Lying for Money

Lying for Money is a branch book that covers financial fraud.

  • Division of labor for trust / checking up on things allows fraud to go further before detection. Consider the scenario in big short where the credit agencies where rating the bankers bonds as low risk because they were captured (the banks were their customers).
  • White collar crime is often about creating a set of transactions that look legitimate. And more troubling, when things go wrong it becomes a question of whether they intentionally went wrong through malfesance or whether the person(s) just failed.
  • Fraud is easier:
    • When dishonesty is a rare exception rather than an everyday rule
    • When trust is outsourced or where there is a “division of labor” for trust (checking up on things)
  • Fraud plays on weaknesses in checks and balances. It is what is not “checked up on” where fraud permiates.
  • A key feature of frauds is that they were stoppable early if someone had taken the time to confirm all the facts.
  • A subjunctive crime - if things turn out well for the underlying business and the wild risks that the control fraudster pay off, the victims may not know that crime ever took place.

This is a crucial theme; a good fraudster will know what the weaknesses in a checking-up process are and attempt to steer through them. But a great fraudster will target the strengths of the system that is meant to catch him, and design his deceit around that

Fraud can be enabled by the difficulty of fraud prevention mechanisms. There may be a cost (time or money) to preventing fraud. The ease of the transaction will be dictated by the relative power of the buyers and suppliers in the market. The “implicit value” of preventing fraud rises in low trust environments and lowers in high-trust environments.

The lesson from the darknet frauds is that you can build technical controls into the system, but fraud will work around them. Precautions are expensive, or inconvenient, or both, and trust is free. This means that people will substitute trust for precautions up until the point at which the “shadow cost of trust”—the expected fraud loss—begins to exceed the direct cost of precautions.

As J. K. Galbraith noted, fraud is unusual in having this time dimension to it, and it means that there is a happy period when the nation as a whole feels richer because the fraudster knows that he has stolen the proceeds, but the victim is not yet aware that he has lost them. He coined the phrase “the bezzle” to describe the aggregate stock of undiscovered fraud in the economy and suggested that its fluctuations might explain quite a few things about the business cycle.

The Canadian paradox

With that in mind, given what we know about the following two countries, why is it that the Canadian financial sector is so fraud-ridden that Joe Queenan, writing in Forbes magazine in 1985, nicknamed Vancouver the “Scam Capital of the World,” while shipowners in Greece will regularly do multimillion-dollar deals on a handshake? We might call this the “Canadian Paradox

Technical controls to prevent fraud will be worked around when the implicit price of those controls is too high.

Chapter 02 : Chapter 2: The Basics

He coined the phrase “the bezzle” to describe the aggregate stock of undiscovered fraud in the economy and suggested that its fluctuations might explain quite a few things about the business cycle.

Patterns of payment

In the first place, it is comparatively cheap to provide. If the alternative to making a sale on credit is letting the goods hang around until the customer can raise the cash, there is a saving to be made by getting them moved out of your warehouse and into the customer’s. This is particularly the case for things like fresh bread, which has a limited shelf life. Second, and related to this, it drives sales

You can tell more about the structure of an industry by looking at patterns of payment terms than you can from any “five forces” or “SWOT” analysis.


As the Liechtensteinian innkeepers knew, holding on to a customer’s valuables (otherwise known as “taking collateral”) makes it a lot easier to extend a loan to a business or a person. It reduces your risk in two ways. In the first place, it is usually easier to check out the quality of things than of people, so if your main risk is the collateral, this is easier to manage than if your main risk is a debtor’s ability and willingness to pay. And in the second, when your collateral is more valuable than the amount of your loan, the debtor stands to lose more by defaulting on the loan than by paying it back

The limiting case here is where you have established a legal right to repossess the house that the debtor lives in, which makes mortgage loans one of the safest forms of credit

Collateral is a substitute for trust. It relies, however, on strong legal institutions to make sure you can actually “realize” it by taking possession of it when a debtor defaults. The activity of “perfecting an interest,” “establishing security,” and similar euphemisms forms a large part of what lawyers send bills to bankers for, because the law on this is tricky at the best of times and damn near impossible when the borrower starts to play games

Wrong way risk

Check Kiting

check kiting4 is the simple practice of stealing money or valuable goods by paying for them with a check that you know (or ought to know) will be rejected because there aren’t sufficient funds in the bank account to honor it. In this form it is known to the specialists as “paper hanging,” and it’s often a crime of desperation or one carried out with stolen checkbooks rather than a calculated commercial decision

The important technical detail here is that because paper checks are particularly common in America, and because the check-clearing cycle is so long, American banks have—unusually in a global context—historically been very generous when it comes to allowing their business customers to make payments out of “uncleared funds,” that is to say checks that have been deposited into their account but that have not yet been endorsed by the bank that they are drawn on

What you do (in the simplest form) is that you open accounts in two banks. Call them Bank A (from which you get a checkbook with pictures of trees in it) and Bank B (which gives you a checkbook full of pictures of sports cars). Pretend for the time being that you put a token hundred bucks into each account. But now you write a check for $500,000 from your “trees” checkbook and deposit it in your Bank B account. That check is going to bounce, for certain. Except… it will only bounce when the check gets presented, and in the meantime, thinking that you have $500,000 in the bank, Bank B will not mind if you write a sports-car check and deposit it in Bank A. If Bank A sees the sports-car check, they will not mind honoring the trees check for the time being, while they are waiting for the sports-car check to clear. If they honor that check, then you can write another check to Bank B, and so on…

Chapter 03 : Chapter 3: The Long Firm

From the lender’s side, it was a perfect storm of the right conditions to get ripped off. Field warehousing was a new financial technology, which had not been thought through. The Amex subsidiary in charge of it had been given tough “stretch” targets and so needed to land big customers. The Bayonne facility was staffed by people who were comparatively cheap to corrupt. And oil floats on water.

  • delegation of fact checking is a weak point for fraud to exploit
  • Economizing on fraud controls “makes sense” because it is a tail risk. For business, since large scale frauds risk is close to zero, it is reasonable to act as though it is zero.
  • lawyers and accoutants do serve as a barrier to bad actors, but it only takes one corruptable one to get past this defense and once inside, there is rarely any further checking.

The government is often a victim of this sort of tactic, because it has some unusual characteristics as a victim (it is large, and has problems turning customers away: see Chapter 11). Many defense procurement frauds work in a similar way. But large private-sector entities are also vulnerable to shotgun/rifle approaches—something like this tactic is usually at the heart of organized insurance fraud, as well as some credit-card and mortgage frauds. Usually at the heart of a shotgun/rifle disaster, you will find a system that somebody has painstakingly engineered to optimize for the average unit cost. This is of course the right thing to do for most industrial processes, but you need to be careful that the thing you’re working on is an industrial process and not a decision-making process. Optimizing for the common case is not the same thing as trying to get the right answer, and a single medium-sized fraud can blow away all the costs saved by shaving a cent off a hundred thousand “normal” transactions

  • Insolvency
    • “commercially insolvent”“legally insolvent”“cash-flow insolvent” occurs when you have a payment come due and cannot pay it.
    • “technically insolvent”“factually solvent”“balance-sheet insolvent” when the liabilities are greater than total assets
  • “Wrongful trading” - when you are technically insolvent, but have not reached the point of being commercially insolvent and you continue to trade.

    Doug Henwood coined a monetary policy rule that “any time Donald Trump is able to borrow money to buy or build anything, interest rates are probably too low.” This reflected the tendency of the Trump organization to load up its properties with debt, extract cash out of them in unusual ways, and then end up in situations where Mr. Trump had extracted multiples of his own initial investment from a project but the bankers and bondholders were left with defaulted debt and collateral worth considerably less than its appraisal value

    • Prosecuting commercial fraud is difficult because everything a fraudster does are similar to the kind of actions a risk taking entrepreneurs would take. The difference between the two is in intent.
    • Fraudsters can also defer blame by having a “front” (knowing participant) or “patsy” (unknowing) to take the fall

What we are talking about is the version of Vincent Teresa’s classic “bust-out” that was seen in the film Goodfellas and common with the New York and New Jersey mob. A moderately successful small business—a bar or restaurant, say—gets a new business partner for its owner. This can come about as the result of a brief period of financial distress and resort to a loan shark, or the bad guys can simply walk through the door with baseball bats. In any case, control has passed from a legitimate owner to a crook, and the new crooked owner can start abusing the trading record of the company to run up fraudulent credit. It makes the simple torch-job antics of the New England Mafia look positively herbivorous

There are even more audacious ways to take possession of another company’s trading reputation. Even when fraudsters pay up front for a company, it can be surprisingly difficult for the selling owner to organize a final board meeting to formalize the transfer of title. The minutes of such meetings have a habit of getting lost, leaving the previous owner sitting around as the only remaining representative of a long firm, trying to convince the police that a bad boy did it and ran away. Never sell a company for cash.

