PDA

View Full Version : checking whether a bank note is a forgery: idea?



WaxRubiks
2009-Nov-14, 06:51 AM
I had an idea, don't know if that is what they do now, where by two numbers could be printed on a bank note, one of the numbers could be put into an complex, secret algorithm that outputs another number. Both numbers would be printed on the note.

Then to check whether the note is a forgery, one would just need to type in both numbers on a forgery-check website and it could tell you if it was a fake or not.


I thought that maybe the one number on the note is made up of two such numbers already...so maybe not a new idea.


I suppose that it could even be done by using a bar code, which a person could photograph with their mobile phone, which would then check with the forgery-check website.

ETA: :doh::doh::doh: I suppose the forgers could just copy such a note, but it might make their job a bit harder

Ronald Brak
2009-Nov-14, 07:42 AM
One possibility: Put a tiny chip inside the note that can store a few bytes. Each time it goes through an ATM or other choke point it gets the number changed and the result is sent to the reserve bank. If the serial number and the number on the chip don't match the records it gets investigated as a possible forgery.

EDIT: Or if it doesn't have a chip at all, that's a pretty good sign it could be a forgery.

Mister Earl
2009-Nov-14, 07:58 AM
As an alternative, you can do away with physical currency altogether and just go digital.

mahesh
2009-Nov-14, 08:04 AM
Didn't some guys in south east Asia try that in the latter seventies?
...except for the digital bit?

HenrikOlsen
2009-Nov-14, 10:43 AM
I had an idea, don't know if that is what they do now, where by two numbers could be printed on a bank note, one of the numbers could be put into an complex, secret algorithm that outputs another number. Both numbers would be printed on the note.

Then to check whether the note is a forgery, one would just need to type in both numbers on a forgery-check website and it could tell you if it was a fake or not.
One problem that needs to be fixed is that your system makes data mining about money transfers possible.
It's one of the axioms on (physical) money that they're anonymous.

WaxRubiks
2009-Nov-14, 11:25 AM
I realised after posting(and hence the ETA) that a forger could just copy the numbers of other notes, but as the forgeries were checked out it would become apparent to the fake-checking website that certain numbers were likely to be forgeries, so state that they might be forgeries.

Moose
2009-Nov-14, 11:26 AM
a) It doesn't prevent one from duplicating an existing number set.

b) License keys operate under this principle. You can't simply enter a random license key code and expect it to work. They're mathed up and compared to a formula. The thing is that reverse engineering these keys isn't all that big a deal. Software pirates release keygen programs as a matter of routine.

Moose
2009-Nov-14, 11:30 AM
I realised after posting(and hence the ETA) that a forger could just copy the numbers of other notes, but as the forgeries were checked out it would become apparent to the fake-checking website that certain numbers were likely to be forgeries, so state that they might be forgeries.

Unless everybody is scanning every bill, every time it's used, who's to say that a bill that was last checked in Surrey, BC might not appear, a week or three later, in rural Ohio? There might be no transactions in between, there might be a few dozen.

Strange
2009-Nov-14, 12:03 PM
They're mathed up and compared to a formula.

I like it. Is that a technical term :)


The thing is that reverse engineering these keys isn't all that big a deal. Software pirates release keygen programs as a matter of routine.


Well, in principle this kind of public key system can be made unbreakable by using a large enough key. But then the private key has to be kept secure. And would you trust any large organization (never mind the government) to keep a secret this valuable?

But ensuring the numbers are unique is the show stopper.

Ronald Brak
2009-Nov-14, 12:29 PM
But then the private key has to be kept secure. And would you trust any large organization (never mind the government) to keep a secret this valuable?

Would I trust an organization that prints money to not release a security feature that makes the money it prints marginally harder to counterfeit? Why yes, I think I would.

novaderrik
2009-Nov-14, 06:43 PM
One possibility: Put a tiny chip inside the note that can store a few bytes. Each time it goes through an ATM or other choke point it gets the number changed and the result is sent to the reserve bank. If the serial number and the number on the chip don't match the records it gets investigated as a possible forgery.

EDIT: Or if it doesn't have a chip at all, that's a pretty good sign it could be a forgery.

there is a large segment of the population that would have huge problems with tracking money like that- and i am among them.
personally, i think money forgery is a minor problem when there is an agency that is "overseen" by the US government that can print as much money as they want whenever they want. there are billions of dollars of counterfeit money floating around- but just this summer they decided to print TRILLIONS of dollars of "real" money with absolutely nothing backing it.

Moose
2009-Nov-14, 06:59 PM
I like it. Is that a technical term :)

Very technical, unlike the much more colloquial term "hash" or "to hash".

