Author

Topic: Combating friendly fraud (Read 2662 times)

donator
Activity: 1218
Merit: 1079
Gerald Davis
April 05, 2012, 09:02:08 AM
#8
So let me see if I got it right: a debt collector buys the debt from you and contacts the customer to get the money in an amicable way (doesn't go to court straight away), right?

Well I wouldn't call it amicable however yes they attempt to collect payment on delinquent debt, possibly setup a payment plan, or accept a partial settlement.

Quote
Pennies on the dollar?! This should mean their own recovery rate sucks (or they have huge ROIs Grin)!

Most debt is never collected however yes.  Buying deliquent debt is high risk, high ROI activity.  I tried to point this out in threads were people were selling Bitcoin debt for 50 cents + on the dollar but I guess people find it hard to believe.   Even high quality delinquent debt (copy of contract, full contact information, ability to sue, high enough balance to make collection worthwhile, credit reporting rights, short time since deliquent, long payment history, etc) routinely sells for <10 cents on the dollar.

Quote
And if they do go to court ultimately, where did you find such a progressive debt collector (who's making a tx in the blockchain look like evidence), if you do not mind me asking.

I work in debt collection industry so I have some contacts.  Most debt will never go to court, it simply isn't worth the cost.  The debtor doesn't know that though and sometimes they don't want to risk it.  I have no intention on using blockchain as evidence.

No need to make it complicated:
Customer purchased digital goods worth $x.
Customer fraudulently reversed charges.
Customer is indebted to me for $y ($x + chargeback fees)
Debt sold to third party and customer is indebted to third party for $z ($y plus max legal collection fee plus max legal penalty interest with clock starting on day of purchase).

Honestly I would sell the debt for nothing.  It won't help my bottom line much anyways.  My goal isn't to profit from debt sale but instead to make friendly fraud punitively expensive for customers.  Steal $100 in BTC from me and I lose $100 but you get are $300 in debt to someone who's business it is to extract value from deliquent debtors.  Plus the fun of your credit score falling 20 to 80 points instantly, plus collection agency calls to your home, spouse, workplace, friends, etc.   Remember as you noticed they are motivated to chase those high ROIs. Smiley

Quote
Outsourcing the debt and successfully dealing with the "black fraud" could be the solution, indeed. The question remains: is it worth the headache?

That remains to be seen which is why I intend to start with physical delivery.
member
Activity: 105
Merit: 10
Say BYE to tobacco taxes and bureaucratized trade.
April 05, 2012, 08:54:59 AM
#7
Given you need to file in the local court where the cardholder resides I doubt it unless your volume is tens of thousands of customers per month.

I am not saying reducing fraud is impossible but of all forms of fraud,  fiat->Bitcoin has to be one of the hardest to defend against..
1) It is new so the teminology and concpets like showing a tx in the blockchain is going to make any claims adjuster or judges eyes glaze over

2) It is irreversible.  Enough said.

3) It is fungible. Other digital goods aren't.  If I steal a game (via friendly fraud) from Steam I risk losing access to Steam forever.  Many gamers may balk at that idea.  1 "free" game vs losing access to all future games on Steam.  Given 1 bitcoin = 1 bitcoin a thief doesn't care about a lifetime ban.   They can just buy (or steal) coins from another source.  Obviously this aspect doesn't apply to "black" fraud but I think we agree there are technical solutions that at least have the potential to fight black fraud.

I intend to start selling Bitcoins but with physical delivery (and contract w/ debt collection company which is advised to customer in advance).  If I can keep fraud costs down I am looking into technology to allow small purchases online with some ideas on customer verification.  So I don't want to be hypocritical and make it look like I am saying it is impossible while perusing it myself.  The bar is pretty high though.  

I have some experience fighting CNP fraud.  I made a good amount of money back in the internet dark ages (1990s) selling phonecards before selling the company (well brand, domain, customer list) to a competitor.  I moved quickly from physical delivery to online delivery.  Now with phonecards you at least have the ability to de-activate fraudulently purchased cards (if they haven't been completely used).  

