Tokenomics in Gaming Economies
What do NFT digital pets, gamers, and crypto degens have in common with Jerome Powell?
Turns out, they’re all dealing with inflation. And more broadly, they’re all governed by principles of economics — to a degree perhaps surprising. In any economy, national or digital, the purpose of economic structure is to provide a stabilizing force for ecosystem balance and growth. An example of this stabilizing force in the real world could be in the form of fiscal policy, involving taxing citizens; in digital economies, this could be a network fee based on underlying user activities that is either sent to the ecosystem treasury for future community use or “burned” (more on this later). On the other hand, monetary policy in the traditional sense involves the Federal Reserve using “QE” or lowering interest rates to stimulate growth — in a digital economy this can be seen in token buybacks and increasing staking rewards or token burn rates.
Now back to that answer of inflation before we get into the gaming side of things. In a net inflationary environment and relatively little to no growth, economic participants lose value over time. We know how this concept works in the US where in 2020 alone the Fed pumped $3 trillion into the US economy through quantitative easing. On a practical level, this served to support the economy short-term in the face of uncertainty and worries regarding COVID-19. However, this action also simultaneously increased M1 money supply and ultimately contributed to increasing inflation as more money in circulation caused the relative value of each dollar to drop. We know that in extreme cases such as in Venezuela, where hyperinflation reached 10 million percent in 2019, inflation can become completely destabilizing to economic participants.
In crypto, though, the primary difference versus a mature national economy is scale; variables that can temporarily mask the realistically long-term unsustainable nature of hyperinflation in crypto are growth of adoption and issuance rates. Many projects utilize immense inflationary rewards to attract new users, all the while depending upon new capital influx to boost token prices. This — what many have termed “ponzinomics” — is reflective of an unsustainable token structure that utilizes substantial token rewards to bootstrap a network and solve the cold start problem but, in the end, is a short-term growth hacking technique.
Wait, what does all this have to do with gaming?
Well, this “growth-hacking technique” was in large-part directly responsible for triggering the explosion in interest in crypto gaming last year that resulted in $5.17 billion in 2021 NFT gaming sales volume. Specifically, Axie Infinity leveraged this technique to go from zero users to an average of 2.5 million monthly users in just over a year. As a result of this explosive growth, Axie was able to provide earnings above the local minimum wage for many of its users including unemployed and blue collar workers in the Philippines.
Despite this phenomenon, going forward the continued promise of the crypto gaming sector will depend on the development of sound and sustainable economic structures and practices. A great example that illustrates this notion of changes towards sustainable structures — particularly as a platform grows in scale — is reflected in the Ethereum merge, which dropped token issuance by 90% and enabled ETH to become deflationary for the first time in October. In this piece I will explore frameworks that can be used to evaluate gaming token economies and also dive into specific projects to reflect on which structures work with various kinds of games. Some of the games I will explore include Axie Infinity, STEPN, Genopets, Skyweaver, NBA Topshot, and The Sandbox among others. As one evaluates gaming tokenomics, it’s important to consider the following three primary topics: 1) LTV/CAC, 2) Emissions rates, and 3) Token Structures (including single token, dual token, and NFT-based economies).
LTV/CAC and Unit Economics
In evaluating traditional or crypto businesses, the LTV/CAC ratio can be an excellent measure of operational sustainability. In crypto, aligned economic incentives enable platforms to forego significant marketing expenses as early users naturally evangelize the platform. Instead of spending on marketing costs to attract new users, platforms can use token incentives to reward users for participation. Therefore, rewards (staking, yield, etc.) should be reflected in CAC. On the other hand, LTV represents the total amount of user spend on the platform over the lifetime of user participation (any NFT purchase plus any in-game spending). If this LTV/CAC ratio is greater than 1x, the token structure is at least approaching sustainable and could serve as preliminary validation that user growth can be attributed to the entertainment and social value that the game provides more than simply rewards alone.
Let’s take the example of STEPN, a “move-to-earn” platform that some liken to a “web3 Strava”. Users must purchase a virtual sneaker NFT in order to participate and then earn crypto rewards (GST) based on their level of exercise and activity as well as by the number of NFT sneakers they hold. As illustrated in depth by Defi Vader, the STEPN 1 year CAC to acquire and retain users is 6.6K GST (worth $12K per user as of June 2022). Therefore, in order to represent a sustainable token structure, the LTV of a user — through spend on marketplace fees, NFT purchases, minting fees (and, at maturity, ad revenue) combined — must exceed 6.6K GST ($12K). Evidently, this was not the case. Users were not spending at the same rate they were getting token rewards and as monthly usage curtailed from 8 million in May to 4.8 million in June, the death spiral between users exiting, token selling, and token price decrease began. By September, monthly users had dropped to 1 million and the token price had dropped from its $7.1 high to a $0.02 low (down 99.7%), as seen below.
