DeFi protocols have too much of a focus on value capture and not enough focus on value creation. As an industry, we have taken the notion of money legos too seriously. Since I came into this space, I was always a believer of the (yet to be proven) fat protocol thesis, but as we evolve, I think I have started to look at it from a bit of a different perspective.

While I love Sushi, I think there is an intrinsic flaw with their protocol that lets the likes of Curve get the upper hand, due to the focus on value capture over value creation. When I look at protocols like Sushi, and xSushi more specifically, it is quite evident what is missing in their model that being their network incentives are in a closed loop specifically within its own ecosystem. What this causes is the opposite of the “money lego” narrative, limiting itself to its own success and more importantly its own failures.

Then there is Curve, which has somehow become the main application of this thesis. Whether this was on purpose or not, does not really matter. The greatest innovations in finance were often done by mistake or at least as an indirect outcome of the Logic of Pragmatism.

Curve’s network ownership is represented in $CRV, while its governance-enabling derivative veCRV addresses the same thing as Sushi’s xSUSHI. veCRV has a second layer of complications tied to it which in turn helped fuel not only this thesis, but also most of my efforts in this odd niche of finance. This complication is the creation of value in governance desirability, unrelated to any value capture at the “application” or “protocol” layer. The notion of governance desirability is of the value that network ownership has created across the entire ecosystem.

First, let's look at what this mindset is based off of - the fat protocol thesis.

“First proposed in 2018, the fat protocol thesis inverses the value accrual model of web 2 into web 3, where in more traditional structures, the “protocols” (TCP/IP, HTTP, SMTP, etc.) produced immeasurable amounts of value, but most of it got captured and re-aggregated on top at the applications layer, largely in the form of data (think Google, Facebook and so on). The Internet stack, in terms of how value is distributed, is composed of “thin” protocols and “fat” applications. As the market developed, we learned that investing in applications produced high returns whereas investing directly in protocol technologies generally produced low returns. This relationship between protocols and applications is reversed in the blockchain application stack. Value concentrates at the shared protocol layer and only a fraction of that value is distributed along at the applications layer. It’s a stack with “fat” protocols and “thin” applications.”

Should liquid-asset supply adhere to an expansionary path that is unresponsive to events in the crypto economy? Or should the money-supply process include a feedback channel? As crypto-economic architects we must choose between open-loop and closed-loop control systems. In the case of a flexible and ever evolving financial layer like DeFi, one has a greater expected value (EV) than the other, as displayed in the case of Curve vs Sushi when you incorporate elements of decentralized decision making and conviction-voting.

What are you doing to yourself when you are revolving incentive mechanisms to your own economy instead of others, or even worse, doing buybacks? Well, you are creating a closed-loop incentive structure, which, if this theory ends up holding any weight, will be the death of many DeFi protocols. You are letting those without the right incentive mindsets dictate an economy that refuses growth and misalign the direction of supply expansion towards fulfilling their own numba go up mentality.

The critical difference between a semi-immutible emissions schedule such as Curve is you eliminate your own self-interested  “feedback” and seek direction directly from the economy the emissions schedule is feeding.

To summarize, Sushi operates within a closed-loop system where the path to market expansion is dictated by elected officials within the team that should in theory have Sushi’s best interest at mind - although that may not be the case in practice.

Curve gets the competitive edge here as its supply expansion is not dictated by those with Curves best interest in mind. This somehow ended up proving to make their model work - instead of focusing on everyone coming through the Curve model for trades and liquidity, the reliance on external actors (eg. Yearn, Convex) to dictate where emissions are pointed allowed Curve to evolve from simply a normal money-lego. This is the baseplate thesis.

Some would consider it very similar to the fat protocol thesis, which in practice is nowhere near perfect, perhaps due to its infancy.

I think when you look at the fat protocol thesis from a value-creation mindset instead of a value-capturing mindset, it has been somewhat proven in practice.

Anyways, beyond price-action, value-flow within DeFi lives in a bubble, and I think in many ways, Curve is slowly proving the validation of the fat protocol thesis, but with some slight deviations from the initial proposal.

