
- Compound's total value locked has declined from a 2021 peak near $12 billion to approximately $1.2 billion, prompting a strategic recalibration away from retail-focused yield incentives.
- Aaron Schnarch, previously chief executive of Coinbase Custody, assumes the role of executive director at Compound.
Compiled by The InvestorSomething wrong?How this is made
Why it matters
Compound's DAO has approved a $52 million, two-year development program, the largest funding commitment in the protocol's history, and installed a new leadership team with Aaron Schnarch, previously chief executive of Coinbase Custody, as executive director [1][5]. The stated goal is to bring institutional credit on-chain [1]. The unstated one is that the retail lending business Compound helped invent has lost roughly 90 percent of its peak size [4][18].
The numbers are not ambiguous. Total value locked has fallen from a 2021 peak near $12 billion to approximately $1.2 billion, a decline of about 90 percent [4][18], and Crowdfund Insider's account of the program says the shift is explicitly away from retail-focused yield incentives [4]. Against that, the protocol has processed roughly $480 billion in cumulative deposits and borrowing since launching in 2018 with a record of zero bad debt [2][3]. Cumulative flow of that size against $1.2 billion currently locked works out to about 400 times [19]. The mechanism functions; the depositors left anyway.
The budget structure carries the argument. About $14 million is available immediately, roughly 27 percent of the total, with the balance released against predefined milestones [7][20]. Some $28 million goes to operations and engineering including Compound V4, and the remaining $24 million to growth [8][9]. Of that growth money, $8 million to $10 million is earmarked for institutional partnership development rather than conventional liquidity-provider rewards [9], which is between a third and 42 percent of the growth line [21]. Read plainly, the DAO has decided that paying for mercenary liquidity is a worse use of treasury than paying for counterparties.
The hires point the same direction. Christopher Donovan joins as chief operating officer from the same role at the Near Foundation [10]. Steven Liu becomes chief product officer, having scaled Maple Finance's assets from $500 million to $5 billion, a tenfold increase [11][22]. Leo Eikelman is chief technology officer [12]. This is a credit-desk bench, not a growth-hacking one.
V4 introduces a hub-and-spoke architecture that centralises capital allocation through a core hub, which the protocol says delivers the tighter risk controls and capital efficiency professional counterparties require [13]. The roadmap also prioritises native real-world asset support, integration tools that let institutions embed on-chain lending in their own products, and compliance features for regulated entities [14]. Protocol representatives say banks, asset managers, exchanges and fintechs want on-chain lending experiences but lack the expertise to build them [15]. More than ten partners have committed, with discussions underway with more than twenty others, and initial institutional-grade products are expected within weeks [16][17]. The announcement was made by the Compound Foundation on 17 August 2026 [23].
Three things to watch. First, whether the milestone gates bind: $38 million of the $52 million is conditional [7][24], and a DAO that waves through tranches has simply written a cheque. Second, conversion. Ten committed partners is a pipeline claim, not a balance sheet; the test is whether TVL composition changes, not whether the headcount of logos does. Third, the hub. Concentrating allocation in a single core hub is presented as a control improvement [13], and it is also a concentration of the thing that has to never fail, against a zero-bad-debt record built on the older design [3].