Hacker News new | past | comments | ask | show | jobs | submit login
In Fifty Days, Payments Innovation Will Stop In Silicon Valley (quora.com)
406 points by thinkcomp on May 11, 2011 | hide | past | web | favorite | 208 comments



In the interest of correctness I should point out that the half a million fee is not a fee but a bond. The state does not take it but it is kept in a trust for people that may be injured by the particular money transmission business that gets a license.

Thus, if a money transmission business steals someone's money you can sue them and when you win you can take your money from the 500,000 bond so you are sure they will not run out on you.

I am not sure whether it is a good law (I have not done enough research to figure that out), but I just wanted to clarify. The quora post made it look like the state was just greedy and taking all the money for themselves.


Thanks for clarifying.

I'd like to add that this is a good practice in the financial industry, and isn't a bad thing at all. It's consumer protection.

And, $500k is actually a pretty low figure for this type of 'bond'.

For example, in Australia I believe you would need a banking license (or a guarantor with a banking license) which requires a deposit of at least $40M in to an escrow account which is managed by the central bank.


there are no consumers and vendors.

there's only traders.

this protects the criminal banking industry by keeping people out. you think these monkey bankers can't afford $1 million?

this is market entry barriers and you paranoid idiots are letting them get away with it.


I'd like to say this should not be legislated, and people should choose who to do business with on their own. If they offer the government a bond to cover losses, then people might go to that business instead of some startup. OTOH, perhaps I don't care when all I'm investing in my "Silicon Valley Facespacecash" bank is $20.

Another reason to use alternative currencies...


The problem with the market judging is a bank is worth your money or not is that by the time everybody decides that bank X is going under it's probably too late to get your money out (e.g. a run on the bank). This is why we have the FDIC.


then use bitcoin. no money to get out it only takes a few minutes.


While we're at it, why have contract law? People should be able to figure out on their own who's likely to keep their word, and the market will punish those who break it.


An argument against specific, outrageously expensive regulations for an industry does not equate to an argument against contract law. Your argument is invalid.


That works in theory. But have YOU read all the fine print when opening an account with your bank? I haven't either -- everything we do is based on some common sense and trust. If we had to verify everything we did with everyone we did it, our way of life would end.


> But have YOU read all the fine print when opening an account with your bank?

Yes. Paraphrased:

Bank agent: “You'll also have to sign this.” (Hands over a sheet of paper.)

Me: “Let's see here… […] ‘comply with the Deposit Agreement and Disclosure’. I need to see a copy of the Deposit Agreement and Disclosure, please.”

Agent: “Of course.” (Hands over a booklet.)

Me: “All right. Well, you may have to wait a bit…” (Starts leafing through the booklet. Cut to analog clock, then to clock fifteen minutes later.)

It was the same thing with signing up for public storage. “… ‘agrees to the Privacy Policy, which is incorporated by reference’. I don't seem to have a copy of the Privacy Policy here.” “Here you go,” and another sheaf of paper appears.

Note that in these cases I'm still basing my interpretation on a lot of contextual and cultural information, because while the alternative of trying to get them to lock down all the definitions separately would be amusing, it would fail. In many cases, every one of the relevant providers is in a position to drop me, whereas I'm reliant on at least one of them accepting me, so the power balance is tipped heavily in their favor.


this amounts to saying people are not competent to manage their own lives. which is true, but don't cloak it.


People choose who to bank with based on marketing and suitability of product.

No one chooses a bank based on their balance sheet.


Couldn't you buy insurance that covers the bond? Typically this is much less than a bonds face value.


You can get a surety bond [see 1817(c) of this law], which functions somewhat like insurance --- you pay percentage point premiums to keep the amount on file. This seems to be how most people handle licensing bonds.


It can be a bond or a cash payment, but either way, the money is locked up and not available to use for operations, or gone.


It will feature on your balance sheet as an asset, though.


Which, if true, only further demonstrates the nonsense that is modern accounting practice.


Huh? Why? It's your money. If you wind down your business and settle all your outstanding transactions --- which, if you're honest, you should be able to do --- the state hands it right back to you. Not only that, but you get to collect the interest on the money. It's a bond, not a fee. Why wouldn't it be listed among your assets? It's part of the liquidation value of your business!


it's part of preventing small fry from getting in.

you could easily base the bond on volume.

the more transactions you do the more you must put in bond.

this again perpetuates the unnecessary division between individual and institution.


Doesn't all regulation stifle innovation? If I wanted to sell a homes in a skyscraper made of papier-mâché, building codes would prevent me from doing that. But that's not necessarily a bad thing: the first time someone leaves their soldering iron on while they're not using it, the building burns down and we have five hundred dead families on our hands. "Innovators" tend to think about the good aspects of their ideas rather than potential downsides, so the government has stepped in to prevent the downsides from being too bad.

In this case, the government is saying, "make any new innovative money transmission system you want, but keep $100,000 in an account so when you fuck up someone's paycheck, they can sue you and you can pay the damages you owe." Not too unreasonable.

Regulation is bad for businesses individually, but it's good for society in aggregate. Progress may be slowed, but it's still happening and happening with fewer horror stories along the way.


I wouldn't go as far as to say all regulation stifles innovation. A completely unregulated market may very well stifle innovation also - i.e., when consumer trust of that market is so low as to discourage economic activity. That is a rather extreme case, though.

I think the more relevant point is that all regulations have overhead - even ones that don't charge a $500K bond. This is something governments sometimes seem to forget. No matter how innocuous your regulation, every single one simply adds to the pile of overhead businesses must deal with, and every single regulation contributes towards preventing more businesses from competing in the market.

The balance of regulation is a tough line to walk, to be sure.


An unregulated market will spawn entities that provide the services it needs. This includes security and trust. Insurance companies, for example, are entities that people trust to protect them against loss. Rating agencies are entities that people trust to provide risk assessment.

The difference between non-coercive (private) entities and government is that non-coercive entities adapt better. So, for example, if you're planning to transfer billions over many years with one bank, you'd want the bank to have all kinds of security. But if you just want a quick, cheap way to transfer 25 cents in a micro payment over the web, you probably don't care if the startup "bank" you use has a $500K bond.


The mortgage crisis says hi. The problem with free-market ideologies isn't that the market won't adjust to consumer need, it's that it causes considerable pain in doing so and will only provide a counterbalance if there's a monetary incentive.

Free market capitalism is, at it's core, evolution. It leads to amazing responsiveness, complexity, and, arguably, beauty. It also has made 99.99% of all species that have ever existed extinct. There are homeostatic forces that keep the aggregate more or less balanced, but evolution isn't good for the species involved, it simply is true.

There's no mechanism in non-random selection with random mutation to prevent Asian Carp from taking over the Great Lakes, Pine Beetles from destroying forests in BC, invasive kelp in the Mediterranean, etc. Similarly, the free market has no mechanism to prevent credit fraud, market collapse, rampant speculation, or any of the other mechanisms of economic calamity we've seen. Your "coercive" government regulation plays the same role that we play in attempting to prevent environmental disasters through the spread of invasive species. Would a global ecosystem entirely managed by direct human intervention be desirable? No. Would be be well-advised to ignore threats to ecosystems and "let nature take its course"? Of course not.

Free market supremacists seem to fail to see the forest for the trees when arguing against regulations. The Free Market isn't an end unto itself, it's an effective means to the ultimate end, which is the improvement of the human condition. It isn't the only means, and it isn't the most effective means in all cases. The role of effective, well-considered government regulation is to harness the benefits of the free market while mitigating its risks. This can only be described as a Good Thing.

