City of Flint Dot Net

Enter your search terms Submit search form   Web www.cityofflint.net www.tedjankowski.com www.thejankowskifamily.com

Home

Home

Where Have All the Hot Goods Gone?

The Role of Pawnshops
Simon M. Fass and Janice Francis
July 15, 2003
Abstract

Recent research argues that because markets for stolen goods act as incentives to steal, police
and criminologists should shift attention from thieves to methods of disrupting demand for the goods.
The underlying research, however, is too thin to support this advice. Effective policy requires
considerably more investigation. Analysis of pawn transaction data from Texas supports this
assessment. It suggests that proposals to disrupt demand are unlikely to succeed, partly because
similar actions already applied to pawnshops have shown limited effect, mainly because hot goods
are invisible in the daily flow of secondhand merchandise through the general retail market. Police
and criminologists should remain focused on thieves and their apprehension, and on pursuing ways to
do this more efficiently, such as through improved tracking of pawn transactions. There may be other
intervention possibilities as well, but much more empirical research is required to identify them.

Key word list: property crime receiving stolen goods pawnbroking

Where Have All the Hot Goods Gone? The Role of Pawnshops.

Recent improvements in public understanding of markets for stolen goods has led some
researchers, Clarke (1999), Kock et. al. (1996) and Sutton (1995), for example, to conclude that these
markets, because they facilitate disposal, act as incentives to steal. From this they argue that police
and criminologists focus too much on thieves, not enough on reducing demand for hot goods. They
then urge more research on the markets and on methods of disruption, such as encouraging stores that
do not want illicit merchandise to install closed circuit televisions, photograph sellers and post signs
announcing participation in crime prevention programs (Sutton, 1998).

More research is certainly warranted. But a shift in focus from thieves to disrupting demand
seems premature. As Freiberg (1997) highlights, all studies to date merely hint at the size and other
features of the market. They offer no basis to justify diverting attention or for expecting that the
markets can be upset in meaningful ways. Indeed, Freiberg stresses, “... public knowledge of [the]
market [for stolen goods] and its dynamics ... is so impoverished as to border on the scandalous.
Good policy cannot be developed on the foundations of ignorance” (1997: 25).

A glaring sign of this poverty is scarcity of research on pawnshops, i.e., pawnbrokers who
make loans as main businesses or as sidelines to other enterprises. Rarely arrested, these brokers have
long been suspected of illicit trade. Scholars recurrently point fingers at them. Interviews with
burglars show their importance for quick disposal. And everywhere there are pawnshop laws that,
among other objectives, aim to reduce traffic in stolen goods.

What makes the dearth of research especially dismaying is that although denunciation of
pawnshops and efforts to halt illegal trade have recurred for a long time, the few empirical studies
done on them, mainly by economists (e.g., Caskey, 1994; Patterson, 1899), legal analysts (e.g.,
Levine, 1913; Nickles and Adams, 1994; Oeltjen, 1996), and others (e.g., Wheat, 1998), ignore or
downplay their involvement. While these scholars act as deliberate or inadvertent pawnbroker
defenders, adversaries, such as Glover and Larubbia (1996) and other investigative journalists,
continue to underscore close connections between brokers and stolen merchandise.

Missing from this arena is dispassionate inquiry that might help estimate pawnbrokers’ share
of the market, and also answer the question of whether new attempts to reduce that share could prove
more effective than past attempts. This question is important because if the answer to it is no, then
prospects for disrupting other, less visible and less regulated markets for stolen merchandise are
remote. There would then be little reason for attention to shift from thieves to markets.

Following a brief look at pawnbroking in the United States, we move toward answering the
question with a review of what has been alleged about the link between brokers and stolen goods.
With data for Dallas, Texas, we then show that the pawnshop customer base holds a small group of
prolific pawners. Containing many people with arrests for thievery, this group is responsible for a
disproportionate number of transactions that, we suspect, involve hot items. These data also show
that even if the items are but a fraction of all things passing through shops, that fraction represents a
significant portion of broker revenue and a non-trivial share of the estimated $40 to $45 billion of
goods that we estimate are stolen every year (see Appendix).

Out of this, our answer to the question is that efforts to disrupt markets for stolen goods are
unlikely to succeed. One reason, for pawnshops, is that proposed disruption methods are similar to
those that have been applied with little effect for a long time. Writ larger, another reason is that stolen
items are invisible in the flow of used merchandise between sellers and buyers. Markets for hot goods
are inseparable from the market for all secondhand wares. Reducing demand for stolen goods,
therefore, implies disrupting the whole retail market for used merchandise. Even if one accepts the premise that this bigger market acts as an incentive to steal, disturbing it serves no fruitful purpose.
Police and criminologists are thus better advised to stick with identification and apprehension of
thieves, and with improving methods to accomplish this, such as finding more efficient means to
track pawn transactions. This does not exhaust the list of possible interventions, but much more
empirical research is required in order to identify the productive possibilities.



