BTC users will always tell you to that LN is the only way to solve the scalability problem and to do micro payments, since they hate big blocks, when you try to use LN and it fails because of routing or liquidity issues, they say that it will be ready in 18 months™, since 2017.

Even if by some miracle, the developers found a way to fix the issues of LN, it still have major flaws because of its design, that make it a bad payment method and also it will hurt BTC in the long term.

LN does not make BTC acessible to poor people

First we need to know how LN works:

1. You and your friend deposit Bitcoins in a multsig address, wich is a address that multiple private keys control.

2. Now you both can send funds to each other without paying fees to miners.

3. You can also connect to a hub, A have a channel with B, and B have a channel with C, so A can send Bitcoins to C using B as a router.

When you are done, you close the channel and deposit your funds back to your on-chain wallet, you need to do 2 on-chain transactions to use LN, now consider that BTC fees can range from $10-30 most of time, that means that to a person to use LN, they need to pay $20-60 in fees, trust in me, $20 is a lot on developing countries.

But hey, Bitcoin does not need to be affoardable to poor people, only people with money deserves to use Bitcoin, BTC users will say this.

BCH avoid this having cheap on-chain fees.

LN becomes more centralized as time passes

Even if you are person that have a good financial situation, you do not want to pay $50 to do every payment you want, so you can connect with a hub, so that you can use LN for a long time before needing to use the blockchain again.

Here is the flaw, to save on fees, you want to connect to huge hub, that is connected to thousand of people, so that you can pay anything you want without opening another channel.

LN incentives centralization, as time passes there will be a lot of channels connected to a few dozens of hubs, this is not ony theorical, this already happening in pratice.

At the moment that i am writing, the 10 biggest hubs are connectd 10,220 channels out of 40,980 that exists, that is right, 10 hubs routes 25% of all LN channels.

And that makes LN weak to censorship, these hubs can stop routing your transactions at anytime, so you will need to close the channel, open a channel with another huge hub, and hope that it will not censor you.

If you have less funds than the transaction fee, you will have to donate all your funds to miners, it is better to just use Paypal and pay things with IOU Bitcoins, much cheaper and less headache.

BCH avoid this by allowing pratical payments on-chain, not forcing users to centralized and custodian hubs.

LN will reduce the security of the BTC blockchain

Users want to pay less fees, miners want to gain more fees, as time passes the amount of minted coins is reduced, so miners will need to receive more fees.

The core devs pushed the narrative of high fees and low volume, with small blocks, a fee market is created, so to BTC survive in the future, the network needs to be broken, wiht users needing to pay $100 fees or more, and waiting days to get a confirmation.

Now the question is: How a fee market wiill grown, if most uses will simply use LN, avoiding miners fees? It is a contradiction, users will do all their transactions on LN, reducing the demand for block space, reducing revenues to miners.

BTC will become less and less secue every halving, until at some point, a 51% attack will be very cheap compared to the value of the network.

BCH avoid this allowing all users to do payments on-chain, so that every user will pay a tiny fee, and there will be a high volume of transactions, keeping the network secure.

Conclusion

LN does not work as a P2P eletronic cash system, it is basically a Paypal kind solution that core devs are pushing on BTC users, it is centralized, custodian and expensive to start to use, also makes the main chain less secure on the long term.