CTO’s Guide to the Basics of Blockchain Fees

Blockchain and the umbrella of technologies that, together, comprise the category of Distributed Ledger Technology (DLT), continue to be a growing presence on the commercial and technical landscapes. Depending on which source of data you trust, the number of digital currencies powered by the various chains on the internet has grown from the original entrant, Bitcoin, which was introduced in 2009, to tens of thousands of different digital currencies today. As more public-distributed ledgers emerge, many of which can support multiple yet distinctly separate cryptocurrencies (also known as "coins" or "tokens"), it's clear that the blockchain industry's growth is remarkable.

smart contracts, which are essentially custom-developed business applications that run on digital ledgers, are attractive because of how they programmatically unlock the unique capabilities of blockchain

In addition to being a foundational technology for the cryptocurrency-based exchange of value where one party transacts cryptocurrency with one or more counterparties, DLT also powers the emerging market for smart contracts. In fact, while many enterprises may have a limited need for cryptocurrency-based exchange of value in their day-to-day operations, smart contracts, which are essentially custom-developed business applications that run on digital ledgers, are attractive because of how they programmatically unlock the unique capabilities of blockchain and other DLTs. For example, the retail giant Walmart uses DLT-enabled smart contracts to automate supply chain tracking of goods from farms and assembly lines to the retail world, where the end products are eventually sold. Meanwhile, insurance companies such as Insurwave use smart contract technology to capture and execute the business logic for its policy administration processes. These are just two of a nearly infinite number of enterprise DLT use cases.

Clearly, DLTs and the applications they enable are paving a new direction for commercial disruption and innovation at enterprise scale. However, as new and innovative as DLT is, there's still an old-school enterprise requirement that needs to be considered before companies adopt any technology, DLT included: Total cost of ownership (TCO).

As the old saying goes, there is no free lunch. Despite their open nature (anyone can use a public distributed ledger without a pre-existing contract or software agreement), using a public DLT such as the smart contract-capable Ethereum blockchain typically incurs fees. Not only might these fees range from acceptable to exorbitant, but the fees on some chains can also fluctuate unpredictably, making it difficult to estimate long-term costs and set budgets.

Thus, to understand the full implications of adopting DLT into any business, C-level executives, IT personnel, and smart contract developers need to have a solid understanding of DLT fees – what they are, how they're incurred, and how to optimize them in terms of TCO. The following sections provide this basic foundation.

Understanding the Nature of Distributed Ledgers

When it comes to understanding the nature of DLT in general (and DLT fees in particular), there are three things to keep in mind.

The first is that by definition, DLT is trustless. This does not mean that a DLT is an insecure free-for-all that's not to be trusted. Rather, it means that instead of relying upon a central, "trusted" authority such as a bank to verify the accuracy and integrity of transactional behavior, a distributed ledger is typically made up of multiple distributed computers that work together across a peer-to-peer network to algorithmically ensure the security, order, and accuracy of the ledger. This work is conducted in a decentralized, automated manner that's coordinated among the ledger's computers according to conventions and mathematical principles of distributed computing that are specific to each particular DLT network, of which there are many.

From one blockchain to the next, there's a range of opportunities for human involvement in these automated processes. Some chains are purely algorithmic, where designers intentionally eliminate the possibility for human intervention. However, other chains may include provisions for people to influence transaction priority and order

The second thing to remember is that the people or organizations who own the computers (often referred to as nodes) that handle the automated workflows of a distributed ledger are typically compensated for their work. Compensation is usually paid using the cryptocurrency that's central or "native" to the given DLT. This means that if you're a company that's considering using a particular chain, for example, the Binance Smart Chain for running your smart contracts, you must keep a certain amount of that platform's native cryptocurrency on hand to pay the fees that are incurred by the smart contract execution. For example, in the case of the Binance Smart Chain, transactions and smart contract fees are paid with the Binance Smart Chain's native cryptocurrency Binance Coin (BNB).

The third thing to keep in mind is that fees are incurred on a pay-as-you-go basis and that fee payment is made immediately as an integral part of a transaction or smart contract workflow. Analogically, it's akin to using a washer in a laundromat. You pay before the machine starts, not after. Payment is made as the service is rendered.

