The last two years in the cryptocurrency space have been largely defined by the spectacular rise and fall of the ICO market. During the rise, projects purporting to do things as niche as revolutionize global dentistry by introducing utility tokens were attaining network valuations upwards of $2 billion USD. In retrospect, it seems as though any project with a compelling-ish vision outlined in a whitepaper with minimal spelling errors could have raised huge sums of capital for its founders with few strings attached. Where could that possibly go wrong? After the fall, many of these projects fell into mainstream and even speculative irrelevancy. However, though some of their network valuations fell as much as 95%, many still commanded nine figure valuations and maintain their high CoinMarketCap rankings today.
The tide that lifted all boats receded and brought them all back down, mostly in the same positions relative to one another. If you were an investor still fundamentally bullish on the asset class, you might be tempted to see this as an opportunity to buy back in at a huge discount, patiently wait, and ultimately realize huge returns from your bet on decentralized dentistry. While this hopeful vision will likely come to fruition for a few of the long run winners, it’s a misinformed investment thesis with regard to most of the top cryptoassets, many of which are restrained by their fundamentally flawed tokenomic designs. If a token’s role as an investment vehicle doesn’t properly function in tandem with the practical success of the platform, then its future price is unlikely to appreciate in increasingly developed, rational, “smart money” markets. If a speculative craze with enough force to make distributed ledger root canals 1000x is the minimal enthusiasm requirement for your token to see new all time highs, perhaps your capital is best invested elsewhere.
This brings us to many investors’ primary question: which high profile coins are currently “on sale” and which are fundamentally flawed and, most importantly, what methodology can be used to tell? By using IOTA, a top 15 coin with a current market valuation of over $820 million and the largest DAG (directed acyclic graph) infrastructure platform, as a test case, we will explore factors that make for promising and worrisome tokenomic design and ultimately outline ways it could be improved for both investors and users.
What is Tokenomics?
To answer this question most comprehensively, I’ll point you to a short essay by Unbounded’s own Jack Laskey. For those of you in a rush, in summary: tokenomics is the purposeful alignment of the incentives of network actors via decentralized tokens. The aim of this alignment is ensuring a healthy relationship between network success and token value. The health of this relationship would manifest as the network grows and facilitates the efficient and seamless clearing of supply and demand for its users corresponding with the token proportionally appreciating in value for investors. Later, we will look at this in application to IOTA specifically.
What is IOTA?
IOTA is a DAG platform that hopes to service the potential future machine to machine economy of the internet of things (IoT) by allowing for scalable data flow optimized for feeless micropayment functionality. When juxtaposed with the market dominant cryptoasset Bitcoin, IOTA is hoping to improve upon its currently low transaction throughput, proof of work (PoW) hardware requirements, mining centralization, and fee market in order to better facilitate a distributed economy where machines are able to send payments to one another in a largely autonomous manner. Imagine, for example, a self driving electric car paying a charging station for the electricity exchanged in the time between picking up customers. This is the type of economy in which IOTA is hoping to play a dominant role. However, by approaching distributed ledger technology (DLT) not through blockchain but instead with DAG, IOTA will immediately bring with it a tokenomic design distinct from all other true blockchain projects. In order to fully vet IOTA as an investment, it will be necessary to consider the project’s likelihood for adoption and success as well as deeply examining the implications of its novel token design. By estimating the output of these variables when considered together we can produce a rough sense of expected value and be better situated to assess if IOTA is fit for investment at current prices.
IOTA Platform Viability
Let’s start the meat of this tokenomic valuation with its most basic consideration: platform viability. At its best, a robust tokenomic design can take a viable platform and bootstrap it with the capital necessary to succeed. Once successful, this strong tokenomic design will help the token appreciate in value alongside the platform and financially reward investors. This will then encourage more innovation and development which will benefit users and ultimately, again, investors. This is the virtuous cycle all network participants desire. However, if our starting point of platform viability is determined to be unlikely, then no amount of savvy incentive alignment can save it. So is IOTA viable or is its position on CoinMarketCap simply marketing and residual ICO bull run hype?
