Diffusion of Fraud White-Collar Crime · Crime Lab 6 · Midwestern State University
Between January 1986 and December 1988, 230 people put about $11.5 million into the oil and gas wells of a small California company that started as a legitimate business and ended as a fraud. Baker and Faulkner asked a question criminologists had mostly skipped: not how the fraud technique spread among swindlers, but how the fraud spread among its victims. Half the investors came in through someone they knew; half answered a cold call or a mailing. Almost none of them told anyone else, and when they grew suspicious, two-thirds never said so to anyone. This lab puts you inside that process. You will see the case, run a model of the diffusion with the levers Fountain's principals and investors actually had, sort the antecedents of CEO wrongdoing that Schnatterly, Gangloff, and Tuschke reviewed, and test Vaughan's claim that mistake, misconduct, and disaster come out of the same organizational machinery. The organizing point is Baker and Faulkner's own: the factors that made the business succeed while it was legitimate are the factors that made the fraud succeed once it was not.
An intermediate fraud
Clinard's term for a fraud committed in or by a business that began as legitimate is intermediate fraud, as against a pre-planned fraud, where the business exists only to steal. Fountain Oil and Gas was run by three brothers from Westlake Village, in Ventura County. The eldest had sold oil and gas leases for years. The company drilled in the Sacramento Valley, organized each well as a separate limited partnership, and sold units to individual investors with a full prospectus that spelled out the risks and stated that the units could not be resold. It struck gas in November 1986. The first alleged misappropriation of investor money came at the end of that year. Money raised for one well was diverted to other wells, to company expenses, and to the president's personal purchases. Contractors went unpaid and filed 39 liens; a federal court put the company in receivership in March 1989; the Ventura County District Attorney's forensic accountant found enough to indict. In 1992 the president was convicted of grand theft by embezzlement, unlicensed sale of securities, excessive taking of funds, and filing false tax returns, and sentenced to five years and eight months; his brothers were convicted of using false statements in the sale of a security and received 16 months and 365 days in county jail.
Who invested. Mostly men (77 percent), white (94 percent), average age 53, median household income $83,750 in 1988, professionals and business owners: the profile of an early adopter in any diffusion study. Three of four had never invested in natural resources before. Investors who came in late did not differ from those who came in early in age, occupation, or experience, which Baker and Faulkner take to mean there were no late adopters at all; the diffusion was cut off before it reached them (pp. 1190, 1198).
Source: Baker, W. E., and Faulkner, R. R. (2003). Diffusion of fraud: Intermediate economic crime and investor dynamics. Criminology, 41(4), 1173-1206 (case history pp. 1183-1188; investor characteristics p. 1190; periods pp. 1190-1192). The article's file name in the course folder says 2006; the citation is 2003.
How they heard about it
Diffusion theory sorts channels into social networks (a tie to the seller, or a tie to someone who already bought) and impersonal methods (direct mail, advertising, telemarketing). In the 1991 national Fraud Victimization Survey, about 30 percent of 1,911 fraud incidents involved social networks and 70 percent impersonal methods; for fraudulent business ventures the split was 24 to 77 (Table 1, p. 1179). Baker and Faulkner surveyed 72 of Fountain's investors, a 30 percent sample that did not differ from the rest on amount invested or timing.
What share of Fountain's investors came in through a social tie, to the company or to a prior investor?
Half and half. Social ties brought in 50 percent (21 to the company, 19 direct to a prior investor, 10 compound), impersonal methods the other 50, and the single largest channel was the cold call, made from purchased lists of affluent people, investors in other speculative ventures, and buyers of big-ticket items. One investor: "The telemarketer got my name from a list of yacht and airplane owners" (Figure 5 and pp. 1192-1193). The company also dressed the part: a prestigious office with maps and core samples, a president with a Lamborghini, a Rolls Royce, and 20 suits at $2,500 each, and a petroleum geologist one investor called "a guru-type guy" (p. 1199).
Fountain asked 61 percent of its investors to refer the company to people they knew, and many said they were pressured to. What share of investors referred anyone at all?
