Q1 2021 marked the beginning of the long-awaited ecosystem wars. With crypto markets continuing to attract mainstream attention, newfound retail entrants pushed Ethereum’s fees to all-time highs. The surge in demand was a net positive for the industry but rendered Ethereum unusable for most retail users, prompting them to look for alternative options since the beginning of the year. In conjunction with the deployment of scaling solutions and the growth of the application ecosystems on adjacent Layer-1s like Binance Smart Chain (BSC), this exodus has become an important factor affecting Ethereum’s dominance in the space.

Q2 2021 picked up where the previous quarter left off. The TVL flows over the last three months prove that Ethereum is still the belle of the ball, but it may not be the only party worth attending.

TVL to the Moon
In a relatively short period of time, total value locked (TVL) in smart contracts went parabolic. Since October 2020, a mere seven months ago, assets flowing in different smart contract platforms increased tenfold from $10 billion (almost all locked in Ethereum) to over $100 billion across several scaling solutions and Layer-1s. The growth is nothing short of extraordinary.

While TVL growth is a strong indication of increasing user confidence and activity in DeFi protocols, it needs to be taken with a grain of salt. The metric is computed based on the USD denominated value of the tokens locked in smart contracts, so in many cases, an increase in TVL doesn’t necessarily mean more tokens are being deposited. An appreciation in token price is also a determining factor. Despite this shortcoming, TVL can be a useful statistic to compare the growth and adoption of different smart contract platforms since it provides a simple apples-to-apples comparison.
The Dawn of the “Ethereuem Killer”
Although the growth of the Layer-1 and Layer-2 sectors could ultimately benefit Ethereum, its dominance in the space is beginning to shrink. Of the $110 billion dollars in TVL, Ethereum controls roughly 77%, a drop of more than 20% when compared to its dominance five months ago.

Despite the recent decline, Ethereum remains home to the most vibrant DeFi and NFT ecosystems. Since October of last year, TVL has ballooned from $11 billion to $85 billion (a 631% increase) as the yield-farming craze infected the crypto markets. The movement, catalyzed by Compound, acted as a propellant for the space as crypto users put more and more assets to work in various DeFi applications.

As a result, Ethereum’s DeFi sector has been the dominant destination for user funds. As of July 5th, roughly 47% of TVL is in six major DeFi protocols: Curve, Aave, MarkerDAO, Compound, Yearn, Uniswap, and SushiSwap. The other 53% is distributed across the remaining 169 applications tracked by DeFi Llama.

Back in October 2020, TVL was mostly concentrated in Uniswap and MakerDAO, which accounted for approximately 43% of the value locked in Ethereum. However, the explosion of new financial primitives that began during DeFi Summer 2020 has whittled away at their dominance as investors moved assets searching for the next hot farming opportunity. This trend has continued throughout 2021, with long-tail applications now housing almost half of all TVL.

The Binance Smart Chain Rollercoaster
In the end, Ethereum’s success also became its burden.

The effects of Ethereum’s rising transaction fees started appearing in the first few months of 2021. Ethereum’s dominance began to decline as developers and users began to look for cheaper alternatives outside of the most popular blockchain. The first Ethereum challenger that began to experience meaningful traction as a result of Ethereum’s fee environment was Binance Smart Chain (BSC).

Propelled by a $100 million support fund and backed by the largest exchange in the world, BSC began to quickly expand its DeFi ecosystem, which featured direct copies of successful Ethereum-based applications. BSC’s proximity to Ethereum in terms of building tools made it easy for developers to quickly iterate and deploy applications. On the user side, BSC’s compatibility with MetaMask lowered the barrier to entry for many Ethereum-native users creating a seamless transition between the two chains. These two catalysts in conjunction with extremely high token incentives led to a large influx of new users who sought cheap yield farming.

By the beginning of May 2021, BSC managed to capture $15 billion in total assets or approximately 13% of Ethereuem’s TVL. PancakeSwap (DEX) and Venus (lending platform) are currently the two most successful applications and the ones that captured the highest amount of assets in the early days of 2021.