What’s essential to understand is that a DAG IS NOT a blockchain. A blockchain is arguably a DAG (in a broad sense) but popular conception of the two are very different in terms of basic design. Let’s break it down: 1) directed, 2) acyclic, and 3) graph.
Directed means that the data is flowing in a particular direction. Imagine a family tree. All new descendants are below their parents and grandparents. The tree moves downward.
Acyclic means that you cannot loop back around in a circular fashion. Imagine again the family tree. Outside of any fun(ny) theoretical hypothetical games, you cannot be your own grandfather: visually represented, you cannot draw a line from you towards your grandfather two blocks above you.
Graph is simply a reference to graph theory which, like a family tree, is a structure used for modeling pairwise relations between objects.
When you consider it broadly, a blockchain can be considered a DAG. If a blockchain were a family tree, it would be an asexually reproducing lineage with exactly one offspring per generation descending down through time. That being said, I reiterate: a DAG IS NOT a blockchain. To understand why not, it’s important to remember the essential benefits of a blockchain. Namely trustless and decentralized consensus which creates the conditions for censorship resistance. Blockchains are extremely inefficient and costly databases but this can be justified when the database is serving something of significant importance that would otherwise be vulnerable to censorship. With Bitcoin, this “thing of significance” is value and money. However, a DAG, like IOTA, that is not also a blockchain is not necessarily well suited to provide these characteristics. Because of its non-linear nature, the IOTA team made the trade off of trustless and decentralized consensus for scalability, speed, and zero transaction fees. With a more traditional blockchain like Bitcoin you either have coins in your wallet or you do not. This is binary. Your faith in the immutability of this binary can be increased with additional confirmations, but barring any malicious activity on the network you either have your coins or you do not. This isn’t the case with IOTA which has no single block to determine binary yes-or-no network consensus and thus relies on a probabilistic likelihood that you have coins. The same wallet may return alternate values depending on when you check it.
Worse still, IOTA’s less than ideal wallet functionality is made even more concerning when you understand the centralized nature of the backend supporting it. What IOTA calls the coordinator (COO) is a centralized node controlled by the IOTA team that is used to validate the network state (remember: probabilistically) with snapshots that determine all token holders’ balances. The source code of the COO and the physical location of the servers are unknown to the public. While the network doesn’t technically require the COO to function, the IOTA team claims its reason for existence is making 33% attacks less likely during the network’s infancy. Presumably, this means that once the DAG grows by onboarding new IoT machines, the COO will be shut off and token holder’s balances won’t be subject to this centralized node’s calculations. While this is indeed IOTA’s stated goal, the exact number of transactions that will act as this threshold isn’t public. Again, the COO code is closed source. Beyond these technical concerns, the messaging from the IOTA team has been less than clear with regard to how essential the COO currently is for network functionality and how susceptible the DAG is to a 33% attack or spam attack.
While these trade offs from a more pure decentralization wouldn’t be acceptable for a digital gold or money, the IOTA team deemed them necessary because of its intended use case: the internet of things (the “IOT” of “IOTA”). The IoT currently exists in relatively trivial form, but the future possibilities of autonomous machine to machine communication and data transfer is potentially great. In 2019, you may have a smart thermostat that automatically communicates with your GPS enabled smartphone which uses machine learning to know to turn the heat on exactly 30 minutes before, and the lights on exactly 5 minutes before, you are due to come home from work on a cold December night. This is already a reality, but as it stands, the transfer of digitally scare value isn’t an essential component. Imagine, however, the aforementioned future self driving car scenario. A robust IoT infrastructure could theoretically allow for things currently impossible like a car and charging station transacting where each are autonomous and own “themselves”. If this were to exist, IOTA’s ability to send micropayments for free and scale theoretically infinitely as new machines join the network would be huge benefits to adoption. The problem is that this economy currently doesn’t exist.