Since companies don’t have fingerprints or passport photographs, letterheads are often accepted as evidence of identity by credit controllers, and if they get into the wrong hands, they can cause serious mischief—fraudsters can write credit references for each other, and even place orders.

Fencing the goods acquired in a long firm is easier than doing so with the proceeds of a robbery simply because the crime involves a time dimension.

Businesspeople expect to be paid for waiting—that’s why they say things like “time is money.” And the time value of money means that while a sum of undiscovered fraud exists, the size of the deception is often growing. This complicates the economics of the thing a lot.

“cool out the mark” - This useful phrase, brought to prominence by Erving Goffman, refers to the practice of con men of all kinds to assign some time and effort to the task of persuading the victim to frame things differently—to see himself not as the victim of a criminal act that requires justice, but as someone who has had a bit of bad luck, or engaged in a failed venture. Goffman contends that, sociologically, lots of noncriminal institutions in everyday and bureaucratic life can, once you have recognized the phenomenon, be seen as having functions broadly similar to the cooling out of marks.

Chapter 04 : Chapter 4: The Snowball Effect

It looked like a free-money machine, and all he needed was more capital. He took down the sign on his rented office for “The Bostonian Advertising and Publishing Company” and put one up for “The Securities Exchange Corporation.”1 He was ready to start borrowing money

The denouement came just as Charles Ponzi was on the point of another ahead-of-his-time discovery in the world of fraud—he had begun to make loans from the Hanover Trust to his own corporation in order to cover cash demands, anticipating the “control frauds” (see Chapter 7) of the Savings and Loan crisis by half a century

First, it can obscure the essential nature of a pyramid; people who would immediately spot a “gifting” pyramid are often less likely to notice the underlying economics if the scheme is dressed up to look like a normal business. There is less of a sense of “something for nothing,” which tends to raise the skepticism of potential recruits.

And second, depending on the product, it raises the possibility of the pyramid becoming a legitimate business. If the recruits are actually selling something valuable to the general public, it is no longer the case that the scheme requires constant new recruits to keep making payments; in a steady state, it is not as lucrative to the upper levels, but the lower levels do not give their money away for nothing. There are actually some legitimate businesses that recruit their sales force on the basis of their being able to earn “downline commission.” And this creates a very strange gray area. It can be quite possible for something to exist and operate and for nobody to be quite sure whether it is a pyramid scheme or not.

But one key difference between fraudsters and legitimate businesses is that compound interest, the driver of growth and returns for the honest firm, is the enemy of the fraud. The reason for this is that unlike a genuine business, a fraud does not generate enough real returns to support itself, particularly as money is extracted by the criminal. Because of this, at every date when repayment is expected, the fraudster has to make the choice between whether to shut the fraud down and try to make an escape, or to increase its size; more and more money has to be defrauded in order to keep the scheme going as time progresses

Consider a really simple investment fraud and take away all the detail to concentrate on the mathematics of the investment returns. You take in a million dollars from your investors, promising them a return of 25 percent on their money. Instead, you steal it. A year goes by, and you aim to keep the fraud going by raising new money from another set of mugs. You get a tame accountant to “verify” that you have made a 25 percent profit, and armed with these amazing performance figures, you go out on the road to raise… how much? At the very least, you have to raise $1.25 million. You need to pay back your investors not only their original stake, but the fictitious profits that you pretended to earn. And it gets worse. Even presuming you never steal any more, at the end of year two, the people who gave you the $1.25 million are going to want $1,562,500 back. If you keep going for five years, then your original million-dollar theft will require you to commit just over $3 million of new fraud

Ponzi schemes show this feature of crime most clearly. But a similar problem tends to show up in any kind of fraud that necessitates keeping two sets of books. As soon as money is extracted, there is a gap between the “real” set of accounts and the “public” set. This gap needs to be filled by some sort of fakery—either an accounting fiddle or a simple lie that the cash is there when it isn’t. And once the gap is there, it will tend to grow

An embezzler, though, or a rogue trader or a tax swindler, has to cover up the existence of the crime itself. That means that it’s very difficult to be a one-time-only embezzler

The need to manage the snowball effect is one of the biggest problems facing a fraudster with ambitions to steal a lot or keep going for longer than a single billing cycle. If you want to steal a lot of money, you have to keep the fraud going. You also have to keep the fraud going if you haven’t figured out your escape route yet, or if you just blundered into it and don’t have a plan at all

the name “Ponzi scheme” is annoying because it doesn’t really define a particular kind of fraud. In glossaries of financial terminology (and, disappointingly, on the SEC website warning investors about pyramid schemes), it is often suggested that the defining property of a “Ponzi” fraud, as opposed to any other kind, is the use of money raised from new investors to make payments to old investors. However, this is basically true of any fraud that aims to maintain a persistent economic existence for any

It’s also important to note that the SEC definition of a Ponzi scheme is actually so wide that it captures many important economic phenomena that aren’t frauds at all

And the claim that Social Security is a Ponzi scheme is itself part of the basis for calls to dismantle it and replace it with something that’s easier for the financial-services industry to charge fees on

There are all sorts of things going on in the background with investment in stocks and bonds, but from the point of view of the customer/investor, the thing just works like a bag of coins—you put money in, then you take money out. This is a “fully funded” approach to pensions, as there is an identifiable balance of savings that is associated with the particular pension liability

The other way of doing pensions is called “pay as you go” in developed economies, but it’s basically the Stone Age model. You work until you can’t work anymore, and then you sit by the fire, call yourself an “elder,” and have food brought to you

The problem arises because at the level of the whole economy, there is no such thing as a fully funded system. Can’t exist. It’s a monetary illusion. People who are retired have to consume goods and services that are produced today by the people who are working today, not goods and services produced thirty years ago by themselves

This makes the accounting of the Social Security retirement trust fund pretty confusing. The liabilities of Social Security are an estimate of the value of its commitments to pay pensions in the future. From an economic point of view, there is no “asset” corresponding to this liability, because those future pension payments are going to be made by the United States of America, using its extremely valuable general-purpose asset called “the right to levy taxes on the biggest economic entity the world has ever seen.” But in order to make the books balance, the trust fund is credited with a series of made-up US Treasury Bonds.

The assets are the same thing as the liabilities. The only sensible way to participate in any debate over the Social Security OASDI Trust Fund is to ignore it. It’s not really a fund; it’s a way of recording a promise to pay pensions in a very public way, to nail down the politicians and make it harder for them to renege on that promise in the future

The use of new victims to refinance old ones is certainly one of those properties. But it’s by no means the only important one

In general and with only a few exceptions, economic things get to be big by starting small and growing, rather than being set up as a big thing in the first place.

So the second key property of the Ponzi fraud is that it’s one that is designed to create an institution that will last for an indefinite period and that will grow; indeed, it will grow exponentially and at a rate that is compounded by the need to generate fictitious returns on money that has been extracted by the fraudster.

A final property of the Ponzi fraud, though, and one that in some way preserves the intuition of the popular definition, is that as with our own taxonomy of long firms, counterfeits, control frauds, and market crimes, the Ponzi scheme can also be defined by the kind of economic institution that it exploits and subverts. As we hinted earlier in talking about the danger of running a pyramid fraud with borrowed money, a key property of many of the things that get described as Ponzi schemes is that they need to undermine one of the “social technologies” that have evolved over the centuries to manage business relationships—the institution of “maturity dates.”

If you’re going into partnership with someone, you need to find out a lot about them, particularly if you need to weigh the possible returns from the venture against your own up-front contribution of capital. If you’re lending money, though, all the interesting questions about the future boil down to one relatively simple one: “Is this person (or this limited-liability company, after about 1811) good for the money?” And the great thing about that question is that it’s often surprisingly easy to answer. Most loan propositions are either “good and look good” or “bad and look bad.” The business of moneylending is to be better than your competition at spotting the minority that are “bad, but look good

What kills you in the moneylending business is not bad loans, but good loans that went bad. Either business conditions change, or you find something out about the character of your borrower that makes you change your mind about whether it was a good idea to lend to them

Equity shares in a company have no maturity

In summary, borrowed money usually comes on terms that mean that it needs to be refreshed. And so any fraud that persists over more than one loan-maturity date needs to in some way subvert the use of rollovers as a means of control and checking. This is the final Ponzi property—a fraud can be called a Ponzi scheme with greater validity, the greater the extent to which the mechanics of the crime revolve around managing the renewal of its financing and convincing the investor and lender communities to keep their money in the scheme rather than demanding repayment in cash

The second option is usually better for the fraudster in the short term, because it economizes on the need to find cash. It was for this reason that Barry Minkow and ZZZZBest always tried to persuade their investors to roll over their restoration jobs—doing so reduced the need for Barry to kite checks. However, in the long term, the greater the extent to which the proceeds are reinvested in the same fraud, the bigger it grows and the faster it will collapse

A slightly more uncertain, but quite logical way of setting the equivalent to a maturity date for a debt is to link it to the conclusion of a business episode that is in some way connected with the funds to repay it

The same scam works with ostriches,12 orchids, fine wines in the barrel, and all sorts of other things that sound as if they might be valuable but for which there is no readily available market price and where the investor is heavily dependent on the promoter for information about when the cash returns should be coming in.