BSC’s TVL has been quite the rollercoaster ride in Q2 2021. Soon after crossing the $15 billion mark, its TVL more than doubled to $35 billion in the span of ten days due to a wild price upswing for BNB and its spawn (like CAKE and XVS). But the euphoria generated by the new all-time highs quickly vanished as a series of negative catalysts hit the market in mid-May, including a liquidation incident on Venus that lost $200 million and a flash loan attack on Pancake Bunny worth $45 million. The rise in exploits drained user confidence and sent token prices and TVL tumbling.

Although TVL across all smart contract platforms contracted, BSC’s was particularly hurt given most of the value locked in its applications was mercenary capital and consisted of assets that had little use outside of incentivizing users. Unlike Ethereuem’s TVL, which has a healthy dose of stablecoins in the mix, the composition of BSC’s TVL was heavily skewed towards the higher end of the risk spectrum, making it extremely sensitive to market swings.

As mentioned above, in the early days of Binance Smart Chain, TVL was concentrated in two of its most successful applications: PancakeSwap and Venus. In October 2020, PancakeSwap was the leading application accounting for 60-70% of all the money flowing in the platform. The dynamic quickly changed when Venus entered the market in December 2020. The lending platform rapidly captured a significant portion of TVL cementing itself as the second-largest DeFi protocol in the BSC ecosystem.

From December 2020 to April 2021, both protocols controlled around 90% of all the assets in the platform. In mid-April, as newer protocols launched, TVL began to spread out to other protocols like MDEX, Ellipsis Finance, and Wault. However, as of July 5th, PancakeSwap and Venus regained their dominant positions within the BSC ecosystem, controlling over 65% of the funds locked in smart contracts.

Polygon’s Rise to Fame
The second platform that started chipping away at Ethereum’s TVL in April of 2021 was Layer-2 solution Polygon. Like BSC, Polygon took advantage of Ethereum’s high fee environment to bootstrap its application ecosystem by offering lower fees and faster transaction times. Polygon’s biggest advantage to date has been its compatibility with Ethereum, which lowered the learning curve for both users and developers. It is currently the third-largest chain in terms of TVL, handling 5% of all assets locked across all platforms.

Polygon’s TVL went from less than $50 million to over $5 billion in Q2. This explosive growth was the result of two distinct factors. The first was the deployment of DeFi majors Aave, SushiSwap, and Curve. The protocols began exploring the scalability capabilities of Polygon at the beginning of Q2 2021 in an attempt to work around Ethereum’s congestion. Given their popularity and strong communities, it’s no surprise that the migration led to an uptick in user acquisition.

The second reason for the nearly exponential growth was the launch of Polygon’s DeFiforAll Fund. The team committed up to 2% of MATIC’s total supply to continuously support the growth of its DeFi ecosystem over the next two to three years in the form of liquidity mining rewards across different applications. Unsurprisingly, the lion’s share of the rewards is currently going to users of the top DeFi applications in the ecosystem including Aave, Curve, and SushiSwap.

As a result, these three protocols account for nearly 65% of all value locked in Polygon. In particular, Aave’s market share of TVL exploded, going from less than 10% to 70% almost instantly due to its liquidity mining program.

Token-incentivized user growth has become one of the most popular strategies used by protocols to bootstrap a community. This strategy, while efficient in the short term, can sometimes prove detrimental if it is not managed carefully. In many cases, the higher yields end up attracting mercenary capital from users who are looking to make a quick buck but have no vested interest in the long-term success of the project. In this kind of scenario, protocols are able to garner millions of dollars only to see it vanish when token incentives stop.

While it might be too early to tell, this kind of user behavior is manifesting in Polygon’s largest applications. The three DeFi giants that account for more than $3 billion dollars of TVL are starting to see capital migrate back to Ethereum. In the case of Aave, which is the protocol offering the highest amount of liquidity mining incentives, it reached a point where TVL between Etherem and Polygon was split 70% and 30% respectively. However, over the past two weeks, roughly 10% of Aave’s total TVL has made its way back to Ethereum.