Rather than rail against all regulation, we should be trying to ensure that the regulations that are created are wise and impartial. Our energies would be much better spent ensuring that legislative bodies are free from undue influence than by trying to remove their power to regulate in the first case.


The Community Reinvestment Act seems to be a pretty likely stimulus for the mortgage crisis - incentivizing "anti-racist" behavior rather than good lending practices. Worse yet, the "considerable pain" you're trying to avoid gets worse over time -- as central bankers reduce interest rates (read: print money) to jack up growth until it gets too fast then pull money from the market (read: increase interest rates) to slow down a "bubble", the yo-yo effect amplifies until pop.

Pine beetles in BC are an interesting point, as it seems like it is precisely the lack of ownership over valuable forest land that is preventing the type of extensive research we would need to protect that resource. While research outcomes are entirely uncertain, it seems that ownership and the forecast loss of value is a pretty solid motivator for solving problems.


Right, that is why We the People create regulations to protect the society that We own. General welfare and common defense and all that jazz.


In your statement:

"The difference between non-coercive (private) entities and government is that non-coercive entities adapt better."

it's not clear if you are equating non-coercive with private entities. If so you are very much wrong.

Private companies have killed, enslaved, tortured, kidnapped, etc. There are a tremendous amount of examples of private companies being coercive.

As to ratings agencies, some spectacularly demonstrated 3 years ago that they can fail and be captured (akin to regulatory capture) by other private entities.


In the short run private entities may do horrible things, it takes time for information to travel through the system. But if we're comparing private to government - government definitely takes the cake on environmental destruction and causing human misery.


> Private companies have killed, enslaved, tortured, kidnapped, etc. There are a tremendous amount of examples of private companies being coercive.

These are instances of private companies acting like governments. What characterizes private companies is that they do not use coercion (except of course when it's justified as in enforcing voluntary agreements and protecting property.)

So for example, the original post was about banking regulation. Private bank regulators could not force banks to operate according to their standards. Instead, they would rely on banks' cooperation. If a bank did not cooperate it would run the risk of being rated badly and shunned by customers relying on regulator's rating. In no case would a non-cooperating, non-conforming bank be subject to being "killed, enslaved, tortured, kidnapped, etc" by a private regulator. On the other hand, a bank which does not conform to government regulation is in danger of killed (dissolved) and its officers "enslaved, tortured, kidnapped, etc." (arrested and imprisoned.)


I like the argument that says that when things are good, they are acting like private companies, but when they're bad, they're acting like governments; ergo, private companies good, governments bad.

Can we officially acknowledge that this part of the thread --- which is notionally about bonding requirements for money transfer companies, but is now discussing torture --- has officially gone off the rails?


dpatru might be using definitions you don't agree with but he/she is still positively contributing to the discussion. Seems on topic to me, just highly abstracted: what are possible systemic solutions to the article's issue? So I've upvoted dpatru's comments that were in the negative.

Thanks for the discussion, very interesting!


I didn't downmod him, but I'm not going to upmod him for relitigating all of libertarianism on a thread about a specific regulation.


How are ideologically-pure statements -- extracted from a fantasyland where abstruse notions matter and all lines are perfectly straight -- considered a positive contribution? Particularly when they are made in willful defiance of all facts.


The whole point of this thread is about whether bonding should be required by threat of force (or torture) by government. Advocates of government regulation claim yes. Opponents claim that there are better, non-violent ways in which consumers are protected. The issue is precisely the use of force. Nobody is saying that bonding is bad or that private rating agencies could not require bonding as a condition of endorsement.


An unregulated market will spawn entities that provide the services it needs.

Information and resource asymmetry are not easily overcome; how is your argument any different than a naive assessment of economics that fails to take into account the impact of information asymmetry on the decisions of otherwise rational actors?


> Information and resource asymmetry are not easily overcome;

The best way to "overcome" information and resource asymmetry is through a free market. In particular prices and word-of-mouth/"the Internet" do this work as well as possible. If a company does a good job for a good price, its fame will quickly spread. If it rips off customers, its infamy will quickly spread. No need for government to get involved a priori, although government can get involved in fraud prosecutions and to help defrauded victims get their money back.

> how is your argument any different than a naive assessment of economics that fails to take into account the impact of information asymmetry on the decisions of otherwise rational actors?

Not sure what you mean by this, but free market competition and innovation through new/better services is a way rich societies increase wealth. Government regulation just slows down the process.


If a company does a good job for a good price, its fame will quickly spread.

So what happens when that company is bought out by new owners, starts skimping on product safety, and 1,000 people die before the market notices?

If 1,000 people die, is the company held liable by the government?

If people band together to enforce preventative measures in the community (even through private means), such that such a thing doesn't happen again, have they created a regulatory government?


> If 1,000 people die, is the company held liable by the government?

Intentional or negligent killing is prosecutable. I doubt, though, that not requiring bonding for a money-transfer startup will cause the deaths of 1000 people. As to businesses generally, tort lawyers as a class make a pretty good living holding businesses accountable. The Ford/Firestone controversy comes to mind. (http://en.wikipedia.org/wiki/Firestone_and_Ford_tire_controv...) In that controversy plaintiffs suing Firestone and Ford alleged that people died because tire failure caused rollover accidents in Ford Explorer SUVs. As a result, both Ford (maker of the vehicle with the defective tires) and Firestone (maker of the tires) had to pay a lot money and their stock price suffered. Also, one of Firestone's tire manufacturing plants was closed and Ford and Firestone no longer do business together. So there are consequences when private companies screw up.

On the other hand, when government screws up and thousands of lives are lost, like when it holds up the sale of a new drug, fails to defend the airspace against terrorist attacks, or goes to war on bad intelligence, there don't seem to be consequences.


I can always tell when a site has gone past the point of no return when comments like this are simply downvoted to oblivion. If you disagree state why.

but whatever, I always call these things. Internet communities refuse to be elitist about quality because they 1. believe in democracy and 2. are made up people who like to be inclusive because they were excluded as kids.


Sorry, a lot of us are just really tired of glibertarianism.


sorry, a lot of us are just really tired of states killing people.


An unregulated market will spawn entities that provide the services it needs. This includes security and trust.

Well of course, how else do you think democratic governments were invented? The market created them.


You're right. Governments have gained the power they have because they offered people a better alternative. I read somewhere that Napoleon was welcomed as a conquerer into parts of Italy because his government was a lot better than the local government. The problem with governments, though, is that they are fairly easy to establish, but, because they claim a monopoly on coercion, they are hard to replace when a better alternative appears.

For example, all of a city's private garbage collectors may do a poor job, so a very capable mayor may convince residents to allow the city to do the collecting. Years later, when situation is reversed and the city is doing a poor job in comparison with private collectors, it will be a lot harder to replace the government collectors with private ones. The correct response to poor private-sector service is not to bring in the government, but to bring in more private competition.


You were alive three years ago; no? AIG, "AAA" rated CDOs, any of that ringing a bell?


You weren't paying attention.

The ratings were issued by companies that had been given a monopoly by govt. Securitized mortgages were a creation of govt.

The idea that you're missing is regulatory capture combined with govt encouraging transactions that didn't make economic sense otherwise. (One of the underappreciated consequences of RC is that it amplifies "private" bad behavior and shuts out good behavior.)