In contrast to virtually every other technology business model, with a DLT, you pay for what you use at the same moment you use it.

This pay-as-you-go model is substantially different from metered or subscription services such as those respectively offered by cloud providers such as Amazon Web Services (AWS) and Salesforce. Metered services keep track of every bit of usage, but payment for that usage is made at the end of a billing cycle, typically every 30 days. Subscription services are typically offered and paid for in consumption bundles, seats, or tiers (e.g., 10 unlimited user licenses for $500 per month). In contrast to virtually every other technology business model, with a DLT, you pay for what you use at the same moment you use it. This includes any additional compute resources that smart contracts might need in order to support the automated execution of custom business logic (more on that little wrinkle in a bit).

In addition to having to pay fees on a pay-as-you-go basis, organizations will also be faced with additional fees for the cost of converting fiat currency (e.g., US dollars) into the native currencies of the chains they choose to work with (and there's a lot more to choosing a chain than meets the eye). This is usually done with the involvement of cryptocurrency exchanges that often charge fees of their own. Enterprises should take the time to understand the differences when it comes to working with the two primary types of cryptocurrency exchanges: centralized (CEX) and decentralized (DEX).

The long and the short of it is that the democratization of a public distributed ledger's services depends on node operators (individuals and/or organizations) whose contribution of computing, storage, and network resources involves non-trivial expenses. At some point, an organization or consumer that depends on those resources will pay for the privilege of using them and the resulting fees will most often be paid for with the underlying DLT's native cryptocurrency.

In addition to considering these three key points when examining the feasibility of using DLT in their day-to-day business operations, there are a few additional challenges that organizations must be aware of when using a DLT and its associated cryptocurrency.

The first includes what to do with the increased burden on the organization's accounting and bookkeeping practices in order to support the use of a DLT and its cryptocurrency. At the very least, this work requires having personnel who are well-versed in currency management. If your organization doesn't have the expertise, it will need to secure such services either by direct hire or third-party employment.

The second of these challenges has to do with cryptocurrency volatility. Unlike a fiat currency such as the US dollar, which remains relatively stable over time, the value of most cryptocurrencies is constantly changing. In fact, relative to a given fiat currency, the values of many cryptocurrencies are changing every minute of every hour of every day. Akin to the stock market, the value of most cryptocurrencies is driven by buyer demand and seller availability. But unlike the Monday through Friday schedule of most stock markets, the cryptocurrency exchanges where buyers meet sellers are open for business 24/7.

This volatility in the value of many cryptocurrencies can make it difficult to predict the cost of a transaction five minutes from now, much less one year from now. For example, suppose a fictitious public distributed ledger called BillyChain has a native cryptocurrency called Billycoin, and the fee for transferring value from one BillyChain account to another is fixed at one Billycoin. If one Billycoin is worth US$1 today, the cost of that transaction today in US dollars translates to US$1. However, one year from now, if one Billycoin is worth US$2, the cost of conducting the same type of transaction one year from now will be US$2. Depending on the use cases and frequency of transactions, just this volatility in cryptocurrency valuation alone can complicate budgeting exercises for businesses that want to estimate the annual TCO for their DLT implementations.

The last of these challenges requires additional attention to issues regarding regulatory compliance, particularly if a company is doing business with various sovereign states. Not only are there issues involving taxation, but there are also concerns about money laundering and the anonymity of scammers. Organizations must pay close attention to Know Your Customer (KYC) and Anti-Money Laundering (AML) regulations as well as jurisdiction-specific laws that prohibit the use of certain cryptocurrencies. The laws and regulations vary from country to country. And, given that cryptocurrency is still somewhat new on the financial landscape, many countries and even US states are changing their laws continuously as the proliferation of cryptocurrencies and the crimes committed with them increase commensurately.

For many organizations, the fear of novel financial and operational complications associated with public distributed ledgers may present barriers to DLT adoption for important stakeholders such as CEOs, CFOs, CIOs, IT executives, and even the general counsel's office. Some of these complications can include:

• The need to keep volatile assets (cryptocurrencies) on hand (and the lack of regulatory clarity for doing so).
• Difficulty in assessing long-term TCO.
• The lack of uniformity in fee calculation from one chain to the next.