But for the time being, it’s just worth noting that laws get written in response to problems that they are meant to address, and that loopholes in those laws often exist because it’s not yet become worth anyone’s time and effort to make a law against something

That, of course, is what Bernard Madoff did. His scandal was based on a considerable refinement of the Ponzi scheme, and on a close understanding of investor behavior. The Madoff fraud relied on a distinction between the legal “maturity” of his investors’ claims on the fraudulent fund and the behavioral maturity. On paper, Madoff’s clients could ask for their money with ninety days’ notice. In the real world, however, they left it in there much longer. And the way that fund redemptions worked, inertia was on Bernie’s side. There was no “rollover” process; on any given day, if no redemption request was received, the fund would presume that none was wanted and the ninety-day clock would reset.

This is quite a general feature of the economics of fraud. Because verification is expensive while trust is cheap, a common strategy for scrimping on checking costs is to use onetime checks, after which a party is within the circle of trust. This system has two failure modes; attacks based on faking the initial verification (which we call “counterfeits” and discuss in Chapter 5) and attacks from trusted parties who later become untrustworthy (generally a subset of the “control frauds” discussed in Chapter

The first step in the game was to control the information flow

The second step was to control the growth of your fund to match your financial needs.

But you need to manage your investors, to make sure that you don’t get people who are going to check up on something that you have not previously thought to fake

Madoff had a few nasty moments in 2005, when Bayou collapsed and everyone looked round to see which other hedge funds used strange, obscure accountants that nobody had ever heard of. But the great crash of 2008 did for him just like it did for everyone else. It wasn’t so much that the outflows accelerated (although they did) as that the inflows dried up entirely

They’re IOUs. That’s all they are. You and your friends might write a chit for twenty dollars, and that’s what the banks do too

Because money is just IOUs, there doesn’t need to be any secret method for creating it. You feel like there ought to be but there’s not. When a bank makes a loan, it doesn’t need to borrow deposits first, because putting the loan proceeds into the borrower’s bank account is just a matter of telling the borrower’s bank “I owe you.” The system works on a version of trust that’s so ubiquitous that it’s almost impossible to break

Big and sometimes important exception to this generally valid point: houses, which do last a long time. Buying a house with a mortgage is an important form of retirement saving. And it’s got some very attractive properties as a retirement-savings instrument, because the value of a house is going to be closely related to one of your biggest expenses in retirement: housing

Chapter 05 : Chapter 5: Counterfeits

First, the Bank of Portugal, despite having the monopoly on printing escudo banknotes, was a private company with shareholders (this was not unusual; so was the Bank of England until it was nationalized in 1945, and the Federal Reserve still nominally is2

The Portuguese Banknote Affair remains one of the most tragic cases in which the weak link in a high-trust society (in this case, notaries) ended up pulling down the whole structure of trust itself. In 1955, Alves dos Reis got an obituary in the Economist saying that his scheme had been good for Portugal on Keynesian principles, which probably ranks as one of the stupidest things that newspaper has ever printed.

Now as then, the key input to the process of carrying out a mining fraud is the manipulation of assay reports and geologists’ studies, and the way that you do it is still to introduce gold from somewhere else into a sample taken from under the ground.

The key to any kind of certification fraud is to exploit the weakest link in the chain; in this case, the initial logging and sealing of the drill cores, which took place at a remote mining camp high up in the Indonesian jungle, where the only people present were either laborers or very junior geologists hired by de Guzman personally and intimidated by him

Gaps like this—between the facts that a certification authority can actually make sure of, and those that it is generally assumed it can—are the making of a counterfeit fraud. Having gotten one set of facts certified, the crook immediately promotes his operation as if the whole thing had a seal of approval.

The central structure of the scam is that you take an industry in which a big up-front investment is needed in order to earn rewards at a later date—the crucial dimension of time that defines the commercial fraud as compared to other kinds of theft. Then you recruit outside investors to pay that cash, and divert the cash to yourself rather than investing it in the project. The counterfeit is the keystone of the fraud as it provides the reason for the victims to transfer the cash.

This “smoke and mirrors” approach seems to make an appearance in nearly every biography of any of the tech giants, and it is not always apparent that the authors understand that faking a nonexistent technology in

What is going on here, of course, is that these professions are considered to be circles of trust. The idea is partly that the long training and apprenticeship processes of the professions ought to develop values of trust and honesty, and weed out candidates who do not possess them. And it is partly that professional status is a valuable asset for the person who possesses it, which can be taken away with a stroke of the pen by a disciplinary tribunal. So the sum of money needed to corrupt a professional is significantly larger than that needed to corrupt a layman; as well as compensating for the risk of being caught and paying the normal penalty, a professional needs to be compensated for the risk of losing his or her professional status, and the lifetime stream of earnings that would otherwise derive from it

If we presume that the premium earned by someone with a professional qualification over the same person without that status is in the order of tens of thousands, and that the duration of a professional career is around twenty-five years, it is not difficult to see that the majority of lawyers and accountants, even if they have no moral sense at all and are looking for opportunities for corruption, are not necessarily going to find it worth their while to get involved in a fraud in which their share of the proceeds is less than about a million dollars

Another way into the circle of trust, of course, is through an affinity group, as we saw with pyramid schemes. In principle, the economics of this ought to work; the more socially connected you are to a person, the more interests and ties you share, and the more costly it would be for that person to rip you off

In many ways, part of the explanation why Californian venture capitalists are so relaxed about fake demos is that their main filter to weed out frauds and flakes is more like an affinity network than anything else

Both sides of the pharmaceutical-fraud problem, however, are facilitated by the fact that certification is supposed to be a onetime process. As with a notary or other professional, one of the roles of a professional is to provide assurance and trust to other parties, so that checking effort does not need to be duplicated. For this reason, strong social norms exist against questioning or second-guessing the judgment or opinions of professionals. When these social norms get intermingled with all the emotional and cultural baggage that is inevitably attached to the concepts of sickness and disease, the end result is a very powerful collective mental block

We can note once more that the crime of drug counterfeiting is essentially an attack on the certification system; even if the counterfeit is chemically identical to the “real” drug, there is still a diversion of profits from what was intended by the patent system, and the fake drug is still fake. The thing is, the certification system for pharmaceuticals is also a safety system. A counterfeit manufacturer cannot be subject to the same audit requirements and checks on manufacturing processes and purity of ingredients, because all the authorities that might be able to certify these elements of the manufacture are also part of the system that underpins the patent

In practice, what happened is what we might have expected from the original insight that fraud is an equilibrium quantity; as the audited chain has become more inconvenient, it has become less relevant. The majority of counterfeit pharmaceuticals bought in the developed world are now sold through unlicensed internet pharmacies. It’s quite possible for people to become the victims of a darknet exit fraud while trying to commit a counterfeit-drug fraud. As the Ghanaian proverb has it, “When a thief thieves a thief, God laughs.”

The lesson of Vioxx is that the concept of a crime against the certification system—a counterfeit in the most generalized sense—is a breach of an overall system of trust, not an isolated attack on a single victim like a long firm.

Chapter 06 : Chapter 6: Cooked Books

But the most common one is that you want to show your crooked accounts to investors so that they give you money. For this reason, any discussion of accounting fraud needs to be put in the context of stock-market fraud, because one is usually the point of the other

A public financial market provides the same service to liars that it provides to honest businesses—it converts stories into cash

The important abstract concept here is that the stock market capitalizes expectations of future profits—literally, it is the place where claims on these expected future profits are exchanged for a capital sum. The story is turned into cash.

it is the institution of buying and selling shares in companies that converts the story into cash, rather than the stock market per se

The first is that the process of checking up on you is distributed across the market rather than being assigned as the responsibility of any one individual. And better than that, it’s distributed across a lot of individuals, most of whom have only a small stake in your story. Like a lot of other areas where counterfeiting frauds work, stock markets tend to operate on a certification basis; there is meant to be a lot of checking up (“due diligence”) at the first occasion on which you enter the market, but once you’re in, you’re in. Compared to the size of the potential opportunity for fraud, there is surprisingly little ongoing monitoring of companies that are already quoted on a stock market. The numbers that a company posts are, for the most part, treated as facts and it is very difficult to gainsay them from the outside

But the real fraud committed by Stratton Oakmont was in facilitating so many companies to go public in the first place—taking money from investors on the basis of false accounts. At base, it was another example of counterfeiting: a certification fraud against anyone who believed that a seemingly reputable brokerage would not get involved with fake companies.