That's not fair. The major ratings agencies were unregulated, all were founded privately, and were only designated NRSRO's after they had captured the market. Meanwhile, companies can apply to be designated NRSRO's, and many companies have been so designated.

The government didn't corrupt Moody's. Commerce did. The government supplied no oversight, and Moody's sold its ratings to the highest bidder.


"govt encouraging transactions that didn't make economic sense otherwise"

More like the companies selling mortgages were desperate to get the loans off their balance sheets so that they could release the capital for new mortgages. As they didn't keep the mortgages for very long the level of risk they were accepting was low so they didn't do much checking on whether people could actually pay or not - they really didn't care.

I remember comparing my experiences of first getting a mortgage 20+ years ago where it was a difficult thing to do with how younger colleagues described things immediately before 2008 and things were totally different - they were giving mortgages to pretty much anyone who asked not because the government made them (I'm not in the US) but because it made them a lot of money!


> More like the companies selling mortgages were desperate to get the loans off their balance sheets so that they could release the capital for new mortgages.

Selling them to GSEs. You do know what the "G" stands for, right?

And speaking of the GSEs, they lied about how many of the loans in their portfolio were subprime. As a result, no one knew that those loans were as common as they were, which threw off everyone's risk calculation.


Actually, I was more thinking of banks like Northern Rock in the UK.


"An unregulated market will spawn entities that provide the services it needs. This includes security and trust. Insurance companies, for example, are entities that people trust to protect them against loss. Rating agencies are entities that people trust to provide risk assessment."

Do you have any examples from reality where this has happened? Are insurance companies and ratings agencies really trusted agencies that would function even better in an unregulated market?

I posit that with no regulation, these entities would become even more unscrupulous than they are now.


All those work on the principle of trust despite not being regulated:

- http://en.wikipedia.org/wiki/Certificate_authority

- eBay feedbacks

- http://www.escrow.com

etc.

Furthermore, magazines, blogs and retail store all act as rating agencies in some ways. Magazines have a strong incentive to only recommend quality products since their reputation is at stake. The same goes with retail store. They make sure they sell quality product because their reputation is at stake and they don't want to lose business.

Of course, the incentive for building those kind of companies is very low given the uncertainty that the government might decide to assume your role, putting you out of business.


CAs suck and are unaccountable. Any of them can (and many of them have) issue a certificate for a site to an attacker.

eBay feedbacks can be gamed and I generally don't trust them.

Escrow services work, but they are generally regulated.

Your retail store example is laughable. I'm sure Walmart really cares about the quality of its products.


> CAs suck and are unaccountable. Any of them can (and many of them have) issue a certificate for a site to an attacker.

> eBay feedbacks can be gamed and I generally don't trust them.

Do you think the government would do a better job at providing CAs or rating eBay sellers? Do you think the government is somehow immune to fraud?

> Your retail store example is laughable. I'm sure Walmart really cares about the quality of its products.

"I'm sure Walmart customers really care about the quality of products they buy at Walmart."

Walmart offers what customers want: cheap prices at the cost of lesser quality. Luckily we are still free to buy low quality products, because the government might "fix that" someday.


Do you have any examples from reality where [an unregulated market spawning entities that provide trusted rating services] has happened?

http://en.wikipedia.org/wiki/Underwriters_Laboratories

http://en.wikipedia.org/wiki/Consumer_Reports


The question was not whether unregulated markets spawn "trusted rating services". It's whether society could function with private rating services in lieu of regulation.


http://en.wikipedia.org/wiki/Moving_the_goalposts#As_logical...

Go back and read your exchange. It looks to me like he answered your question, and you're changing it now.


Actually, he is not moving goal-posts, he is just asking for the second half of his originally stated question to be answered. I know it is silly, and probably sounds like nitpicking to you, but please, don't make false claims, particularly when they are just veiled ad-hominem.


But it does appear that he has a legitimate question (whether it is different from his original question or not)


These are both government regulated non-profits. Compare to the BBB.


Why? The regulations erect great barriers to entry for would-be competitors, solidifying profits of the current market players and allowing them great latitude for slacking off and sub-par performance.

We're seeing the same thing now with the money transmitters.


This is true -- the hardcore capitalist would argue that the plaster industry has been hamstrung by the quaint requirements from building code to use wood, steel and concrete in construction.

There's a give an take here... the government has a perceived duty to protect the citizenry against bad actors, that that introduces a bias against sudden change aka "innovation". "Innovation" isn't always good -- just like to some people, certain "freedom fighters" are "insurgents".

On the other hand, there are entrenched business interests like Western Union (and payday lenders in "red" states) who use their influence to keep the status quo around, even if it hurts consumers. (ie. the poor and ignorant who spend $10 to send $100 to someone)


I think stifles competition is more concerning than stifling innovation per se.


The money-moving startup that can't cough up $25k to post a bond isn't scaring PayPal or Western Union.


Regulations can also be good for individual businesses, bad for their competitors and bad for society in aggregate.


Government regulation is in just about every case either not needed at all or better provided through non-coercive institutions. Since government innovates more slowly than private industries, government regulation tends to become outdated, going from common-sense prudence to arbitrary burden on innovation. For example, right now the Mississippi River in the US is flooding, destroying many homes. This is a huge loss partly because homes are very expensive. Maybe it would make sense to have some housing constructed of paper-mâché, not try so hard to resist disaster, but instead concentrate on making it cheap to rebuild. Government building codes prevent this kind of innovation.

Regulation reflects a lack of creativity. It says, "There is no other way to do this." But maybe there is another way and the politicians just can't think of it. They shouldn't be allowed to limit other, more enlightened people.

The above assumes that regulations are honestly designed to promote the public good, a dubious assumption. Often I suspect that regulation is designed to protect influential but inefficient businesses against competition.


You wrote:

"Since government innovates more slowly than private industries, government regulation tends to become outdated, going from common-sense prudence to arbitrary burden on innovation. "

The space program spawned a tremendous amount of innovation. Public research universities throughout the nation innovate on a grand scale. The internet is an example of government innovation.

Regulation does not reflect a lack of creativity. It reflects an acknowledgment of a problem and steps to address the problem. Sometimes side effects occur and are bad and the regulation needs to be reanalyzed. Sometimes there is regulatory capture (by private entities).

Government had to mandate the use of seatbelts. This led to airbags because after the government made safety an issue innovations were made in this area. They probably would have been made without government intervention but government got the ball rolling and the innovations occurred sooner as a result of government.

Regulations can be good, bad, or neutral. But talking about "non-coercive institutions" makes your view look extreme.


The air force actually developed seat belts from scratch. http://en.wikipedia.org/wiki/John_Stapp demonstrated that the human body could withstand vary high G's for short periods, and because more pilots where dieing in car accidents than airplane accidents the air-force was willing to do some vary important basic research.

Which brings up the basic issue, the incentives of car makers / home builders / bankers to create safe products does not line up with society's benefit from safe products without some external input.



That's an interesting thesis, but can you name one good building material that you can't actually use to build a house? People talk about regulation in the US like it's hindering any form of innovation, but there is surprisingly little terrible regulation that's actually enforced. Which is not to say you can't point to regulation that harms individuals but rather it's much harder to find things that harm society over the long term.

PS: Not to side track to situation; we are talking about building materials and you can use a wide range of things including straw and dirt in the US so what's missing?


