This is correct, All currencies have transaction costs, but compete on having lower transaction cost than the alternative.
The theoretical maximum is zero transaction cost for a transaction. It cant go positive. Components of transaction cost are triangulation, transfer, and trust. Currencies attempt to lower the cost of the transfer and trust components.
Any positive value a currency claims to have as method for settling transactions is just a differential with another method.
There is a cost to going to an ATM and carrying USD cash, but it is lower than carrying around 100lbs of cabbages to barter with.
Don't forget externalities. E.g. you can get a fee-free bank account with fee-free cash withdrawals, courtesy of the bank being able to loan out that money under fractional-reserve restrictions. In that sense transactions can pay for themselves. Presumably nobody would print money if it wasn't being put to use. Hence, novel currencies can enable novel business models and potentially do more than just lower transaction costs in a zero-sum way. (Admittedly the value of such business models at present, e.g. crypto scams, is somewhat debatable)
Economic activity. Switching from free banking [1] to a centralised currency made trade easier in America which increased the amount of trade. It's why monetary unions work [2].
Bitcoin, similarly, has enabled certain modes of trade that previously didn't exist.
I think that is more of consequence resulting from lower transaction costs, rather than an economic externality. That said, I understand the definitions of externality is changing in popular usage.
> that is more of consequence resulting from lower transaction costs, rather than an economic externality
An externality is "an indirect cost or benefit to an uninvolved third party that arises as an effect of another party's (or parties') activity" [1]. If Visa reduced its fees, the merchant's additional profit is an externality.
> understand the definitions of externality is changing in popular usage
Third-party economic consequences have always been the definition [2].
As I see it, I think the distinction between consequences and externalities is that externalities are applied to cost and pricing, but not concepts.
Wikipedia doesn't go too deep into when externality analysis is appropriate or useful, but hints at this in the following sentence to the one you quoted.
>Externalities can be considered as unpriced components that are involved in either consumer or producer market transactions
Something like the advent of centralized currency isnt a priced good or action, so it doesnt make sense to say it has a positive externality (e.g the transaction price of "centralized currency adoption" is not a priced object).
This is why it is more appropriate to say something like the advent of centralized currency has consequences instead.
Regarding the Visa example, the merchant isn't a uninvolved 3rd party, so the profit is not an externality. Value on the table during an economic exchange is not an externality. Similarly, the value created or lost for each interacting party is not an externality.
One more thoughts on common externality misconception.
The general connotation is that externalities describe pricing failure and market failure (e.g. pollution). The usefulness externality analysis breaks down quickly when you move from realized external costs or benefits to opportunity costs or benefits.
If a $1 pill saves a worker's life, is the value that person generates over the remainder of their life an externality. Technically yes. If you sell them a pill again a year later, the same is true. That doesnt mean the pill is priced incorrectly.
This illustrates how externalities can be duplicative and should not be confused with pricing errors.
> the transaction price of "centralized currency adoption" is not a priced object
Of course it is. Joining a currency union carries costs.
Broadly speaking, you're correct: the term has ambiguous meaning. My point is that isn't something new, but an element that has always been with the term.
I think that still runs afoul of the "uninvolved 3rd party" part of the definition.
Two actors can both generate value from a transaction due to a difference between price and their respective utility value for what is traded. This producer and consumer surplus is explicitly distinct from externalities.
If there is a currency deal between the US and Argentina, the consequences to those countries are not an externality. However, if this deal produces a 2nd order change in the Chinese RMB, some would call that an externality, although I would call it a consequence (because externality implies mispricing)
Any positive value a currency claims to have as method for settling transactions is just a differential with another method.
There is a cost to going to an ATM and carrying USD cash, but it is lower than carrying around 100lbs of cabbages to barter with.