From Robert Hettinga
Robert Hettinga is a financial cryptography strategy and policy consultant in Boston. He is founder of the First International Conference on Financial Cryptography (FC97), the International Financial Cryptography Association, the Digital Commerce Society of Boston, and the e$ and e$pam mail lists. He is also financial cryptography editor of JIBC.
Visit for more related articles at Journal of Internet Banking and Commerce
On Fri, 13 Jun 1997 [email protected] wrote:
> Hi all,
> I suspect that one of the principal things that the Feds are
> worried about is the potential for money-laundering.
This is a loaded statement. Money laundering is only a concern in so far as it means government control over the economy is diminished. (And to the extent that it allows one to seize the funds a their title converts the to United States at the instant of commission).
Money laundering is a "tack on" offense. (Much like, say, mail fraud). The number of original cases which derive from actual money laundering investigation is vanishingly small. Instead it is usually added on to an indictment when the defendant is or has been under investigation for something else.
Because money laundering statutes are generally phrased something like "knowingly concealing the proceeds of a criminal act," usually you find the criminal act first and then look to see if attempts were made to conceal the funds. Professional money launderers are rarely caught.
> At the moment, conversion of money from illegal sources (drug
> sales, extortion by terrorists, major theft etc) into the legal
> economy (equities, bonds, property etc) is difficult because
> any financial institution is obliged, in most parts of the
> world, to obtain proof of identity of its clients and toreport
> suspicions of wrongdoing.
I disagree rather strongly. Currently the favorate method is to hand the cash, in bulk, to the professional money launderer who, on the spot, cuts a clean bank check (perhaps from a reputable import/export or realestate company) for the cash amount minus fee (5-20% usually). The launderer takes all the risk in the process, including smuggling the funds out, hashing them through iterations and (usually) returning them right back into the United States as legitimate overseas investment. It's like the separation of capital and management skill. The money launderer is free to concentrate 100% of his time to managing his extensive laundering empire, the hundreds or thousands of shells and webs of accounts and maintains the liquidity to drop 5 million on the notice of a phone call.
> Hence, I suspect, the $750 limit.
> The reason for this check is that it is otherwise very easy to
> shuffle funds back and forth between financial instruments to
> confuse the trail and defeat the cops.
The $750 limit is going to do about nothing for the problem of money money laundering. It will inconvenience the casual launderer, and that is about all. What it will do is put a significant cost on the head of the consumer. A CTR costs a bank between $5 and $15 to file today (according to the ABA). $17 if you listen to the Report of the Financial Action Task Force on Money Laundering.
In 1993 the 368 largest banks (assets over $1 billion) filed 4.5 million CTRs. The cost was estimated at $72 million dollars. (John Byrne, General Counsel, American Bankers Association). 10,765,000 CTRs were filed in 1994. About .5% are marked "suspicious."
Now the $750 limit? The number of reports to be filed is staggering and .5% is beyond government to police properly without 5,000 new hires. No, clearly the $750 limit is not to catch money launderers, but to create and perpetuate detailed transactions record keeping.
FinCEN is much more useful to link transactions to defendents in non-money laundering cases. "What do you mean you weren't in California in May? Our records show you accepted two wire transfers there on the 15th and the 16th."
And consider this. If I build a machine which has a 95% accuracy rate in detecting money laundering, that is to say that it will identify a given transaction as money laundering or legitimate with 95% accuracy, I still have a serious problem. Given 10,000 transactions, with .2% (20) representing money laundering we find the following figures: 19 (95% of 20) money laundering transactions will be flagged as illegal 1 (5% of 20) laundering transaction will be incorrectly flagged as legal. 500 (5% of 10,000) legitimate transactions will be incorrectly flagged as illegal.
For every one money laundering transaction flagged there will be 26 legitimate transactions flagged and only about 3.6% of all the flagged transactions will actually be illegitimate. Now consider all this in context. There are over 700,000 wire transfers a day amounting to over $2 trillion. About half go through FedWIRE and CHIPS. SWIFT is harder to count.
The most pro-government figures have somewhere on the order of $400bln a YEAR being laundered. The figure I used above (.2% of transactions are money laundering) is high by orders of magnitude. Depending on who's figures you use it's more like 0.008%.
> I can't subscribe to the full libertarian view that there's no
> such thing as right and wrong, only freedoms and restrictions.
> Because of this, I think we have to accept that the Feds have a
> valid concern, as instant and frequent movement of large sums
> of untraceable, impersonal e-cash would mean that the only way
> to stop money laundering would be to check the identity of
> anyone converting real-cash into e-cash. Now that does perturb
What you do not address is what is wrong with money laundering. Money laundering was never a crime until 1986 in the United States, [Money Laundering Control Act of 1986 (Codified @ sections 1956 and 1957 of Title 17 of the U.S. Code)] and then only made so because it fit in the "war on drugs."
The result: Total financial transparency in the U.S. Money laundering is a created offense. Also consider that becuase current efforts to curb money laundering are basically useless, you may see identity checking anyhow. (The $750 limit is basicially an identity checking provision already). Money laundering is a crime because the drug market in the United States is so substantial it is impossible to police it without following the money. You can't do that unless you have near financial transparency.
> Given the response of governments to anything untaxed (eg
> running drugs that aren't tobacco), these regulations could
> stay with us for years to come. Has anyone got another
> foolproof scheme to foil e$-laundering?
Deep down even you seem to recognize that Money Laundering has nothing to do with catching criminals. It's all about control and revenue. The feds would do better to concentrate intelligence assets (more so than now) on counter narcotics and do their best to schedule raids to seize cash when it's sitting around than to track every transaction on the planet. I don't have it here, but if anyone is interested I had a report on the costs of Money Laundering enforcement I can try and dig up.
One of the interesting figures was the number of dollars spent to stop one dollar from being laundered (Around $35,000 as I recall) and the number spent to seize one dollar of laundered money (Around $110,000).
> Martin Farncombe,Ernst & Young UK
> (views represented are personal, not corporate)