One final thing worth knowing about the way that stock markets capitalize future earnings—they don’t always do it very efficiently. In principle, everyone ought to focus purely on cash, and do so with a very long-term forecast horizon. But that’s extremely difficult, so people who actually have to trade shares will often look for shortcuts. One such shortcut is to look only at short-term earnings and assume they’re a reasonable proxy for long-term cash flows. This means that manipulating a single year’s earnings, even at the expense of future years, can often create enough of a bump in the share price to extract some fraudulent cash. And another important shortcut is that for fast-growing companies, people often look at revenues (or sales) rather than earnings, presuming that these are a better guide to the long term than a profit number that might be weighed down by start-up costs. This can be useful to an accounting fraudster: revenues are often easier to manipulate than earnings

There are two species of false accounting. It is tempting to call them “accounting fraud” versus “accounting manipulation,” but really they are both forms of fraud. Just think of them as “things you have to lie to the auditors about” and “things that the auditors will help you with

Completely fake sales As in sales that never happened, but that you record in a ledger as if they did, and present the fictitious ledger to the auditor. Usually best to make these fake sales overseas, perhaps in an emerging market, but in any case somewhere where there’s a substantial time-zone gap or an equally substantial language issue and where company registration is opaque.

Fake sales by economically meaningless transactions

Two property developers, for example, might sell small parts of their developments to each other at an inflated valuation in order to create a “recent transaction” that would allow them to value their whole business higher, and persuade their banks to keep lending against what they believed to be valuable collateral.

You find another company that the auditors don’t know is controlled by you, either by directly being owned by a connected party, or by being an actually independent company that has agreed to take the role of a “front” for this transaction. Then you lend this company a sum of money (either as an ordinary loan, or by buying newly issued shares in it), on condition that it uses the money you lend it to buy goods from you. In this way, you can effectively buy from yourself, recycling money raised from investors and making it look like profits

Up-front recognition of revenues

the “Kutz Method,” after the accountant who managed to persuade their auditors to agree to it. In a long-term contract, like a lease, the usual method of accounting is to report the payments made by the customer as revenue, spread out over the life of the lease.5 The “Kutz Method” was to record all the payments as revenue on the day the contract was signed

Delayed recognition of costs

Completely fake assets

although the biggest fake-asset frauds tend to involve offshore bank accounts

Unreported debt

You don’t want to show too much debt on the balance sheet. But you do want to borrow too much—borrowing money in a corporate vehicle you control is a basic step in getting cash out of a fraud. So what do you do? Get a separate company to borrow the money, telling the bank that you guarantee the debt, then (because the separate company is controlled by you) use the loan for your own corporate purposes

Overvalued or undervalued assets

Auditors, analysts, and other disappointments

All of these frauds should have been prevented by the auditors, whose job it specifically is to review every set of accounts as a neutral outside party and certify that they are a true and fair view of the business. But they weren’t. Why not? The answer is simple; some auditors are crooks, and some are easily fooled by crooks

But as a crooked manager of a company, churning around your auditors until you find a bad’un is exactly what you do, and when you find one, you hang on to them. This means that the bad auditors are gravitationally drawn into auditing the bad

companies, while the majority of the profession has an unrepresentative view of how likely that could be.

there is the problem that even if an auditor is both honest and competent, he has to have a spine or he might as well not be. Fraudsters can be both persistent and overbearing and not all the people who went into accountancy firms out of college (and then ended up doing audit rather than a more glamorous specialty) did so because they were commanding alpha-type personalities

As with several other patterns of behavior that tend to generate frauds, the dynamic by which a difficult audit partner gets overruled or removed happens so often and reproduces itself so exactly that it’s got to reflect a fairly deep and ubiquitous incentive problem that will be very difficult to remove.

Wall Street Confidential By way of a second line of defense, investors and brokerage firms employ their own “analysts” to critically read sets of published accounts

A set of fraudulent accounts will often generate “tells”—in particular, fraudsters in a hurry, or with limited ability to browbeat the auditors, will not be able to fake the balance sheet to match the way they have faked the profits. So inflated sales may show up as having been carried out without need for inventories, and without any trace of the cash they should have generated.9 Analysts are also often good at spotting practices like “channel stuffing,” when a company (usually one with a highly motivated and target-oriented sales force) sells a lot of product to wholesalers and intermediaries toward the end of the quarter, booking sales and moving inventory off its

The independent, market-driven watchdogs suffer exactly the same flaws as those with statutory status and backing. The problem is that spotting frauds is difficult, and for the majority of investors not worth expending the effort on. That means it is not worth it for most analysts too. Frauds are rare. Frauds that can be spotted by careful analysis are even more rare. And frauds that meet both the preceding criteria and are also large enough to offer serious rewards for betting against them come along roughly once every business cycle, in waves.

The private-equity industry”—people, and pools of savings that they manage, who do deals buying stakes in companies but not on the normal stock market. Two main kinds—management buy-out merchants, who do what the name implies, and “venture capitalists,” who invest in companies that don’t have enough profits of their own to raise money from the public. The former kind tend to get defrauded by accounting misstatements, the latter to be victims of companies that are simply frauds from top to bottom. Neither of them is keen on admitting they’ve been taken for a ride, as they need to portray themselves to their own clients as well-connected geniuses.

Matching the time period in which you report revenues and costs to the time period in which the services or goods are provided is known as the “accruals principle” and accountants fight the most unbelievable holy wars over how it should be done properly

Fraudsters are also highly reluctant to dig into their own pockets to pay tax on fake accounting profits with no corresponding cash, so an anomalously low average tax rate is also often a red flag

Chapter 07 : Chapter 7: Control Frauds

Sometimes, the fraud is coming from inside the building. The person best placed to steal from a company is often the person in charge of it, and the person who is best placed to nullify the systems and controls intended to detect and prevent fraud is the person who is in charge of administering them. A “control fraud,” in the broadest sense, is a fraud that depends on the delegation of management and control from someone who bears the risks and rewards of success and failure, to someone who does not ultimately bear those risks, but who can arrange the business so as to benefit from the rewards

There are few control frauds in the Bible or in classical literature, because they can’t exist in a world in which trade and industry take place on the basis of small workshops and warehouses, managed by their owners

The most common white-collar crime committed by insiders, of course, is simple embezzlement, or hand-in-the-till fraud. In second place would probably come the “kickback”—a side payment made to a purchasing or sales manager by a supplier or customer of the company in return for preferential treatment on a contract. Neither of these, though, is really a “control fraud” in the strict sense of the term; they are straightforward thefts that happen to be committed by employees of the company

The really economically interesting kind of control frauds are the ones where having control of the enterprise is essential to the nature of the crime, rather than just being the means by which the opportunity to steal presents itself. Someone with management control has unique abilities to defraud in ways that others simply cannot. In particular, control of an enterprise allows you to commit a subjunctive crime

This is the nature of the control fraud; as well as the other dimensions of white-collar crime, it rests on the ability of the manager to take action on behalf of someone else, and therefore to take much bigger risks than that someone else would ever have regarded as reasonable if they were acting on their own behalf. This is part of the reason why the most egregious examples of control fraud tend to be found in the financial sector, because that’s the part of a modern industrial society in which there is most delegation of risk-taking authority.

The principle that “the failure of your employees to obey your company’s rules is not my problem” is known as the “indoor management principle,” or more poetically “Turquand’s Rule,” after the liquidator of a nineteenth-century railroad who discovered that the directors had illegally run it into bankruptcy and failed in his bid to get the fraudulent debts struck out.

you. Then, on the basis that 10 percent of something morbidly obese can be worth more than 100 percent of the same thing at its healthy weight, you start taking measures to inflate the size of the thing you control, letting the rewards naturally flow to you. That’s why it’s such a pernicious fraud and so difficult to detect; the individual mechanics are all innocent and the crime is in the overall scheme

The way that investment banks pay bonuses5 as a percentage of trading revenues turns the compensation system into something of a control-fraud construction kit. Traders get a share if they make a profit, but typically don’t have their bonuses reduced below zero if they make a loss

There is a second line of defense for a bank that has screwed up and missed someone breaching their risk limits. And that is that trading is a cash business. Every day at close of business, trading books are “marked to market” and the daily profit or loss calculated. If there is a loss for the day, you need to post that amount of cash with the exchange

This brings in the third element of the rogue trader—the abuse of market conventions to generate fraudulent funding. Although most positions are “margined” every day in the way just described, some trades generate cash up front. As we saw in the description of the Brazilian straddle, one such trade is the sale of an option, or any other insurance-like contract

This is important, and we introduced the conventional “incentives” theory of rogue traders because the channel from the size of the trading book to the size of the bonus is the engine of the control fraud

When you take into account the likelihood of losing future years' bonuses—which resembles a kind of downside risk from losing too much—the calculation isn’t close. In the real world, there is much more of a problem with traders being too risk averse and “banking” profits if they have a good start to the year. Normal people responding to the structural incentives of the investment-banking industry don’t act like Nick Leeson.

