Transaction Fees and the Adoption of Bitcoin | Hacker News Reader

Transaction Fees and the Adoption of Bitcoin

disruption.vc

13 pointsfredoliveira13 comments

this is ignoring several key costs associated with bitcoin:

1) knowledge (the average person needs to invest time to learn what bitcoin is, how to use it, how to protect themselves from fraud, etc.)

2) hidden cost of buying and selling bitcoin (bid ask gap) that is, no one really uses bitcoin, they barter with it between currency transactions with real currencies like USD or Euro, this cost is appx. 2-3%

3) the cost of loss due to fraud, theft, etc. and zero recourse when that happens (there have been so many examples, I'll just leave you to google specific cases)

In the end, when you calculate these implicit and explicit costs, bitcoin comes out to cost about the same or more than existing transaction options. But with the added 'advantage' of having zero recourse if anything goes wrong (libertarian wet dream) and requiring your Aunt Judy to learn fun new things like blockchains, hashes, encryption, etc.

Yay!

I believe the point of the article is largely to raise questions and discussion about a couple of things:

1. Potential disruptive impact of alternative currencies with respect to transaction fees -- people want easy ways to exchange money, for commercial or personal reasons, that aren't cost-prohibitive. Can alternative currencies present a legitimate challenge to traditional payment providers in this arena? Can traditional providers learn from it by finding (sustainable) ways to better serve consumer/merchant demands?

2. Whether payment providers, of any kind, can profit and compete on consumer behavior data as an offset to higher transaction fees, and how much consumers are willing to tolerate it. We haven't seen this play out yet for Bitcoin and its ilk. jusben1369 pointed out LevelUp here.

The article does not propose a conclusion that Bitcoin is appropriate for your Aunt Judy today. That's probably been beaten to death by coverage all over.

The bid-ask spread will shrink as (if) the currency stabilizes, and you won't need to round-trip as (if) the currency becomes more widespread. It's not really any different in these respects than any other currency - if, say, I decided I wanted to pay someone in Baht or Rubles or something, it's just that there is not presently much of a market dealing primarily in bitcoin.

I'm not going to push any notion that bitcoin has no issues, and these are issues in the short term, but these particular issues - cost especially - aren't anything like insurmountable.

Completely agree that my post is ignoring those things. The post is an extremely forward looking one. Bitcoin could crash and burn before anyone gets a chance to implement something like this. It could be replaced by a second generation cryptocurrency that doesn't have a public ledger. There's a lot of scenarios here where nobody ever gets to do what I was talking about.

I opted to focus entirely on the costs of transmitting the money. If we did a full analysis, we'd also have to look at chargebacks, figure out how to set a value on the speed of the transaction, and deal with the fact that sometimes Western Union just 'loses' your money and isn't sure where it is. Not all of these costs would be paid by the same person, but they do all add up to a systemic cost.

But that's not actually the primary point. There's plenty of reason to believe (besides transaction cost) that cryptocurrency and smart contracts have a long future ahead of them. I'm simply trying to point out that there are business models here that aren't yet being explored.

Those credit card fees are really high. Ie in the US they're closer to 1.5% for any merchant doing decent volumes.

Many of the fees associated with non Bitcoin transactions reflect the cost of regulation. So until Bitcoin is regulated these comparisons are misleading. (and it will need to be subjected to some types of regulations if it wants broad adoption vs niche adoption)

Lastly, LevelUp is trying to do the "very cheap to no fee processing" and making it up with mining data and providing BI services for a fee back to their merchants. Not sure if they'll succeed or not.

I used credit card transactions rates published by Stripe for the table. Square's pricing is pretty similar, though they don't have a '+ X cents' for non-manually entered transactions, which does make them a very good choice for buying a $3 coffee. Certainly Walmart is going to get better volume rates than anybody using Square or Stripe, but most people are not large retailers.

Bitcoin is actually regulated already, however the costs of Bitcoin regulation cannot be reflected in the protocol's transaction fee. An additional fee could be assessed by the service provider, but so far these fees typically happen right now at currency exchange points, rather than as a bonus 'sales tax'. BitPay has chosen to charge a flat rate monthly payment for their non-free plan while Coinbase charges 1% to cash out to USD.

Also I'm well aware of LevelUp. Point was not that other payment mediums cannot do the same but that Bitcoin can potentially provide valuable insight for a much larger set of transactions once you're able to make sense of a big enough chunk of the blockchain.

Stripe's pricing is ludicrously expensive. And Square is trying to be more than just a payment processor. These companies are competing for markets that seriously care more about "pretty website" than "efficient business". I mean, even PayPal had much better rates than Stripe: that $1 charge would have a $0.10 fee if done via PayPal (even using PayPal's direct credit card billing solutions). For larger price points, Stripe finally added some volume discounts, but they are nowhere near as good as their competitors (like, they require so much volume as to be unattainable for most merchants). Using Stripe as a comparison for this kind of purpose is thereby just silly...

Fair enough. That said, I know of small businesses that are charged vastly larger %s going through a merchant account, etc (like nearly double what Square/Stripe charge) so I don't think the rate is entirely unfair. I'd rather aim for a median rate than 'lowest possible' for a table like this, but I haven't seen anybody post average rates that are anything but estimates and I am reluctant to publish a number in a table like that that I'd just pulled out of thin air. I believe median card-not-present rate is ~2.5%, but that's a thin air number.

The issue at hand is that you've pulled a ludicrous flat overhead fee for the low-end numbers, leading to a 33% charge... even normal un-negotiated merchant accounts are usually not that expensive (the percentage gets higher, but the fixed fees are normally lower). PayPal (which as far as I know anyone could get) is 5%+$0.05, which works out to 10% on the $1 charge.

The typical fee for a simple transaction is currently 0.0001 BTC (~$0.07), not $0.38.

I also do not think it is fair to say off-chain bitcoin transactions are free. There is some cost in implementing, running and securing an off-chain database of balances.

That makes more sense. I was wondering why the fee for $1 worth of BTC was listed so high.

I used 0.0005 BTC (~5kb), but you're absolutely right that a simple transaction is substantially cheaper. My assumption here is that as the economy grows, due to the underlying technology of Bitcoin, more BTC will need to be 'gathered' together in a wallet before being sent out. This results in a larger transaction size and a higher fee to exceed the priority threshold.

I suspect that in the future, this will be a fairly common case. I suppose I might liken the theory here to defragging a hard drive. An unused hard drive stays in a nice neat, contiguous state. But use it a whole bunch and things get messy. And in BTC's case, you have to pay a bit extra for the miners to come along and tidy everything up.

I could be totally wrong on whether that will ever become common, and the developers could just as easily reduce fees over time as well, but such is the nature of taking guesses at the future.

Feb 24th we're coming out with a way to make payments at $.18 a payment (no fees under $2) in under 30 seconds without making a new account. I hope everyone reads this article and keeps it in mind when we launch!

Reply on news.ycombinator.com