SushiSwap followed a similar pattern. TVL in Polygon began to grow at a rapid pace starting in May when the token incentives were announced. Consequently, SushiSwap’s TVL split between Ethereum and Polygon immediately went from 95% and 5% to 75% and 25%. Similar to Aave, the narrative changed over the past two weeks as Ethereum’s portion of TVL started to climb up.

Lastly, among the three protocols, Curve is the one that saw the least amount of funds flowing to its Polygon branch. Roughly 15% of Ethereum’s TVL migrated to Polygon over the course of six weeks right after the liquidity mining rewards were announced. In the same fashion as the previous two, Ethereum’s TVL started increasing in June as users flocked away from the side chain. After the recent relocation, Polygon’s share of Curve’s aggregate TVL shrunk from 15% to 7% in one month.

There are few potential reasons driving this kind of behavior. As mentioned earlier, the first one could be the result of users locking funds for a short period of time to take advantage of the exorbitant yields offered across the entire ecosystem. One piece of evidence supporting this point is the relationship between the number of incentives and the amount of TVL captured by Polygon.

To illustrate, Aave’s liquidity mining program was broken down into Phase 1 and Phase 2. During Phase 1, 1% of MATIC’s total supply (roughly $40 million at the time of writing) was allocated to Aave’s liquidity mining program. About 2 months later, Phase 2 was rolled out, extending the incentives program until January 17th, 2022, and increasing the rewards from $40 million to $85 million (assuming a MATIC price of $1.7 USD). As a result, and without much of a surprise, Aave experienced the most substantial inflow of assets among the three DeFi protocols. On the contrary, Curve’s liquidity mining program was a speck of Aave’s amount. The program, which launched on April 22nd, allocated a mere $6 million to match CRV rewards over a 16-week window. Consequently, the amount of value that moved from Ethereum to Polygon in the case of Curve was relatively small.

Another reason for the retreat could be due to the asset composition of the two chains. Similar to BSC, if the assets locked in Polygon’s applications are riskier compared to the ones locked in Ethereum, then during market drawdowns the dollar value locked in Polygon will shrink relative to Ethereum. While this might be a possibility, it’s worth highlighting that most of the value fleeing to Ethereum through Polygon’s bridge is in USDC.

However, the root cause of the USDC decline might not be entirely driven by fleeing mercenary capital. The drop in USDC supply locked in Polygon’s bridge peaked on May 18th which is the same day as the Iron Finance/Titan debacle. The bank run triggered by investors offloading their TITAN tokens could be the culprit of the great migration of USDC back to Ethereum.

Lastly, a third potential reason for Polygon’s ephemeral TVL could be due to Ethereum’s improving fee environment. The average gas fee per transaction has dropped down to December 2020 levels from levels as high as $30 dollars per transaction in mid-May. The drop can be attributed in part to the increased activity in adjacent smart contract platforms like Polygon and BSC.

Although this could be one of the drivers behind the retreat, it’s hard to believe that users would simply go back to Ethereum due to lower fees. At the moment, Polygon’s liquidity mining programs are still active, meaning users can enjoy low-fee transactions while getting a boatload of rewards. From a utility maximization perspective, low fees in Ethereum shouldn’t be enough incentive to leave “free money” behind unless the market’s overall perception of Polygon has deteriorated in recent weeks.
A Field of Dreams
BSC and Polygon aren’t alone in this new multi-chain world. They’ve captured most of the attention outside of Ethereum so far by mimicking Ethereum’s development environment and end-user tooling, a growth hack that almost negates any switching costs. But the likes of Solana and Cosmos have gained traction despite their radically different designs.
Solana Ecosystem
Solana has optimized for speed over Ethereum compatibility. It uses a different virtual machine and several new approaches for managing transactions and state changes to achieve best-in-class transaction times. Solana’s speed enables it to support a new class of crypto use cases, such as the on-chain matching engine sported by Serum. But shoehorning Ethereum compatibility into Solana’s new environment has proven to be tricky and could be detrimental to performance. In response, the project has forgone the EVM route and relied on hackathons and funds like Alameda to incentivize developers to build on Solana in its current state.

