Current cryptocurrencies don’t satisfy the functions of money, limiting their adoption on a wider scale. Specifically, current cryptocurrencies don’t serve as a good medium, measure, or store of value. These limitations stem from the technical and economic realities of current cryptocurrencies. Specifically, they have slow transaction times, volume limitations, and high volatility. This post surveys plausible technical solutions to these problems and proposes the creation of a new cryptocurrency based on this analysis. It concludes with an analysis of the possible socio-economic impacts of such a system, were it created.
Keys To Practical Use
In order to solve the issues presented above, we need a coin with,
- fast transactions
- high volume
- price stability
In order to encourage wide adoption of a new currency, performance at these tasks must as good as existing alternatives. Therefore, we should use the dominant exchange systems to set minimum goals for performance.
The VISA network is capable of 24,000 transactions per second1 with a transaction speed of 10 to 20 seconds. This provides good baseline goals for transaction speed and volume capacity. Compare this to bitcoin’s maximum 7 transactions per second.
We can use the price stability of the US dollar as a standard for any new currency. Given the US dollar’s status as the number one currency of international trade, we can assume it’s price stability is sufficient for practical use.
Directed Acyclic Graphs (DAGs) are one solution to improve cryptocurrency transaction speed and volume limits. This architecture rejects the traditional blockchain for the DAG structure, enabling superior scaling that increases transaction speed as the network grows.
The first cryptocurrency to make use of this structure was DagCoin. The abstract from their 2015 whitepaper explains the concept:
DagCoin is a cryptocurrency design that attempts to be highly decentralized by merging the concepts of transactions and blocks and making each user that transact a miner. Each transaction carries a proof-or-work and references one or more previous transactions. The resulting authenticated data structure is a Direct Acyclic Graph (DAG) of transactions where each transaction “confirms” one or more previous transactions.
Similar efforts include the more recent IOTA, Byteball, and NANO. Depending on the design, the attainable transactions per second (tps) can surpass the VISA network. This analysis cites a theoretical maximum tps for IOTA around 70k tps, 46k tps higher than needed to surpass VISA capacity.
Of course this capacity is theoretical. Current load as of May 2018 hovers around 100 tps. Without increased stress testing it is difficult to know what the true current maximum capacity of the IOTA network is, though future development will hopefully make this clear.
Regardless, the utility of the DAG architecture in providing a decentralized solution to increase transaction speed and processing capacity makes it a reasonable choice for new currencies aiming for high performance.
In fiat currencies, price stability is ensured, in part, by the action of central banks that dynamically adjust the money supply based on current economic conditions. Being decentralized, cryptocurrencies require an alternative solution. This post details several methods. I give a brief overview of these methods below.
1. Fiat Collateralization
The simplest way to ensure price stability is to collateralize the coin with fiat currencies. For instance, we could set our coin to a 1:1 exchange rate with the USD, where every coin is directly exchangeable for one USD. This scheme is less an independent currency and more a digital representation of existing fiat currencies. As such, this coin would have the same price stability characteristics as the underlying backing asset. Unfortunately, this solution has a significant flaw: it requires a centralized holder of fiat reserves, undermining the purpose of using a cryptocurrency in the first place.
2. Crypto Over-Collateralization
A more decentralized approach is to collateralize the coin with other cryptocurrencies. However, because simple collateralization gives the coin the same price stability characteristics as the underlying asset, a naive implementation would do nothing to reduce the volatility of current cryptocurrencies.
One solution is the over-collateralization of the coin. For instance, lets say we have a price stable coin pegged 1:1 with USD. For every $100 worth of our stable coin, we collateralize with $200 (or more) of an underlying cryptocurrency asset, like Bitcoin or Ether. In this case, even if the price of the underlying asset falls by half, we can still liquidate our stable coin at it’s pegged ratio of 1:1.
There must be some method to channel exchange rate information from exchanges into the system. This can be achieved through a trusted third party. However, this increases centralization, decreasing security.
A decentralized alternative is Schelling Coin where the accepted rate is the medium of a large number of proposals and each user submitting a proposal between the 25th and 75th percentile gains a reward of N coins.
Crypto-over collateralization is an improvement over fiat-collateralized coins, but still has drawbacks. The first is the need for large amounts of backing capital, making this method very capital intensive. This method also requires some method to incentivize providers of the backing asset, likely through interest payments. This creates an expense just to hold the price stable coin. Another issue lies in the complexity of coordinating interest payments and collecting data on current exchange rates in a decentralized way.
Although none of these issues are untenable, they do present serious challenges to practically implementing this method. That being said, DAI seems to be making tremendous progress in this front with their ether-backed stable coin.
3. Crypto Baskets
A related approach is the collateralization of the price stable coin with a market basket of cryptocurrencies. Approached naively, this method would likely not be sufficient to ensure price stability, since the coin would simply reflect the stability of the cryptocurrency market as a whole. However, in conjunction with the over-collateralization approach discussed above, the use of crypto-baskets may reduce the capital requirements of that method (at the cost of more complexity). To my knowledge, the DAI team is currently developing the technology to allow collateralization by any ERC20 token.