Pawnbroking in the United States

Pawnbroking is the business of lending cash for a fixed term against the security of a deposit,
or pledge, of personal property. The loan is typically 30 to 75 percent of the market value of the
pledge. If borrowers repay the loan on time together with interest and other fees, the broker returns
the goods. If obligations are not paid, then borrowers forfeit their property. The broker earns revenue
from interest and fees or from sale of the forfeited items.

The business has almost always been associated with exorbitant interest rates and with
facilitation of traffic in stolen goods. As a result, pawnshops are everywhere subject to specific state
and local regulation. Some laws set maximum limits, or ceilings, on nominal (i.e., official) interest
rates and on storage and other administrative fees that brokers use to push effective rates that
customers pay to levels higher than nominal rates. In addition to licensing, bond, land use zoning and
like requirements, other regulations focus mainly on identifying stolen goods and limiting their flow
through shops. Among other things, these require that brokers submit records of all transactions to
police in a timely manner, that the records describe every item pawned, and that borrowers supply
personal identification, at times including fingerprints or photographs.

Laws were highly restrictive during the first half of the 20th century, dampening growth of
pawnbroking, or at least that of visible, licensed shops. Levine (1913) counted 1976 of them in 1911,
after which the number dropped to 1509 in 1929 and 1374 in 1948 (U.S. Bureau of the Census,
1953). As they still do, laws also varied greatly across states. With profitability harder in some places
than in others, this variation contributed to big differences in spatial distribution, for example, from
2.6 shops per 100,000 urban population in Kentucky to 17.9 in Florida (Levine, 1913).

Numbers exploded during the second half of the century, due in part to regulatory reform,
especially liberalization of interest ceilings (Johnson and Johnson, 1998), and in part to rapid growth
in demand for used merchandise.1 Based on a count of telephone directory listings, Caskey (1994;
1995) estimates that the number of shops jumped from 4,850 in 1985 to near 10,000 in 1994. We
count 12,300 unduplicated listings in 2002. With allowance for those that do not list themselves, this
suggests the presence of perhaps 15,000 brokers. 2 As we show in Table 1, pawnshops have become
as convenient to their customers, be they borrowers or criminals, as the 12,500 McDonald’s fast food
restaurants are to their hungry patrons. 3

Another change, in recent years especially, is the structure of the industry. Independent
outlets have been absorbed or displaced by regional and national firms, such as Cash America, which
in 1987 became the first publicly-traded pawnshop corporation in the country (Scott, 1992), and
EZCorp, Express Cash, and First Cash, which traded on Wall Street a few years later (Caskey, 1995).

These corporate players have moved aggressively to acquire merchandising respect, to
replace the pawnbroker’s disreputable image with that of virtuous entrepreneurship (Drysdale and
Keest, 2000). Some efforts focus on transforming seedy shops into smart-looking retail stores (Berg,
1991; Breyer,1995; Ruisseaux, 1995). Others involve strategic use of advertising in local media, and
encouraging these media to report positive trends in the industry, notably the move of brokers into
suburbia and corresponding widening of the customer base to include more middle- and higherincome
individuals (e.g., Auster, 1997; Calkins, 1987).

What Scholars and Journalists Say About Pawnshops and Stolen Goods

These exertions may have succeeded in removing some of the stigma attached to use of
pawnbrokers, especially in higher-income strata. But they have yet to undermine the widely-held
conviction that shops serve what Glover and Larubbia (1996) call the modern thief's automatic cash
machine. The perceived link between theft and pawnbroking, as mentioned earlier, is indirectly
supported by some scholarly research. Cromwell’s (1991) interviews of thirty apprehended burglars
in Texas showed that 18 percent used pawnshops as a primary method of disposal while others used
them irregularly. Our analysis of transcripts from interviews by Richard and Decker (1993) of one
hundred burglars in Missouri suggests that even though required to have their picture taken, 42
percent used pawnshops for goods disposal, half of them regularly.

More commonly, the perception is sustained by recurrent newspaper articles concerning
pawnshop crackdowns, arrest of operators, proposals to computerize pawn records, and the like (e.g.,
Gryta, 1998; Krause, 1998; McGeevy, 1997). Grounded or not, belief that shops attract criminals
continues to fuel opposition to their presence in residential areas (e.g., Ingrassia,1995; Lundy, 1994).