What makes one a protocol and not an application?

In the case of DeFi, what makes one a baseplate and not a lego?

I think Curve shares much more similarities with Ethereum than most think. In the case of baseplate, protocol and application distinction is not binary and more of a spectrum, and Curve is closer to the protocol side of the spectrum. Using the open/closed-loop example from earlier, Ethereum is a open-loop system. Extracting the technicalities, Ethereum has no other goal other than to operate a decentralized smart contract protocol and has a monetary policy that third-parties (miners) fight over in order to satisfy their own self interests. In turn, that keeps the economy (the blockchain) afloat. On the other hand, the applications building on top of it (DeFi) are only powering the protocol beneath it by causing network traffic and therefore creating the perfect intersection of supply (ETH) and demand (hashrate).

Lets apply this same logic to Curve, another open-loop system. Curve has no goal other than to operate a decentralized exchange and has a monetary policy that others (DeFi) fight over to satisfy their own self interests. This in turn keeps the Curve economy (liquidity) afloat. The expansion path is an undictated monetary policy which emits $CRV to essentially those who emit the most energy towards it (Yearn/Convex) - similar to that of a miner fighting over Ethereum emissions.

When one can achieve this ability to not only integrate you into their economy, but make their whole application revolve around you the way Curve has, you have graduated from an application to a protocol.

Again, ignoring price appreciation and simply looking at influence, Curve evolved from keeping its network ownership in a closed-loop into an open-loop which indirectly made it the baseplate for legos to build on top of. It created value - it did not fight to capture it.

I think the baseplate thesis is different from the fat protocol thesis on a few other fronts. The main difference being that instead of a focus on value flows making the protocol fat and the application thin, the micro-economies we are building in DeFi require you to focus on governance desirability to make sure that your DApp lives as a protocol, not an application.

In DeFi’s short lifespan, composability has proven to be a spectrum. I think it is fair to say that most protocols react and evolve based on a competitor’s successes rather than focusing on getting the composability edge. A clear example of this was the industry’s move from liquid → illiquid governance in the form of vote-escrow tokens and conviction staking.

In all actuality, if you are trying to build the foundations of our paradigm shift instead of a simple money-lego, your incentive mechanism should be sourced from an open loop system instead of one that is closed loop. Embracing the PvP nature of DeFi, let’s compare our market with some slightly-altered economic competition theories and see how they relate to the baseplate thesis.

Cournot competition in boring TradFi terms is an economic model describing an industry structure in which rival companies offering an identical product compete on the amount of output they produce, independently and at the same time. It is named after its founder, French mathematician Augustin Cournot if anyone cares. The degenerative translation of this is rival DEXes offering a near-identical product competing on the amount of liquidity they capture, independently and at the same time through their emissions.

DeFi fundamentals; DEXes, lending platforms, stablecoins, etc. (in the grand scheme of things) operate in markets with limited competition, where most market-leaders have 3-4 real competitors - oligopolies. They often compete by seeking to steal market share away from each other. One way to do this in DeFi is to alter the monetary policy of your money-supply process through a feedback channel.

According to the law of supply and demand, higher token output (emissions) drives down token prices, while lower output raises them. As a result, applications must consider how much liquid-supply a competitor is likely to churn out to have a better chance of maximizing profits and liquidity retention.

In short, efforts to maximize profit are based on competitors’ decisions and each DAOs output decision is assumed to affect the DApps market value. The idea that one DAO reacts to what it believes a rival will produce forms part of the perfect competition theory.

Unfortunately, this is a story we are all too familiar with, others trying to outdo each other eventually driving themselves and profits to the ground due to mass-dilution. In the case of DeFi, this is something that applications constantly have to deal with. However, in the case of the protocol living beneath it, they only stand to benefit, case in point, the stablecoin market and Curve. The playbook is relatively simple, instead of fighting with other DEXes for liquidity, offer another competitive and niche market (stablecoins) the ability to get the “Cournot edge” on the market by acquiring your token, this is what made Curve the baseplate, not the lego. Looking at DeFi for 2022, my thesis is simply to do what Curve did and create value for others to fight over, not fight to capture it.