The temptation is always there, but it is only tempting to people for whom the structure of the bonus system is a psychological hunger rather than a rational calculation

The key to the fraud is the ability to control information about the size of the position, and to keep funding the losses by having a sufficient position of trust to persuade the bank to keep making payments. It is not so big a step from the rogue trader to consider what might happen if an entire bank came under the control of fraudsters and if the rogue was at the very top.

The crisis had its roots in the 1970s and the attempt to tame inflation by raising interest rates

As interest rates went up, the rate paid on deposits went up with them while the interest on the mortgages stayed the same. The 3-6-3 rule became a 12-6-12 rule: if you’re paying 12 percent on deposits but still only getting 6 percent on loans, you’ll be bankrupt within twelve months.

The first one removed size limits on S&Ls, in the hope that they would be able to “grow out of trouble.” Immediately, a wave of mergers began

So the regulations that kept them out of business loans, commercial-property loans, and—crucially—direct ownership of speculative property developments were also relaxed, on the basis that higher risk meant higher return, and that was what was needed.

Before long, the S&L industry had moved on from its history of small local banks, and was increasingly characterized by quite large organizations, often owned or controlled by property developers, and several of which were connected to the junk-bond financier (and later convicted securities fraudster) Michael Milken. It was at this point that the second phase of the S&L crisis can be said to have begun.

However, although appraisal fraud starts with a corrupt appraiser, it usually cannot end there. Bank examiners would have an easy time if all that was backing up a crooked bank’s books was a set of subjective opinions of nominally independent “professionals.” To make the most obvious point, an appraiser only has a certain amount of leeway in his or her valuation to make assumptions that depart from the price paid in relevantly similar recent open-market transactions. If you want to systematically fiddle the valuations, you need to fiddle the market too.

t is at this stage that the “control” element of a control fraud starts to become the key to the operation. The more things you control, the greater your ability to create fake evidence to justify the fake valuations that underpin the fake profits earned from your fake assets. Control of the top of the organization allows you to undermine all the controls and completely corrupt the infrastructure of trust and checking that is meant to protect the system against external attackers. So a Charles Keating figure has a surprisingly wide menu of options to choose from in creating fake “open-market” transactions

Cash for trash. A borrower is desperate for a $10 million loan against a property that is probably worth $8 million. You refuse to lend it to him. Instead, you tell him that you will lend him $30 million, as long as he uses $20 million of it to buy another exactly similar property from one of Charles Keating’s real-estate companies. This is a wonderful scheme for Keating; he has extracted $20 million in cash in exchange for an $8 million property—a profit of $12 million. And it looks good for the S&L too; the supplicant borrower’s

frauds. Like playing chicken by running back and forth across the train tracks, each individual project might be considered “high risk” but the eventual outcome is certain. The difference between good and bad borrowers in this context isn’t one of probability—it’s one of quality

The S&L crisis was an example of the classic control fraud. A crook took control of an economic entity and used it to create fake profits, which could then be extracted by “legitimate” means (in the sense that large fake profits justified large salaries, bonuses, and dividends, which are all normal ways to distribute earnings, as opposed to theft). It’s intellectually interesting because of the change in the method of theft, but it’s still a case where a single controlling mind took a decision to steal money

The control fraud also, however, permits of an even higher degree of abstraction. In the run-up to the 2000s financial crisis, we saw the development of what might be called a “distributed” or “self-organizing” control fraud

One important aspect of a control fraud is that the method of extracting cash is usually not intrinsically criminal. You set up an organization such that it will legitimately pay out a percentage of its value to you through dividends, bonuses, and commercial transactions with other controlled entities, then you blow up the size of that entity to an absurd level by taking on a load of debt, and let the normal and legitimate mechanisms of the corporation transfer the fraudulent value into your pocket

rather than organizing the fraudulent inflation of the corporation yourself, you simply set up a system of incentives and (non-) checks and balances, such that other people were likely to inflate it for you? In other words, rather than committing crimes yourself to inflate the value, you just created a criminogenic10 environment in the firm, and let nature take its course

One way to think about what happened was as the result of two interlocking control frauds. In order to get the mortgage securitization market off the ground in the first place, the investment banks had to bend the rules, if not commit outright fraud. The system that it put together facilitated the commission of another massive fraud12 on vulnerable borrowers. And the consequences of this fraud folded back to the initial fraud, with the investment banks attempting to extricate themselves from the consequences of the second fraud by turning the first system of sharp-but-probably-legal practice into a definite and massive fraud. Let’s start with the organizational and legal disaster area that was called MERS

The banks decided that they would create a central database, called MERS (the Mortgage Electronic Registration System). They would also create a nominee company, also called MERS

The intuition I’m trying to establish here is firstly that in order to understand these things you have to adopt an analytical approach that a lot of people find to be pretty terrifyingly amoral. And secondly, that everyday consumer financial products are often much trickier than the stocks, bonds, and derivatives of high-end finance, because they have all these little features built into them to account for consumer behavior. And you have to analyze them based on the way that they are actually used, not by looking at the interest rates and maturity dates on the paperwork. The key to understanding the subprime-mortgage distributed control fraud is to think about early repayment

But in a market in which incomes are stagnant, interest rates falling, and house prices rising, that prepayment penalty starts becoming crucial to the economics of the thing.19 It was typically set at an amount of money roughly equal to six months’ interest payments. So if you got into a situation in which borrowers almost all refinanced after two years, then the lender effectively got five mortgage payments for the price of four. That’s quite a sizable bonus, particularly if it happens over and over again.

You want to make sure that the loan is affordable at the teaser rate, but completely unaffordable at the “go-to” rate. If this happens, then the borrower is effectively stuck and forced to refinance whether their income has gone up

And so, like a minimum-payment credit-card borrower in the sweatbox, the subprime-mortgage customer would be making their monthly payments on a loan that grew rather than shrank every two years. This was, not wholly unfairly, characterized by banking-sector critics as “renting a customer to sit in the house you are betting on

It’s the same answer every time we ask about the typical victims of fraud. Greedy people, desperate people, and people who didn’t know what they were doing

This was the criminogenic incentive. Nobody really decided to do this, but if you pay over the odds to buy a particular type of loan product that only suits a very small proportion of borrowers, then that loan is going to be sold to people whom it doesn’t suit. And the structure of the Option ARM lent itself to misleading sales practices

Texas didn’t have a housing bubble in the 2000s. This came as a surprise to some people as the Lone Star State had historically been quite prone to real-estate booms and busts; it had been the second most important locus of the S&L scandal and Don Dixon of the Vernon S&L was almost as big a crook as Charles Keating. But during the particular distributed, self-organizing control fraud that characterized the bubble of the 2000s, a few pieces of banking law that had been hanging around in the state constitution since the turn of the century suddenly became useful. Texas did not allow loans that had negative amortisation. It restricted mortgage loans to 80 percent of the appraised house value and prohibited the act of taking out a mortgage to pay other debts. And finally and most importantly, Texas prohibited the charging of prepayment fees on “high cost” (which is to say, subprime) mortgages. A few other states (Vermont, for example) had similar laws and saw similar effects

falsifying documentation to make it look as if the mortgage had been transferred years ago. The term robosigner was coined to describe a bank employee who had his or her signature on so many loan agreements that it was obviously impossible that they were originals; they were just facsimile loan documents that had been put out in a row on a table and signed, one after another, for hours on end. And then delivered to the court system as if they were originals

So we have two big deals here—the system of incentives that created the Option ARM market, and the tragicomedy of errors that produced MERS. Nobody ever sat down and decided to do them, but they happened. And while they were happening, a very large amount of money flowed in bonuses and stock options to the C-suite executives. There’s a strong feeling that somebody should have gone to jail over this.

Goodwill” is the key here. If you have a bunch (say $100 million) of mortgages paying 5 percent interest, and the going rate is 10 percent, then your mortgages are actually only worth about $50 million. If you write them down to that value, you’re bust. But if someone takes you over for $100 million, they can say that the mortgages are clearly only worth $50 million, but your unique franchise, reputation, etc. is worth another $50 million. It sounds crazy, but even honest auditors will usually go along with it—as the acquirer is paying $100 million and he is presumed to know his business. If interest rates subsequently fall to 8 percent, the acquirer can sell your mortgages for $80 million, locking in a real loss of $20 million but declaring an accounting profit of $30 million. The way in which goodwill is accounted for has changed since the 1980s, but it’s still a big problem area.

