In his talk, Joel Monegro from Placeholder took us on a round trip from macro to micro then back to macroeconomic framework analyzing market equilibrium, to arrive at a perspective of understanding value from economic costs.
Joel addressed the confusion between economic costs and accounting costs, and the distinction between value capture and investment return. He applied value capture philosophies onto protocol vs. application in the context of DeFi stack, ended with insights for future work:
- Distribute cost to distribute value;
- Protocol increases total value by flattening out the cost curve;
- Look for new cost centers for business model at application layer.
Direct link to slides here. WTFdao has edited the abridged transcript below.
Today I will talk about value. I may share some of my frustrations with regards to how sometimes we think or talk about value. What I want to do today is share a little bit of a broad macro framework for how to think about economic value.
This will be related to the question of value accrual and crypto more broadly, and then we’ll try to massage it and talk about DeFi a little bit, but it applies all the same. Then we’re going to talk about the distinction between economic value and investment returns, which is something that can draw a lot of confusion in different contexts. And then finally, we’re going to talk a little bit about different value capture philosophies between crypto applications and protocols, with some applications to DeFi. Again, they’re a little universal.
DeFi is actually pretty interesting to look at as a sector to test ideas, because the stakes are higher. And so for example, we’re sitting down designing governance systems for DeFi, if it works for DeFi, where the stakes are high, then we can transpire a lot of things to other areas and so on.
1. How Costs Determine Value
We’re going to start with how costs determine value. And I’m going to rely on something that they teach at econ 101, which is the concept of economic equilibrium. Now, we only have about 10–15 minutes here. So we’re not going to dive deep into a lot of these concepts. But if you’ve studied econ before, then all of this is going to make sense, this is going to be familiar if you haven’t. What I want to leave you with is some basic heuristics or concepts and then you can go and study independently.
- Market equilibrium
This is what equilibrium looks like in most econ textbooks, basically the x-axis is quantity level, the y-axis is the price level. This may already be triggering things around, for example, MV=PQ and other economic models, that have been applied to crypto in different contexts. Basically in the macro sense, economic equilibrium happens when supply matches demand. That’s pretty easy to understand. What we can look at here is what happens at different price levels and different output levels in the economy, whenever supply does not match demand.
The green curve going down is what we call the demand curve, and it can take many different shape or form depends on the model. The red curve going up is the supply curve. The point of equilibrium is the price level and the output level at which supply matches demand. The reason why that is the optimal point, why that is called equilibrium is that at a lower price level you’re under-producing, at a lower price level and the equilibrium level, there is more demand than there is supply, so you’re leaving a lot of value on the table, you’re not maximizing value.
When the price level in the market is higher than the price of equilibrium, then you have a situation where supply is greater than demand. Then you have another kind of market failure. And so, whenever prices find themselves either above or below equilibrium, then you have these unhealthy dynamics in the market, which then drive the price levels or the output levels back to equilibrium. So in the case of where supply is greater than demand, you can see that across the output axis where if you have at P₁, you have the demand curve showing basically people are demanding Q₀. But if you follow the supply curve, the supply side is producing Q₁ and so you have a lot more supply than demand, the economy falls out of equilibrium.
- MB=MC at equilibrium
So we can go down a deeper level, and take the same framework and talk about marginal analysis, which is taking the same concepts, the same ideas and apply them to microeconomics. And so here we’re starting to talk about the marginal cost of a thing, or the marginal benefit or value of the thing. In economics, marginal benefit is really the value of something. It’s called benefit, because it’s very broad. And it’s got a bunch of different nuances to it, which we will cover today. But it’s basically the way to think about benefit is the value of a thing, or an activity, or an object, or an economic output as measured by its price. And so we go back to the y-axis with the price level. Once we go down to a deeper level to micro, using the same kind of general economic framework, we find that: at equilibrium marginal benefits equals marginal cost, and this is another kind of fundamental principle of economics.
Some of the same principles apply. Now the nuances are a little different. What happens here is, whenever you have the price level being below the marginal costs of something, then it’s not profitable. So the economy or company can’t function, if the value of the things it outputs is lower than the cost. Similarly, when the price of a thing is too far above the cost to make it, then there’s too much value on the hands of the consumers and not enough value in the hands of the producers.
So, the theory here is that economic equilibrium or value is maximized at the point where marginal benefit equals marginal cost. And we can observe that or kind of study that by looking at two things: consumer surplus and producer surplus. Producer surplus is basically how much value there is for the producer, as represented by the space above the marginal cost. And so whenever the price finds itself above the marginal cost, then the producer has a benefit because it costs them x. And if the price is higher than x, then there’s a surplus there for the producer. Similarly for the consumer, whenever the price level sits below the marginal benefit curve, then that’s a surplus or an extra benefit to the producer that the consumer is getting, because they’re getting value for greater than what they’re paying for, or the value of the good is greater than the cost they’re paying.
But from the economics perspective, what markets try to do is maximizing value. And so what happens is that total value is maximized when marginal benefit equals marginal cost, which is when there is no consumer and no producer surplus. This is when the producer is able to essentially break even on all of their costs, and they’re not operating at an economic loss. And the consumer is basically paying the lowest possible price that they can pay for it going to that point where marginal benefit equals marginal costs. I’m going to be almost done here with the economic theory.