> can you name one good building material that you can't actually use to build a house?

http://www.youtube.com/watch?v=PKH0qoaXR88 Discusses how building codes hinder new and "natural" building methods. In particular, the speaker mentions that older codes have failed to keep up with better techniques for strawbale building.

The deeper issue is not whether new materials can be used to meet code requirements, but whether new ideas of what constitutes a safe dwelling are allowed. Building codes are based on assumptions about what a dwelling should be like. For example, the floor should be relatively stiff. If a floor is too bouncy, it won't meet code even if it's strong enough to support the weight. Walls can't be too flimsy. The idea of a house that just safely collapses in a windstorm or floats away in a flood is, I suspect, outside of most building codes' conception of a safe building. The code focuses on the idea of a strong, permanent structure that will resist nature's forces. If houses were plants, the code would only recognize big trees. However, there are also grasses, which much cheaper to produce and also resistant to nature's forces in their own way (they bend). Yet building codes don't recognize structures that are inspired by grasses, only big immovable trees.

I don't know if grass-like houses are a good idea or not, but I'm sure that if we expand our thinking we can come up with better shelters. The problem is that it's hard to innovate and government regulation only tends to make it harder.


Houses are not websites when they fail people often die.

In that video he is trying to build long term structures and he complains not about the building codes, but the fact that there is no data to support the safety of new building ideas. The whole point of a building code is you get to avoid running the numbers, if you build a structure using these methods with these materials it's safe. There are rules for temporary structures like tents, but if you want it to last for 50 years you need to demonstrate it's safe. And, if he was capable of demonstrating he could build a safe house out of toothpicks and spit he could have done so because the only limit on his construction was providing number so an engineer could demonstrate the safety of the structure. And, the compromise of "build a load baring structure from well understood materials and fill it however you like" is vary open.

PS: You can build a floating house.


No I don't think most regulations are good for society in aggregate, including building codes. What if we had software codes to make sure we write secure software, would that be a good or a bad thing?

Take this payments law for example. Its intended objectives are admirable but I don't believe it will accomplish them nor benefit consumers in the long run.

By artificially imposing a high barrier to entry (licensing fee), the consequence will be to encourage the formation of monopolies.

Furthermore, payment companies will have a lesser incentive to secure their service or build a reputation of trust since consumers will be led to believe that "all payment companies must be secure since they are all approved by the government". What was once an important competitive advantage will lose a lot of its importance.

There are probably other perverse effects I could think of but my point is that things are much more complex than they might seem.


> No I don't think most regulations are good for society in aggregate, including building codes.

Nearly every large earthquake proves you wrong on building codes. Compare deaths for a given magnitude quake in areas with strong building codes vs. areas with weak or no building codes.


That's not the point. Alcohol prohibition reduces alcohol related deaths. Doesn't make alcohol prohibition a good thing for society.


Alcohol has some societal benefits. Some people find it pleasurable, it helps some people overcome anxiety in social situations, and it helps programmers cope with the horrors of programming in Java.

I don't see what societal benefits structurally unsound buildings provide.


It doesn't impose a large fee. It requires a bond. The details matter.


You're right but in this case, the detail doesn't really matter. The point is you need a lot of capital to enter the market.

Once you are licensed, nothing prevents you from committing a fraud except for the potential risk of losing your bond. This risk could be insignificant given a large amount of money to defraud.

All this law does is guarantee that from now on, only rich people are allowed to commit fraud.


Is the bond too small? Is that what you're arguing with that second sentence? I can't follow.


What I mean is that this law which is intended to prevent fraud does nothing of the sort. All it does is guarantee that the people who commit fraud were able to pay the bond. What's the point? If they do commit a fraud, they'll probably steal much more than the value of the bond anyways. All this regulation really does is stifle competition, I'd eliminate it altogether.


The risk that the bond mitigates is not that your money transfer enterprise is a criminal conspiracy. It is that you are incompetent. The concern is that after the first or second instance in which you lose a 5-figure sum of money for a client, you'll pack up and leave town.

I think we can agree that's a far more likely scenario than premeditated fraud.

Incidentally: $500k is the floor of the bond value needed. It scales up to $7MM with transaction volume. Personally, I think they should uncap it altogether.


His point is that the bond is too large, and that's _just_ for California. In the linked quora post, it was specifically pointed out that there are 43 other states where one has to do the exact_same_thing where the bonds vary from $10k-$1M.

Edit: Check out PayPal's list: https://www.paypal-media.com/licenses


You don't actually have to pay the $500k. A new business with no history (good or bad) might pay $25k to post a licensing bond.

I don't have anything to say about the 43 other states that want bonds to conduct money transfers in them, but some of the coverage here seems a tad breathless.


tptacek: Perhaps it is breathless, but I think many of us find it interesting. States noticeably absent from Paypal's licenses: New Mexico, South Carolina, Georgia (heavy banking industry), Rhode Island, New York (heavy banking industry), and Nevada (heavy gambling).

I might put together a spreadsheet this evening if I have time.


> All this law does is guarantee that from now on, only rich people are allowed to commit fraud.

So you're fighting for the little guy to be able to commit fraud too? FWIW the rich guy is not just risking losing the bond, but also going to jail. That's the threat, the bond is just so customers can get paid.


> So you're fighting for the little guy to be able to commit fraud too?

Absolutely. Given that this regulation doesn't help with fraud, why not at least encourage competition.


Why require bonds for anything? Bonds aren't going to keep construction companies from committing fraud. They won't keep movers from making off with people's goods, or folding up shop when they get in traffic accidents. For that matter, why require millions of dollars of insurance coverage to work on infrastructure projects? It won't keep me from committing fraud, if that's my real goal.


I have nothing against private insurance. I have something against laws which make it mandatory. I think people should be free to decide for themselves what risk they are willing to take. For instance, the choice to do business with a relatively unknown payment service which offers low transaction fees or great customer support, at the risk of losing money.


I have a hard time getting too worked up about the prospect of squelching the money transfer company that can't afford 1/10th of 1 FTE to post a bond. That same company can't afford to secure their software (software security for a money transfer application is almost certainly more expensive than the cost of a 500k surety bond).

Meanwhile, if you're against basically all licensing and bonding, you're naturally going to be against this one too. Personally, I think that if we're going to require bonds to move furniture, it seems sane to require a bond to move cash.


I agree with you on most points.

I wouldn't do business with a payment company that can't afford a 500k expense. Requiring insurance for moving furniture is also important to me, I wouldn't let a company move my furniture without them offering a solid warranty.

What I question though is whose role it is to impose those requirements, the government or the customer? I believe it should be the customer's role.

> Personally, I think that if we're going to require bonds to move furniture, it seems sane to require a bond to move cash.

It is sane, but why not let companies choose whether or not they want to get licensed and let customers choose whether or not they want to take the risk of doing business with an unlicensed company. Note that I do not object laws that deal with misrepresentation, lying, breach of contract, etc.

Anyways, as you said, it's more a question of principle than anything particular about this specific regulation.


Ok, meanwhile, we're howling at the moon about requirements that are, in the scheme of how the government already regulates mundane businesses, totally business-as-usual.

If you don't believe in regulations at all, you don't believe in this regulation. Fair enough!

But if you're basically happy that we have an FDA and an FDIC and an NTSB and an FAA (as artificial examples; substitute your favorite California regs bodies): how is it unreasonable or surprising that California would want money transfer companies bonded? You can't build back porch decks without bonding. You can't move pianos without bonding. You can't sell cars without a license bond. But we want people to move cash without them?