What I learned is that organized crime is just that ... organized.  I estimate less than 5% of actual customers were fraudulent but if they found a hole in the fraud "net" they would spam it with hundreds of orders before that hole was patched.  So everything can look good w/ low chargeback rates for weeks even months and then a tidal wave of fraud just start rolling in.   I had one month were 37% of orders were disputed.  

The other difference is markup.  In the 90s, phonecards were marked up 70%+.  So I would buy 1 million minutes @ 0.4 cents per minute and mark them up to 5 cents (effective w/ connection fee).  With fatter margins you can absorb more fraud.  Saw a lot of copycats try it with resale accounts (like what stores use) and the smaller margins (10% to  20%) means they couldn't absorb the fraud and died.

Obviously the margin on Bitcoins has to be much much much lower (or you only attract the thieves).  Less fat means you have to be nearly perfect.

Long winded but ....
TL/DR I don't think legal is ever a viable option.  Selling the debt (even for pennies on dollar litterally) may have some value as the friendly frauder gets stuck w/ full charge + $50 to $200 collection charge (max allowed by the State, not sure about international) plus penalty interest charged by debt collector plus negative hit on their credit report.

It is my hope if I can weed out the "black fraud" that the risk of hundreds of dollars in collection charges, negative credit report action, and just the downright hassle of fighting collection company will make it unattractive to friendly frauders.   ..... Maybe. Smiley

Thanks for your elaborate response!

Let me get this straight: a debt collector buys the debt from you and contacts the customer to get the money in an amicable way (doesn't go to court straight away), right?

Pennies on the dollar?! This should mean their own recovery rate sucks (or that they have huge ROIs Grin)!

And if they do ultimately go to court, where did you find such a progressive debt collector (who's making a tx in the blockchain look like evidence), if you do not mind me asking.

Outsourcing the debt and successfully dealing with the "black fraud" looks doable, indeed. The question remains: is it worth the headache?
donator
Activity: 1218
Merit: 1079
Gerald Davis
April 05, 2012, 08:20:52 AM
#6
Is it possible to minimize the legal costs by disputing friendly chargebacks in batches maybe? Grin

Given you need to file in the local court where the cardholder resides I doubt it unless your volume is tens of thousands of customers per month.

I am not saying reducing fraud is impossible but of all forms of fraud,  fiat->Bitcoin has to be one of the hardest to defend against..
1) It is new so the teminology and concpets like showing a tx in the blockchain is going to make any claims adjuster or judges eyes glaze over

2) It is irreversible.  Enough said.

3) It is fungible. Other digital goods aren't.  If I steal a game (via friendly fraud) from Steam I risk losing access to Steam forever.  Many gamers may balk at that idea.  1 "free" game vs losing access to all future games on Steam.  Given 1 bitcoin = 1 bitcoin a thief doesn't care about a lifetime ban.   They can just buy (or steal) coins from another source.  Obviously this aspect doesn't apply to "black" fraud but I think we agree there are technical solutions that at least have the potential to fight black fraud.

I intend to start selling Bitcoins but with physical delivery (and contract w/ debt collection company which is advised to customer in advance).  If I can keep fraud costs down I am looking into technology to allow small purchases online with some ideas on customer verification.  So I don't want to be hypocritical and make it look like I am saying it is impossible while perusing it myself.  The bar is pretty high though.  

I have some experience fighting CNP fraud.  I made a good amount of money back in the internet dark ages (1990s) selling phonecards before selling the company (well brand, domain, customer list) to a competitor.  I moved quickly from physical delivery to online delivery.  Now with phonecards you at least have the ability to de-activate fraudulently purchased cards (if they haven't been completely used).  

What I learned is that organized crime is just that ... organized.  I estimate less than 5% of actual customers were fraudulent but if they found a hole in the fraud "net" they would spam it with hundreds of orders before that hole was patched.  So everything can look good w/ low chargeback rates for weeks even months and then a tidal wave of fraud just start rolling in.   I had one month where 37% of orders were disputed.  The other challenge is markup.  I was buying 1 million minutes wholesale @ <1 cent per minute and mark them up to 5 cents (effective w/ connection fee).  With fatter margins you can absorb more fraud.  

Obviously the margin on Bitcoins has to be much much much lower (or you only attract the thieves).  Less fat means you have to be nearly perfect.