(Or maybe I got those backwards. :shifty: )


Well, in principle this kind of public key system can be made unbreakable by using a large enough key.

Not really. The problem here isn't that the key is breakable. This isn't an encryption task. The problem is that you need to somehow protect the algorithm. Security through obscurity _never_ works. And once you have the algorithm details, you can fake valid keys.

And either way, it doesn't prevent you from getting a few other dollar bills and duplicating them. If you're passing your duplicated bills in the same general area, all you'll see by data mining is that the bill is circulating.

The task is tamper-evidence, and you can't do that by means of an easily duplicated feature/label.

Ronald Brak
2009-Nov-14, 10:56 PM
personally, i think money forgery is a minor problem when there is an agency that is "overseen" by the US government that can print as much money as they want whenever they want. there are billions of dollars of counterfeit money floating around- but just this summer they decided to print TRILLIONS of dollars of "real" money with absolutely nothing backing it.

Wow, you must have Zimbabwe like hyperinflation.

Strange
2009-Nov-14, 11:15 PM
The problem here isn't that the key is breakable. This isn't an encryption task. The problem is that you need to somehow protect the algorithm. Security through obscurity _never_ works. And once you have the algorithm details, you can fake valid keys.

But... in public key cryptography (which can also be used for signing which is effectively what we are talking about here - where the two numbers would be the plain text serial number and the encrypted version) the algorithm can be public. In fact it has to be.

Mmmm... algorithm; such a nice word.


And either way, it doesn't prevent you from getting a few other dollar bills and duplicating them.

Agreed.

publius
2009-Nov-14, 11:16 PM
For those that don't follow the Black Monday thread, here's the Fed's M0, the monetary base from "FRED", the St. Louis Fed's economic data system:

http://research.stlouisfed.org/fred2/graph/fredgraph.png?bgcolor=%23B3CDE7&chart_type=line&drp=0&graph_bgcolor=%23FFFFFF&height=378&mode=fred&preserve_ratio=checked&recession_bars=On&txtcolor=%23000000&width=630&id=BASENS&transformation=lin&scale=Left&range=5yrs&cosd=2004-11-04&coed=2009-11-04&line_color=%230000FF&vintage_date=2009-11-14&line_style=Solid&mark_type=NONE&mma=0

The monetary base is over $2T now, compared to about ~850B before the crisis started. That's roughly a 130% increase. However, the larger monetary aggregates aren't exploding themselves. The basic relation is the famous equation of exchange:

MV = PQ, where M is the money supply, V is the velocity, P is the general price level, and Q is the quantity of goods/services output. Taking the differential and dividing through by the quanties themselves gives you the inflation relation:

dP/P = dM/M + dV/V - dQ/Q

And that expresses Milton Friedman's famous dictum that inflation is too many dollars chasing too few goods. M is the money, V is the chasing, and Q is the goods.

What is happening is velocity is collapsing as well as the broader monetary aggregates inspite of all the base money printing. (The dollar is falling relative to other currencies like a stone due to the printing, though).

And, a bit of technicality here. This new money is not actually physical, but electronic. Banks maintain electronic reserve accounts at the Fed. All the newly "printed money" is just these digi-dollars. The physical currency has not increased much at all.


-Richard

Strange
2009-Nov-14, 11:21 PM
The basic relation is the famous equation of exchange:

MV = PQ, where M is the money supply, V is the velocity, P is the general price level, and Q is the quantity of goods/services output.

Interesting. How is V defined?

publius
2009-Nov-14, 11:27 PM
Interesting. How is V defined?

Ah. Well, the equation itself serves to define it:

V = PQ/M.

It is the total economic output in dollars over the money supply. Think of V as the number of times a unit of currency "turns over" over the period of interest.

Say you and I are doing business. I have one dollar and buy something from you for a dollar. You turn around and buy something from me for a dollar. PQ there is 2 dollars. M is 1 dollar. Thus the velocity is 2.

You can see we can increase V without changing M.

The question is what "M" to use. M2 or MZM is usually used, not M0, with different velocity figures. Velocity has been collapsing during the crisis.


-Richard

publius
2009-Nov-14, 11:49 PM
If you want to see pretty pictures, charts, and graphs, here's the St. Louis Fed's "Monetary Trends", a monthly newsletter. This is a 1.2MB download:

http://research.stlouisfed.org/publications/mt/20091101/mtpub.pdf

They have a graph of MZM and M2 velocity.