DLT is similar to other revolutionary technologies such as web, mobile, cloud, machine learning, and artificial intelligence (AI) that initially presented perceived barriers to adoption while offering the promise of significant business disruption, reinvention, and innovation.

Although it’s a subject for a future article on Blockchain Journal, keep in mind that DLT is similar to other revolutionary technologies such as web, mobile, cloud, machine learning, and artificial intelligence (AI) that initially presented perceived barriers to adoption while offering the promise of significant business disruption, reinvention, and innovation. But, in the same way that organizations managed through those barriers to unlock the game-changing business potential of those now-mainstream technologies, successful approaches, and strategies for dealing with the challenges of working with blockchain already exist.

Still, if there's one area of DLT to delve into to solidify the foundation for everything discussed so far, it would be how distributed ledgers charge fees to their users. The differences and implications can be significant.

The Three Categories of Transaction Fees

As mentioned above, practically every DLT implementation, whether it's strictly a cryptocurrency platform like Bitcoin or one of the many smart contract-enabled chains such as Algorand, Avalanche, Cardano, Hedera, or Solana – charges fees based on usage. Also, in truly coin-operated terms, fees are typically levied at the same time as the services are provided according to three categories:

• Cost of computing
• Cost of incentives
• Ancillary costs

Let's take a look at the details of each of these categories and then examine how the fees are determined.

Cost of Computing in General

Every transaction that takes place on a distributed ledger consumes electricity for CPU utilization, memory usage, network I/O, and disk storage. For some ledgers, that power consumption can be infinitesimal. For other ledgers, it can be astronomical. It depends on the intensity and length of computational activity, two variables that can differ wildly from one chain to the next. But, regardless of the degree of computational labor involved, the electricity used is paid for by those who own the computers doing the work on behalf of the network. To make ends meet, the owners of the hardware performing the computational labor typically look to be profitably compensated for doing that work. The resulting fee typically covers the computational cost of the transaction and, depending on the network, some or all of that cost is passed back to the party that initiated the transaction (known in Blockchain Journal terms as the originating account).

For example, imagine a scenario in which Bob sends Alice 1 Ether (Ether a.k.a ETH is the native cryptocurrency of the Ethereum network). Bob will be charged a fee by the Ethereum network for executing the 1 ETH transfer of value to Alice. The total amount of the transaction will consist of the 1 ETH that Bob sends to Alice as well as a fee for processing the transaction.

Usually, this transaction fee is but a fraction of the amount being transferred between accounts. For example, if moving 1 ETH between two accounts incurs a fee of 0.00002394 ETH, 1.00002394 will be deducted from Bob's account. But only 1 ETH will be paid to Alice.

If Bob has sufficient funds to cover the new transaction total (the 1 ETH transfer plus the 0.00002394 ETH transaction fee), the transaction will continue through Ethereum's validation algorithm. By the time the transaction is finalized, the new total will have been deducted from Bob's Ethereum account.

The 1 ETH intended for Alice will be disbursed to Alice's account and the remainder is used to cover the fee charged for doing the computational work on the Ethereum network. In terms of Ethereum, the transaction fee is split between a base fee and a tip that's paid to the operator of the Ethereum node that added that transaction to the Ethereum blockchain.

Once the transaction reaches finality, the base fee is burned and that amount of value is removed from the network's total circulation of ETH. The tip is credited to the Ethereum account belonging to the operator of the Ethereum node that serviced the transaction. The division of the total fee into a base fee and a tip is somewhat special to Ethereum. Other distributed ledgers have their own formulas. The details about Ethereum transaction fees are covered in the article Understanding the Details of Ethereum Transaction Fees.

Could They Make It Any More Complicated?

Beyond the issue of how fees are potentially divided, there's even more to transaction fees than meets the eye. This is particularly so for enterprises that must comply with complex accounting laws and regulations. In many cases, when it comes to legal compliance, a given fee might need to be recorded specifically to satisfy a particular government regulation. This can make even a simple set of cryptocurrency transactions convoluted and complex. Let's look at an example.