Vader Research concludes that many prominent crypto gaming platforms do not provide anywhere near to a 1x (or more) LTV/CAC ratio, estimating an average payback period (for the gamer, not the platform) of 1–4 months…in other words, token inflow growth would essentially need to provide each user with a complete return on upfront investment every 1–4 months. It doesn’t take a financial engineer to see that this is unsustainable after a while. To be clear, these structures aren’t inherently negative but do realistically necessitate large amounts of user and capital inflow for the token value to stay afloat or grow. It’s important to understand the risk inherent with a “ponzinomic” or unsustainable structure — they can work effectively to stimulate growth, but can also act reflexively in the opposite direction in a downturn. As a result, we should all be wary of games that offer extremely high incentive rewards but lack evidence of similarly high user in-game spend to match it.
Emissions Rates and Sinks
Remember, good tokenomics should reflect token value increase in response to network and game usage growth. So how can the protocol design measures to actually accrue value to the token and combat the inflationary minting of new tokens as rewards? It does so with sinks, which act as a deflationary tool toward the token supply; essentially, every time the in-game currency is used in a transaction (in the marketplace or on a skin, item, or upgrade), tokens used are “burned”, reducing circulating token supply and decreasing token sell-side pressure. Through sinks, the more the in-game currency is used, the more tokens are burned, the less currency supply in circulation, and the more value accrues to tokenholders. For example, in DeFi Kingdoms, every in-game transaction involving the JEWEL token results in 1% of the JEWEL fee being permanently burned; in this way the token burning mechanism aligns incentives of JEWEL tokenholders (increasing token price) with DeFi Kingdom usage (via in-game transactions).
In the context of in-game sources and sinks, one of the more important considerations is the ability to adjust on the fly, where the “central bank” game developer (often decided by the DAO) will alter elements of token structure. For example, “move-to-earn” app STEPN announced they would combat its inflation problem with a token “buy back and burn” program, using a portion of its $122 million quarterly earnings to decrease its token supply. Just as there may be advocates and detractors of decisions made by the Federal Reserve, web3 policies are no different. Although this move by STEPN promises to boost price and accrue value to tokenholders in the near term, critics may claim that money going toward buybacks could be better used as a direct reinvested back into game development. Another example of game developers altering tokenomic policies can be seen when in late 2021 Axie Infinity altered breeding costs, a “sink” that involves both the in-game currency token (SLP) and governance token (AXS). Amidst accelerating AXS price increase (up 80% in a day), Axie decreased the AXS portion of breeding costs to maintain a low barrier to entry for users (as seen in the figure below). In evaluating a gaming economy, one must weigh the benefits and drawbacks of game developer involvement, how much they want users to have a say in these policies, and how these policies are implemented into token sinks and emissions.
Token Structures
Finally, besides the relative levels of incentive rewards and deflationary measures, it’s important to understand the specific form of token economic structures that are deployed within games and the implications they have on gameplay, governance, economic activity, and value accrual. Just as a country’s economic structure should reflect both the values of economic participants and makeup of that particular country, token models should likewise be context-driven; while there are a number of different models that can work, projects should align their model with the type of game they offer as well as the user that they are targeting. I will explore each of the three most prominent structures: the single token model, dual token model, and NFT-based economy.
Single Token: The single token model is utilized by a number of prominent platforms such as The Sandbox (SAND) and Decentraland (MANA). In a single token model, all economic activity revolves around one token of (usually) a fixed supply including in-game transactions, sinks, inflationary rewards, governance, and proportional distribution (investor cap table ownership). As exemplified in The Sandbox, the single token model has been successfully deployed for metaverse platforms that offer third-party developers of partnering brands (ex. The Walking Dead, BAYC, Atari) to buy plots of land where they can create games, items, and UGC to engage potential users and customers — all priced in the platform’s native token (SAND).
Compared to its dual token counterpart, the single token model is more vulnerable to token speculation and price volatility directly impacting the game via price of in-game items. Taking the Sandbox example, the price for in-game assets such as fees for playing games, buying equipment, and land sales have all fluctuated significantly along with the hype and mania that led to a SAND high price of $7.4 in November 2021 to the current SAND price of $0.78. Because of this all-encapsulating token, if token price increases substantially, early users are heavily rewarded with in-game influence as, on the flip side, new users over time can inherit a substantially higher barrier to entry as in-game asset prices in some cases become prohibitive, potentially limiting the overall growth of in-game usage and adoption. While this model offers the most all-encompassing form of utility in a token, perceived limitations around token sell-side pressure, lack of flexibility, and regulatory concerns led to the development of the dual token model.
