ThePendleEffect

The "Pendle Effect" refers to the demand surge a token experiences when Pendle Finance launches PT/YT markets for it. By tokenizing future yield, Pendle enables fixed-rate strategies, leveraged yield farming, and points amplification that attract capital and new holders. Critically, Pendle creates two-sided demand: YT buyers seeking leveraged points exposure drive up YT prices, which pushes PT to trade at a deeper discount — attracting fixed-income seekers who buy PT for guaranteed yield at maturity.

This report explores the Pendle Effect using on-chain data from Dune across five stablecoins and yield-bearing tokens. We measure holder counts, total supply, the share of each asset's supply held within Pendle's SY (Standardized Yield) contracts, and PT deposit activity on major lending markets (Aave and Morpho).

The data is consistent with the Pendle Effect being real, but reveals that it manifests very differently depending on the asset's starting conditions — specifically its pre-existing scale and holder distribution — and the degree to which PT collateral becomes composable with lending protocols.

USDe (Ethena)

USDe is the strongest case for the Pendle Effect by absolute scale. Pre-Pendle (Jan 22, 2024): 85 holders, $131M supply. The asset reached a peak of 58,271 holders and $16.5B supply (Oct 2025), before supply contracted to the current ~$6.8B while holders continued to grow.

Growth came in multiple waves. The initial surge was driven by Pendle's YT structure enabling ~7x leveraged shard exposure for Ethena's airdrop — without borrowing or liquidation risk. Ethena's Sats Campaign (5–30x multipliers on sUSDe, USDe, ENA) compounded the incentive. A second wave in mid-2025 coincided with PT-sUSDe's acceptance as Aave collateral, creating the recursive looping strategy that sustained demand.

PT lending deposits for USDe alone (excluding sUSDe, which is treated separately) peaked at $4.4B on Sep 15, 2025. Combined with sUSDe's own $2.8B peak a few days later, the USDe complex accounted for ~$7.2B of PT deposits on Aave and Morpho at its height — orders of magnitude larger than any other asset in this study. Deposits have since unwound to $186M for USDe specifically (Mar 2026), a far steeper decline than either holders or supply, consistent with the recursive loop compressing as PT yields fell.

USR (Resolv)

USR tells a different story. For 16 weeks before Pendle integration, USR had ~30 holders with $9–14M in supply — essentially flat. Then it exploded: from 30 holders pre-Pendle to 4,563 on launch day (Dec 23, 2024) and 10,476 one week later. Holders continued to climb, reaching 69,806 at peak in September 2025 — a ~2,300x increase from the pre-Pendle base.

The more interesting finding is what happened after the peak. Despite supply collapsing from its $721M peak (Feb 2025) down to $104M today — an 86% drawdown — the holder base barely moved. Latest count is 69,111, a -1% change from peak. This holds even after the USR exploit on March 22, 2026, which wiped out much of the asset's TVL. Immediately following the hack, Resolv paused minting, normal redemptions, and other protocol functions, meaning users could not freely exit their positions for an extended period after the event. The addresses remained, in part because they could not leave.

PT lending deposits peaked at $55M (Mar 18, 2025), concurrent with the supply peak, but decayed to effectively zero by mid-February 2026 — well before the hack and the subsequent pause. The deposit collapse therefore reflects yields compressing and leveraged capital unwinding through normal channels, not exit forced by the exploit. What's striking is that holder retention persisted through both dynamics: the leverage unwound, the protocol was hacked, redemptions were paused, and the holder count stayed near its all-time high.

This makes USR a more nuanced case than the original "receded as yields compressed" framing: Pendle drove a large initial acquisition, the capital associated with that acquisition was indeed transient, but the addresses were sticky in a way that the dollars were not — though the post-hack pause means we should be cautious about reading the address stability as fully voluntary. Whether the residual holders represent dust, dormant claims awaiting future redemption, or genuine residual engagement is an open question.

USD.ai

USDai, a GPU-collateralized stablecoin for AI infrastructure financing, had just 32 holders before Pendle Boost Vaults launched on Aug 25, 2025. The protocol ran aggressive 25x Allo Points multipliers on Pendle, driving rapid adoption.

Peak holders (6,348 on Nov 21, 2025) and peak supply ($698M) coincided precisely with PT maturity expiry and the Auto Vault airdrop. After this, holders and supply declined steadily: -60% in both by late April 2026. USDai follows a similar pre-hack trajectory to USR, where capital appears to have been primarily incentive-driven.

Notably, PT lending deposits were just ~$6M at the holder peak, then surged to $143M by February 2026 — but by then holders had already declined ~26% from peak. Deposits arrived from a different cohort and too late to reverse the outflow. Latest PT deposits: $7.5M.