24 percent, one referral each, and only after being asked. One respondent in 72 put someone in touch with Fountain without a request from the company. The reasons for not referring (Table 3, p. 1195): too risky, might not work out for others (31 percent of mentions); a private matter, "I keep these things to myself" (30 percent); "I didn't trust Fountain" (23 percent); never considered it (14 percent). A physician who lost more than $200,000 explained the mix: he would not involve his colleagues, he found the principals "a little slimy," and "my greed and their smooth superficial successful exterior overcame my suspicions" (p. 1194).
Why the network was there and not used. Baker and Faulkner give two reasons. The units could not be resold, so unlike a stock, which investors talk up to raise the price, a Fountain well was a congestible good: telling others could only crowd you out of the next one (pp. 1197-1198). And financial transactions happen inside a culture of privacy and self-reliance, so investors who were geographically concentrated and demographically alike, and therefore very likely to know one another, did not know one another as investors. One discovered a friend and colleague had invested only at the receiver's meeting: "You got screwed, too?" (p. 1200).
Who lost, and who said anything
The two-stage pattern in the losses. Only one exploratory well in seven pays, so some investors would have lost money in an honest company. But the loss rate climbs from 48 to 61 to 95 percent by year of entry, tracking the period in which the diversions took place, and the 1988 wells were not worse: four produced gas, including the company's biggest producer (Table 4 and note 3, pp. 1195-1196). The people recruited last, when the company most needed new money, were the people with almost no chance of getting any back.
Closed awareness. Those who did talk mostly talked to family, not to other investors. The reasons for not going to the authorities: did not know it was fraud (38 percent), "felt powerless" (25 percent), feared they would "look bad" (15 percent). Baker and Faulkner's conclusion is the one the model on the next tab is built around: Fountain had encouraged investors to talk to each other while it was legitimate, and once it turned, the fact that they did not is what kept the fraud alive. "I didn't talk to anyone about my suspicions," one investor said, "because there's no one to talk to" (pp. 1196-1197, 1201).
Source: Baker and Faulkner (2003), Figure 5 (p. 1194), Table 3 (p. 1195), Table 4 (p. 1195), and pp. 1192-1201 as cited. Percentages are of the 72 investors surveyed.
A model of the diffusion
The model below is a simplification built to approximate the Fountain figures, not the data themselves. There are 400 potential investors in 16 overlapping social circles, alike in the way Fountain's investors were alike. Each month for 36 months the company recruits through three channels: the principals' own ties (fixed), cold calls to purchased lists (you set how many calls a month), and referrals, where an investor who is asked refers one person they know with the probability you set. Recruitment is slower in the startup months and faster after the gas strike in month 11. The diversions begin at the end of year one, and an investor's chance of losing everything follows the year they entered (48, 61, or 95 percent). From month 20, investors begin to grow suspicious. A suspicious investor tells another investor they happen to know with the probability you set, and only if they know that person is an investor. When enough investors are talking to each other, a complaint is filed and the company goes into receivership; otherwise the liens close it in month 39. Run the three named settings, then at least one of your own. Every run is logged for your submission.
Cold calls per month: 30
Chance an investor who is asked refers someone: 24 percent
Chance a suspicious investor tells another investor: 33 percent
The network at the end
Each cluster of dots is one social circle; the faint lines are the ties people have to one another, including a few across circles. The three circles at the top left are the ones the principals themselves belong to. Pick one thing to see at a time. Hover over any dot for that person's history.
New investors by month
Channels and losses
Run log
| Run | Setting | Calls | Refer | Talk | Investors | Exposed | Lost all | Year 3 entrants | Year 3 lost |
|---|
Model parameters were set so that the observed setting lands near Baker and Faulkner's figures: about half of investors through ties, a referral rate of 24 percent conditional on being asked, loss rates by year of entry from Table 4, exposure late in year three. Random draws differ on every run, so two runs of the same setting will not match; run a setting more than once before drawing a conclusion from it.
Pressure, opportunity, rationalization, applied to the top of the firm
Schnatterly, Gangloff, and Tuschke reviewed the management research on CEO wrongdoing published since 2005 and sorted its findings with Cressey's triangle from Lab 3. Their definitions: pressure "reflects the necessity to commit wrongdoing ('have to')," opportunity "suggests the ability to commit wrongdoing with the belief that the act will not be detected ('can do')," and rationalization "is the ability to explain an act of wrongdoing as morally justifiable ('it's okay')" (p. 2409). Each side has internal sources, inside the firm, and external ones. Ten findings from the review are below. For each, pick the side of the triangle the authors file it under. The reference gives the page.