If, for the sake of argument, we imagine that it does exist, we are still posed with huge problems that IOTA would need to overcome in order to become a success. Imagine that we have a vastly interconnected IoT ecosystem all communicating with, and sending value to, one another. To achieve this with IOTA’s DAG design, each machine would need to be able to perform proof of work to validate previous transactions on the network. This is how IOTA theoretically scales. Each new transaction needs to validate two prior transactions. While maybe it would be simple for Elon Musk to add a computer (hardware) with the relevant PoW hashing functionality (software) to a Tesla of the future, this may be more difficult for the smart thermostat and smart light bulbs to accomplish. Lets again assume this hardware limitation is solved over time thanks to Moore’s law. Even still, IOTA would need to become the standard in both hardware and software to create the network effect they are banking on. The further down you dig into this winner-take-all IoT drake equation, the less and less viable IOTA seems as an underlying platform. There are other major concerns with IOTA’s fundamental design, but for the sake of brevity I will leave it at this. For more check out Multicoin Capital’s IOTA valuation or the other sources linked at the end which all served valuable while researching this piece.
Tokenomic Implications on Token Value
To examine the tokenomic implications of IOTA we need to assume that all the moon shots from the previous section pay off. Against the odds, we live in the IoT future and IOTA’s DAG is the industry standard. Now, how does the token function and does it reward investors? If this answer to this question is “no” or “uncertain”, then even the extremely optimistic futurist/technologist should steer clear from investing in 2019.
Let’s start with IOTA token basics: circulating supply, inflation model, inflation mechanism, and initial fundraise structure.
Circulating supply: IOTA has 2,779,530,283,277,761 coins in circulation. In words this is ~2.78 quadrillion. However exchanges list them as MIOTA which are a higher unit of IOTA such that circulating supply is 2,779,530,283 to adjust for unit bias.
Inflation model: since there is no mining on the IOTA DAG, all coins that will ever exist are currently in circulation.
Inflation mechanism: Since there is no inflation there is no block reward as there is with Bitcoin. The benefit of joining the network comes with the cost of having to perform PoW to mine on it and no additional benefit is included to incentivize engaging in network consensus. Because of this all network actors are nodes, users, and miners. Each faces the same incentive structure.
Initial fundraise structure: with “fair mined” coins like Bitcoin, and more recently Grin, there is no fundraise. Capital organically flows into the network by incentivizing potential users with inflation produced block rewards. Because of this, there are no “angel investors” in Bitcoin but instead early adopters who were handsomely rewarded for their foresight. Grin is an interesting example since its initial mine was much more competitive than Bitcoin’s since an industry had already been created around blockchain and cryptocurrency prior to its launch, unlike with bitcoin which preempted the industry. IOTA’s fundraise sold all of its coins to ever exist for roughly $600,000 in BTC at the time of the raise. The implication when network valuations appreciate to upwards of $14 billion USD is that there are likely several huge mega whales that have the ability to flood the market and dry up all liquidity.
From the tokenomic 101 first pass, this is bad on several levels. Most obviously, holding a token that is likely disproportionately held in the hands of a few individuals is highly risky. In immature markets with thin order books, you run the risk of pump and dump schemes and other malicious tactics by whales. Or, perhaps less conspiratorially, you run the risk of a whale becoming fundamentally bearish on their initial investment. If they own a large percentage of the coins how will their trade affect the market? These short term price concerns would make an intelligent investor wary.
