Kenya's mPesa vs. Bitcoin | Hacker News Reader

Kenya's mPesa vs. Bitcoin

pymnts.com

30 pointsjusben136947 comments

Except his main point (of Bitcoin failing to grow exponentially) is just false. Look at the "number of transactions per day excluding popular addresses" (located at https://blockchain.info/charts/n-transactions-excluding-popu...). You see a clear straight line on a logarithmic scale. So while Bitcoin may not be growing as fast as mPesa did, it is definitely growing exponentially.

Agreed -- this is what I signed in to post. I have been waiting for a long time, with growing frustration, for Bitcoin's transaction rate to start taking off[1]. For several years, the transaction rate was near flat. But look at it now! Clearly, people are using Bitcoin for to buy and sell.

In fact, my thesis is that increased use of Bitcoin as a currency is partly what is driving down the price. Most of the highest-volume Bitcoin-accepting vendors (Overstock, Gyft, Tiger Direct, etc.) change their Bitcoins immediately to USD, which creates significant selling pressure.

1. If anyone is wondering why the most popular addresses are excluded, it's because those addresses are almost invariably the exchanges, and so they skew the data upward.

Can you explain why you think you have any visibility into the trades from those retailers?

Probably because they use coinbase or others that change the money immediately (or guarantee the current price which usually amounts to the same.)

That just pushes it off a degree. How do you have any sense of that volume?

(volume being critical to deciding whether the trades are interesting)

Dude,

Your chart is really quite identical to "chart 2" lower down on his page (if anything his chart looks "more optimistic" and it covers a longer timeframe).

Sure, bitcoin transactions are growing, maybe exponentially if you squint (and maybe not). Just at a rather lower rate than the virtual currency he spotlights and that's most of his argument.

Basically, he is looking at a payment system that is used primarily for, like, payments and showing it's contrasts to bitcoin.

I mean, does anyone really see bitcoin as primarily for payment in today's world?

Bitcoin has no appeal as a payment system compared to fiat currency for legal transaction. It might be fun and novel but that alone won't make it successful.

Erm near zero transaction costs and irreversibility are a HUGE appeal, anyone who sells online and had to deal with Paypal and Credit cards would tell you plenty of nightmares about frozen accounts, chargebacks and chargeback fees and rampant fraud.

Why on earth would I as a purchaser want to buy something with bitcoins (which offer me no protection) instead of using a credit card (which offers me extensive protections).

Cheaper prices would be one.

As a merchant when a customer chooses to pay with bitcoin I pass on the savings (that would instead go to middlemen like paypal and credit card companies) onto the customers, and they love it judging by feedback received.

Credit card companies and paypal etc are parasites who siphon billions from global commerce, their business models are based on old money transmission methods.

There used to be a time when people would ask as to why they would use credit card over cash, yet people moved on and accepted new concepts and technology

That's illegal according to the contracts you have to agree to in order to accept credit cards or paypal. That's why almost nobody is doing that.

This isn't true anymore.

http://www.cardfellow.com/blog/charging-customers-a-credit-c...

Perhaps you don't want the purchase to be connected to your credit card, or perhaps the purchase is small enough for it not to matter. I've bought a few indie games with Bitcoin, admittedly mostly for the novelty, but it also turned out it was actually easier than typing in my credit card information. It also requires me to place less trust in the website; sure credit cards have much more protection, but it's still a pain if your credit card number winds up stolen.

However, when it comes to physical goods costing more than a trivial amount of money, I'd certainly want to use a credit card.

"it was actually easier than typing in my credit card information"

What about if you amortize over the time it takes to set up your bitcoin wallet, etcetcetc. (and to be fair, it takes time to set up a credit card, but there's also a much larger set of purchases to amortize over).

I setup my bitcoin wallet in the early days when there was much less infrastructure, and back then it took a while! However, a few months ago I tried a peer-to-peer bitcoin website and I was surprised at how quick it was. I think went from sign-up to bitcoins in my account within about 15 minutes (+ confirmation time).

There's still a significant learning cost, of course, but I'm encouraged by how much better the infrastructure around Bitcoin is compared to just a few years ago. The infrastructure looks pretty professional now, and the security practices are noticeably better (though that isn't exactly hard, considering how insecure the earliest services were!)

When I was holidaying overseas my credit card was left in a taxi. The taxi driver then proceeded to spend over $5000 in 3 days. I spoke to my bank and within a week the money was refunded to me. Those merchants should have had better processes to handle locals using foreign credit cards. And likewise they have the ability to work with local authorities to recoup the lost money.

These processes work every day to protect consumers. People using Bitcoin may use it once but the first time they are burned they will never use it again.