Chapter 08 : Chapter 8: The Economics of Fraud

Once you have written down the list of things you would need to do in order to turn your workplace into a successful fraud, sit down and have a look at it. Isn’t this a useful document? It shows you: What the key indicators are that show whether your business is doing well or badly. What a really good set of numbers (and maybe even nonnumerical performance indicators) would look like. How growth and compound interest are expected to affect the legitimate business over time. What questions you should ask of a really good set of numbers to make sure that they reflect a good reality rather than someone manipulating them. In other words, to understand how to defraud something is to understand how to manage it.

of, “thinking like a fraudster” might be a way to generate new insights

In fact, we can see that there’s almost an inverse relationship between our imaginary pairs of templates (management/fraud and fraud/management). The easier something is to manage—the more possible it is to take a comprehensive view of all that’s going on, and to check every transaction individually—the more difficult it is to defraud. Vulnerability to crime, in other words, tends to scale with the cognitive demands placed on the management of a business.

It also gets easier with increasing uncertainty with regard to what a “normal” or valid transaction looks like; that’s why so many big frauds occur in brand-new business lines where there has been no time to establish a baseline of common practice.

Modern crime is driven by the cognitive demand placed on managers because it exploits the technologies used in an industrial society to manage that cognitive demand. Fraudsters parasitize the economy by attacking the systems we use to economize on knowledge, information, and attention.

As Hayek noted, the benefit of a market system is that it economizes on information gathering, by allowing aggregate production and consumption decisions to arise organically out of lots of small transactions rather than one big plan

That might be solved by technological progress. He thought it was insoluble, because most of the information that you would need to plan an economy was “tacit”—embodied in personal experience, spread out across the production units themselves, and not available, even in principle, to any information-collecting authority. Although this very strong version of tacit knowledge is controversial, a weaker version of the idea has been very influential indeed

It is worth contemplating for a moment a very simple and commonplace instance of the action of the price system in order to see what precisely it accomplishes. Assume that somewhere in the world a new opportunity for the use of some raw material, say, tin, has arisen, or that one of the sources of the supply of tin has been eliminated. It does not matter for our purpose—and it is very significant that it does not matter—which of these two causes has made tin more scarce. All that the users of tin need to know is that some of the tin they used to consume is now more profitably employed elsewhere and that, in consequence, they must economize tin. There is no need for the great majority of them even to know where the more urgent need has arisen, or in favor of what other needs they ought to husband the supply

large cluster of these long-term contracts is what we call a “firm,” and Ronald Coase’s contribution to this strand of intellectual history was to set out the circumstances under which firms would form, and how the economy would tend not to the frictionless ideal, but to be made up of islands of central planning linked by bridges of price signals.

The price mechanism and decentralized markets work to use private information at the firm level, but within the firms, managers are as blind as Soviet central planners ever were. The problem of trying to make sure your desired outcome happens when you can’t directly monitor the person responsible for doing it is, in the most general terms possible, known as the “principal/agent problem,”

The basic idea is usually to create something that works a bit like a price signal, to allow the private information to be revealed, and then to design a contract based on this price signal that aligns the incentives of the “agent” (the employee) with those of the “principal” (the manager or owner) as much as possible. So if you wanted someone to trade LIBOR futures on your behalf, but you couldn’t tell how good they were at it or how hard they were working, you might design a contract based on something you could measure (like their trading profit) and align their incentives with yours (by paying them a bonus based on it). I have picked this example intentionally to warn the reader that the approach often doesn’t work very well

And it could almost as fairly be said that a very great proportion of management theory since Taylor has been made up of calls to measure different things, in order to correct for the biases introduced by the previous round of changes

Management theory tends to cycle between trying to deliver efficiency, quality, and customization. If you mainly measure cost indicators, quality tends to suffer. If you mainly measure quality indicators, the cost tends to drift up. If you refuse to compromise and demand both low cost and high quality, you tend to find you have concentrated too much on your own production process and not enough on what the customers want. If you try to achieve all three goals at once, you tend to go mad

The underlying problem is that most of the time, we are trying to manage or administer things that are too complicated to be aware of every detail at every time, so we need to choose what we hope will be a representative subset of all the information that we have

At the highest level of abstraction, what we’re looking at here is a specific application of a general problem in the field of engineering: the problem of designing control systems, known as “cybernetics.” One of the foundational ideas of cybernetics is the “Law of Sufficient Variety,” coined by the psychiatrist and robot builder W. Ross Ashby. This states that: In order to ensure stability, a control system must be able, at a minimum, to represent all possible states of the system it controls.

If the system you are trying to control has more variety than your control system has ability to keep track of, then you have three choices: expand the variety of the controller, reduce the variety of the system being controlled, or give up on the attempt to control. Most management measurement techniques can be seen, cybernetically, as variety attenuators. They’re methods of taming the detail and uncertainty of the underlying system, to reduce it to something that can be represented in the head of a responsible manager—to, literally, “make it manageable.”

they talk about “managing what you measure.” The underlying variety has not gone away; it’s just been hidden under a set of simplified metrics. At best, they maintain the broad structure

branch of economics called “public choice theory” is devoted to the question of how people who are embedded in organizations tend to pervert the control systems to their own benefit. At one end of the scale, this shows up as slacking, taking excessive risks, empire building, and so on. At the other, it shows up as fraud. Adding more capacity to manage a system involves adding more people whom you have to trust.

There are only two ways of addressing a problem of insufficient variety in the control system. One of them involves reducing the information set, which makes you vulnerable to fraud from the things you are not monitoring. And the other one involves introducing new people to trust, which makes you vulnerable to fraud if they turn out not to be trustworthy

fraud is one of the classes of events that fall outside the direct information set, and that the control system needs to be able to respond to. The way in which businesses and control systems usually deal with these sorts of events is to treat them as random events (“risks”) and manage them on a statistical basis. “Risk management” is the measurement approach of assuming that your unpredictable events come from a probability distribution that can be guessed or estimated

there’s a clear definition of what a poor-quality project is going to be: it has inaccurate information, unrealistic or completely unanalyzed assumptions, and important details unspecified or left to chance.

Risk-management approaches have this tendency to drift off in the direction of optimizing for the total cost; the cost of checking minus the cost of fraud losses

If you’re trying to minimize costs, you will be drawn in the direction of industrialized and standardized approaches that make you very vulnerable to rifle-then-shotgun approaches. Fraud losses fundamentally aren’t random variables; they are someone else’s choice variables and their likelihood is quite likely to be determined by your own actions in trying to prevent them. In general, cost systems should be optimized for cost and decision systems need to be optimized for decisions. Failing to observe this principle gives you outcomes like Medicare and MERS.

The very big fraudsters designed their crimes specifically around the weaknesses they had identified in the control system itself. They cannot be usefully modeled as random events or defects; to do so is to lose the important structural facts about how they happen

Homomorphism is a useful mathematical term. It has a rigorous definition that is difficult to understand, but if you use it in contexts where you’re looking for something meaning “a simplified summary that loses a lot of potentially material detail but hopefully captures the important structural features,” you will be broadly correct and sound like a hell of a science nerd.

Chapter 09 : Chapter 9: Market Crimes

But in the USA, it is treated as a theft of the intellectual property of the company; information can only be illegal to trade on if it has come from a closely defined “insider” and has been acquired in exchange for some sort of payment

But it is clear on looking at the historical development of these laws that, for all that they are passed under the rhetoric of “investor protection,” they actually look much more like straightforward commercial decisions taken in order to improve a country’s market share of global investment

it’s a reflection of the gradual understanding on the part of capital-market operators that investable wealth was becoming a mass-market phenomenon, and that if you stopped robbing people blind with stock pools and takeover rumors, you would attract more of them into the capital markets, to make more money overall by robbing them through trading commissions and management fees.

This makes retail orders very valuable to the market. One of the reasons why stockbrokerage commissions are so cheap these days is that retail brokers have finally realized how valuable they are. They charge a quite substantial fee to players like the high-frequency traders for the privilege of dealing against their order flow, and they rebate some of this fee to their customers. But the retail orders would eventually dry up if the customers lost too much or felt that they weren’t being given a fair chance. And without a steady flow of “dumb money” lubricating the wheels, the professionals would find it a lot harder to trade, as they’d always suspect each other’s motives for buying or selling.

It’s not so much that the law creates the crime as that the defining characteristic of market crimes is that they’re crimes against people’s legitimate expectations, and that means that context matters a lot

The prohibition of cartels really is an example of a market crime as distinct from any other type of fraud; it is a rule passed in order to make the overall economy work better, which draws its legal force from a general interest of society rather than a specific right of the injured party.

