The case against BNB
This article should be seen as a continuation of my previous post in which I laid down the case for ETH.
While I covered the gas issues with ETH then, they were largely waved off. There was also an example laid down which compared cities and how building infrastructure and network effects takes time and how this would work in favor of ETH.
Chains are like super markets, you want to go to the one which has the most amount of shops, most choices. You do not want to go to one mall to buy one thing and another to do other stuff. Some people enjoy spending time doing shopping and in other cases the value proposition might be so high that it is justified going through the hassle. But those are the exceptions rather than the rule.
The chain with the biggest community, which in turn is a function of the most options in terms of dapps built on it wins.
What was not obvious before (although it is now in hindsight) is how easy it is to fork and build replica dapps on a different chain.
This is especially true if it is an EVM (Ethereum Virtual Machine) compatible chain.
Given that the code to all dapps is typically open source, creating a fork and deploying it is extremely easy.
Going from 0 to 1 is hard, but going from 1 to n is not as much
The question then becomes where will the users come from?
And the answer is a well resourced sponsor such as a centralized exchange like Binance or HECO can direct a large amount of traffic to your replica app on that chain.
Fees on Ethereum were always non trivial, but of late the fees have become absolutely crazy. A typical Uniswap transaction will set you back by about $20–30 and if you interact with money markets such as compound you would spend hundreds for setting up a vault and so on.
Some of this fee rise is linked to the increase in Eth prices, but most of it is the network reaching the limits of its throughput capacity and demand being more.
A comparable swap on Pancakeswap is 100x cheaper and 100x faster. This leads to greater transaction volumes, greater liquidity and more user adoption.
No wonder BSC has been gaining traction.
The key question we need to assess is will this continue in the long run and whether BSC will supplant Ethereum.
My position is No, it is quite unlikely.
Let me explain why.
Blockchain has a trilemma where speed, cost and decentralization come at the cost of each other.
You cannot be fast and decentralized, and cheap. Something has to give.
The bet BSC is making is that people dont care about decentralization as much as they care about speed, costs and being quasi permissionless.
USDT is the biggest stable coin in the world. There is hardly anyone in the DEFI space who does not know about the concerns the industry has about USDT.
Yet almost everyone uses USDT. It has the most number of offramps and a large number of OTC providers support it.
This is evidence enough that the perception of being permissionless trumps actual mathematically provable permissionless.
A similar phenomenon is going on with BSC.
Almost everyone understands that BSC is not decentralized. But the day to day interaction shows no direct evidence of manipulation by the validators.
And users have super fast speeds and extremely low costs relative to Eth.
On the flip side, Eth costs are on an increasing trajectory. And even still some of the composability in Ethereum is about to break down due to Uniswap v3 using NFTs and the L2 solutions being largely not interoperable.
So is BNB, the native token of BSC a sure bet?
One should understand that there is no real innovation in BSC.
This is not a moral judgement call, but a statement of fact. This matters because Ethereum has the potential to solve its problems.
There are a few different solutions that are being worked on that promise to relieve the gas costs problem
- ETH2 and move to Proof of Stake
- Sharding
- Layer 2 roll ups such as Optimistic, ZkSync, ZkStarks etc
On the other hand BSC has no such publicly known plans. This is because gas costs are not a problem on BSC in the first place.
However the lack of decentralization will come to bite BSC, it is a question of When and not If.
When you have too much power, it is tempting not to use it. And more importantly even if CZ was a sage or an automaton who cannot be tempted, a small number of validators that are closely related to each other present a single (or small number) point of failure.
It is way too tempting for a state actor to decide one day enough is enough and use leverage on the validators and bring everything to a grinding halt.
Or the Russian mafia to kidnap a bunch of people and take control.
The possibilities of failure here are endless and only limited by your imagination.
We have seen this story before with EOS.
However users will keep on patronizing BSC till Eth gets its house in order or some other chain that solves for decentralization and speed together comes to fore.
BSC claims it is working on increasing the number of validators and increasing decentralization. But this is not in their own best interest and the reality is Ethereum is likely to solve for the speed and decentralization problem before BSC.
Innovation happens better when there are more heads working on a problem and they are operating in an unfettered manner.
And as BSC gains more adoption, it will run into the same problems that Eth has. This is especially true if BNB goes higher in value. Its higher price translates to an increase in gas costs on BSC, which in turn destroys the value proposition BSC currently has.
The only way for BSC to solve for this without innovating is by reducing the number of validators even further.
However not all is doom and gloom about BSC. They do not have just Eth refugees, they have also widened the pool by giving brand new users their first taste of pretend Defi.
These users will eventually realize why decentralization matters and make rational choices to protect their wealth long term.
The key question that remains is when will an inflection point come, when does the trend of users leaving Eth for BSC reverse or at a minimum some other chain that solves for decentralization emerge.
We will address that in a future piece.