Reasonable people, I suppose, can disagree about whether the bar for accepting and moving cash from people should be as high as the bar for re-siding a garage. But I don't think reasonable people can call the bar a conspiracy against the public.


Don't mistake me: I do not think there is any sort of conspiracy going on and I actually believe the government's intentions are good. I simply believe they are wrong in how to achieve those intentions.

> But if you're basically happy that we have an FDA and an FDIC and an NTSB and an FAA (as artificial examples; substitute your favorite California regs bodies): how is it unreasonable or surprising that California would want money transfer companies bonded?

I might seem pretty emotional about this, but in fact I'm not even American ;). Among the things you mentioned, I only know about the FDA and I do think Americans would be better off without it, for pretty much the same reasons I outlined previously (as a side note, I believe Health Canada bases its own regulations on the FDA). Milton Friedman explains it better than I can here: http://www.youtube.com/watch?v=OazixMEY9I0


So you're against the requirement to carry car insurance?

What happens when someone runs over you with a car and they have no liability or personal injury protection with which to compensate you for your medical bills and they have no money and therefore are judgment proof? Too bad, so sad? Shouldn't have been walking down the street?

Yes, this happens now, but now it's a criminal act to drive a car without insurance so you're breaking the law by potentially putting other people at risk.


I'm curious why someone thought this was an invalid point (I just modded it up). Aren't the principles involved in staking money for insurance, because you might cause damages that you can't repay personally, pretty much the same as those involved in staking a bond because your business might damage its customers?


How does requiring a bond not help with fraud? Consumers who have been refunded from frauds with bond money certainly disagree with your statement.


>Doesn't all regulation stifle innovation?

there are regulations and there are regulations.

>"make any new innovative money transmission system you want, but keep $100,000 in an account so when you fuck up someone's paycheck,

just for example, a reasonable regulation would just limit the per-transaction(or per-month per-person, etc, ...) amount to, lets say, $100 for unlicensed/unbonded companies.

On the other side, the regulation that have the same $500K bond for any company is just very favorable for big established players. The $500K is nothing for AMEX or PayPal, and provides virtually no coverage for the risks these companies presents to consumers. Ie. if any of these companies fucks up the paychecks in their system, the $500K would be just a drop in the bucket.

Even current banking regulation is designed way much better - basically it forces to keep reserves as percentage of the liability amount.


> Regulation is bad for businesses individually, but it's good for society in aggregate. Progress may be slowed, but it's still happening and happening with fewer horror stories along the way.

This is not necessarily the case. Often times, regulation is put in place to address some high-profile case, but ignores the consequences of the regulation.

Currently, infants less than two years of age can fly in an airplane without a separate ticket. Years ago, after some instances where such children were injured/killed in accidents, there was talk of forcing parents to buy separate tickets for infants and require them to use an approved child seat. However, had this regulation gone into effect, the extra cost of flying would have forced many families to drive instead of fly, resulting in a greater amount of injuries/deaths.


Nope. Sometimes regulation spurs innovation, forcing businesses to go from "business as usual" to "how do we deal with this?". Air pollution regulations in California helped drive a lot of innovation around cleaning up car exhausts, for example. Without that restriction, movement in that area would have been very slow.


People should have full information.. I should know if the building I move into is made from plaster of paris. However, if I still want to move into it... I shouldn't be passing on my genes.


They want companies that operate like banks, such as PayPal, to be licensed. I'm not sure I see the downside. PayPal has a shocking record of abusing it's customers, and it can only do so because it isn't bound by the same laws as banks are.

Personally, I think if you rely on PayPal/etc for your business, you're asking to get financially raped.


If they wanted companies to be licensed they could just require a license. Instead they require a license and a bond worth millions of dollars in some cases, nominally to protect consumers. Yet if you have a million consumers in a state and a $1,000,000 bond, that means that if the company defaults, each consumer gets back a dollar.

This doesn't sound like consumer protection to me. It sounds like an artificial barrier to entry.

Also, PayPal has had licenses in all states that require them since its IPO, and clearly there's no connection between this issue and customer service.


Well, there is a (flawed) line of reasoning that says that if you can come up with $1M then you must be more trustworthy than someone that can't.


I don't think it's that flawed.

Would you elaborate?


Trustworthiness is not tied to money, the ability to make large sums of money, or the ability to convince others to loan you money/invest in you.


Depends on context; in some cases it absolutely is.

Courts often "trust" people to go free temporarily if they can post bail pending trial. Lenders will "trust" you with their money in the form of a loan if you can post collateral.


Yeah, but "ability to pay a court-ordered settlement" correlates pretty highly with having some cash reserves.


Yes... but you don't usually have 1 million customers and owe them all money at once. That's just daft.


Anyone that has a balance in PayPal is owed money by PayPal. So, any account that does more than just send out funds to other accounts (that is, purchaser-only accounts) is owed money.

Same for a bank. Any customer with an account that's greater than zero is owed money by the bank.

It's not a stretch at all to have a million customers owed money.


My apologies. The excess over your reserve per customer is unlikely to equal the account value per customer. Wait, are paypal under the auspices of the financial regulator? Who approves their reserving methodology?

My spaghetti monster, do they even have to hold a reserve, or can they just spend all your money on the derivatives market?


Sounds to me like the bond needs to be a couple tens of millions, then.

If you want to get into banking, you had better have the chops to be able to treat your customers correctly. It's not a hard concept.


This won't affect PayPal in any way. They have plenty of money to cover the licenses and bonds and whatever else. They're part of the status quo and are not considered a threat by Visa/Mastercard in any way, shape, or form. In fact, PayPal is one of their best customers.

If you have a bone to pick with PayPal, you ought to be wishing for more competition, not supporting barriers to entry for their competitors.


This is the financial services industry cranking down regulations to prevent competition as the coming onslaught of digital payments is about to take place.

This is nothing but downside as it relates to innovation in the payments space. These companies like their business models and don't care for it to be disrupted. They love regulations at this point.


>PayPal has a shocking record of abusing it's customers, and it can only do so because it isn't bound by the same laws as banks are.

Some might argue that they can only continue to do so due to a lack of competition, competition which is stifled by the very legislation ostensibly aimed at helping consumers.


I would agree, but the price is steep for no reason other than apparently to keep new players out, and there isn't any indication that the license, if acquired by e.g. Paypal, would do anything to curtail Paypal's abuses (they are a shitty company I agree).


Here in Europe, PayPal (which is acting like a bank) had to get a banking license (like any other bank). As far as I know they didn't go bankrupt yet, so it can't be a wholly bad thing.

If I drive a vehicle that acts like a car, I need a drivers license. If I run a service that acts like a bank I need a banking license.

I really don't see a problem here.


> I really don't see a problem here.

I tend to agree. Surety bonds have been effective for a long time to ensure completion of services in case of fraud or failure on the part of a contractor or service provider. States have been requiring surety bonds for many kinds of licensed professions for years, building contractors being a big one.


They are talking innovation; sure Paypal can get licenses like that, but a startup can't. Most (by far) startups don't have 500k to lock away. And most don't have the money to pay the license fees. So yes, it's an issue as far as innovation of payment services goes; now suddenly you NEED to get millions in investment and that all doesn't go into the product but instead into government crap.


And most startups don't go into banking. Seriously, if you're going into that kind of business you should have that kind of money behind you.