Pawnbrokers insist that this is mis-perception, countering that shops are in fact good
neighbors serving the needs of cash-strapped populations. They illustrate this with genial stories in
the media of the benefits obtained by individual borrowers in financial distress (e.g., Marino, 1997).
They claim, further, that most if not all pawnbrokers are honest, hard-working individuals who try to
reduce risk of receiving stolen property by working closely with police (Chuang,1998; Stroer,1997).
As a result, they say, pawnshops accept few stolen items.

Some data seem to bolster this assertion. A review of 150,000 pawn transactions by Cash
America found only 34 stolen items, or 0.02 percent of the total (Kleinfield, 1989). A 1991 check of
65,000 transactions in Dallas County, Texas, found 250 stolen items, or 0.4 percent (Scott, 1992). In
Oklahoma, 873 of more than 1.5 million items pawned in 1995, or less than 0.1 percent, were
identified as stolen (Wheat, 1998). Police in Los Angeles, California, recover about $700,000 a year
in stolen property, also a tiny share of all transactions (McGeevy, 1997). Similar figures, respectively
.01 and .07 percent, are reported in Florida for Collier and Palm Beach counties (Florida Committee
on Criminal Justice, 2000)

But these data are not convincing. They stem from efforts to match goods in pawn against
lists of stolen items with serial numbers, engraved ownership markings, or other unique identifying
features that victims report to police. Most stolen goods do not have such unique features. Most
thefts, as we note in the Appendix, are not reported.

Pawnbrokers also find support from scholars studying the business. Caskey (1994), after
interviewing six brokers, argues that they do not traffic in stolen goods because, risking arrest or
suspension of their licenses, it would be foolish for them to do so. Nickles and Adams (1994), to
buttress their argument that possession of property by pawnbrokers should imply legal right to the
property, claim that virtually all pawned goods involve true owners. Oeltjen (1996), reacting to a
broker’s claim in court that 5 percent of transactions involve stolen goods, tries to discredit the
assertion in a footnote by saying, disingenuously, that it is not substantiated by fact.

Evidence to challenge the idea that hot goods are an insignificant share of all pawned items is
often just as anecdotal as that which supports it. Much of this evidence flows from ad hoc law
enforcement actions, such as a sting operation by police in Manatee County, Florida, that was filmed
and then broadcast on national television during the NBC network’s “Dateline NBC” program on
May 11, 1999 (Florida Committee on Criminal Justice, 2000). Here, undercover officers created a
pawnshop to investigate the extent to which thieves would use this type of facility. Hidden cameras
then recorded how several frequent pawners came in to fence wares that were obviously hot, a few
even revealing how they burgled homes and businesses. The state requirement that pawners leave a
thumb print at every transaction did not deter these individuals.

An earlier undercover operation in Fort Lauderdale, Florida, filmed and then aired on
December 21, 1997, during the CBS network’s “60 Minutes” program, involved detectives posing as
homeless people trying to pawn computer equipment with stickers indicating that they belonged to a
prominent local business (“Quick Cash”, 1997). Notwithstanding overt signs that the goods might be
hot, staff at three of five shops accepted them, one even asking for computer monitors the next time.

Luring CBS to Fort Lauderdale was a series of investigative reports by Glover and Larrubia
(1996) in a local newspaper claiming that city pawnshops routinely accept suspect merchandise.
After sorting through 70,000 pawn tickets to identify and examine backgrounds of the 50 most
prolific pawners, the journalists found three common characteristics: most were unemployed, 78
percent had arrest records (half of them for property crimes, most others for drug offenses), and all
possessed a seemingly endless supply of things to pawn. A police survey of frequent pawners
produced like findings in Portland, Oregon. It noted that 90 percent were chronic drug users with
long criminal records, and that most were unemployed (Hammond, 1997).

The combination of high arrest and unemployment rates among prolific pawners implies that
pawnbrokers have a correspondingly high probability of receiving stolen goods from such people.
However, without an indication of the proportion of prolific pawners in the whole population of
customers, or of the share of their goods relative to all pawned items, it is hard to gauge the
significance of Fort Lauderdale and Portland findings. Johnson and Johnson (1998), for example,
make clear that frequent pawners are not representative of the general clientele. Their interviews with
1100 randomly selected borrowers in 1997 show that most are employed males, usually high-school
graduates, without bank accounts (Johnson and Johnson, 1998).

In general, research by scholars and journalists suggests three things. First, pawnbrokers do
have some role in recycling stolen goods. Second, frequent pawners present the highest likelihood of
acting as main agents through which pawnshops acquire hot goods. Third, the volume and value of
these goods may be substantially greater than the tiny fractions that have been proposed. Pawn data
from Dallas, Texas, provide circumstantial evidence to support these suggestions.

