DeFi Yield,
Explained

Section I

Where does DeFi yield actually come from?

Yield does not appear from nothing. Behind every percentage a protocol displays lies a real economic source — or, in some cases, an artificial mechanism that eventually runs dry. Understanding the distinction is the foundational literacy move for any serious DeFi participant.

Source 1: Borrowers paying interest

The most straightforward form of DeFi yield works exactly like traditional banking: someone needs to borrow an asset and pays interest for that privilege. Protocols like Aave and Compound function as decentralized money markets where depositors supply liquidity and borrowers draw on it, paying a rate that is distributed among lenders.

This yield has clear economic logic: there is genuine demand for credit and the market prices it. The rate rises when demand outstrips supply and falls when capital is excess. It is sensitive to market cycles, but its existence depends on no artificial incentive. When borrowing demand is low, yields compress — that is the system working correctly.

Source 2: Traders paying fees to liquidity providers

Automated market makers (AMMs) like Uniswap and Curve have no traditional order book. Instead, users contribute pairs of assets to liquidity pools and receive a fraction of the fees paid by every trader who uses that pool. Yield here is directly proportional to trading volume routed through the pool.

This mechanism creates a direct link between real economic activity and return. A heavily used pool generates substantial fees; a pool with thin volume produces very little. Fee-based yield, where the pair has genuine demand, is arguably the most organically grounded form of return in DeFi.

Source 3: Protocols emitting their own tokens to bootstrap liquidity

This is where analysis becomes critical. Many protocols, especially in early phases, display elevated yields not because they have sufficient organic revenue, but because they distribute their own tokens as an incentive. The user deposits assets and receives, in addition to any real yield, an allocation of the protocol’s native token.

This practice is legitimate as a bootstrapping strategy: it attracts initial liquidity and builds a user base. The problem arises when the APY visible to the user is composed 80% or 90% of these emissions. If the token price falls — as it tends to when emission schedules are aggressive — the real yield evaporates. This is the single most underappreciated risk in DeFi for participants without experience.

Section II

APR vs APY

These two acronyms appear constantly in DeFi and the difference between them is not trivial. Confusing them leads to incorrect comparisons and miscalibrated expectations.

APR (Annual Percentage Rate) is the simple annualised return without accounting for the compounding effect. APY (Annual Percentage Yield) incorporates that effect: it assumes that earned interest is periodically reinvested, and that each period therefore earns interest on a larger base.

In practice, DeFi tends to display APY because the figures are more attractive. Before comparing yields across protocols, verify which metric each one uses and how frequently it compounds. A protocol showing daily-compounded APY and another showing simple APR are not directly comparable — the spread can be meaningful at higher rates.

The question to ask: if a protocol shows you APY, how often does it compound? If there is no clear answer, or no automatic compounding mechanism, the APY is a theoretical number that depends entirely on your operational discipline to materialise.

Section III

The yield landscape — five primary mechanics

DeFi is not monolithic. There are five structurally distinct categories of yield, each with its own economic logic, risk profile, and market behaviour.

Section IV

Impermanent loss, made intuitive

Impermanent loss is the most counterintuitive concept in DeFi. Many users encounter it too late — after withdrawing their funds and discovering they would have done better simply by holding.

Imagine you deposit into an ETH/USDC pool when ETH is worth $2,000. You put in 1 ETH and 2,000 USDC — $4,000 in total, 50/50. Now ETH rises to $3,000. Arbitrageurs automatically rebalance the pool: you now hold less ETH and more USDC than when you entered. If you withdraw at this point, you will have captured less value than if you had simply held 1 ETH and 2,000 USDC separately.

The loss is “impermanent” because it only materialises upon withdrawal. If ETH’s price returns exactly to $2,000, the loss disappears. But markets rarely return to precisely their starting point, and price divergence in volatile pairs can be substantial — the diagram below shows how holding outperforms LP provision as the price ratio drifts.

The practical conclusion: AMMs with volatile pairs (ETH/BTC, ETH/USDC) are only profitable for LPs when accumulated fees exceed the impermanent loss. Curve and other correlated-asset AMMs solve this by operating in pairs where prices do not diverge significantly. When evaluating any AMM yield, the fee yield and the impermanent loss risk must be assessed together.

Section V

Real yield vs token emissions

This is the single most important distinction in DeFi. It determines whether the yield you see on screen is economically real, or a promise that depends on new buyers sustaining the price of an incentive token.

Real yield derives from genuine economic activity: trading fees, loan interest, network validation rewards. If the protocol shut down its incentive programs tomorrow, this yield would continue to exist because someone is paying for a service they are receiving.

Emissions-based yield is different: the protocol creates new tokens and distributes them to depositors. The rate looks high while the token price holds. But the selling pressure from those farming those tokens tends to erode the price, and with it, the real return in base-asset terms.

To distinguish them in practice: look for the protocol’s yield source breakdown. The documentation of serious protocols — Aave, Curve, Lido — publishes exactly how much of their return comes from organic fees. If a protocol does not disclose this information, that itself is a signal worth weighing.

Section VI

Reading risk — three quick heuristics

You do not need to audit smart contracts to form a reasoned first approximation of a protocol’s risk. These three heuristics cover the majority of common risk vectors.

Glossary

Reference terms

Editorial definitions. Technical terms remain in English as used across the industry.