Bitcoin Fee Price Elasticity

This was the topic of a post here in 2014, which got 1 (one) comment.

Price elasticity means how much demand will be reduced if price goes up.

That’s an important point now. Some people think that Bitcoin fees need to be low or zero, or users will switch to something else, and Bitcoin will be doomed.

The question if that is right depends on the question if and how much demand is elastic.

Right now empiric data doesn’t point to high elasticity. Fees are way up, but there are still enough users to keep block size fluctuating around 850k.

My best estimate for average transaction size is somewhat more than $1,000. Bitcoin is moving around $300 million a day, with around 260,000 transactions.

For an $1,000 transaction, obviously saving a couple of dimes in fees is not a strong consideration. So it makes sense to expect low elasticity at the present level (average fee of less than $1).

So the answer would be: Yes, some users will be priced out of Bitcoin without larger blocks. But right now, there are enough users left to keep the seats in the airplane full. No urgent need to admit twice the number of people (without particularly caring if that will cause the plane to crash in mid-flight) to save $0.37 on fees.

Bitcoin fees need to get higher to deliver empirical data on exactly when demand drops because of prices.

When and if that happens, smaller demand means that fees go down immediately over market rules. No need to push for a Bitcoin fork in that situation either.

(Cross-posted from Bitcoin reddit post here).