Dual Token: Pioneered by Axie, now a number of prominent games deploy a dual model (Axie, Genopets, STEPN) that divides the economy into two tokens to create varying roles and incentive alignments. The utility token typically serves as the primary form of in-game currency with an uncapped and inflationary supply to adjust for user growth as well as participant rewards (SLP in the case of Axie). Alternatively, the governance token serves as voting rights over protocol decisions and typically has a capped or deflationary supply (AXS in the case of Axie). Outside speculators often hone in on the governance token to gain exposure to platform upside; game users largely do so both for financial upside and voting rights.
Splitting these tokens into two enables the in-game utility token (SLP in Axie) to be less vulnerable to price fluctuations caused by speculation; a “separation of powers” of sorts so that outside investors don’t have an outsized and direct impact on the cost of purchases and items in the game. It also allows for fixed supply and certainty over governance token allocation and at the same time still maintains flexibility for the utility token where the community can have a say over dynamic incentive rewards, pricing, and emissions over time. Finally, governance tokens should have some utility beyond voting rights. For example, in Genopets, while KI is the utility token for most in-game activities like crystal refinement and terraforming, the GENE governance token is also needed in-game to craft valuable items and purchase NFTs. In evaluating a dual token model, one should ensure that spending on valuable in-game items are accounted for in utility token sinks and look for use-cases that tie the governance token to in-game activity instead of voting rights alone.
NFT-based economy: While the projects mentioned above implement NFTs into their game as a part of their economy (LAND and ASSETS for Sandbox and Axies for Axie Infinity) paid for in their native currency, some applications rely on NFTs as their sole economic design function. Particularly prevalent within trading card games, this design is based purely on the supply and demand for in-game items, or NFTs, lessening the need for a governance or utility token. Skyweaver, a blockchain-based trading card game, deploys this model where each card, essentially an NFT, has a varying degree of rarity and use-cases within the game; a card can be “Base” level, “Silver” level, and “Gold” level, and possesses a particular element (air, dark, earth, fire, etc.) that determines certain abilities such as healing powers or abilities to damage the opponent. As a result of these varied levels of rarity and in-game utility, the core economy that drives Skyweaver is reflected in the levels and buying and selling of these different cards through the in-game marketplace. As such, there is no need for a potentially volatile in-game currency token to overcomplicate the game; instead all Skyweaver transactions are made with USDC (a dollar-pegged stablecoin), which serves to minimize friction particularly towards first-time crypto users as purchases are essentially denominated in US Currency terms.
Another great example of this market-based NFT approach is NBA Topshot, a digital sports trading card platform. Slightly different from Skyweaver, the vast majority of Topshot users actually pay directly via their credit card. Created by Dapper Labs, Topshot offers primary drops (which go directly to Dapper Labs) and a secondary marketplace for cards (where Dapper gets a 5% transaction fee). The combination between easy onboarding and major IP (via the NBA) has paved the way for a consumer friendly approach, leading to $1 billion in sales and 21 million transactions since inception in 2020. Ultimately, because there is no native token* that denominates in-game assets in either Skyweaver or Topshot besides that of the US Dollar, the only way for users to “earn” on the platform is in the appreciation of NFT asset value as well as earning other NFTs as rewards for performing certain in-game actions. For most speculators looking to gain exposure to platform upside, there is a higher barrier to entry versus the single or dual token model as NFTs must be purchased outright versus fractional investing allowed by governance tokens. In this way, this model can to some extent diminish the influence of outside speculators on those that actually use the assets within the game. For games in which a trading element is core to their offering and ones that target first time crypto gamers, this strategy can be particularly effective.
Conclusion:
In summary, there are a multitude of factors that should inform the structural and incentive reward designs implemented into digital gaming economies. And we’re just reaching the tip of the iceberg (we haven’t even breached the controversial topic of digital land yet). While the quality of a crypto game itself will determine the absolute economic value to users through entertainment and social value, long-term economic success will necessitate proper implementation of tokenomic reward structures. In analyzing these economies, one should implement the tools that can provide insight into risk levels (LTV/CAC and emissions rate) as well as continue to learn in real-time from the practices and failures of the iterative token models continuously brought forth. At a high level, the following takeaways encapsulate the primary considerations brought forth in this piece:
- Identify the quality of token sinks and measure their counterbalance toward inflationary incentive rewards
- Assess whether user and capital inflow growth and in-game spend meets the standard demanded by growth-hacking inflationary rewards
- Determine whether token structure and design reflects the type of game and user it is targeting
*Note: While the majority of trades on Topshot are in fiat, platform transactions are executed on the FLOW blockchain