Imagine that your company purchases 5 units of the aforementioned fictitious cryptocurrency Billycoin. Let's say the total value of that Billycoin at the time of purchase is US$100, where 1 Billycoin = US$20. Also, the transaction fee charged to make the purchase is 0.10 Billycoin, which is deducted from the total purchase. Simple, right? Not really.

In order to accurately record the purchase of Billycoin, your company also needs to record the relative value in US dollars as determined at the time of purchase. Table 1 below describes the scenario according to a purchase date of November 1, 2022 at 11 AM EDT.

NOTE: This fictitious scenario is for demonstration purposes only. Billycoin is not a real cryptocurrency and the 1 Billycoin transaction fee charged to make the purchase of the cryptocurrency is also imaginary.

Now, here's where things start to get convoluted in terms of bookkeeping.

By purchasing the 5 Billycoin, your company's cryptocurrency assets will have increased by 4.9 Billycoin (the total purchase minus the transaction fee) or a relative value of US$98. However, your company paid 0.10 Billycoin (US$2) as the transaction fee to execute the purchase. This 0.10 Billycoin (US$2) expense is a cost of doing business. Hence, depending on local laws, it may not be taxable. Neither should the net amount 4.9 Billycoin (US$98) you purchased because it is neither income nor a gain. Rather, it's a simple currency swap. Nonetheless, not only did you incur a business expense of 0.10 Billycoin (US$2) to do the purchase, but you also have some increased overhead costs due to additional bookkeeping efforts. If your company was only doing business in US dollars, you would only need to keep books in terms of that one currency. Now that you've purchased Billycoin, you have to track both Billycoin and US dollars, not to mention the impact of both on organizational taxes. And just because there are existing laws in place for working with multiple currencies does not mean those same laws automatically apply when a cryptocurrency is involved. Depending on the jurisdiction, those laws may not have been written yet.

But Wait, There's More

Now imagine that on December 31, 2022, the conversion rate increases so that 1 Billycoin = US$22. The value of the 4.9 Billycoin you have on hand is now worth $107.80 (4.9 Billycoin x US$22). If your company decides to reap the profit and sell its 4.9 Billycoin, you will have incurred a financial gain. Whether that gain is considered a "capital gain" or some other type of gain is a matter of local laws (many of which have yet to be written). Of course, you'll first need to pay another 0.10 Billycoin transaction fee to execute the sale. Table 2 below describes the scenario:

Depending on the current laws and regulations in your jurisdiction, your company may have to pay tax on that gain. Perhaps the taxable gain will be calculated on the gross profit of the sale, which is $5.60. Or maybe it will be calculated according to net profit, which is the gross minus the two transaction fees incurred to purchase and sell the cryptocurrency. The calculation as well as the rate of the actual tax on the gain will be made according to current, new, and potentially retroactive laws in the jurisdiction where the gain was made.

However, depending on the current state of regulations or who you ask, there can even be a question as to whether any cryptocurrency being held in this manner is really a taxable asset. What if the cryptocurrency is purchased as "fuel" to power a company's operational use of a particular DLT as described in this article? For example, buying ETH so that a company can run its smart contracts on Ethereum. In such a case, is the purchase considered a swap of one asset – USD – for another (ETH)? Is the purchase a tax-deductible business expense similar to that of gasoline for a fleet of trucks? What about the fees themselves? Are they treated differently? And, if the ETH is purchased to pay for smart contract activity, what if the value of that ETH appreciates in terms of USD? Is the increase in value taxable and if so, when? After usage for fees (how is this any different from gasoline that may have appreciated in value while it sits in the tanks of that fleet of trucks)? Only after reconversion back to fiat currency? These are significant questions that could have a material impact on an organization's bottom line.

Given that most governments are still grappling with how to handle taxation and other legal implications of cryptocurrency and DLT, today's answers to the questions posed above could change tomorrow. Judging by the headlines coming out of Washington, DC alone (never mind the European Union and other international jurisdictions), the situation is very fluid. Still, no matter what, organizations that intend to use DLT in their day-to-day business operations will need to gingerly navigate an ever-changing legal and regulatory landscape.