One of the first things one learns to prove in price theory is that in a competitive market, the price of a single unit of output will tend to equal the marginal cost—the cost to an established producer of increasing output by one more unit

This is a rare example in which elementary economics works, by the way; companies in industries with high fixed costs and low marginal costs (like airlines and media) go bust a lot because they can’t resist competing prices down to levels that don’t cover their overheads

Even in industries where the problem isn’t so drastic, there is always a huge, unbearable tension between objectives of market share and profitability. It’s not really all that much of an exaggeration to say that managing the trade-off between these two objectives is at least half of the skill of strategic management

The illegal dumping of toxic waste hardly seems like a minor or technical offense—it’s one of the most serious corporate crimes of violence that there is, and given the orders of magnitude of people affected, it almost certainly significantly exceeds the worst excesses of the Mafia in terms of the number of deaths caused. But it is a kind of fraud (and one that often involves other frauds in counterfeiting safety certification), and as a kind of fraud, it is essentially a market crime. If anything, the inclusion of this category of corporate violence under this heading ought to disabuse the reader of any sense that market crimes are “technical” or “victimless”; they include some of the most callous and despicable actions ever to be carried out under the heading of crimes of dishonesty.

In fact, the population affected does not stop at the edge of the toxic cloud. The market crime inherent in a pollution case is an attack on the overall framework by which we trade off environmental costs against economic benefits. It’s a crime against the control system of the overall economy, the network of trust and agreement that makes an industrial economy livable.

In biological contexts, a part of the system that has escaped the normal self-regulatory mechanisms and begun to grow without constraint is called a “cancer.” Unless they are controlled, fraudulent business units tend to outcompete honest ones and drive them out of business. In doing so, they generate profits, and those profits can be redirected into financing the corruption of the whole system. Runaway corruption is something that does happen, and that can undermine entire societies.

As we mentioned when looking at accounting fraud, stock markets like clean narratives; it’s their way of managing the almost infinite variety of the information they have to deal with every day. This often means misfortune to companies that look like they ought to have a simple story (“Buy Piggly Wiggly, the chain of self-service stores that’s growing across the country!”) but actually don’t.

Shares are bought and sold on credit. When you make a deal on the stock exchange,6 the buyer has a few days to deliver the cash and the seller has a few days to deliver the shares. This arrangement is quite like trade credit, isn’t it? And trade credit is both a necessary administrative convenience and an opportunity for all kinds of misbehavior, as readers will have noticed

The answer, and the source of Clarence Saunders’s misfortunes, is that the brokerage community maintains a pool of shares available to be borrowed,7 and makes them available to scrounge up by people who have made short sales. Since, on any given day, buy and sell orders in any given share tend to more or less even out,8 the pool of shares available to borrow tends to be a reasonably constant quantity from day to day. Unless something unusual happens.

like professional traders launching a “bear raid” on a company that is fundamentally sound, perhaps because of headlines combining its name with the word bankrupt. Short sellers jumped onto the Piggly Wiggly share, borrowing stock from the pool. Their aim, of course, was to wait until the price was much lower, then buy back the shares and return them, meeting their obligations to deliver the shares

What does reduce the size of the pool is if someone—say Clarence Saunders—starts buying shares, and instructing his brokers not to make them available. When this happens, you can get a situation where the stack of IOUs obliging short sellers to deliver a share is greater than the pool of shares they can borrow to meet them. If you push things really hard, the pool can be totally exhausted and shares can no longer be scrounged up. When that happens, a short seller is in a very nasty place—he has obligations to deliver the shares, but no way of doing so. Rather than borrowing the share, he has to buy it, and the holder of all the shares that used to be in the pool can literally name his price. This is called a “corner,” the fairly obvious metaphor being for the one that the short seller is trapped in

This is the quintessence of a market crime; all Clarence did was buy shares at the going price, and everyone who dealt with him did so willingly and transparently. He did not deceive anyone; he literally took out advertisements in the newspapers saying what he was doing. But the market wanted to protect itself, and his conduct was disruptive to a set of economic institutions that other people rely on

Although Bob Diamond was at least partly separated from his job for insinuating that the Bank of England had told him to underreport his funding cost, it’s certainly true to say not only that the global regulators were pleased to see lower rates, but that they were correct to do so. It would be exaggerating a bit to call LIBOR “the fraud that saved the world,” but it certainly saved the central banks a lot of trouble

So why the rage? The undermining of public trust was not just the banking sector’s punishment; it was the crime. Comparatively few people were actually involved in transactions in the LIBOR market. Substantially more had some exposure via the hundreds of trillions of dollars of derivative products linked to the LIBOR survey. But nearly everyone has an interest in maintaining the convention that financial markets operate with a degree of fairness; they may be highly unequal, they may be full of aggressive players acting in their own interests, but it was crucial to the self-interest of the society that we had made to believe that markets were not structurally rigged.

Chapter 10 : Chapter 10: Cold Cases

A summary of the book so far might be that fraud is what happens when you can’t check up on everything. And the economics of fraud are all about the best ways to organize the process of checking up on things, given the state of the world you find yourself in. It shouldn’t be surprising, then, that both fraud itself and the techniques with which we combat it have grown up and developed alongside the development of the capitalist economy

Fraud is possible to the extent that people are prepared to trust strangers, or to leave valuable objects out of their immediate control; the ancients had much less occasion to do this than we did. There was less to check up on

The main frauds in the sagas involve things like inheritances, because the right to an inheritance was one of the first abstract stores of value to emerge as legal systems became codified

Being a phony princess was about the only way a woman could pretend to be rich for much of recorded history, and the social structure of European nobility meant that you could get quite a lot of formal and informal credit based on the assumption that your royal family would pay it back. If nothing else, you might be able to secure a good marriage

Unlike a land caravan, a ship’s cargo is very difficult to monitor in between leaving its point of origin and arriving at its destination—even if the merchant were to personally accompany the cargo or send an agent to do so, he is still wholly at the mercy of the captain. Even in the absence of dishonest dealing, all sorts of bad things can happen to a cargo at sea. It is for this reason that the maritime industry was significantly ahead of the rest of the economy in history, both in the development of mechanisms to handle risk4 and in the parallel development of fraud. Shipowners formed the first syndicates or “companies” to raise capital and pool their risks. They invented insurance and were the first people to seriously think about forms of financing.

A bottomry was a loan secured against a ship, which had the particular provision that if the ship was sunk, the loan did not need to be repaid. This made it a very unusual kind of credit; for most of recorded history there was no such thing as bankruptcy or limited liability and a loan once agreed upon had to be repaid whatever the cost

The largest and most advanced human society to have existed without material fraud may have been one of the pre-Columbian American societies such as the Incas, who appear to have organized their economy entirely on the basis of directed labor and shared consumption

It took the Crash of 1929 sixty years later to make people change their minds. Back in 1868, investors had to put up with the Erie Railroad Company, a corporation whose affairs were so bad, even by the standards of the day, that it was nicknamed “The Scarlet Woman of Wall Street.

Rollo’s book ends with a set of recommendations for licensing and regulation of the inquiry-agent system, to govern conflicts of interest and to ensure that shoddy or dishonest agents cannot drive good ones out of the market by underselling them. He identifies the key problem of the industry—that the business of providing credit ratings is marginally profitable, and that therefore there is always the temptation to cut corners. He also notes that the inquiry agents of London were subject to severe conflicts of interest and would tend to give favorable ratings in order to gain business. It is fascinating and disturbing to notice that not only were none of his recommendations adopted, but that with only small updates to the language, all of them could have been published in the last ten years as a critique of the credit-rating agencies that gave AAA recommendations to worthless mortgage-backed securities. Once more, the same economic conditions generated a similar balance of market power, with similar consequences in terms of the kinds of frauds that were made possible. To a large extent, what society gets in terms of large-scale commercial fraud is determined by what it is prepared to legislate for.

Jonathan Levy notes in his book Freaks of Fortune that “risk,” as an everyday word to describe the general consequences of uncertain chance, is a relatively modern adoption of a specialist term drawn from the shipping and insurance industries, simply as the name of the commodity that they bought and sold at Lloyd’s Coffee House and similar insurance markets.

Chapter 11 : Chapter 11: Fraud Against the Government

The government has one attractive property as a target for someone aiming to commit a fraud—it is prepared to take on nearly anyone as a customer

it is the economic actor that cannot, by definition, be part of a small personal trust network, but that has to, also by definition, get involved in the economic life of nearly everyone.

Public-sector bodies are usually very big, and they have senior-management functions delegated to nonspecialists who were usually elected to office for reasons unrelated to their competence in financial control. This tends to matter more in big defense contracts than in paper-clip procurement

The distinction between tax avoidance and tax evasion is a commonplace—“avoidance” is usually taken to be the (usually legal) use of technicalities and corporate structures to minimize a tax liability, while “evasion” is the (usually illegal) act of providing false information to the tax authorities in order to conceal the existence of a liability in the first place. Of course, the two categories overlap. If you are a really aggressive avoider, you may believe you have no tax liability to declare, but the taxman may disagree and consider you to be an evader.