I wish your taxi driver good luck spending stolen Bitcoins without the owner's private key. The fact that your credit card worked for 3 days for $5000 in unauthorized transactions is a bug, not a feature.

What if you left your phone or laptop in the taxi? You could say that at least for the laptop, your wallet should be encrypted with a password... but you're depending on everyday people picking good passwords then, which we all know they don't do.

Storing your coins in a multisig address will mitigate this threat. Look for multisig service providers to offer MFA address security this year. I expect several will tie into major wallets and be very easy to use.

If you're keeping serious amounts of cash in a physical wallet, it's on you to remain vigilant about not leaving it out in the open. Same principle with Bitcoin, although I'd say losing your phone with a Bitcoin or two in it won't be quite as devastating since the digital wallet can exist in multiple locations simultaneously. If someone steals your phone, you can restore a backup of the wallet and transfer out the coins just to be sure.

> Storing your coins in a multisig address will mitigate this threat.

Assuming you mean a 2-of-3 address, where one key is on the laptop, one is a password-derived key, and one is... stored at home or with a friend, I suppose, in case the laptop is lost?

We must assume that (a) your laptop is never stolen while the wallet is decrypted, which implies making absolutely certain that end-users never leave their laptop unlocked, and (b) that nobody ever threatens violence in order to get the password, aka the rubber hose attack.

The only way to prevent this from happening in an irreversible payment system is to ensure that the end-user does not have access to all their money at one time, especially while on holiday. This is an absolute downside in comparison to credit cards.

You're right, people looking to steal bitcoin invariably target the private keys.

(Less antagonistically, different things are different and considering a few of the differences usually isn't super useful)

You've missed the point by concentrating too much on the detail that it was a lost credit card. Just change it to "I bought $5000 worth of equipment online and the seller shopped cheap junk instead and then disappeared".

Bitcoin: too bad for you.

Credit Card: chargeback. Either you get the $5000 back from the seller, or from his merchant account provider if their risk assessment folks didn't set a high enough amount to hold back in reserve in his account. (Or rather, you get it back from your card's issuing bank, and they get it back from the seller or his merchant account provider).

Credit cards are exactly that, credit. Bitcoin is not a credit system.

If you had had a debit card and money was fraudulently spent, good luck getting that back from your bank. Debit cards are used to make hundreds of billions $ worth of purchases online each year.

> Those merchants should have had better processes to handle locals using foreign credit cards.

Isn't this victim blame?

My debit card has similar protections as credit cards. And I'm not sure why the lack of credit in Bitcoin is a feature.

Yes to be clear, at least in the US, most debit cards have this protection. Although it's not quite as simple as a chargeback.

near zero transaction costs

Ignoring the sizable 3% or greater costs incurred in acquiring the currency, plus the transaction costs associated with prioritizing a transaction for processing.

irreversibility

For merchants. It's a definite killjoy for customers, and customer uptake is far more important than merchant uptake for a new currency.

anyone who sells online and had to deal with Paypal and Credit cards would tell you plenty of nightmares about frozen accounts, chargebacks and chargeback fees and rampant fraud.

Nope. The only nightmares I've heard are from people who did things wrong. Paypal and merchant services make their rules pretty clear, and they even offer assistance to merchants to figure out what they need to do. If you're having problems with Paypal or your merchant service provider, it's because you messed up.

I avoid eBay and use Craigslist for selling stuff because I don't want to risk a PayPal nightmare (due to scams which I always hear about). I'd rather have cash in my hand that I can verify is not counterfeit.

There's a huge cost to doing business on Craigslist, to wit, you have to deal with people who do business on Craigslist. (there's good, but dang there's a whole lot of flakes and wackos on CL)

The fact that buyers can't do chargebacks is a good thing. It matches the real-world model (established sellers who can be assigned a reputation, unestablished buyers who can't) much better than credit cards. For eBay-like situations, m- of-n transactions allow for reversibility.

It costs about 1% to acquire Bitcoin using USD. If the merchant wants to receive USD instantly without exchange risk, that's another 1%. Not 3%, which is the typical cost of doing business with credit and debit cards. The last taxi driver I talked to said he pays 5% because he couldn't get approved for Square.

Chargebacks are important for credit and debit cards because you hand over your payment credentials many times each day. You never hand over your Bitcoin payment credentials. No one complains that cash doesn't have chargebacks because it just isn't a problem.

It'll be fun to speculate on both sides of this issue throughout 2014, but there are tens of millions of dollars floating around to convince merchants that accepting Bitcoin is a good idea. If it's cheaper and usable, it's going to work out. Let's see what the world looks like in July.