As if this isn't already enough to keep track of, another important issue is how the computation fees themselves could fluctuate according to the way each chain determines its computational costs. For example, as explained in Understanding the Details of Bitcoin Transaction Fees, the way Bitcoin determines its cost of computing differs from that of Ethereum (see Understanding the Details of Ethereum Transaction Fees). Thus, understanding each chain's method for fee calculation, particularly on chains where those fees can be dynamic, is important when comparing TCO across chains.

But Wait, There's Even More: Smart Contracts Have Their Own Fees

Depending on the chain, another factor that's relevant to determining the costs of transactions might have to do with smart contracts and the way their programming code is structured independently of any transactional activity. Unlike the single computational fee charged for a simple transaction (e.g., transferring funds between an originating account and a target account), a smart contract charges computational fees that are frequently based on how the code of the smart contract itself was written by the organization's software developers. How those fees are computed depends on the programming language used and the DLT upon which the smart contract runs. But, in general, Blockchain Journal has seen two patterns emerge in its multi-chain research; the way a smart contract uses data types and the structure of its functions.

The impact of data types and function structures are discussed in detail in another Blockchain Journal Rabbithole (see Why To Optimize Smart Contract Code For Blockchain Fee Reduction). But in a nutshell, the total cost of running a smart contract is often related not only to the number of times the smart contract must execute but also to the burden of computing resources required to work with its variables and functions.

For executives who need to understand the complexities of blockchain fees and costs, the high-level takeaway is to make sure the organization’s developers and architects have the necessary experience and know-how to ensure that smart contracts are coded for cost optimization.

For example, with many chains, it takes fewer compute resources when multiple variables in a smart contract that are of the same exact type are declared in a contiguous group instead of randomly ordering declarations of different types. Similarly, it's possible for smart contract developers to optimize function declaration in the given smart contract in order to maximize the utilization of computing resources while minimizing cost. For executives who need to understand the complexities of blockchain fees and costs, the high-level takeaway is to make sure the organization's developers and architects have the necessary experience and know-how to ensure that smart contracts are coded for cost optimization.

More importantly, just because the business logic behind a smart contract might result in the equivalent of a single ordinary DLT transaction (and many do!), it's possible that it required hundreds of lines of code to reach that conclusion. In such a case, the cost of running the program will very likely exceed the cost of executing a single transaction. This is why that code needs to be optimized to incur the lowest fees possible. Fortunately, smart contract programmers have been publishing best practices for minimizing the financial cost of running a smart contract for a while. These quality standards are available. They just have to be followed.

Just When You Thought "No More Costs," It's Like Your Cable Bill

There is still yet another factor to consider about transaction fees that goes beyond calculating the cost of computing and optimizing smart contracts. This cost has to do with the second item listed above: incentives.

Incentive fees correlate directly to network congestion, which is a problem found on some DLTs. Depending on which public ledger you're using for transactions or smart contracts, the more congested a DLT network is in terms of transactions that are waiting to be processed, the higher the fees might be to process each of them on a timely basis.

In a perfect world, transaction processing on a DLT network would be done on a first-come, first-served basis, in which all transactions are processed in the exact order they are received and at lightning-fast speeds. However, with some ledgers, the real world is anything but perfect. First-in, first-out queues can be inefficient. Few people want to wait in line if they don't have to. That's why some air travelers will pay an additional fee to board an airplane before others. Airlines figured out a while back that more value can be extracted from customers who are willing to pay extra to advance to the head of the line.

For some ledgers, the same concept of maximum extractable value (MEV) can be true with transactions.

The way that many, but not all, public ledgers work is that incoming transactions are organized into a common area; a waiting room (known as the "unordered transaction pool" or "mempool") from which the DLT's network of nodes selects pending transactions for processing and finalization. However, in many (again, but not all) cases on chains where such a pool exists, transactions are not randomly selected from the pool. Rather, some transactions are prioritized according to prioritization incentives that are optionally offered with any transaction. The higher the incentive (the more value that a node operator can extract from the transaction), the higher the probability that a transaction will be processed ahead of others in the pool. It's a market-driven paradigm that's subject to the law of supply and demand on a momentary basis. Again, depending on the network, this means that total transaction fees could fluctuate depending on a variety of factors, including the originating account's prioritization needs as well as the degree to which the ledger is congested. Naturally, with chains that allow for congestion-related incentives and fees, the transaction costs tend to go down when the ledger is less congested.