-Richard

Salty
2009-Nov-15, 03:06 AM
Hi, Frog march -

Something that early this year hadn't changed from my cab driving days was a simple and quick way to check for counterfeit American currency. Look at the picture of the president and check the background. If the background is solid, it's counterfeit; and if the background is comprised of teeny-tiny dots it's real.

Cheers,
Salty

Ronald Brak
2009-Nov-15, 03:46 AM
(The dollar is falling relative to other currencies like a stone due to the printing, though).

I don't know what would have happened if money supply hadn't been increased, but I find it hard to believe that the US dollar wouldn't have fallen anyway. After all the US has been hit harder by the financial crises than other developed countries and all else equal that would result in a fall in the dollar.

(Yes, okay, Iceland was hit harder.)

Ronald Brak
2009-Nov-15, 04:11 AM
I'll also mention that the Australian dollar has gone up while the US dollar has gone down, which appears to make sense when the their predicted economic performance over the next few years is compared, but the amount of stimulus each country had was roughly the same in terms of percentage of GDP.

The Backroad Astronomer
2009-Nov-15, 10:51 AM
Here in Canada a strong american dollar means they buy more of our stuff while a strong Canadian dollar means we can buy more stuff.

Strange
2009-Nov-15, 11:48 AM
The question is what "M" to use. M2 or MZM is usually used, not M0, with different velocity figures.

Riiiight... Thanks. I think. :confused:

mugaliens
2009-Nov-15, 12:28 PM
I had an idea, don't know if that is what they do now, where by two numbers could be printed on a bank note, one of the numbers could be put into an complex, secret algorithm that outputs another number. Both numbers would be printed on the note.

I have to had it to you - this is the very beginnings of the modern financial transaction system that's been in place for more than a decade!

publius
2009-Nov-15, 09:04 PM
I don't know what would have happened if money supply hadn't been increased, but I find it hard to believe that the US dollar wouldn't have fallen anyway. After all the US has been hit harder by the financial crises than other developed countries and all else equal that would result in a fall in the dollar.

(Yes, okay, Iceland was hit harder.)

This gets very complex. The funny thing is the crisis caused the dollar to initially rise. The dollar had been declining before, reaching an all time low back in 2008 before the crisis (and oil reached an all time high, of course). Then the crisis hit and the dollar skyrocketed.

What happened was all the debt destruction and deleveraging caused a mad scramble for dollars. This was a worldwide phenomenon because of the dollar's status as the reserve currency.

That mad scramble was relieve by the Fed through the massive printing, initially through swap lines with other central banks that peaked at $500B outstanding. As things normalized the swap lines were drawn down, but the Fed's QE purchases of T's and MBSes made up for it, keeping the newly printed dollars in the system.

That is shoring up the big banks here (the Too Big To Fails, TBTFs), raising their capital ratios. They are not lending out this "hot money" to the broader economy -- heck no, the real economy sucks, but are simply using that hot money to go into the stock market and T's (help fund Uncle Sam's record deficits), and .....

There is also a dollar carry trade going on. Borrow dollars for nothing from the Fed, then exchange it for foreign denominated assets (like say Australian dollars, where interest rates have been raised nicely). Make a free return thanks to Helicopter Ben.

For those who doubt this is going on, look no further than the correlation between the S&P 500 and the dollar index. There has been a perfect inverse correlation, even down to the interday level. Dollar goes down, market goes up. Dollars goes up, market goes down.


-Richard

publius
2009-Nov-15, 09:16 PM
Riiiight... Thanks. I think. :confused:


Sorry, I forget not everyone is up on the monetary jargon. The progression of M's represent the broader monetary counts. Commerical banks, through fractional reserve lending "create" money above the base money supply. M0 is the base. M1 is the next level, consisting of currency in circulation plus checking deposits. M2 is the next, consisting of some savings accounts. They used to track an M3, the broadest one of all, but the Fed quit, saying the cost of collecting the data wasn't worth the information benefit.

MZM stands for "money of zero maturity" and is sort of a hybrid. It counts all accounts, including money market funds, that have no time limit -- that is, it is supposed to be all liquid funds in the system. So, it includes some stuff in between M1 and M2, yet some more stuff that was in M3.

-Richard

HenrikOlsen
2009-Nov-15, 10:27 PM
But what has this to do with making unforgeable dollars?

mugaliens
2009-Nov-16, 12:09 AM
If you want to see pretty pictures, charts, and graphs, here's the St. Louis Fed's "Monetary Trends", a monthly newsletter. This is a 1.2MB download:

http://research.stlouisfed.org/publications/mt/20091101/mtpub.pdf

They have a graph of MZM and M2 velocity.


-Richard

Thanks - that is a very interesting read!

Ronald Brak
2009-Nov-16, 12:54 AM
But what has this to do with making unforgeable dollars?