In fairness, you do tend to hear horror stories about charity drives and such getting shut down by Paypal (and it would be nice if there were actual non-shady competition in that space).

Fraud and theft is much more rampant in the BTC universe. By now there have been dozens of large-scale thefts when services were compromised. And unlike with established payment solutions it is next to impossible to get your money back if it happens.

As a consumer it is also impossible to pay with BTC ad hoc unless you already own some and thus expose yourself to its volatility. Of course I could see something that I want to buy, then log onto a BTC exchange (which requires me to already have an account) and buy enough BTC to then pay with those BTC. But why would I subject myself to that additional effort for no practical benefits for myself?

4-5% transaction cost (http://blockchain.info/charts/cost-per-transaction-percent) is certainly not "near zero", and in many cases is higher than credit card transaction costs.

just because the current transaction costs are paid with freshly printed money does not mean they dont exist, and as the rate of minting goes down as we near 2080, those transaction costs are going to start coming more and more from the users instead of the mines.

I enjoyed this article in that it seems fairly balanced, unemotional and included hard data. It does raise the question about how much of Bitcoin's interest in the press is driven by it's usage as a currency vs its appreciation as a financial asset.

Unsurprisingly pymnts.com is not positive about payments going completely P2P and wiping out the middle men from the picture entirely.

Well you do have to wonder as to who pays their bills

Well Gentlemen the arguments presented either hold water or they don't. Best to stick to the points presented if a rebuttal is what you have in mind.

They don't. The comparison makes no sense whatsoever.

mPesa has been around for longer, and has a bigger momentum. I don't think there's anything to get out of this besides that. Also, how many people there have at least smartphones, if not PCs, to make their Bitcoin payments? With mPesa you can do it through SMS if I'm not mistaken. So again, of course the market for mPesa is bigger right now.

I also don't think it's fair to look at Bitcoin's growth "over the past 5 years", and just conclude Bitcoin "has been flat", when it's just starting to take off now in some countries, and the fact that those regions don't have a ton of smartphones doesn't help.

Maybe this will help BTC take off? https://coinbase.com/sms_service

That one depends on how long Coinbase manages to stay around. They are registered as a money services business with FinCEN, but they do not seem to have registered with the states.

It's possible that they did FinCEN registration first (as that is the easiest and I believe is free, whereas the states require that you have a bond), and state registration is in the works, but I've not seen any evidence of that. I'd expect that if they were actually in the midst of getting that done, they would be trumpeting that fact widely, as it would be a great way to make themselves stand out from all the illegal-with-no-intent-to-become-legal Bitcoin money transmitters.

Perhaps that's true in general but it wasn't true of the internet:

http://www.zakon.org/robert/internet/timeline/#Growth

For an apples to apples comparison, you must start with the premise that Bitcoin is a platform. M-Pesa, built on top of multiple platforms, is a company. If you want a starting point as a comparison between between the two, you should start the clock on M-Pesa back to the origin of cell towers (platform #1) followed by SMS (platform #2). It was only when these two technologies matured that M-Pesa (a company) became possible.

I feel like you're stretching pretty hard here. Taking your second point first, why wouldn't we turn back the clock for bitcoin then to the "start" of the internet? And the author here, while sighting M-Pesa, references other new payment types too like PayPal which feels closer to Bitcoin (independent exchange mechanism that can be pulled out in a hard currency) when stating that Bitcoin is well behind the curve of where it needs to be.

Paypal is another company, based on existing platforms. Their success was directly tied to their ability to predict fraud and take real-time decisions to block or, at times allow, a high risk transaction to process (assuming their algorithm determined the cost of blocking the transaction was greater than allowing it to go through and issuing a refund). People believe their differentiator was payment processing, but its this very belief that allowed them to kill off their competition.

If you view Bitcoin as similar to PayPal, then that is at the heart of why we view this issue differently. I would compare Bitcoin much more closely to HTTP, or other protocols -- and PayPal as being akin to BitPay. When Bitcoin is compared to companies built on top of existing protocols and platforms, then I believe the premise is misguided. M-Pesa's success is directly tied to cheap mobile airtime. In Malawi, I purchased a new cellphone with minutes for $12 USD, and I was paying the airport rate. The same principal goes for Kenya -- there were M-Pesa booths everywhere I went in Nairobi, but only because of the amazing uptake of basic cell phones mostly used for SMS. Without that mature platform, the business could never survive.

Bitcoin could ultimately fail, but not due to the adoption curve as outlined in the article.

pymnts.com ...I was so hoping for a site on grape meads. Alas.

Kenyans are Keynesians, eh? Go figure...

Reply on news.ycombinator.com