Long winded but ....
TL/DR I don't think legal is ever a viable option.  Selling the debt (even for pennies on dollar litterally) may have some value as the friendly frauder gets stuck w/ full charge + $40 to $100 chargeback fee + $50 to $200 collection charge (max allowed by the State, not sure about international) plus penalty interest charged by debt collector plus negative hit on their credit report for the next 7 years.

It is my hope if I can weed out the "black fraud" that it will be possible to make friendly fraud punitively expensive and thus not so "friendly". .... Maybe. Smiley
member
Activity: 105
Merit: 10
Say BYE to tobacco taxes and bureaucratized trade.
April 05, 2012, 08:01:07 AM
#5
In court? You are kidding right? Legal filings alone would be more than you would ever recover.  You must also PROVE that friendly fraud occurred.  Between the cases you can't prove, the ones you just lose because judges are stupid, your court costs, and your legal team costs you are talking a major cost likely 5x to 10x anything you recover from the thief.

So how do you pay for that cost?  Raise your prices?  Ok well you just drove away the honest customers (who will be price sensitive) and packed your sales book with dishonest customers (who don't care what the price is because they have no intention of paying anyways).

Is it possible to minimize the legal costs by disputing friendly chargebacks in batches maybe? Grin

Minimizing the "black fraud" should be fairly easy (e.g. confirmation of small random charges to the card, tight monthly limits, etc.).

These guys here are trying to "deter" friendly fraud. I'm curious of the results!
donator
Activity: 1218
Merit: 1079
Gerald Davis
April 05, 2012, 07:53:00 AM
#4
I was wondering, why can't you hunt all the friendly chargebacks down, one by one, providing evidence you have sent the Bitcoins (in the court of law, if necessary)?

Is the process of combating friendly chargebacks prohibitively expensive?

Yes.  
Your merchant bank doesn't matter.  It is the customer's merchant bank who determines is chargeback stands.  Their customer (i.e. who they profit from) is the potential scammer.   The proof must be airtight for them to risk losing their eternally indebted cash cow.  Your merchant bank does like the cost of chargebacks so it just looking to drop you for business that has higher revenue:cost ratio.

In court? You are kidding right? Legal filings alone would be more than you would ever recover.  You must also PROVE that friendly fraud occurred.  Between the cases you can't prove, the ones you just lose because judges are stupid, your court costs, and your legal team costs you are talking a major cost likely 5x to 10x anything you recover from the thief.

So how do you pay for that cost?  Raise your prices?  Ok well you just drove away the honest customers (who will be price sensitive) and packed your sales book with dishonest customers (who don't care what the price is because they have no intention of paying anyways).
member
Activity: 105
Merit: 10
Say BYE to tobacco taxes and bureaucratized trade.
April 05, 2012, 07:47:03 AM
#3
And I do intuitively understand there are many "underwater rocks" in doing this, but can someone please pinpoint them to me?
member
Activity: 105
Merit: 10
Say BYE to tobacco taxes and bureaucratized trade.
April 05, 2012, 07:34:14 AM
#2
Or, is the pie maybe not big enough to be worth the hassle of starting such a service?

SWIFT transfer would simply die as a funding method, in my opinion, and if I think of potential API's for merchants, I'm getting even dizzy! Think about all those high-risk merchants (like myself) who are forced to switch their credit card processors every month. It's such a pain in the ass, trust me. I would much rather re-direct ALL of my customers to a 2-step service to purchase Bitcoins (or, even better, integrate a legal binding API) than going through this credit card acceptance hell!
member
Activity: 105
Merit: 10
Say BYE to tobacco taxes and bureaucratized trade.
April 05, 2012, 07:23:48 AM
#1
We all hate friendly chargebacks (no carding involved), and if you think about it, it's the only reason why Bitcoins cannot be bought via plastic cards up to the present time, a service that would dramatically increase the proliferation of Bitcoin.

I was wondering, why can't you hunt all the friendly chargebacks down, one by one, providing evidence you have sent the Bitcoins (in the court of law, if necessary)?

Is the process of combating friendly chargebacks prohibitively expensive or is it just impossible to find a merchant bank that would be flexible enough?

Waiting for your opinions.
Jump to: