Scalability of Bitcoin

Question

1. What will incentivize miners to keep processing transactions after bitcoin production slows down and or stops?
2. I understand that miners could work off fees, but are these costs per transactions for the consumer cheaper than they are now with the current credit card system?
3. If the suggestion is just for the consumer to send bulk transactions instead of individual payments, what stops someone from buying coffee and then not sending it?

Answer 1

1. What will incentivize miners to keep processing transactions after bitcoin production slows down and or stops?

We don't know. The hope is that fees can take over the function of subsidy over time (and to an extent, that seems to be happening), but the incentives in the system become more complicated to reason about in that case (search for fee sniping attacks).

2. I understand that miners could work off fees, but are these costs per transactions for the consumer cheaper than they are now with the current credit card system?

It is important to realize that not every payment in BTC-the-currency needs to happen on Bitcoin-the-blockchain:

• Transactions happening within exchanges do not hit the blockchain, and probably wouldn't fit regardless.
• If you need to pay me, and I trust you, you could just give me a private key with associated funds.
• Payment channels (including bidirectional ones like Lightning) can result in many small payments being aggregated before making the whole world validate it.

Bitcoin-the-blockchain is good as a currency:

• Fair, verifiable, beyond control, resistant to theft and seizing.
• Being natively digital, it hopefully simplifies building systems on top (for example, systems with cryptographic proofs of reserve).
• Going beyond the traditional forms of ownership: money that inherently requires multiple people to sign off on, or even further, money that can be tied to the exchange of knowledge (look for ZKCPs).

For all these properties, it is not good at being a payment system:

• It's slow at transferring (it takes hours for transactions to confirm).
• It is expensive for small payments (its fees are proportional to transaction complexity, not amount)
• It lacks consumer-oriented features like disputing claims.

All this means is that for different purposes, different technologies may be more appropriate. Nothing prevents existing (or, hopefully new) payment solutions to be brought to BTC-the-currency.

3. If the suggestion is just for the consumer to send bulk transactions instead of individual payments, what stops someone from buying coffee and then not sending it?

Lightning does not have this problem - it processes transactions between two parties instantly, without needing the whole world to verify it for them. The downside is that it requires channels to be setup beforehand, and loaded with some money from both parties (which returns to them after closing the channel, or can be taken by the other party when you cheat).

Answer 2

After mining rewards slow down enough, the miners will rely on fees more and more. As you have predicted, it would result in horrendous fees that would render bitcoin unusable for ad hoc transactions.

These issues has been addressed since the original Bitcoin whitepaper and alternative solutions were proposed later. The original solution is to remove [EDIT: or push further] the block limit of 1 MB that has been put there as a temporary solution only. It would let the miners to verify more transactions at once, collecting more income from fees without the fees having to be higher.

Another solution is a network such as Lightning Network, which, once operational, will allow users to make secure transactions that doesn't need to be synced with the blockchain every time the transaction takes place, but only from time to time, without any trust necessary between the users.