0 of 10 sorted.
What the sort shows. The pressure findings are almost all about money that flows to the CEO or expectations that flow at the CEO: options, pay gaps, analysts, activist owners. The opportunity findings are about who can see what: ownership stakes, complex firms, boards that are friendly or inexpert. The rationalization findings are the thinnest, which the authors say outright; management research measures pressure and opportunity from the outside and has only begun on the mind of the person at the top. Put this next to Lab 5: the accounts you coded there are the visible residue of the side of the triangle the management literature can least observe.
Back to diffusion. One of the review's external rationalization findings is Baker and Faulkner's Type 1 diffusion in action: options backdating spread among firms between 1996 and 2005 through board interlocks, directors who sat on more than one board and carried the practice with them (p. 2418). The fraud among Fountain's victims spread through purchased lists and a few family ties; the technique among perpetrators spread through the network of the people who govern corporations.
Source: Schnatterly, K., Gangloff, K. A., and Tuschke, A. (2018). CEO wrongdoing: A review of pressure, opportunity, and rationalization. Journal of Management, 44(6), 2405-2432, at the pages cited on each item.
Three outcomes, one machinery
Vaughan's review takes the sociology of organizations and asks how things go wrong in them as a matter of routine. Her abstract states the claim: "routine nonconformity, mistake, misconduct, and disaster are systematically produced by the interconnection between environment, organizations, cognition, and choice" (p. 271). The three outcomes are distinguished by whether a rule or law was violated on behalf of organizational goals (misconduct), whether the adverse outcome was unexpected and its harm contained (mistake), or whether it was large in scale, public, and unexpected (disaster). The distinctions matter to lawyers and regulators, who respond to each differently. Vaughan's point is that they are produced by the same structures, so that an organization built to avoid one is not thereby protected from the others, and one outcome often arrives with another. Sort the five events below by the category that fits the outcome itself, then read the reference on what else was present and what the legal system chose to name.
0 of 5 sorted.
Fountain at three levels. Environment: an informal capital market with no institutional investors, an oil price of $1.70 per thousand cubic feet, a one-in-seven hit rate, and an industry in which, as one experienced investor put it, diverting funds between wells is common and "fine if the DAs do not become aware of it" (p. 1188). Organization: a family firm of 15 to 20 people in which the same three brothers sold the units, controlled the accounts, and decided which well got the money, with each well walled off as a separate partnership so that no investor could see across them. Cognition and choice: investors who read a prospectus that told them they could lose everything, saw a Lamborghini, and treated the outcome as a failed business rather than a crime. None of these is unusual. That is the argument.
Source: Vaughan, D. (1999). The dark side of organizations: Mistake, misconduct, and disaster. Annual Review of Sociology, 25, 271-305 (abstract p. 271; assigned pp. 271-280 and 287-292). The event outcomes are from the court records and agency reports cited under each event.
Lab 6 response sheet
Answer the four questions below in complete sentences. Then use the button at the bottom to assemble your answers, your run log, and your sorting record into one block of text, and paste that text into the Lab 6 submission in D2L before you leave class. Your answers stay on this page and are not sent anywhere until you paste them.
About the model and the sources
The Fountain figures are from Baker and Faulkner (2003) at the pages cited on the first tab. The diffusion model is the lab's own construction and is described on its tab; its parameters were chosen to reproduce the observed channel mix, referral rate, loss rates by year, and timing, and no result from it should be cited as a finding of the article. The CEO wrongdoing items are from Schnatterly, Gangloff, and Tuschke (2018) at the pages cited. The event outcomes on the fourth tab are from the Rogers Commission report (1986), the Boeing deferred prosecution agreement (N.D. Tex., January 7, 2021), the BP guilty plea (E.D. La., November 15, 2012), the Wells Fargo resolution (February 21, 2020), and Baker and Faulkner for Fountain.