Lets pick a random token with no biases, BTRFLY. I think the goal should be to follow in the steps of clear winners and seek to ignore the market battle for liquidity opting for the path of creating value instead.

Like Curve and others trying to follow in their steps, the goal should be to create governance desirability, the competitive edge this token has is the ~$100M or so meta-governance rights the treasury holds across DeFi, something new market participants need if they want to grow their liquidity for them to compete. How do we transform our USP into that of one which appeals to DAOs more than it does to retail?

One way of doing this would be to migrate future emissions and incentives into an illiquid governance token of the users choosing, governance-locked (gl) or revenue-locked (rl) BTRFLY. Reward programs from the scheduled pulses are advertised as extremely high APY as is done in the regular (3,3) narrative. However, there is a catch, the rewards are emitted as illiquid and non-transferable BTRFLY, depending on the users-end goal and incentive alignment, they would choose whether they are minting the gl or rl derivative of BTRFLY. Usage of the network, whether it is on the harberger tax level or the protocols built around the ecosystem (M&As, Hidden Hand, Racket) grants ownership of the DAO rather than simply a liquid incentive. Incentives are better aligned this way), capital inflows are coming from long-term aligned capital and protocol ownership is thus distributed to those same players.

r/gl btrfly holders still share in fees of the platform, so users are ultimately rewarded in liquid reward as the platform grows (entitlement to future revenue). One could even take from the ve(3,3) model and put a valve on emissions correlated to % supply staked. Emulating LooksRare in regards to their rewards program is a poor move, it incentivized wash trading and ultimately does not get any eyes on the project. In this environment, you need to make bold moves to stand out / take market share from moated competitors - adding the second layer of complexity through the two lock token mechanics makes the model more reflexive. In sum, one is left with a new open-loop economic model in DeFi, enabling those to leverage not only the protocols incentive structure (emissions) for their own best interests but also tokenizing its underlying influence in DeFi from its treasury, a baseplate for those to build on top of.

disclaimer: i hold every asset mentioned in this article and this is not financial advice

DeFi protocols have too much of a focus on value capture and not enough focus on value creation. As an industry, we have taken the notion of money legos too seriously. Since I came into this space, I was always a believer of the (yet to be proven) fat protocol thesis, but as we evolve, I think I have started to look at it from a bit of a different perspective.

While I love Sushi, I think there is an intrinsic flaw with their protocol that lets the likes of Curve get the upper hand, due to the focus on value capture over value creation. When I look at protocols like Sushi, and xSushi more specifically, it is quite evident what is missing in their model that being their network incentives are in a closed loop specifically within its own ecosystem. What this causes is the opposite of the “money lego” narrative, limiting itself to its own success and more importantly its own failures.

Then there is Curve, which has somehow become the main application of this thesis. Whether this was on purpose or not, does not really matter. The greatest innovations in finance were often done by mistake or at least as an indirect outcome of the Logic of Pragmatism.

Curve’s network ownership is represented in $CRV, while its governance-enabling derivative veCRV addresses the same thing as Sushi’s xSUSHI. veCRV has a second layer of complications tied to it which in turn helped fuel not only this thesis, but also most of my efforts in this odd niche of finance. This complication is the creation of value in governance desirability, unrelated to any value capture at the “application” or “protocol” layer. The notion of governance desirability is of the value that network ownership has created across the entire ecosystem.

First, let's look at what this mindset is based off of - the fat protocol thesis.

“First proposed in 2018, the fat protocol thesis inverses the value accrual model of web 2 into web 3, where in more traditional structures, the “protocols” (TCP/IP, HTTP, SMTP, etc.) produced immeasurable amounts of value, but most of it got captured and re-aggregated on top at the applications layer, largely in the form of data (think Google, Facebook and so on). The Internet stack, in terms of how value is distributed, is composed of “thin” protocols and “fat” applications. As the market developed, we learned that investing in applications produced high returns whereas investing directly in protocol technologies generally produced low returns. This relationship between protocols and applications is reversed in the blockchain application stack. Value concentrates at the shared protocol layer and only a fraction of that value is distributed along at the applications layer. It’s a stack with “fat” protocols and “thin” applications.”

