Joshua Gans has recently proposed that bank account numbers should be portable, just like mobile phone numbers. I was initially sceptical, but his FAQ neatly addressed most of the concerns that I had. (See also a writeup in the Canberra Times from Peter Martin.)
Advertisements
This is a brilliant idea. The banks will say it will cost too much but that is untrue. There are simple technical solutions.
Another thing to make banking more efficient is to give instant transfer of funds. If you can prove who you are then why shouldn’t you be able to transfer funds immediately? There is no technical, legal, or administrative reason why this cannot be done today. If this was done then you would find that a large number of payment systems would arise – like mobile payments, ATM machines by other vendors, smart card tickets – and the cost of bank transaction would drop.
The banks will never have portable numbers and will never allow instant transfers unless forced to because both help prevent competition.
I’d love to see that.
Gans misses the point completely. Even if you can switch bank accounts (those where the bank owes you money) at the press of a computer key, you still won’t be able to switch loan accounts, and that’s where the problem is, if there is one.
This isn’t a new idea.
It’s been tossed around financial circles in various forms since BPay was introduced in 1997, but no-one can come up with a business case to pay for it.
The Dutch enjoy some banking portability and have done so since 2004 (see http://www.nvb.nl/scrivo/asset.php?id=22956).
And back to the phone analogy, number portability doesn’t make switching any easier if you are in an onerous contract or cannot get credit with a different provider, a situation analogous to that described by Spiros above.
The Dutch enjoy switching assistance not number portability.
I read all of the bloggers of standing who are of the view that portable bank accounts is a good and sensible idea and I conclude that dementia has set in for me. I clearly don’t understand anything I read. However some of the stuff that made me a no.
First what a large number of European banks, consumer organisations and government administrations said, NO. Why cost, technical etc. Didn’t find any that said yes get it going now.The Dutch thought about it and limited themselves to switching, which excludes mortgages, the cause of the uproar. Estimate of cost of number portability for Dutch banks only, 300 to 500 million euros. And then some of what the computer boys at Free University of Amsterdam had to say.
“Bank account numbers (BANs) are not easy, and the more you know about them, the more it looks like an esoteric doctrine or a mysterious art: the cabala of numbers…
Our colleague Andy Tanenbaum once said that the good thing about standards is that there are so many to choose from. The occult world of bank account numbers is no exception to this truismâ€
That seems to make telephone numbers pears and bank account numbers apples for me. And
“almost all banking software is written in Cobol. Cobol has special properties, or better: lack thereof. From our 223-benchmark project, and other projects we found that Cobol only allows for hard-wired constants and data-types [9]. So these systems are chockful of hard-coded prefixes, up to entire BANs, hard-wired BAN-related data types, hard-coded internal booking numbers that start to interfere with new 10-digit BANs, and so on. Also it is a Herculean effort to detect all the variations and their bank-specific hidden semantics: how to decipher the meaning of certain prefixes, and whether it is exploited. In one system, e.g., we found that the hard-coded prefix 666 implies that the surname of the client starts with the hard-wired character X2. We found that other systems exploited this, for instance, to schedule batch jobs for printing bank statements. Or hard-wired bank-codes are exploited to identify a physical location. These bank-codes are used in other systems to streamline certain business processes. For instance, to print bank statements at a central shared service center, to internally ship them near the customers, and then mail them out for quick, accurate, and cheap delivery. There are many more examples, and implementing number portability can have devastating effects on the business due to the semantical mismatches with alien numbers. So the policy of domestic BAN portability should be banned. It is close to impossible to implement, let alone at no additional cost.â€
I would have thought it was obvious that all of Australia’s financial institutions, let alone the banks have a range of number lengths, unlike the Netherlands which at least has a standard length for numbers. Then there is what information is included in those Australian numbers, including some times check digits, and how the institutions use that information. Easy peasy to implement number portability? It took 10 years to implement SEPA (IBANS EU)
Anyway they have all had their way, it isn’t happening in the EU, or the Netherlands at this point in time. But Australia should run with it, on its own. Even if some computer scientists think it is close to impossible to implement.
Not that anyone is obliged to help me out but I am clearly right in outfield with no idea what the matter being discussed is. So I would appreciate some greater detail on how this grand solution of number portability to stop banks increasing their interest rates would work. The Amsterdam boys call the issue another IT-Soap.
There are many ways to solve the problem. For the system to be efficient you do not change the internal bank systems. They all remain the same as they are.
One way is to move your bank account number from bank a to bank b is for bank b to create a new account with a different number. Now put the mapping in a global database that contains the mapping of one bank account number to another. When anyone does a transfer of money from one bank account to another then they should check to see if the number has changed.
This has other advantages such as enabling an individual to easily amalgamate bank accounts.
Another solution is to have low cost automatic instant transfers of money between accounts no matter where the money resides but that leaves you with bank fees on two accounts and unnecessary transaction fees.
Both the above solutions are simple to implement and easy to do but the banks will not do it of their own violation because they lose their competitive advantage of their existing client base.
I have just been through the process of trying to move from one bank to another because the second bank offered me better service on a different product and asked me to move my day to day banking to them. This turned out to be a nightmare and was not worth doing. What happens is that you have to notify everyone who knows your number to pay to the new account. The only practical way to achieve the change is to leave the old account open for at least a year and to see the activity on the account and then notify those who are still using the old account to cease and desist and people paying you money tend to be a bit annoyed at having to change their systems to accommodate your changes.
Mortgage accounts are different to bank trading accounts. They are places where you owe the money and not places where you store your money. The transfer of a mortgage requires more than the transfer of a bank account.
Given then that you want to take advantage of a new service (like a bank that offers lower interest on mortgage accounts) that a bank offers. You will find that the bank may require you to open up accounts with them. In practice what most people do is not to look for the best deal but to stay with their existing bank. This reduces competition and it enables banks to increase fees and interest rates without losing existing customers.