Why is banking (specifically money transfering) different than other kinds of business? Seems to me that money transfer is a particularly easy business for the market to regulate. A failed money transfer is a lot easier to spot than, say, a dangerously defective physical product.


A fraudulent one isn't though.

You're also dealing with especially sensitive data, life ruining data if it gets into the wrong hands.


You realize Google Checkout is often lamented on HN because it doesn't operate in many European countries? It is exactly for these reasons.

Much like copyright laws are abused for business protection, so are regulations. Certainly within the US it would be better to have a federal license framework in order to minimize friction.


What, that Google doesn't have money to put up bonds or buy the necessary expertise? I don't get your point.

More likely it's Yet Another Google Product They Can't Be Bothered To Finish (TM).


It's not about having enough money, but whether it is financially worth it. It's an ROI calculation that involves navigating several regulating bodies with what appears to be minimal payback. It certainly isn't a technical limitation.


If money transfer isn't lucrative, it seems weird to blame regulations for the fact that Google isn't staying involved with it.


If your revolutionary banking startup can't dredge up $500k in the land where Color managed to get away with $41 million, then your banking startup probably isn't such a good idea to begin with.


I'm not saying the idea is totally pointless, but the $500k needed regardless of the scale of the business seems a bit excessive. Maybe there should be a sliding scale, so that if your transaction volume is $10k, you need a bond for $10k, etc, up to a ceiling? That would at least make it possible to start out testing whether it works without dumping a huge amount of cash.


It does scale with transaction volume, but the floor is $500k. But it's explicitly payable via a surety bond, which works like insurance; applicants with good credit history might pay just a few thousand dollars to get one.

There is also a class of money transfer company ("A licensee that engages in receiving money for transmission") that only needs a $250k bond.


Here is the problem: PayPal was born in the US. Among other things, probably because it was easier for them to do business there. PayPal couldn't have been concieved in Europe.


> I really don't see a problem here.

I do: it prevents innovation. People should be allowed to run an unlicensed banking service, and customers should be allowed to accept the risks that go with it (by all means, require customers to sign a form saying they understand it's an unlicensed service and they could lose all their money.)


Well, losing all your customers' money is certainly an innovative way to run a bank.


But does PayPal getting a license mean that their actions towards customers have to change? Or do they just have to pony up some dollar amount to continue screwing their customers in the same way they always have?


I completely agree to this. PayPal somehow has been evading this problem for a decade but finally the government is stepping in to square away this mess. How many horror stories have we heard about PayPal locking up people's accounts for no good reasons?


So, a new startup comes along, and things are going well. The have new, innovative ideas that satisfy the needs of consumers. They manage to get a few hundred thousand dollars from a VC so that they keep operating for another 6 months. This startup offers a service that is cheaper, safer, quicker and generally more efficient that what is currently available. All of a sudden they'll be breaking the law unless they hand over half a million to the government. They can't find the money. They're done. You never hear about it. Other new companies that might have added value in the marketplace come along, look at the regulatory hurdles, and don't even try.

They might have provided a service that would be beneficial to the consumer and they might have hired a bunch of programmers once they had established themselves. The established companies already in the market would have had to lower prices and become more efficient to compete. Instead, the competition is crushed, you never hear about it, never know about it.

This is what this law is designed to accomplish. Nothing to do with Paypal. Stop talking about Paypal.

Many regulations serve the interests of politically well connected powerful oligopolies in this way. This is just one example. The reason put forward is always that this is some way of protecting you. That, if this regulation didn't exist, these companies wouldn't have to be responsible. In reality, it strengthens the hold on the marketplace of the oligopoly already in place and makes you more vulnerable to them.

Keeping companies out of the marketplace stifles competition. This is not good for you, the consumer.


I generally agree. What most people often don't realise is that a lot of regulations that are allegedly in place to protect to consumer are actually there to prevent established interests.

This is particularly true of licensing. The standards for licensing and the like are often co-opted by the industry/profession being licensed. As such, they use licensing to raise the barrier to entry, effectively limiting competition.

One example that comes to mind is that the aestheticians in Texas at one point were fighting to require people that do threading to get a license. I believe said license would require thousands of dollars and hundreds of hours to obtain. Ultimately, there was absolutely nothing about threading the license at all. I am not sure if this law was ever passed, but it is merely on example of licensing works in practice to hurt competition, raise prices, and hurt the consumer.


> All of a sudden they'll be breaking the law unless they hand over half a million to the government.

Or put up a few coins for a surety bond. If you're a financial startup and your investors don't trust you enough to put up bond money, you should probably find a different niche (or investors).


So, we assume they're going to commit some type of fraud or crime up front so they have to submit to posting bond, but the investors are supposed to trust them with another half million. Or, as you put it, a few coins. Wow, in your world a half million is a few coins. You live in a universe different than mine.


Fraud or criminal elements are not required. Old fashioned screw ups and security lapses do the job of losing other people's money just fine too. The bond requirements are just a method of basic consumer protection. Money transfers are a serious business with serious consequences, if your investors don't trust you with either the funds for the bond or the payments on a surety bond, you really are in the wrong business.

A surety bond means you only have to pay a small portion of the total. It's similar to insurance. You pay a company with larger resources to vouch for you in case you have a problem. Instead of ponying up (the refundable) $500K you pay something like $25k. You don't get it all back if you close up shop, but you didn't need to put much capital out there. This is how a lot if not most licensing bonds work.


If said company was truly innovative and satisfies the interests of consumers, I'm sure it won't have much trouble raising the cash necessary to continue being a law-abiding citizen.


The implication of what you are saying is that the half million for this startup company will present no extra hurdle and will have no impact on their ability to compete and get started (as you stated it "won't have much trouble").

This begs the question: what amount would be a barrier to entry into the market, 1 million, 10 million? Is there any amount of regulation and cost that you would consider a barrier to entry into the market? If as you imply a half million is not a barrier ("no trouble"), why is not a barrier, why is not a hurdle?


Part of the problem is that people here are so used to cheap software startups that their frame of reference is skewed. We're talking about at $500k escrow or a (as some claim) $25k bond. $25k is peanuts in the cost structure of starting most businesses.

It takes upwards of $500k just to open a McDonald's franchise but they're still growing like weeds.


Reminds me of mortgage lead generation.

If you want to take people's names and then pass them on to mortgage brokers, LowerMyBills style, you have to register as a mortgage broker in almost all fifty states. Requirements vary widely from state-to-state but included obnoxious things like high fees, local residency requirements, and annual written tests.

The end result - mortgage lead generation didn't stop, but it became something that was difficult to legally bootstrap. You needed one hell of a GC, and he needed to have experience. Venture-backed companies had a big advantage in the space.

As long as the rewards are high enough, I suspect the same will be true here. Nothing great, but not the apocalypse.


His argument makes no sense. He says that 43 other states have such regulations, so that if you want to do business nationally you have to get licensed in each, at costs ranging from less than California's new cost to more than California's new cost.

So what is the significance of California making it 44 states? I doubt a new payment system that only works in a handful of states has much of a chance, so if regulation kills payment innovation it would have have been killed long ago, when the majority of other states adopted regulation.


The difference: if you start a business in California, you have to care about California law, and federal law, but not necessarily any other states' laws. Similarly, Internet businesses don't legally have to collect sales taxes for states other than their home state (and most don't make the mistake of putting their HQ in a state with sales taxes), though various states have attempted to (illegally) claim otherwise.