These efforts have led to some early signs of success. Solana’s DeFi sector now has just under $1 billion in TVL, up around 4x since the start of Q2. This total is only a fraction of the assets locked in Ethereum, BSC, or Polygon due to the relatively small size of Solana’s current DeFi ecosystem (DeFi Llama tracks eight projects). But the network has an extensive pipeline of new financial products set to launch on or migrate to it over the coming months.

If Solana can continue to grow its developer base and application pipeline, its uniqueness may become a strength. Projects won’t be able to port their contracts and liquidity to secondary networks easily. With a burgeoning economy, an inherent developer lock-in, and plenty of funds to spare, Solana could become a popular platform for DeFi users to park their assets in the not-too-distant future.
Cosmos Ecosystem
Cosmos’ multichain DeFi ecosystem is gradually coming to life. Terra users locked over $2.2 billion worth of assets in its two applications, Mirror Protocol and Anchor Protocol, by the end of Q2. THORChain reached almost $140 million in liquidity in its first three months despite rolling out in a guarded launch. Kava also quietly accumulated $250 million in TVL (flat growth in Q2) across its lending protocol called HARD Protocol and built-in crypto-collateralized stablecoin system. Even upstart AMM platform Osmosis attracted almost $80 million in TVL within a few weeks of its mainnet launch.

Aside from Osmosis, these chains have operated independently of one another. A few have established bridges or on-ramps to external networks (Terra’s Shuttle Bridge to Ethereum and THORChain’s Bifrost gateways to several Layer-1s), but none have implemented IBC, the Cosmos ecosystem’s defining interoperability feature that allows these otherwise sovereign chains to communicate. Terra, Kava, and THORChain will likely change their IBC status before the end of 2021. But the question remains if the Cosmos Hub will be able to capture any of their asset flows by facilitating cross-chain transactions.

The only real case study for IBC transaction flows is Osmosis. It leveraged various IBC connections (and incentives) to bootstrap liquidity and trading quickly. While its launch led to an immediate spike in IBC activity across the Cosmos ecosystem, the ultimate beneficiary was Osmosis, not the Hub. If Cosmos’ more prolific zones create connections with each other instead of routing through the Hub, the Cosmos Hub may miss out on cementing itself as a central component of its self-titled ecosystem.

Attracting capital flow is precisely why the recently released Gravity DEX will be critical to the Cosmos Hub’s long-term strategy. The exchange gives users a reason to deploy capital to the Hub or use it as a pit stop before moving to the next zone. It could also tie the Hub’s value to the ecosystem’s growth. If Gravity becomes the primary trading venue with the deepest liquidity pools within Cosmos, it could solidify the Hub as the center of the ecosystem’s economy, capturing fee revenue from each passerby.
Closing Thoughts
One year ago, the concept of a multi-chain world was simply an abstract idea. Back then, the data available pointed at a single truth: Ethereum was the sole destination for DeFi applications. In the span of only six months, that truth has drastically changed.

We now live in a world where several blockchains outside of Ethereum have significant traction and a substantial amount of developer activity. While Ethereum may remain the focal point, it’s become undeniable that the cryptoeconomy will not reside in one single chain but will equal the aggregate economic activity taking place across many Layer-1s and scaling solutions.

While attracting capital is a mandatory first step, these newer Layer-1s and scaling solutions must figure out how to retain new users. As indicated by the shift in dominance from BSC to Polygon in May, incentivized liquidity pools attract mercenary capital that will quickly exit when the hot ball of money finds a new home. Retaining capital will require networks to offer differentiated use cases and ancillary services that complement (and don’t cannibalize) existing applications.