Some countries (most importantly Switzerland) have recognized a third category of offense—that of “tax fraud.” The distinction is between a mere neglect to inform the tax authorities of a source of income, and the actual presentation of a fraudulent document

There was no particular reason why the Swiss offshore tax-evasion franchise collapsed—it wasn’t a cumulating, snowballing fraud like most of them. It was just not useful anymore. The Cold War was over, and so the governments of the developed world had less interest in maintaining plausibly deniable bolt holes with which one could fund shady things. Top rates of tax had been falling for decades, so there was less urgency felt by the global elite. And so tax evasion is a bit of a boring business these days; there are hardly any Birkenfelds left, to the chagrin of Ferrari dealers and the joy of people who like peace and quiet in restaurants

Modern tax evasion, then, occupies a borderline between a market crime and an old-fashioned long firm against the government, with the credit extended by the fact that tax is paid in arrears

So the government’s ability to use the banking system and its records as a tool of law enforcement is an important part of its overall power to enforce its laws. It is also, to return to a theme, a significant check on the growth of criminal enterprises; a drug ring cannot grow faster than its cash-handling capability. Undermining this system and trying to make dirty money appear clean is a fraud against the government, and the name of the crime is “money laundering

The coronavirus epidemic, by the way, was an absolute disaster for the money laundering industry—it became immediately impossible for a neighborhood pizza restaurant to report $50,000 of cash takings every week without anyone noticing.

In the context of money laundering, the objective is to get cash transformed into a deposit held with a reputable bank. Once that objective has been achieved, the money is effectively “clean” and can be spent or invested; a criminal may have other problems in explaining to the authorities how he is able to support a lifestyle well out of proportion to his honest income, but the payments system will not generate red flags.

diseases. If the purpose of a trust system is to avoid the need for checking, then you need to take into account that when you allow a new entity into the circle of trust, you are also, implicitly, making the decision to trust everyone that they trust. The way this works for money launderers is that good banks don’t accept large deposits of dirty cash. They also tend to check up on transfers from banks that are known to be “high risk” (in context, a euphemism for “probably crooked”). But there is a tier of mediocre banks, who are not so bad as to be the subject of aggressive checking by the good banks, but neither are they so good as to be unwilling to deal with the bad banks. Plotting a path through a chain of financial institutions from one that is prepared to take a suitcaseful of cash to one that will allow you to conveniently use the money is a big part of the skill of money laundering

The other crucial component of a money-laundering strategy is a list of the ways in which something valuable can be owned, bought, and sold without the identity of its owners being a matter of public record

And more than this, a jurisdiction where verification is difficult or time-consuming is one that changes the trade-off between checking and trust, the trade-off that is at the center of the crime of fraud.

Money laundering is basically a market crime. It has been decided that, because the global financial system is so ubiquitous, it makes sense to use it as a tool of law enforcement. That decision having been made, the creation of a new set of crimes was an inevitable consequence. Airlines are not prosecuted for helping criminals flee justice4 and the sellers of guns and balaclavas are not held liable for robberies, but financial institutions have, in return for their many privileges in the wider economic system, been delegated certain investigative tasks

Of course, one of the bodies that commits the most fraud against the government is the government. Public-sector bodies tend to have people in senior positions with major financial responsibility who have become used to a lifestyle totally out of proportion to what they might be able to enjoy if they were not an elected representative, and who have ambitions for a lot more. And the conversion of political power into money is not difficult; if you have the ability to put people in jail or destroy their businesses and you’re not getting rich off it, I’d say you’re not trying

It’s not necessarily a correct analysis to look only at the amount of corruption and assume that all corruption is bad; you need to take a holistic view of the relationship between corruption and the rest of society, which incorporates the overall level of trust that the corruption is exploiting. Fraud and dishonesty in a high-trust society with strong institutions can be evidence that something good is happening, so quickly that people don’t find it worth their while to waste time minimizing their fraud losses. This attractively contrarian idea that fraud can be a positive sign should not be pushed too far, though. Ideas

idea was that corruption was a rough-justice form of deregulation and that in the absence of effective state institutions, things might go better if people simply stepped around what institutions there were. It didn’t work at all. What actually happened was first, that the corrupt economy swallowed the honest economy. And second, the level of stealing got big enough to have macroeconomic effects of its own

Chapter 12 :

The socialization of the professional and business classes is not totally dissimilar to that of Polynesian islanders, and fraud is the ultimate taboo. Anything that touches it is unclean. Obviously, there are solid reasons why this social and psychological barrier exists; it’s an evolved phenomenon, reinforced by training and ritual, which is meant to encourage aggressive and sharp practice up to the limit of the law, while inculcating a terror and revulsion toward anything that is marked as a step too far

the trouble about taboo places is that nobody goes there. Ironically, the strong social barriers that are put in place to stop people from committing fraud are also strong barriers that stop people from thinking about fraud at all. One of the strongest defenses in a criminal’s arsenal is the ability to look someone in the eye and indignantly ask, “Are you calling me a liar?”

In passing, at the end of Chapter 8, we started thinking about the extent to which frauds could be thought of as random events that happen when a bad person bumps into a structural control weakness

Donald Cressey’s model of the “fraud triangle,” set out in his 1953 classic, Other People’s Money, has never been surpassed

Need. Bankers commit acts of dishonesty for the same reason that heroin addicts do; they have been put in a position where they need to come up with larger sums of money than they can generate by honest means

Opportunity. An opportunity to commit fraud is a weakness in a control and checking system—either one that exists because of the way in which the intrinsic variety of the system has been reduced to make it manageable, or one that the fraudster himself has created by exploiting a position of control.

Rationalization. White-collar crime is committed by people who have been trusted with something, and there are psychological barriers to breaking a trust

The fraud triangle does not distinguish between frauds launched against targets of opportunity, and frauds that are designed against specific targets by people who know what they’re doing. And that’s a really important distinction.

The big fraudsters create their frauds, design and sustain them. They build them around the cognitive environment of their victims and take pains to manipulate the environment so that they can remain undetected and grow

We need to make a distinction between “incidental” fraud, the kind that can be controlled by a risk-management system, and “entrepreneurial fraud,” which can’t be controlled by any kind of system, because the control and information system is precisely what it is made of.

The amount of money you lose in the event of your becoming a victim of a crime is dependent on two things: how long it takes you to discover the fraud, and the rate at which the fraudster can extract value from you while the crime is going on

In fact, catching a big, “entrepreneurial” fraudster is usually done by accident. The most common form of accident is that the fraud grows too fast for its perpetrator’s ability, has a cash liability falling due that it cannot pay, and collapses. In bankruptcy, people check up on things that they had been happy to take on trust when they were dealing with a solvent counterparty, and once it has been discovered that there is not enough money to pay all the creditors, the story comes out pretty quickly.

As a fraud grows, it usually needs to expand the number of people exposed to it, because its significance increases in its local economy. After a while (or even quite early on, if the fraudster was unlucky or careless), it grows beyond the domain of its initial planning, and the fraudster can no longer be confident in his initial work designing around the checking and control processes to which the fraud would be subject. The fraud reaches the front door of someone who is in the habit of doing things a little bit differently. That person carries out their standard checks—they ask for a site visit on a weekend, rather than a weekday, for example, or call up the insurer of a ship rather than the port to find out when it set sail. And suddenly everything is apparent. The essence of fraud is stagecraft and the creation of facades that look utterly convincing, but only when viewed from the right angle.

In the first place, it implies that diversity in methods makes a system less vulnerable to deceit

There actually were some state-level medical insurers during the 1980s that had claims expenses less than a twentieth of Medicare’s, and they achieved it by largely ignoring cost efficiency and concentrating on processing their claims correctly, rather than efficiently. And they tended to be staffed by loss adjusters and appraisers who were given a large degree of independence and allowed to do things their own way—if you managed to get a fraudulent claim past Bob, that wouldn’t necessarily mean that your next similar claim would get through when it was processed by Sally.

Secondly, we can combine two insights. The defining characteristic of entrepreneurial fraud is growth, and an entrepreneurial fraud is one that has been designed for a particular control environment. So we know that candidate frauds have to be things that demonstrate the growth characteristic, and we know that fraudsters are vulnerable to a change of perspective. Or to put it in simple terms: Anything that is growing unusually fast, for the type of thing that it is, needs to be checked out. And it needs to be checked out in a way that it hasn’t been checked out before. If there is a Golden Rule for fraud detection, that’s my best guess at it.

  • The optimal level of fraud is not zero, optimal progress and with appropriate risk checking will dictate the level of fraud

Links to this note