You know... If you look like a bank, walk like a bank and quack like a bank, you should obey the same regulations banks do.

Some laws exist to protect the establishment. Others exist to protect society from it. I am not sure where this one sits.


How does this affect SaaS billing companies such as Chargify, Spreedly, Recurly, et. al. since they're merely passing along information to a gateway? I assume not at all, if I understand what I am reading.

What about something like Stripe? It sounds like they may be affected?

This does not sound particularly fun or exciting.


Credit and debit cards are issued by banks and banks are exempt from money transmission laws.

This is why you see most startups gravitating toward niche markets that revolve around the existing infrastructure, but those niches are becoming awfully narrow.

In contrast, I contend that the system itself is flawed and needs to be replaced.


Ah, thank you for explaining that to me. I some how missed that from reading the link. Though, it was quite verbose and I am, admittedly, partially into a celebratory post-coding-session bottle of wine.

I agree. The system most definitely appears to be flawed.

Isn't Paypal forced to operate as a bank in Europe as opposed to the way they exist in the U.S.?


Yes, Paypal operates with a Luxembourg banking license in Europe (http://en.wikipedia.org/wiki/PayPal#Bank_status ).

Note that banking in Europe is in part regulated by national law of the member states. To operate as a bank in Germany, you need not only 0,5-5 million € as starting capital (depending on the type of financial transactions you are getting into), but you also need two reliable* executive managers who have at least three years experience leading a bank of the intended size (or similar experience).

Thus, the barrier to entry is rather high in Europe, too. But after that, if you get big enough, the tax payers will pay all your bills, like in the US.

* no criminal record and no other facts known that would cast doubt on their trustworthiness.


Yes, Europe is full of regulations, and I don't think anyone that's read HN for long enough isn't tired of hearing about a lack of innovation in Europe vs. the US (especially SV).


Yes, Paypal is a bank in Europe. They have a bank license in Luxemburg and have devised some pretty nifty technique's for porting that around the rest of the EU.


Can't I just live as a shareholder in silicon valley, basing my limited company and shill directors in Alabama, and leasing the limited company software produced by my other company in California?

Requirements:

2 limited companies.

Licensing agreement.

Alabama-based Non-executive directors (cousin Ed and Chuck).


Then your other company in California have to post the bond. The software has nothing to do with it.


The company in California produces software components. It is not involved in any financial transmissions whatsoever. That is my point.


Then as long as you don't transfer money in California, you're fine. If you do, you need a license. Note the part of the law that refers to the travel costs incurred to get an inspector to whatever part of the country you domicile the business-end of your business in.


>Oh, and if you want to do business nationwide, you'll need 43 more of those licenses from almost every state.

Does this mean California is just another state in a long line of similar laws? Or is there something unique about California's situation in this aside from it containing Silicon Valley?


California is also the home of many tech-savvy people that buy stuff online ... if you can't get a license to operate in California and maybe New York, you may as well not exist.


While I agree with other commentators that the prose is a bit over the top, its an interesting move on the part of financial interests. (there was a great report in the Murky News that special interests are writing the laws [1])

Clearly financial transactions have provided a means for the state to both detect and to a lesser degree mitigate criminal activity. The phrase 'follow the money' works because such activities are at their heart economic in nature (trying to enrich the criminals) and understanding the path to that enrichment often allows law enforcement to disassemble the criminal enterprise behind it.

That being said, I would not be the least bit surprised (although I did not read through any floor debate if there was any) if the people who voted for this law did so believing that they were addressing a great threat to society, which is the proliferation of payment processors which are not accountable to the state for tracking or even reporting their transactions in any form.

That does cut into 'innovation' as thinkcomp points out but it doesn't preclude it. It changes the conversation from 'build it and see what works' to something more along the lines of a slow plod from idea to implementation while bringing legislators and law enforement along. The bad news is that criminals are motivated to stay ahead of 'the system's abilty to track their money, consumers need a value proposition and a trial period before they switch so adoption rates are probably (not data driven, just intuition here) lower.

Perhaps the next innovation is a payment processing company that provides the regulatory shield for innovative sub companies.

[1] http://www.mercurynews.com/politics-government/ci_15517816


Does contractor licensing and bonding cut down innovation in contracting? Aren't most contractors firms of less than 10 people? Are there not hundreds of them? How can you defend the argument that a yearly bond premium payment that is less than many companies pay for liability insurance is a serious drag on innovation?


Hmm, I'm curious how you equate building a contracting business is equivalent to building a new payments system?

I agreed that the article's portrayal of the fees as extortionate was hyperbole but agreed that raising the barrier to entry would slow down the creation of new payment systems. If you accept that the rate of innovation is related, at least linearly, to the creation of new companies in a product space, then reducing the creation rate would necessarily reduce the rate of innovation.

An alternative look at your question with contractors would be to ask "Has the licensing and bonding requirements cut down on innovation in contracting business models?"

I don't know how to answer that question without first removing the licensing and bonding requirement and then observing the resulting market. It certainly mitigates crime in the contracting market to some extent by making it more difficult for a criminal to pass themselves off as a bonafide contractor. But the current state of affairs don't say about whether or not the contracting market would be more innovative without those restrictions.


What's the jurisdiction? Some freshly made illegal company can operate to California residents and be located outside of it (even the USA.) Are they going to block the website to California residents?


And so we have one more reason to use Bitcoin.


... whose exchange rate against the dollar has gone up by a factor of 6 over the past weeks? If you had any debt nominated in Bitcoin, you'd be screwed.


...which is why you should define debts in terms of some weighted (USD, EUR, Gold, BTC, ...) currency basket and then pay in Bitcoin using its "exchange rate" at that time.

One option: a volume-adjusted average of MtGox.com trades over the last K hours

Another: use 0.001 of the amount to buy options to convert X Bitcoin into the agreed upon amount of (USD, ...). Market-making HFT bots will compete, making X the most fair amount.

All this complexity could be handled by user-friendly software...


But then you're effectively pinning the value, or debt value of BitCoin on the USD. I think that is counter to the idea of Bitcoin.


You wouldn't be screwed if your business operates in Bitcoins, and your customers pay you in Bitcoins. If you take out a loan for 100 Bitcoins, you will still owe 100 Bitcoins, regardless of the USD exchange rate.

If you aren't dealing in USD, it doesn't matter how much value the USD loses.

Similarly, if you had debt denominated in USD, it wouldn't matter how much value the Zimbabwe Dollar or Weimar Republic Mark lost in hyperinflation. Your debt burden would be the same amount of USD. It just depends on your frame of reference.


The dollar has not reduced in value by a factor of six in the last few weeks. Nearly all of the change in the USD/BTC exchange rate is due to a reevaluation of the real value of bitcoins.

So even if you did all your business in them, it's likely prices would have fallen substantially in the last few weeks, and you would be screwed, per the grandparent.

That said, the source of BTC's volatility right now is precisely the fact that few people are doing business in them, and nearly all demand for them is speculative.


> You wouldn't be screwed if your business operates in Bitcoins, and your customers pay you in Bitcoins.

At some point, you need actual money to do things like pay taxes. Last I checked, no government accepts Bitcoins for this.


How much larger does the Bitcoin economy have to get ( I believe it's currently ~$30M) before the same money transfer companies will start pushing for them to be regulated.

Remember the cab companies going after UberCab?

