Scaling Problem For Lightning Labs Taro – Bitcoin Magazine
This is an opinion editorial by Evan Price, a software engineer of 15 years and advocate for privacy rights.
Taro is a new protocol being developed at Lightning Labs that promises to enable creation and transfer of digital assets on the Bitcoin blockchain and specifically on the Lightning Network. It is being hailed as a revolutionary advance in cryptocurrency tokenization. I am skeptical of any proposal aiming to transfer non-bitcoin tokens on the Bitcoin network, but Bitcoin is a permissionless network and if Taro fans are intent on building and deploying it no one can stop them. This is the magic of Bitcoin: it is a truly neutral arbiter. Bitcoin only enforces the protocol rules; it does not pass judgment on how those rules are used.
Taro’s design is very clever. It hides a data structure called a sparse Merkle sum tree inside of the Taproot scriptpath, which is itself a Merkle tree that lives inside every Taproot address. It’s Merkle trees all the way down! However, I believe this design places a fundamental limitation on the scale that can be achieved with any asset issued using the Taro protocol. The crux of the problem is that every time a Taro asset is issued or transferred it must happen inside a Bitcoin transaction that will eventually be committed to the blockchain. Bitcoin’s block space is intentionally limited in order to minimize the resources required to run a Bitcoin node. This keeps the network decentralized and is a fundamental pillar of the bitcoin security model. Blockspace must be scarce in order for bitcoin to remain secure.
I believe that any protocol that requires a bitcoin transaction to move another asset will be inherently limited by the block space market. We are currently in a period of persistently low fees, so these protocols should work fine for now. But if bitcoin use spreads to most of humanity, as I believe it will, this low-fee period will be definitively over. As the fee market grows the cost of bitcoin transactions will become increasingly high. When this happens all other assets will be priced out of the Bitcoin blockchain. In the long run, successful monetary assets will be better served on a single purpose blockchain, or even better, a non-blockchain database where fees will be lower and transactions will be more affordable.
A lot of hype around Taro is focused on its use in Lightning channels. I have many concerns about the complexities involved in this design, but let’s assume everything works as intended. This will scale the protocol beyond what is possible exclusively with on-chain transactions, but I don’t believe this will reduce total on-chain transactions for two reasons. First, Lightning is optimized for small-value transactions. This is because the value of a Lightning transaction is limited by the amount of liquidity committed to Lightning channels. On-chain bitcoin transactions have an unlimited maximum value and are usually a better choice for large transfers of wealth. Second, moving small value transactions onto Lightning won’t decrease congestion in the long run due to induced demand. People will consume the additional capacity until a new equilibrium is reached. That equilibrium is determined by how much congestion people are willing to tolerate. On a blockchain congestion equates to fees. This phenomenon is not exclusive to Bitcoin, it applies to any blockchain that integrates with the Lightning network such as Litecoin or Blockstream’s Liquid sidechain.
If Taro is deployed and used it will increase bitcoin fees. Paradoxically, this decreases the utility of Taro. This negative feedback loop will limit the scale that Taro assets can achieve in the short term. In the long term as people flee weak currencies for the safe haven of the strongest currency, bitcoin, the fee market will organically grow from native bitcoin use. At this point the writing is on the wall for monetary assets issued on Taro.
Another use case for Taro is NFTs. Side note: Lightning Labs carefully avoids the term NFT in their official communications, but I struggle to find an alternative meaning for the phrase “unique and non-unique assets as well as collections.” I have my issues with NFTs, as many Bitcoiners do, but their existence and use is undeniable; they are here to stay. NFTs may see some traction on Taro but I’m not convinced that Bitcoin is good for existing NFT use cases. Do you really need unstoppable censorship-resistant displays of conspicuous consumption? In any case, I think some NFTs may find a niche on Bitcoin using the Taro protocol. NFTs are designed to benefit from artificial scarcity so I don’t believe they will be sensitive to high prices caused by the growth of the fee market. It’s likely that once they gain a foothold on the Bitcoin blockchain they will become very difficult to dislodge, to the detriment of users of the Bitcoin asset.
I do not mean to give the impression that Taro is worthless. In fact, I think it may end up being a tool that supercharges Bitcoin and Lightning use all over the world, just not in the way most maximalists dream about. The name is a subtle hint at the goal of the protocol: taro is a popular root vegetable and staple food across large swaths of Africa, Asia, and the Pacific islands. Stablecoins are the most popularly-used cryptocurrencies the world over. Stablecoins marry the speed and borderless nature of cryptocurrencies with the most popular unit of account in the world, the dollar. Many stablecoins are designed to operate on a multitude of blockchains and Taro seems poised to open the gates for stablecoin use on bitcoin. The increased reliability and security of bitcoin will only improve the value proposition of these coins. I believe this will be a bootstrapping phase in the transition from the old global currency, the dollar, to the new global currency: bitcoin. What is not clear at all to me is how carrying stablecoins over bitcoin rails will incentivize more of the world’s population to use the most trustless, decentralized, secure, and inflation-proof money ever invented.
Credit to Ruben Somsen for introducing me to these ideas and helping me refine my argument.
This is a guest post by Evan Price. Opinions expressed are entirely their own and do not necessarily reflect those of BTC Inc. or Bitcoin Magazine.