# Gavin Andresen
# 2015-02-01 23:35:10
# https://bitcointalk.org/index.php?topic=941331.msg10330613#msg10330613

@s{quotedtext}
@s{quotedtext}
 @p{brk}
What's wrong? @p{par}

Lets say those million dollar transactions are 250 bytes.  That is 4,000 of them in a 1MB block. @p{par}

So $40,000 total reward to the miner @p{--} about eight times current block reward. @p{par}

BUT YOU ARE SECURING TRANSACTIONS WORTH SOMETHING LIKE 2,000 TIMES MORE VALUABLE THAN TODAY'S TRANSACTIONS (estimated average transaction USD value for today's average transaction is about $380). And I GUARANTEE that attackers would have a much easier time pulling off a double-spend of one million-dollar transaction than 1,000 $1,000 transactions. @p{par}

The math for "large value transactions will generate enough fees to secure the chain" just doesn't work. @p{brk}
The math for "lots of small transactions will generate enough fees to secure the chain" might. @p{par}

Also: @p{par}

I still haven't heard a coherent argument on why large value transactions are necessarily also high-fee transactions. @p{par}

I'd suggest you go research existing high-value-payment networks and see what typical fees are for multi-million dollar transactions. FEDWIRE is running at 6 transactions per second, average transaction value over $6million, with fees per transaction @p{(bf}UNDER ONE DOLLAR.@p{bf)} @p{par}

Why? Because if you are giving somebody one million dollars for something, you almost certainly have built up real-world trust, and probably have a longstanding relationship, signed contracts, etc etc. @p{par}

If you think Bitcoin is different, please explain the scenario where I send a stranger who I don't trust (so have to rely completely on the blockchain) $1million for something. @p{brk}