How about the Louisiana Funeral Directors going after the monks for making caskets[1]?

[1]http://www.techdirt.com/articles/20100825/11152810773/funera...


I'm still new to the bitcoin thing but how can it be regulated?


With difficulty. The most likely way I can see is regulation of exchange markets (which could actually be beneficial to the markets, as a way of showing credibility).


True, but an exchange market could really pop up anywhere at any time and disappear just as quickly. People could liquidate through Craigslist if they really needed to. If the BTCs have enough distribution and momentum, it'll be around a long time.


A bit of a red scare here. All the time I was reading I was led to assume that companies like PayPal and Square (imperative mechanisms for freelance these days) were to be affected, ultimately though, this was dismissed toward the end. I can understand the frustration here, but like a lot of people are saying, those are the rules. Also, for the FaceCash folks, if you intend to disrupt the financial industry in any fashion (in this case circumventing traditional forms of cash transfer) you should be ready for a significant amount of red tape and "fuck you."


We're ready for it, but that doesn't mean we have to take it.


so what if my company uses paypal to process payments and send payouts to users? do i have to care about this stuff or its just paypal's headache now?


What does this mean for systems like Facebook Credits?


Nothing. Facebook credits are not a money transfer system, it's whats known as a 'closed loop' system. If they opened it up so that anyone can take a payment with a facebook credit and receive dollars then they would start to fall under the money transfer arrangements and would require these restrictions.

I don't think $500k is much of a barrier to them getting into this business however...


Can't anyone build a Facebook app that takes payment in Facebook credits, and then cash out that payment into dollars?


I wonder how BankSimple is doing now.


Probably just fine; I'm sure they had to apply for a regular banking license anyway, and banks are exempt from money-transmission regulations.


I'm not sure about that. From their website:

However, BankSimple is not a "bank." We partner with chartered banks who provide FDIC-insured products, leaving us free to concentrate on designing the complete consumer banking experience, via the web and your smartphone.

What implications this has on their regulatory requirements, whether banking or money transfering, would be interesting to discover.


Yep, BankSimple doesn't have to apply for a regular banking license because they don't hold the actual deposits. (See SmartyPig for an example of a startup in the market that uses the same 'we don't hold the deposits' strategy.)

Piggybacking off of someone else's banking license isn't trivial, but it beats the delays and capital requirements associated with starting a bank.

Of course, they still have to deal with all the internal industry regulations in order to offer things like ATM cards that work across ATM networks. Those aren't regulatory requirements, but when you're dealing with VISA-size entities they might as well be.


Every company in Silicon Valley is actually a Delaware corp anyway..


I think it's a relevant distinction. Clearly the California company doesn't apply to all corporations in the world or even country. The point is there's likely a loophole that for an interstate or international software corporation will render the regulation toothless.


Every company doing business in California needs to have a license to operate here (and obey State laws). No easy loopholes here.

Companies mainly choose Delaware because Delaware is efficient and well-known for processing of paperwork and legal matters.


What does this mean for freelancing sites? Or for services like AdSense where credit accumulates in a user's account till it's withdrawn? Do they have to get a license too?


Here goes one more reason to use Bitcoin, that's where the real innovation is.


Bitcoin is operating in a legally very grey area, you can bet your bottom-dollar that as soon as they get real traction laws like this are going to come down on the ecosystem extremely fast.


Im quite sure most of what Bitcoin is used for is already illegal, but that doesn't mean anything. It's just being used for what people have always done, this is the government outlawing perfectly normal things. Thanks to bitcoin (and it's growing network of services that operate from cypherspace) we nolonger have to dance to the governments tune.


I'm sure similar claims can be made for Linden dollars, which have been used for far more transactions over the years than Bitcoin has.

Why single out Bitcoin? Because it's new?


Bitcoin is much more powerful then Linden dollars, Egold etc. because it allows transactions online, anonymously, and has no central authority. No other payment method has ever had all three of these characteristics.


No?

I can create an anonymous Second Life character at any time, linked only to my account info (an email - Mailinator solves that). I can build something in-game and sell it, creating money with no person bound to it. Or be given money in-game for a real-world purchase, leaving me with money that still has no connection with me. There's barely a central authority, as there's nothing stopping you from passing money around in-game, which would be the major barrier to illegal trading. And what is there is rather impotent, as everything can be done just as anonymously as in Bitcoin - they can get your IP address, but only if they're watching, and that's far from an identification. Just go to a coffee shop.

Bitcoin just makes it easier, and fault/attack-tolerant. It's been possible for quite a while. Though perhaps the most useful aspect to this is that transactions aren't revokable, where they are in nearly any other non-physical system.


Sure is! A great reason to use a rapidly inflationary currency with no promises of stability or reliability.

Oh, wait. Those are reasons not to use it. My bad.


You've mixed up deflation and inflation.


Slow deflation at first, rapid inflation eventually. That tends to be how bubbles work.

(Not to be confused with physical bubbles, which are the other way around...)


Exactly. The deflationary period will come when most people liquidate their Bitcoin holdings because they realize it's essentially useless. Or try to liquidate, anyway. At that point there won't be many buyers.


Deflation is an increase in the value of a currency. You're describing an eventual decrease in the value of Bitcoin, which would be inflation.


You're right, I apologize. Sorry, should not post under the influence of burritos.


Can this affect Bitcoin, and if so, how?


no bitcoins in CA


Today I learned Hacker News is filled with libertarian ideologues.

It's amazing that people who pride themselves in critical thinking and problem solving (hackers), can actually be so naive as to believe this nihilistic Ann Rand crap.

HN is starting to look more like Reddit every day.


HN has always had its share of libertarian partisans, some who were reasonably cogent and some who drank the Randian kool-aid and did not notice when it dissolved their brain. The difference is that on HN you will see a lot more pushback from people who are not going to let the second class of ideologues get a free pass on their BS (e.g. the responses to various "just let the market sort it out" tropes being dragged out by the usual suspects.) On Reddit the moderates would be downvoted into oblivion and this discussion would have turned into a libertarian circle-jerk within fifteen minutes.


If you run a node on Bitcoin, aren't you technically a payment processor?

I wouldn't be surprised if this is step 1 of a 3 step plan on Bitcoin takedown. Clever too I might add, so props to whoever came up with it.


Who is John Galt?


Some obnoxious bint beloved by people who clamor to be his oppressed, destitute underclass.

(What's that? They all think that they'll be him? Statistics would like to have a word, folks.)


So is paypal going to shut down after this goes into effect?


I don't think they'll go anywhere.

I assume they can easily afford any cost involved. I wouldn't be surprised if they already have the licenses while existing as a non-bank.


Boohoo, cry PayPal a river.

In most countries outside the US, PayPal is regulated like a bank. This means PayPal can't just lay claim to your money and never give it back to you.


Actually, I wouldn't be surprised if PayPal supported the law. Many regulations protect large, established players by raising the cost of entry for new market participants. PayPal can afford to post this bond, but many bootstrapped start-ups probably can't. This is just one of the ways regulation _can_ increase the price of products and services; they decrease competition, allowing established players to raise prices.


Boohoo, cry PayPal a river.

I don't understand what this has to do with PayPal. PayPal is large enough to easily jump through whatever hoops are required to keep doing business. This is a problem for new companies, that would compete with PayPal.


They can, however, (and arguably should) put a hold on your account if they suspect fraud.




Registration is open for Startup School 2019. Classes start July 22nd.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact

Search: