Lending and borrowing on an vanilla DeFi money markets has a PvP element to it. Borrowers and lenders both want to extract as much returns as they can. Sometimes, it can take a turn into an outright PvP situation. We’re currently seeing that play out on Aave, where the protocol is considering changes to the CRV pool parameters to combat Curve’s founder from “pseudo-selling” CRV by instead taking loans against it. At the moment, these loans are collateralized by an amount of CRV that would be near-impossible to liquidate.

While the pooled lending model certainly isn’t going anywhere, it does open up the design space for alternative solutions. One of those solutions is Stella, launching on mainnet on June 21.

Stella is a re-branding of Alpha Finance Lab or Alpha Homora. If you were around for the last DeFi cycle, chances are you heard of them. I always thought Alpha was cool product, but it failed to muster long-term usage in its previous incarnation. Stella is essentially a completely new product, but builds on Alpha Homora’s goal of trying to enable leveraged strategies. It introduces a new lender compensation model, termed “pay as you earn” or PAYE.

Essentially, Stella will have strategies that people can leverage up and deposit into. The leverage, as expected, comes from lenders on Stella. But rather than charging a floating interest rate that is determined by the pool size and utilization, there will be a return split between users and lenders. As the strategy reaps returns, a portion of this earned return will go to lenders.

The mechanics here are relatively simple. A user obtains leverage from the protocol’s lender; that loan + posted margin are deposited into the strategy. An LP token is minted against these assets and kept with the protocol. By virtue of combining the posted collateral and the loan, this is genuine leverage (under-collateralized loan) from the perspective of the trader. But for the protocol, it’s over-collateralized since they own the LP token and can liquidate the position if things turn south.

Now, it’s important to understand that this just a new design. Nothing is ever perfect, so there are bound to be trade-offs.

With the PAYE model, lenders do not receive any compensation until the borrower’s position generates a positive return. While their principal is not in major risk here as Stella has pre-determined risk parameters and a liquidation model, their returns are contingent on the success of leveraged strategies. Lenders, by virtue of their role, take on risk from the moment they give their capital to a borrower though. So this model seems to naturally skew towards borrower incentives, as they are effectively given 0% loans until they start generating returns.

Assuming a situation where the leveraged strategy generates 10% APY, lenders receive roughly 3.67% per annum on their capital — which is not exactly the most lucrative passive investment strategy, but one could argue is justified for the level of risk. Additionally, its worth noting that when a borrowers position is liquidated, lenders receive a liquidation fee on top of their principal.

In conclusion, I think Stella offers a pretty good deal for borrowers who want to lever up into a strategy. And while it would be disingenuous to say lenders aren’t compensated, I would argue they are perhaps not being compensated enough for the risk they take in certain situations. One thing I won’t deny is that this seems to have more stability in terms of cost of leverage, which is a welcome advancement. And while there is no utilization-driven incentive to repay loans (like borrow rates skyrocketing), there is a 30-day maturity to each loan after which it has be repaid or is forced to repay.

If you’d like to dig further into Stella, you can find the full documentation here.