USDG (Global Dollar)

With ~44,386 holders and $1.58B pre-Pendle (via Global Dollar Network partners including Robinhood, Kraken, Galaxy), USDG was already well-distributed. Multichain holder growth has been modest — +38% to 61,031 by +6 weeks — broadly consistent with pre-existing organic trajectory.

However, holder counts understate Pendle's role. Within seven weeks of the Feb 23, 2026 launch, Pendle's SY-USDG contract became the single largest holder of USDG on Ethereum. The chart below overlays daily total USDG on Ethereum against daily USDG held in the Pendle SY contract — both on the same dollar axis, making the share visually explicit.

Three observations:

1. Pendle SY grew from $0 to $121M in seven weeks. After a brief ramp from launch to March 10 (~$1.4M), SY balance jumped $1.4M → $24.6M in a single day on March 11, then to $107M by March 19. This is not organic accumulation; it is consistent with a large concentrated deposit event.
2. Total USDG on Ethereum grew concurrently. From a pre-launch baseline of ~$355M, total supply on Ethereum climbed to $474M. About two-thirds of that net growth ($118M of $119M) can be attributed to the SY contract filling up. The rest of the non-SY USDG on Ethereum has been roughly flat.
3. SY peaked at 27.9% of Ethereum supply on April 17, 2026, and currently sits at 25.4% ($121M of $474M). PT-USDG deposits on lending markets reached $28M by +7 weeks.

Whether this concentration represents net new demand or reallocation of existing holdings is an open question the data alone cannot answer. What is clear is that Pendle quickly became one of the most significant venues for USDG on Ethereum, capturing roughly one in every four USDG tokens.

sUSDe (Staked USDe)

sUSDe was already a substantial asset by the time Pendle launched its first PT-sUSDe:

• At Pendle launch (Sep 16, 2024): 7,753 live holders (quarterly snapshot) and $1.16B supply. sUSDe was already a billion-dollar asset at the time of its Pendle integration.
• Peak live holders: 40,010 on April 1, 2025. Peak supply: $5.22B on September 23, 2025.
• Latest (Apr 23, 2026): 35,771 holders, $2.09B supply, $403M in PT deposits.

So the Pendle Effect on sUSDe is approximately 5x on holders and 5x on supply from launch to peak, not the 85x / 1,351x multiples the earlier narrow data suggested. The supply data shows two distinct cycles: a first peak in the low-$3Bs in late 2024, contraction, then a second and larger peak at $5.2B in September 2025. Through both cycles, the holder base continued climbing until Q2 2025 and has held relatively flat since.

What remains remarkable about sUSDe is the share of supply that flowed through Pendle. PT-sUSDe deposits on Aave and Morpho peaked at $2.81B on Sep 19, 2025 — more than 50% of sUSDe's total supply at the time. The Aave recursive loop on PT-sUSDe was the single largest composability dynamic in this dataset. Deposits have since unwound to $403M, a steeper decline than supply.

“

Most of the supply and holder data in this report was pulled directly from Dune's stablecoins dataset, which tracks balances, supply, activity and labels across more than 200 stablecoins on every major chain.Explore it at dune.com/stablecoins

The Pendle Effect varies by starting conditions

For assets with small pre-existing holder bases, Pendle integration coincides with explosive holder growth — three- to four-figure multiples. For already-distributed assets like USDG, holder growth is modest and the effect manifests instead as supply concentration, with Pendle's SY contract becoming the single largest holder venue on-chain within weeks.

sUSDe sits at the intersection. It was already a billion-dollar asset at launch, so its multiplier story is moderate, but its supply concentration through Pendle exceeded that of any other asset in this study.

Pendle SY capture coincides with net new supply, not reallocation

The USDG data speaks directly to a question that's hard to answer with holder counts alone: when Pendle's SY contract fills up, is that capital being freshly issued or redirected from existing wallets?

For USDG on Ethereum, total supply was flat for weeks before Pendle launched, then climbed as SY filled up — and the growth in total supply almost exactly matches the growth in SY balance. Non-SY USDG on Ethereum barely moved. The supply that filled Pendle SY appears to have been issued specifically to flow there, suggesting Pendle is acting as a net demand driver rather than a redistribution venue.

This is one asset and one observation window, so it shouldn't be over-generalized — but it points at a specific mechanism worth watching in future Pendle launches: whether SY capture coincides with concurrent supply expansion on the chain the SY contract lives on.

Lending market composability correlates with supply growth

PT deposit data from Aave and Morpho shows a consistent pattern: lending deposits track supply more closely than holders, and inflate and deflate faster than either. For USDe, USR, and USDG, deposit growth was concurrent with supply and holder growth. USDai is the exception — deposits surged only after holders had already begun declining, suggesting the lending-driven cohort and the incentive-driven cohort were distinct, with little overlap in timing.

Composability with lending protocols and the ability to execute looping strategies appear to be a powerful complement to a Pendle listing — though scale and timing of that composability seem to matter as much as its mere existence. These are correlational observations from a small sample; causality cannot be determined from the data alone.