Centralization of token ownership is one thing, but let’s revisit the technically centralized nature of IOTA’s back end. When starting a decentralized network there are reasonable concerns about 100% pure decentralized infrastructure from the get go. Sometimes “training wheels” like IOTA’s COO are necessary to allow for the initial development and launch of a network. It is, however, essential for both network stakeholders and token holders that the process of decentralization post “training wheels’ is made known and structured in a way such that the initially centralized party has an incentive to cede this powerful position. Are investors buying a token for a centralized network or a decentralized network? Often times this powerful position manifests as a way to extract rent through playing the role of some network intermediary. In IOTA’s case with the COO, they are not gaining rent per transaction since the DAG is feeless. However, they are potentially patching up an otherwise unreliable network and in effect buttressing the IOTA token price on exchanges, a token of which the IOTA team is no doubt a large holder. If the network were to function in its final state without the COO, would it work reliably? If not does IOTA warrant the valuations it has and will have? Approaching this tokenomically is difficult since the incentive alignment issue isn’t directly a token design concern, but rather a network functionality issue. However, a minimally intrusive token related solution could reassure investors and users that the network will become decentralized when ready and drop its training wheels. Something as simple as locking up a portion of IOTA tokens owned by the founding team until the COO is disabled. Perhaps this oracle that determines this “state of the COO” could be delegated to a vote by IOTA token holders proportional to holdings at various snapshot dates over time along some previously determined, and public, schedule. While this is a somewhat post hoc remedy it would improve the token as an investment for all future investors who are otherwise forced to hope that this network will one day no longer rely on the COO. Considering a preemptive solution built and implemented on the technological layer that involves token mechanics more directly be an even better place to start. For this reason, tokenomic design should be considered alongside the technical design of the platform and not be left as a value capture afterthought.
Further troubling still, the tokenomic logic of the stakeholder incentive alignment is fundamentally unsound. In a blockchain like Bitcoin there are two groups: miners and users. Miners effectively play both roles but users are not incentivized the same as miners. This is helpful because these groups of people have different goals which come into conflict and play out in a self correcting way. Miners want to mine as many blocks as possible filled with high fee transactions and receive the highest possible block rewards in return. Users want to use a network that sends transactions seamlessly, cheaply, and quickly, in a trustless manner. What mechanism is there to prevent miners from accepting fewer transactions per block to increase the network fees? If some miners attempt this, they will be undercut by other miners in the highly competitive environment who are happy to accept more transactions per block. Suppressing the number of transactions per block is an intentional decision to not maximize profits (more transactions leads to more cumulative transactions fees). Over time this will result in lower profits and thus less capital to reinvest into mining infrastructure, a fate that won’t fall upon their full block mining competitors. This would make it uncompetitive for miners with this strategy in the long run. Conversely, if user volume grows sufficiently high and blocks are full but users are still seeking feeless and quick transactions their transactions will remain unconfirmed until network congestion falls or they include a competitive fee payment. This push and pull facilitates a network that is optimally useable for end users and minimally profitable for miners. In IOTA, all nodes have the same incentives. This could cause problems in an IoT economy where a user (either machine or person) may not necessarily have the same economic goals as a fully incentivized validator looking to secure the network.
To Buy or Not to Buy?
After considering both platform viability and tokenomic design, the IOTA project seems to be fundamentally flawed. Its highly volatile price movement over the past year appears justified by fundamentals on the bear side and inspired primarily by hype and irrational exuberance on the bull side. Even if an IoT ecosystem matures in the coming decade and the IOTA hardware/software solutions become its standards, it is still unclear what value would accrue in the token to allow for the desired virtuous cycle of token and network value creation. If ALL of these concerns are somehow remedied and the platform and token are a success, buying in today still seems like a risky venture since token custodianship and security options are less than ideal. If you buy tokens now, you need to ensure that you still have them later in order to sell after they have appreciated in value. Most signals from the IOTA team indicate that they have spent the majority of their efforts working on backend infrastructure. This, they claim, is why many of the front end user interfaces are still lacking. If we take IOTA at their word on this, it would indicate as though they view tokenomic design as a front end concern. If early investors are unable to frictionlessly interact with the platform, their capital will likely be diverted to other projects. If this happens, a negative feedback loop will be created that draws away from token price potential. Waiting for an irrational bull crazy larger than the ICO boom of 2017 is not an appealing investment thesis. IOTA, and other projects, would greatly benefit from a deeper examination of the tokenomic mechanics of their platform at an early stage in development. It is essential that decentralized token economies that rely on decentralized tokens prioritize the token’s design and functionality as highly as its technology and business components. When they do, all aspects of the platform are more likely to find success.