- TV=TC at equilibrium
Now we’re going to go back up a level again, and think about Total Value versus Total Costs. And the way we come in here from thinking about marginal analysis is the understanding that total value is basically the sum of all marginal benefits in economic context in the market. And total costs is the sum of all the marginal costs.
The more valuable insight is getting to this formula here is: at equilibrium, total value equals total cost. And we can observe the same behaviors whenever the price level sits below TV=TC. It’s sub-optimal because there’s more demand for a good or service than there is supply, so you can increase production to meet that demand. And similarly, if you find yourself in a situation where the price level is much higher, or higher in general, you’re in a situation where production isn’t profitable and that’s also an undesirable state of the market. Understanding total value equals total cost gives us an interesting tool to start thinking about value more broadly.
- Understanding costs
Now let’s talk about this again. So one thing is that we consider costs from the point of view of economics, not just a costs from the point of accounting. This is a big confusion, economic costs includes all kinds of costs.
For example, in marginal analysis, something that can confuse people is the fact that in the real world, businesses have margins and have profits. And so the notion that prices are at equilibrium when marginal benefit equals marginal cost can be a little confusing. That’s because if you look at it through an accounting context, in a business context, it doesn’t really make sense. But an economic context, everything can be a kind of cost. And so for example, if you have a business that produces income and reduces excess revenue or excess income that it can distribute back to shareholders in this framework, that is the kind of costs which we would call the cost of capital because it’s the cost that it took for that company to basically go buy the capital that it needs to operate and so on. So that is the one thing that we have to understand about these models as you go out and explore them is the full scope of costs is what drives at these insights.
- Value goes to where the costs are
Second, is the understanding that value tends to go to where the costs are. While we are not able to go into all the nuances about equilibrium, what we can pull out from this basic notion that total value at equilibrium equals total cost, make some educated assumptions about the behavior of value in a macro context, so that can help us understand the question of value accrual more broadly.
Trying to figure out how value works is abstract and difficult in predicting future value is particularly difficult, but observing costs is a very practical and objective thing that we can do to estimate the behavior of future value.
- Applying TV=TC to cryptoeconomics
So some ways in which we can use this insight is, we have this post (Web Vs. Crypto Service Models, Cost Structures And Value Distribution) that we wrote comparing web versus crypto service models, from the point of view of the cost structure. For example, understanding that value and costs tend to be related, value tends to go to where the costs are, and that you can’t have value without some form of costs, allows us to study things like how different service models distribute value in different ways.
In this piece of analysis, we looked at the web service model, which has a concentrated cost structure in the sense of a single company typically takes on all of the cost of production to produce a service. So the value that comes from that activity gets concentrated in the entity that is taking on those costs. We can now see how different distributed cost structures like in crypto that use decentralized open production models where it’s no longer a single company taking on the costs, but every miner, every investor, or even every user who is taking care of their own private keys in some ways, taking on some of the costs of production. Therefore the value is more distributed when the costs are more distributed. So, that is one thing that we can glean from total value equals total cost.
We can also use it to figure out traditional value between protocols and applications. This is probably my biggest frustration with fat protocols: the confusion between value capture from an economic standpoint, and the notion of investment returns, which are two very different things.
2. Economic Value Capture vs. Investment Returns
A lot of people took the observation that the protocol layer captures more value to mean that the protocol layer captures all the returns. Total value capture as a metric is more of an input to Total Addressable Market (TAM) that has nothing to do with investment returns, which has a lot more variables to it. So at a high level, value capture is more of a function of how much value there’s available at a different layer. Returns have a lot more to do with: cost basis, meaning how capital efficient or how costly it is to operate at one layer, the concentration of ownership in a particular investment when you’re looking at it from an investor’s point of view, and also the growth rate.
And the way this can play out in protocols vs. applications is that at the protocol level, you have most of the costs because the cost of production and the data and the functionality is handled at the protocol layer. And therefore, if you follow TV equals TC, most of the value has to go to the protocol layer. But the return profiles between an application and a protocol can be completely different.
If I put $10 million to work on ethereum today and buy a few basis points of that network, and I could get a venture-scale 5x return on it, but in order for ethereum to give me a 10x on $10 million, today it has to add $72 billion in market value in order for that to happen. Whereas if I take a million dollars and invest in a seed stage equity company building an application business and I can buy 20% of that business, I can get higher than the 5x but get the same 50 million return.
The company doesn’t have to add so many $2 billion in value to itself in order to provide me with the same dollar return, it only has to add something like $500 million in value. So understanding the difference between economic value capturing being available and investment returns is key to understanding what’s worth investing in and what’s not worth investing in, or what’s worth working on, what’s not worth working on.
3. Value (DeFi) Protocols vs. Applications
So just to end it with some philosophies around value capture mechanisms between protocols and applications. We don’t have time to go deep into a lot of these things.
Protocols tend to build defensibility through self-sovereignty and liquidity as network effects. These are somewhat under-developed thoughts. But there are two things that we keep coming back to as drivers for defensibility in a network and they’re related to each other.