Should liquid-asset supply adhere to an expansionary path that is unresponsive to events in the crypto economy? Or should the money-supply process include a feedback channel? As crypto-economic architects we must choose between open-loop and closed-loop control systems. In the case of a flexible and ever evolving financial layer like DeFi, one has a greater expected value (EV) than the other, as displayed in the case of Curve vs Sushi when you incorporate elements of decentralized decision making and conviction-voting.

What are you doing to yourself when you are revolving incentive mechanisms to your own economy instead of others, or even worse, doing buybacks? Well, you are creating a closed-loop incentive structure, which, if this theory ends up holding any weight, will be the death of many DeFi protocols. You are letting those without the right incentive mindsets dictate an economy that refuses growth and misalign the direction of supply expansion towards fulfilling their own numba go up mentality.

The critical difference between a semi-immutible emissions schedule such as Curve is you eliminate your own self-interested  “feedback” and seek direction directly from the economy the emissions schedule is feeding.

To summarize, Sushi operates within a closed-loop system where the path to market expansion is dictated by elected officials within the team that should in theory have Sushi’s best interest at mind - although that may not be the case in practice.

Curve gets the competitive edge here as its supply expansion is not dictated by those with Curves best interest in mind. This somehow ended up proving to make their model work - instead of focusing on everyone coming through the Curve model for trades and liquidity, the reliance on external actors (eg. Yearn, Convex) to dictate where emissions are pointed allowed Curve to evolve from simply a normal money-lego. This is the baseplate thesis.

Some would consider it very similar to the fat protocol thesis, which in practice is nowhere near perfect, perhaps due to its infancy.

I think when you look at the fat protocol thesis from a value-creation mindset instead of a value-capturing mindset, it has been somewhat proven in practice.

Anyways, beyond price-action, value-flow within DeFi lives in a bubble, and I think in many ways, Curve is slowly proving the validation of the fat protocol thesis, but with some slight deviations from the initial proposal.

What makes one a protocol and not an application?

In the case of DeFi, what makes one a baseplate and not a lego?

I think Curve shares much more similarities with Ethereum than most think. In the case of baseplate, protocol and application distinction is not binary and more of a spectrum, and Curve is closer to the protocol side of the spectrum. Using the open/closed-loop example from earlier, Ethereum is a open-loop system. Extracting the technicalities, Ethereum has no other goal other than to operate a decentralized smart contract protocol and has a monetary policy that third-parties (miners) fight over in order to satisfy their own self interests. In turn, that keeps the economy (the blockchain) afloat. On the other hand, the applications building on top of it (DeFi) are only powering the protocol beneath it by causing network traffic and therefore creating the perfect intersection of supply (ETH) and demand (hashrate).

Lets apply this same logic to Curve, another open-loop system. Curve has no goal other than to operate a decentralized exchange and has a monetary policy that others (DeFi) fight over to satisfy their own self interests. This in turn keeps the Curve economy (liquidity) afloat. The expansion path is an undictated monetary policy which emits $CRV to essentially those who emit the most energy towards it (Yearn/Convex) - similar to that of a miner fighting over Ethereum emissions.

When one can achieve this ability to not only integrate you into their economy, but make their whole application revolve around you the way Curve has, you have graduated from an application to a protocol.

Again, ignoring price appreciation and simply looking at influence, Curve evolved from keeping its network ownership in a closed-loop into an open-loop which indirectly made it the baseplate for legos to build on top of. It created value - it did not fight to capture it.

I think the baseplate thesis is different from the fat protocol thesis on a few other fronts. The main difference being that instead of a focus on value flows making the protocol fat and the application thin, the micro-economies we are building in DeFi require you to focus on governance desirability to make sure that your DApp lives as a protocol, not an application.

In DeFi’s short lifespan, composability has proven to be a spectrum. I think it is fair to say that most protocols react and evolve based on a competitor’s successes rather than focusing on getting the composability edge. A clear example of this was the industry’s move from liquid → illiquid governance in the form of vote-escrow tokens and conviction staking.