4. Seniorage Shares
A more radical approach is Seniorage Shares, a method that rejects collateralization altogether. Instead, this method operates more similarly to a traditional central bank, algorithmically adjusting the supply of the coin itself in order to affect changes in exchange rate.
In a traditional price-pegged fiat currency, when the price of the currency is too high relative to the peg, the central bank will increase the money supply by issuing new currency. (Technically it’s through a more complex system involving reserve ratios, interest rates, and government securities, but the effect is the same)
Conversely, if the currency is too low relative to the peg, the central bank will decrease the money supply. This can be enacted through various tools of monetary policy, but is most directly achieved by the central bank holding currency, removing the stored currency from circulation and effectively decreasing the money supply. This implies some source of income for the central bank, which in a centralized system could be provided by tax revenue.
Adapting this to cryptocurrencies requires some adjustment. Increasing coin supply is simple and can be achieved by issuing new coins. However, lacking the ability to collect tax revenue, a cryptocurrency must come up with a new method to decrease coin supply.
Enter Seniorage Shares. In order to collect the currency required to decrease supply, the system sells ‘seniorage shares’ on the open market. These shares entitle the owner to a fraction of the coins minted during periods of growth where the coin is valued higher than the peg.
It’s an novel idea that’s gathered significant criticism, typically on the basis that price stability of the coin is dependent on the future growth of the coin’s value – if users no longer expect growth, the coin risks collapsing entirely as the worth of shares plummet to zero.
5. Dynamically Mined Supply
The biggest issue with Seniorage Shares is it’s approach to decreasing coin supply. While innovative, it lacks robustness in handling sustained periods of negative pressure on coin price. I propose an alternative system called Dynamically Mined Supply (DMS).
DMS operates directly on coin supply, increasing or decreasing the number of coins depending on the current exchange rate. Adjustments are made by modifying the reward to miners when confirming a transaction.
When a user creates a transaction, they attach a small transaction fee. In a simple implementation of DMS, the system increases coin supply by rewarding miners with more coins than were attached as a transaction fee. In order to decrease coin supply, the system simply rewards the miner with less coins than were attached in the transaction fee.
The ability to reduce coin supply is limited by the difference between mining rewards and transaction fees. The finite value of transaction fees means that the speed of coin supply decrease can only be increased by increasing transaction fees.
Unlike Seniorage Shares, DMS is robust against sustained negative pressure on coin price. Under DMS, the worst case scenario is a temporary decrease in coin price below peg value without the risk of total collapse, unlike Seniorage Shares. Because the deflationary mechanism can continue to operate without an expectation of future growth, DMS is more resilient to sustained periods of coin value decrease.
Drawing from the above analysis, I propose a new cryptocurrency that implements the DAG structure and makes use of DMS in order to ensure price stability. Instead of pegging the value to any one fiat currency, I propose pegging to the SDR, a basket of the currencies used by the world’s five largest economies. Implemented properly, such a currency would be able to achieve the three goals defined at the beginning of this post: fast transactions, high volume, and value stability.
The creation of a cryptocurrency with the capacity to satisfy the three functions of money could have large ramifications on the world economy.
For some developing economies, inability to ensure the stability of the indigenous currency leads to the adoption of a foreign currency as legal tender, in a process called dollarization. Often the currency adopted is the US dollar, though not always. This process has many advantages for developing countries, including increased economic stability and international legitimacy. However, the idea of adopting a foreign nation’s currency, especially if that nation is an enemy of the adopting nation, presents political challenges. Moreover, many nations are opposed to letting a foreign nation determine their domestic monetary policy.
An international cryptocurrency, not controlled by any one country or pegged to any one currency, has the potential to enable dollarization without the downsides of adopting traditional fiat currencies. In a process some call ‘cryptoization’ countries can adopt a price stable cryptocurrency as legal tender for domestic and international trade. With increasing adoption, such a currency could encourage increased cross border economic activity, leading to more efficient markets and lower average prices for all goods.
1 To be more fair, comparing what VISA does to what bitcoin does isn’t entirely honest. This post provides a decent discussion of the differences, but in short, although VISA claims to handle 24k tps, the typical load is closer to 1.7k. Still significantly higher than bitcoin, but not as dramatically. Moreover, a VISA transaction isn’t truly ‘finalized’ just because the transaction says approved after 10 seconds or so. In reality, a VISA transaction may take as many as 30 days to fully run through the various banks and clearing houses required for traditional financial transactions. Because bitcoin transactions correspond to actual transfers of value at the time of the transaction, they are more comparable to bank wire transfers, which cost around $30 and take days to process. By this metric, bitcoin already far surpasses traditional transfer methods in both speed and affordability.