Holders are stickier than supply or deposits

Across the dataset, holder counts decline more slowly than supply or lending deposits, and in some cases don't decline at all. USDe holders kept growing through a multibillion-dollar supply contraction; sUSDe held ~90% of peak holders through a $5.2B → $2.1B drawdown.

USR is the most extreme case but also the noisiest signal: in the immediate aftermath of the March 22 hack, Resolv paused minting, normal redemptions, and other protocol functions, meaning users could not freely exit their positions. Some portion of USR's holder retention should therefore be read as captive rather than voluntary. USDe and sUSDe are the cleaner tests of the underlying pattern, since both saw their contractions through ordinary market mechanisms.

Subject to that caveat, the pattern suggests Pendle-driven holder acquisition, while often initially incentive-driven, leaves a residual base that persists even after leveraged capital unwinds. The addresses are more permanent than the dollars they once held.

As stablecoin dominance shifts toward a multi-stablecoin landscape, DeFi composability is increasingly the path to distribution. This analysis suggests Pendle plays a specific and significant role in that stack — as a launchpad for emerging assets and a capital aggregation venue for established ones — with lending protocol composability acting as a powerful amplifier. The data supports the view that Pendle functions as both a distribution layer and a capital aggregation venue, even if the sustainability and causality of that demand depend on factors outside Pendle's control.

Data sources

• Holder counts and supply: Dedicated per-asset Dune queries using stablecoins_evm.balances where available, plus per-token deduped multichain queries. Daily granularity for USDai, sUSDe, USDe supply, USR, and USDG; weekly anchors are used when charting weekly-binned assets.
• sUSDe live holders: Quarterly snapshots of addresses with net_balance > 0, TOTAL across all chains (forward-filled onto the daily grid for charting).
• Pendle SY capture (USDG): stablecoins_evm.balances on Ethereum. Total USDG on Ethereum vs. balance held by SY-USDG contract 0xc1799CaB1F201946f7CFaFBaF1BCC089b2F08927, daily.
• PT lending deposits: Combined PT deposit TVL on Aave and Morpho, time series of cumulative net deposits, bucketed by asset family. USDe and sUSDe are now measured separately rather than as a combined "USDe family" bucket.

Pendle launch dates

Defined as the first week PT tokens show transfer activity on-chain:

• USDe (Feb 12, 2024, coinciding with Ethena Season 2)
• sUSDe (Sep 16, 2024)
• USR (Dec 23, 2024)
• USDai (Aug 25, 2025)
• USDG (Feb 23, 2026)

Limitations

The on-chain data in this report is consistent with anecdotal evidence that Pendle integration — especially when in combination with Aave looping strategies — has been a meaningful driver of demand for the assets studied here. That said, this analysis measures correlation, not causation. Stablecoin demand is shaped by many factors beyond Pendle — broader market conditions, exchange listings, issuer incentive campaigns, and lending integrations — and many of these coincided with the Pendle launches in our dataset, making it difficult to isolate Pendle's specific contribution. A few asset-specific caveats are worth flagging:

• USR holder retention partly reflects Resolv pausing minting and normal redemptions after the March 22, 2026 exploit. Some portion of the residual base should be read as captive rather than voluntary, and may include dust or dormant claims.
• The USDG SY-capture finding is based on one asset over a short window. The data shows net new issuance to Ethereum tracking SY balance, but cannot rule out that the same supply was relocated from other chains.
• Holder-count methodology varies by asset. USR and USDG use deduped multichain unique addresses; sUSDe uses quarterly snapshots; USDe sums per-chain holders and double-counts cross-chain addresses. Directional trends hold, but absolute holder counts are not directly comparable across assets.

Appendix: Dune queries

All charts and figures in this report are derived from the following Dune queries.

USDe (Ethena)

• Daily supply across all chains (per-chain rows summed for chart): dune.com/queries/7361963
• Holders + supply at weekly anchors (covers USDe alongside USR, USDG, USDai): dune.com/queries/6854688

USR (Resolv)

• Daily deduped supply and unique holders across Ethereum, Base, BNB: dune.com/queries/7362037

USDai (USD.AI)

• Daily holders and supply: dune.com/queries/7360446
• Holders + supply weekly anchors (combined query): dune.com/queries/6854688

USDG (Global Dollar)

• Daily deduped supply and unique holders across Ethereum, Ink, Solana, xLayer: dune.com/queries/7362193
• Pendle SY capture: total Ethereum supply vs. SY-USDG contract balance: dune.com/queries/7362267

sUSDe (Staked USDe)

• Daily supply (mint − burn across all chains): dune.com/queries/7361208
• Holders + supply from decoded ERC-20 transfers (Ethereum): dune.com/queries/6860946

PT lending deposits (Aave + Morpho): dune.com/queries/7354921