So far in 2022 most of the largest stakeable cryptocurrencies don’t compound staking rewards automatically. Most methods I’ve found online for maximizing staking rewards lean heavily on brute force approximations and estimates based on very specific conditions. In truth, an effective strategy is highly dependent on what is staked, how much is staked, how quickly rewards accrue and the cost to compound the rewards earned.
“Compound interest is the most powerful force in the universe,” Albert Einstein.
On the surface, staking is similar to earning interest in an account at a bank where carrying a balance earns a yield relative to the balance carried. The math works like compound interest, but staking rewards are not bank account interest and there are risks to staking that don’t apply to most bank accounts. Binance Academy’s article on staking is a helpful place to start for anyone trying to learn what staking is or how it generally works.
Know Your Crypto, KYC
Maximizing staking rewards requires at least a basic understanding of the coin or token being staked. Cryptocurrency attributes also vary widely so asking good questions upfront will help inform a solid strategy.
Does this crypto compound automatically? If a crypto compounds automatically the most effective strategy could be as simple as letting the principal accrue rewards. Cardano ADA stakers are in luck because Cardano is one top cryptocurrency that compounds staking rewards automatically. Some cryptos like CAKE provide multiple options and can compound either manually or automatically. Knowing a crypto is compounding automatically can also help in other ways, like avoiding unnecessary costs if a wallet interface implies action is required. I made this exact mistake claiming rewards early last year when I was testing staking ADA in Atomic Wallet.
How much does it cost to compound rewards? Compounding more frequently will always earn a higher yield when there’s no cost to compound. Osmosis’s OSMO token distributes rewards daily that don’t compound automatically. The good news is that transactions to claim and stake OSMO are possible at the time of writing for a nominal cost of just 0 OSMO. Any crypto that has no cost to compound will earn more when rewards are compounded each time they are available. When there is a cost to compound, and there usually is, the cost must be considered. For example, the .001 ALGO cost to compound Algorand is low enough that until recently it could have been practical to compound regularly by creating transactions for 0 ALGO. In contrast, gas fees on Ethereum could make staking an ERC20 token prohibitively expensive.
Do I influence the cost to compound? It stands to reason that a strategy can be improved if the cost to compound can be reduced. On supported networks it’s possible to influence the cost very directly by using Web3 wallets like Metamask and Keplr that permit users to specify gas fees. Transactions fail when the specified gas is too low, so attempting to lower gas fees can be challenging and can still result in paying more gas than necessary. Some cryptocurrencies like Cosmos ATOM can be delegated to multiple validators and require gas per validator. When claiming or staking requires gas for each validator the cost can be dramatically lowered by delegating to fewer validators.
Is there a minimum delegation? Not all cryptos have a meaningful minimum delegation, but not knowing a crypto has a minimum delegation can thwart an otherwise good strategy because rewards can end up idle and unable to compound. The minimum delegation for Binance BNB is currently 1 BNB which means at least 1 additional BNB is required to compound rewards. At the current price of around $430 and approximate yield of 8.21%, $65,000 in BNB can only be compounded about once per month. ONE Harmony’s minimum delegation is 1000 ONE. Even though, at the time of writing, ONE is only 17¢ and has a yield of about 7%, it would take an investment of almost $30,000 just to be able to compound monthly. An effective strategy might convert idle staking rewards into something else that can earn a return.
Determining When to Compound
When rewards don’t compound automatically and there is a cost to compound, earnings are optimized by determining the best time to compound. There are two main ways compounding intervals for staking rewards can fall short:
- Compounding too infrequently leaves rewards idle and undeployed rewards aren’t earning a yield. The resulting principal increases more slowly and has lower earning potential.
- Compounding too often incurs unnecessary costs also resulting in lower principal and fewer rewards. This problem can be much worse than the former because there’s no ceiling to the costs.
If the cost to compound is greater than the rewards claimed, attempting to compound results in a net loss.
Compounding too infrequently and too often are natural opposites and both can be avoided by determining where they intersect. The best time to compound is when the return on the compounded rewards would pay the cost to compound in the same length of time. That way rewards don’t sit idle unnecessarily, and the act of compounding literally pays for itself.
All methods for determining compounding intervals are not exact because the true yield of cryptocurrencies changes over time based on factors that can’t easily be anticipated. Actual rewards earned can also be different from the expected yield for many other reasons like validator fees or the number of blocks a validator is selected to add to the blockchain during the reward interval. This approximation is the best balance of simplicity and effectiveness that I’ve found:
principal - I’ll use principal to denote the amount that is actively earning staking rewards.
yield - I’ll use yield to denote the percentage of the principal earned in rewards annually.
cost - I’ll use cost to denote the sum of the amounts which need to be paid to stake or compound staking rewards.
- The principal multiplied by the yield is the amount the principal would earn in one year if staked.
- The cost divided by the amount the principal would earn in one year if staked is the length of time, in years, it would take the principal to earn the cost.
- The staking rewards multiplied by the yield, simplified above as ( principal × yield2 ), is the amount the staking rewards would earn in one year if staked.
- The square root of the cost divided by the amount the staking reward would earn in one year if staked is the length of time, in years, it would take the staking rewards to earn the cost.
This approximation works well for all positive values of principal, yield, and cost. To put it into perspective, a starting principal of 100 units growing 9% annually with a compounding cost of .003 would grow to approximately 1470 units over 30 years. Using this method the resulting value after 30 years would be less than 0.0015% shy of its maximum earning potential. The yield is unlikely to be exactly the expected rate on most networks, and the true yield has a substantial impact on compounding intervals. In this example, if the yield changes from 9% to 8% after the first year the resulting value would already be more than 1% lower than the maximum earning potential in under 5 years.
Any method for determining compounding intervals needs to be recalculated when the principal, yield or cost to compound changes.
Calculating in Realtime
I’ve taken a heuristic approach when planning because I don’t know how to predict the future. In practice, the time elapsed, rewards earned and average yield are known or can be calculated at any point in time. Phrasing the above method as a question, it’s possible to determine whether to compound rewards in realtime using only basic math:
Would the net earned ( rewards - cost )earn at least the compounding cost in the same amount of time? Please feel free to refer to this spreadsheet template that incorporates these methods.
Chief Technology Officer, Treasure
Disclaimer: The views and opinions in this piece are just the author's own, offered to the public at large and not to any one particular investor.