- Self-sovereignty is basically the degree to which a crypto network is independent by itself. So when you have systems that have some degree of centralization, that makes them less self-sovereign. But self-sovereignty is a multi-dimensional object, we have a post (Sovereign Cryptonetworks) on it that describes the different ways in which the different levels of self-sovereignty that we inspect when we assess the degree of independence of protocol, or a network. But it is a driver of defensibility because it has to do with trust and scale.
- Liquidity has network effects is something that’s kept coming back as we continue to making investments. Sometimes we find that the networks that have more liquidity, not only on the financial market side, but also liquidity in mining, liquidity in infrastructure, liquidity in usage. That is a new form of network effect that we’re observing. It’s interesting for us it’s a very different dimension from the kinds of network effects that we see on other service models like of the Web.
Finally, the application business models remain unclear. We just talked about scenarios in which an application layer business or opportunity may generate investment returns and we may even find ourselves in situations where there are more returns on the application layer rather than the protocol layer, you kind of take into account all the conditions.
But the question still remains how do they end up actually adding value and creating business value for themselves. We bring a lot of interesting innovation in terms of what is the business models that application companies are pursuing. In general, we think that a lot of regular businesses will probably remain taking transaction fees or charging subscriptions, but may present themselves at a much lower cost because the costs are lower in a crypto services architecture, technically speaking. So we may find a lot of the same, but just lower, which is good.
I am particularly interested in the notion that if an application is building on top of a number of protocols, that application should have a stake in the protocols on top of which it builds and that it can derive great economic value for itself by basically adding functionality or adding value to the underlying protocols. This is starting to show up in an interesting way. When we look at the wallet business, for example, we’re now seeing wallets starting to take into account how much ETH should they be having on balance sheet because buying ETH at a point in time kind of gives you an option on gas prices in the future. And so how do you end up managing your internal treasury to basically work with all the underlying protocols that you’re a part of?
I have three more closing notes.
1. Considered the distribution of costs. You want to design the distribution of value to match generally, the distribution of costs. And this is one way in which TV equals TC or marginal analysis or these other models that we look at, can kind of lead us in that way. If we do understand that for there to be equilibrium, our costs and value need to be equated then when we’re designing token distribution models or inflation models and deciding those that value should go, then you want to kind of draw the line along the lines of costs. And so one way in which people can fall into a trap is we have seen systems where the economic model over-advantages capital over labor, in the way that systems are launched and constructed. And so then that can end up having consequences in the system can fall out of equilibrium.
2. Protocols increased total value in the market by essentially flattening out the cost curve. So if we go back to the kind of model charts, what protocols do or open source in general, is they tend to flatten out the cost curve in a way that ends up lowering prices, but creating greater overall output. Because then Q essentially increases as you drop P, P*Q ends up being a larger number when you flatten out the curve. And so that’s a kind of philosophical principle about where value is being created broadly in protocols vs. applications.
3. Finally, define business models at the application layer, look for new cost centers. And so there’s another way in which we can use the same models to figure out what business models to build the application layer. If it is indeed true that it is difficult to capture value without incurring costs, then we can start using that insight to figure out okay, where are the costs at the application layer? And then how do they become value? Or how do they become business models.
The example I like to use for that is Coinbase. So Coinbase is a wallet, but they’ve taken on a lot of costs in building integrations that the traditional financial system and KYC and operating as very expensive infrastructure to make Coinbase run the way that it does. And because they have incurred those costs, then they have both motive, reason and a venue for then building a business on top of that. And so really, if you think about the moat that Coinbase has built around itself, it has to do with having incurred all of those costs. Having made all of those investments, and then that allows them to then capture value on the cost centers. But they are, for example, not really able to charge you a transaction fee on Bitcoin greater than the Bitcoin network, because at that point, it doesn’t really cost them anything. So they can’t really charge you anything for that at equilibrium.
When we’re dealing with applications or DApps that outsource all of the cost of production and functionality to crypto networks below, it can be really hard to figure out where are the cost centers at the application layer that can allow those companies to build businesses.
We’re seeing interesting iterations in DeFi where you have DApps like Zerion that we are invested in, and competitors like InstantDapp and so on. You can fight that feature war and then end up at a place where all of the DApps offer the exact same functionality. But when it comes to each of them, building businesses on top and they’re going to have to figure out how to create additional layers of functionality that have costs that are not covered by the underlying protocol there. That’s it.
Please contact@tzhen on Twitter for corrections.
DeFi.WTF Osaka was the first of a series of WTF events, traces the emergent DeFi stack and explores its implications through in-depth conversations with the space’s main actors.
About WTF — the Power of Asking: a modern resurrection of the Greek Pnyx, to build a supportive space for open, inclusive, informed, and thoughtful exchanges of ideas, for the practice of the ancient art of asking the right questions to move beyond hype and arrive at knowledge.
Stay in Touch:
- Website: www.DeFi.wtf
- Twitter: WTFdao
- Telegram: https://t.me/defiwtf
- Donation (ENS): defiwtf.eth
- Support us via “A DeFi Billboard” Social Experiment