In all actuality, if you are trying to build the foundations of our paradigm shift instead of a simple money-lego, your incentive mechanism should be sourced from an open loop system instead of one that is closed loop. Embracing the PvP nature of DeFi, let’s compare our market with some slightly-altered economic competition theories and see how they relate to the baseplate thesis.

Cournot competition in boring TradFi terms is an economic model describing an industry structure in which rival companies offering an identical product compete on the amount of output they produce, independently and at the same time. It is named after its founder, French mathematician Augustin Cournot if anyone cares. The degenerative translation of this is rival DEXes offering a near-identical product competing on the amount of liquidity they capture, independently and at the same time through their emissions.

DeFi fundamentals; DEXes, lending platforms, stablecoins, etc. (in the grand scheme of things) operate in markets with limited competition, where most market-leaders have 3-4 real competitors - oligopolies. They often compete by seeking to steal market share away from each other. One way to do this in DeFi is to alter the monetary policy of your money-supply process through a feedback channel.

According to the law of supply and demand, higher token output (emissions) drives down token prices, while lower output raises them. As a result, applications must consider how much liquid-supply a competitor is likely to churn out to have a better chance of maximizing profits and liquidity retention.

In short, efforts to maximize profit are based on competitors’ decisions and each DAOs output decision is assumed to affect the DApps market value. The idea that one DAO reacts to what it believes a rival will produce forms part of the perfect competition theory.

Unfortunately, this is a story we are all too familiar with, others trying to outdo each other eventually driving themselves and profits to the ground due to mass-dilution. In the case of DeFi, this is something that applications constantly have to deal with. However, in the case of the protocol living beneath it, they only stand to benefit, case in point, the stablecoin market and Curve. The playbook is relatively simple, instead of fighting with other DEXes for liquidity, offer another competitive and niche market (stablecoins) the ability to get the “Cournot edge” on the market by acquiring your token, this is what made Curve the baseplate, not the lego. Looking at DeFi for 2022, my thesis is simply to do what Curve did and create value for others to fight over, not fight to capture it.

Lets pick a random token with no biases, BTRFLY. I think the goal should be to follow in the steps of clear winners and seek to ignore the market battle for liquidity opting for the path of creating value instead.

Like Curve and others trying to follow in their steps, the goal should be to create governance desirability, the competitive edge this token has is the ~$100M or so meta-governance rights the treasury holds across DeFi, something new market participants need if they want to grow their liquidity for them to compete. How do we transform our USP into that of one which appeals to DAOs more than it does to retail?

One way of doing this would be to migrate future emissions and incentives into an illiquid governance token of the users choosing, governance-locked (gl) or revenue-locked (rl) BTRFLY. Reward programs from the scheduled pulses are advertised as extremely high APY as is done in the regular (3,3) narrative. However, there is a catch, the rewards are emitted as illiquid and non-transferable BTRFLY, depending on the users-end goal and incentive alignment, they would choose whether they are minting the gl or rl derivative of BTRFLY. Usage of the network, whether it is on the harberger tax level or the protocols built around the ecosystem (M&As, Hidden Hand, Racket) grants ownership of the DAO rather than simply a liquid incentive. Incentives are better aligned this way), capital inflows are coming from long-term aligned capital and protocol ownership is thus distributed to those same players.

r/gl btrfly holders still share in fees of the platform, so users are ultimately rewarded in liquid reward as the platform grows (entitlement to future revenue). One could even take from the ve(3,3) model and put a valve on emissions correlated to % supply staked. Emulating LooksRare in regards to their rewards program is a poor move, it incentivized wash trading and ultimately does not get any eyes on the project. In this environment, you need to make bold moves to stand out / take market share from moated competitors - adding the second layer of complexity through the two lock token mechanics makes the model more reflexive. In sum, one is left with a new open-loop economic model in DeFi, enabling those to leverage not only the protocols incentive structure (emissions) for their own best interests but also tokenizing its underlying influence in DeFi from its treasury, a baseplate for those to build on top of.

disclaimer: i hold every asset mentioned in this article and this is not financial advice