Errrrrrm... By putting speedometers in notes we can record their individual velocity?

Hmmm. Maybe I should go talk to Publius in that Black Monday thread.

Ronald Brak
2009-Nov-16, 04:22 AM
Well, I went to the black monday thread, but I dont' think I'll comment there. There's too much confusing stuff there for me, which means I'd write simpler and simpler sentences in an attempt to reduce my confusion and that just seems to annoy people.

Ivan Viehoff
2009-Nov-16, 11:41 AM
Interesting. How is V defined?
Actually not interesting. One takes three numbers and combines them in a formula to obtain "velocity of circulation". This is only of interest if one has some prior reason to suppose that "velocity of circulation" should have some properties that might affect the behaviour of the things that make it up.

But there is nothing, I am aware, that says "velocity of circulation" means anything other than those other numbers combined. So I think it is a vacuous concept.

Ronald Brak
2009-Nov-16, 02:08 PM
Interesting. How is V defined?

Divide the dollar total of all transactions by the total amount of money. So if the total amount of transactions in Dopetopia amount to $10 a year and the total amount of money in Dopetopia is $2 then the velocity of money is 5 and the money on average gets cycled five times. If people in Dopetopia start working overtime then the velocity of money can increase. If people get laid off it can decrease.

NASA Fan
2009-Nov-16, 04:45 PM
I did a research paper last summer about counterfeiting. I was pleasantly surprised to learn that counterfeiting is not as big of a problem as I thought.

I pulled this from my paper

According to the Federal Reserve Bank of San Francisco, as of 2004 the amount of counterfeit currency was minimal enough to not have an effect on today’s economy. The U.S. treasury estimated that in fiscal year 2001, that less than .01 percent of the “approximately $600 billion in U.S. currency in circulation was counterfeit” (That was information pulled from the federal reserve's web-site).

One way to further reduce US counterfeiting is to use bills similar to what many European countries do
Have different size bills for the different denominations
Have different color bills for the different denominations

That would limit/eliminate counterfeiters cutting the corners of a high bill and pasting them to a lower bill--relying on a naive cashier not to notice the difference. It would also limit/eliminate counterfeiters from whitewashing a lower denomination bill and printing a higher denomination bill on it (this practice means that the paper is right, passing the counterfeit pen test).

They could also do what the Australians have done and develop a "non-fibrous bank note made of polymer substrate" According to the article I read Australia has seen a reduction in their counterfeit bills since introducing the new notes because "it is more time consuming, more difficult, and more expensive to counterfeit polymer notes, and it is easier to recognize a counterfeit note"

TrAI
2009-Nov-16, 05:11 PM
In many cases the problem is not in the systems for detecting fake currency, but in the operation of checking it. Modern bills can have a large set of security features. I am holding a 200 NOK note here now, it has, as far as I can see:
- watermarks
- register marks(geometric figures that must match between the two sides), - microtext
- metalic strip with text embedded inside the note,
- Eurion constalations,
- patterns that tend to create interference/moire patterns when copied,
- a strip with holographic print, made from what looks like metalized plastic, bordered by bars of UV fluorecent color.
- a vertical line of repeated "200"s printed in some sort of metalic/iridescent color
- fibers that fluorece in under UV embeded in the paper
- a figure representing a comet printed in UV fluorecent ink
- each color seems to be actual colored ink, not mixed by subtractive color mixing and patterns.

Also, Each value of note have different color schemes and size.

It should be rather difficult to fake it, but then, it is all up to the one recieving the note to check it, and that is the trouble, at least in shops using the traditional registers with a money drawer, the security features is useless if the person recieving the money hardly looks at the thing. Many shops have started using machines that the operator feed the money into, instead of just putting it in a drawer, and I suppose these machines do check some characteristics.

publius
2009-Nov-16, 07:35 PM
Actually not interesting. One takes three numbers and combines them in a formula to obtain "velocity of circulation". This is only of interest if one has some prior reason to suppose that "velocity of circulation" should have some properties that might affect the behaviour of the things that make it up.

But there is nothing, I am aware, that says "velocity of circulation" means anything other than those other numbers combined. So I think it is a vacuous concept.

Actually, science does this all the time. The notion of monetary velocity is an important one, and I think intuitive. If you have a pile of rusty money in a vault and are not spending it, it has no effect, giving lie to the old quantity theory of money (in the short run). What is important is not the quantity of money alone, nor the velocity alone, but the product, the monetary momentum if you will. That is very important to appreciate.

M2 velocity correlates well with the employment fraction, that is the fraction of the working population that is working.


-Richard