Some organizations might reject the idea that transactions can be arbitrarily prioritized

To the extent that DLTs have provisions for such charges, this means that enterprises need to understand how those fees on a given DLT work in order to make sure transactional activity is prioritized according to the needs of the particular use case. Some organizations might reject the idea that transactions can be arbitrarily prioritized in this way and might therefore choose to work with chains that have no provisions for discretionary or incentivized transaction reprioritization. But, when working with chains that allow for prioritization incentives, it's important to understand which transactions need to be processed very quickly and which ones can wait. Differentiating between the two and having policies and procedures in place for optimizing transaction expenses is the way to go. Operationally, this means that businesses need to be vigilant in terms of monitoring their transactional activities and the incentive fees they're willing to incur.

Ancillary Costs

The third type of fee that goes with using a DLT platform are the ancillary costs. Just as currency exchanges in airports charge fees to exchange US dollars into Euros, there are also costs associated with the conversion of different fiat and cryptocurrencies into one another. As mentioned above, when a company decides to use a particular DLT, in most cases it will need to hold some amount of the native cryptocurrency of that DLT. Either the company will have to buy the cryptocurrency using fiat money such as US dollars or Euros, or the company can do an exchange using another cryptocurrency it already holds; for example, exchanging Bitcoin (BTC) for Ethereum's ETH. Especially for enterprises that prefer to work with centralized cryptocurrency exchanges, these conversions will take place on a platform such as Binance and that platform will invariably charge a fee to do the work just as the currency exchange kiosk at the airport charges a fee to convert USD to Euros.

The cost of ancillary fees is far from a trivial matter. A retail speculator who is buying Bitcoin as a long-term investment using US dollars might incur nothing more than a one-time, maybe a two-time charge, paying the fee to acquire the BTC with US dollars and then paying another fee to exchange some or all of that BTC back to USD later on. However, if your organization intends to use a variety of cryptocurrencies to support a multi-chain strategy (e.g., to conduct international trade) and you intend to power that strategy by way of smart contracts, those ancillary fees can add up quickly. The same is true for organizations that are looking to optimize their cryptocurrency holdings to support ongoing chain usage without significant financial exposure to cryptocurrency volatility. Hence, an understanding of the true cost of those ancillary fees and being cautious about how they're incurred is not just good to have. It's essential.

Moving Forward

The goal of this article is to provide a fundamental understanding of how DLT works in the context of how fees are generally incurred among DLTs. Using examples to demonstrate the varying degrees of fee complexity, this article covered the basics of computational fees, incentive fees, and ancillary fees; however, it cannot begin to comprehensively enumerate the exact fee structure for every ledger in existence. Also, the article examined how DLTs relate to both cryptocurrency in general and smart contracts in particular.

The content of this article is high-level. The intended reader is a business person or IT executive who wants to understand the mechanics and more importantly, the costs that can go along with using a DLT to support a business's strategies as well as its day-to-day operations. However, as deep as this Blockchain Journal rabbit hole has gone when it comes to describing the various fees associated with DLT usage, organizations may still encounter other sources of fees; for example, when engaging in more complex transaction workflows that interoperate across the boundaries of two or more public distributed ledgers. It is therefore beholden on any organization to fully understand the fee structures and formulas of the ledger(s) they are considering to power their next blockchain application.

Therefore, this high-level understanding of blockchain fees is a good place to start. Yet, there's more when it comes to comparing and selecting a public ledger that's worthy of your business requirements. DLT is complex and full of details. In articles to come on Blockchain Journal, we'll be examining those details from an enterprise perspective and in a way that brings the general concepts presented above to life. As always, Blockchain Journal welcomes your feedback. You can email the editorial team at [email protected]. 

Editorial Disclosure: This article mentions the Hedera public distributed ledger. At the time this article was published, Swirlds Labs, Inc. sponsored the operation of Blockchain Journal and is also part of a multi-party coalition that contributes to the success of the Hedera ecosystem. For more information about Blockchain Journal's policies regarding editorial independence and neutrality, please read our welcome to Blockchain Journal.