Data from Dallas, Texas

In addition to data from interviews with nine imprisoned property offenders (confirming the
findings of Cromwell (1991) and Richard and Decker (1993) on use of brokers), eleven pawnshop
managers and a dozen police officers in pawn units in Dallas and surrounding municipalities, these
data comprise three related components. First is a primary database of all pawn transactions recorded
by the Dallas Police Department (DPD) during the six-year period from January 1, 1991, through
December 31, 1996. Each transaction shows a pawn ticket number, a pawner identification number,
shop identification number, transaction date, and classification code for items pawned. 4

The second component, devised to help us calculate transaction values, comprises 1000
randomly selected pawn tickets issued by 102 pawnbrokers during the first half of 1993. In addition
to items in the primary database, each ticket shows the pawner's zip code address, age, sex, and race;
the loan period, amount of loan, finance charge, description and quantity of items pawned, and
whether the transaction involves a loan or sale. With these data we estimate mean dollar values for
twenty-one items that together represent almost 70 percent of all pawned goods.

The third data component, designed to examine arrest histories, is sample data on 2000
pawners. These show pawner identification number, street address, age, race, gender, and state arrest

record, if any. To create the sample we stratified the primary database into ten frequency-oftransaction
classes (i.e., one pawn transaction during 1991-96, two transactions, etc.), and then
randomly selected 200 individuals from each category. The selection provided a list of identification
numbers that we then used to search through state and county public records for individual arrest
information. Glover and Larrubia (1996) used a similar method. The difference between the two
approaches is that we cover all pawners for six years, not just the prolific ones for one year.

The Evidence Connecting Pawnshops and Stolen Goods

These data show that Dallas pawnshops received more than 5.5 million items in pledge
during the six-year period, or a daily average of 23 items per shop. 5 Table 2 indicates that about two
dozen items made up more than 70 percent of the total, most in the categories of jewelry, electronics,
tools, office equipment, and firearms. This distribution pattern is roughly similar to that reported by
Caskey and Zikmund (1990) and also, as noted in the Appendix, to the composition of stolen goods.
This similarity between types of items pawned and stolen is expected because the features that make
some things better candidates for theft than others, i.e., easily concealed, removable, valuable,
enjoyable and disposable (Clarke, 1999), also make them good things to pawn.

The 5.5 million items were pledged by 523,000 different individuals during the course of
nearly 2.9 million transactions (Table 3). 6 We estimate the total loan value of these transactions at
about $208 million in 1999 dollar terms, or an average of $73 per transaction. 7 This is near the $70
that Johnson and Johnson (1998) report for 1.2 million pawn tickets reviewed by the National
Association of Pawnbrokers in 1997, and the $75 for 1.6 million tickets issued in Illinois in 1998 and
1999 (Illinois Office of Banks and Real Estate, 2001).

Frequent pawners generated a disproportionate share of this activity. The 14,500 people
pawning 30 times or more, though only 2.7 percent of the total, were responsible for 29 percent of all
transactions and goods, and 24 percent of total loan value. 8 This group, as in Ft. Lauderdale and
Portland, also held a higher proportion of individuals with arrest records. Its members were two to
three times more likely to have been convicted for theft, larceny, burglary or robbery than those who
pawned once or twice. Nearly two-thirds of the 1100 individuals within the group who pawned more
than one hundred times had arrest records, more than half of them for some kind of stealing.

Looking at the most prolific pawners, the top 100 individuals who each pawned more than
250 times, Table 4 shows 83 with arrest records. Of these, 58 had accumulated 300 convictions for
property as well as other offenses, or an average of 5.2 arrests per individual, Most property crime
arrests, 74 percent, were for theft, 11 percent for burglary of vehicles, 7 percent for burglary of
homes or businesses, 5 percent for robbery, and the rest for forgery and car theft. Other infractions
mainly involved drug possession (23 percent) or driving without a license (23 percent).

Among the 42 persons not apprehended for property crimes, 17 had no records while 25 had
49 convictions for non-theft infractions, or an average of 2 arrests per person. Very similar to the
previous group in age, sex and race composition, arrests here were mainly for drug possession (40
percent) and driving without a license (14 percent). But it would be premature to assume that these
people committed no property offenses. As we note in the Appendix, nearly three-fourths of thefts
from households, half of burglaries, and a third of robberies were not reported in 1999. Employee
theft, a prime source of stolen goods, usually goes unreported too. And for reported crimes, clearance
rates are low in large cities: 15 percent for theft, 12 percent for burglary and 23 percent for robbery
(U.S. Federal Bureau of Investigation, 1995-2000). That is, 85 to 95 percent of property crimes are
not solved. This suggests that many thieves are able to evade arrest, a fair share of them are likely to be in the group of frequent pawners that do not have property-related arrest records, and a goodly
proportion of things pledged by members of this group are unlikely to be their own property.

By the same token, however, it is equally premature to presume that all frequent pawners, or
even just those with property arrests, systematically dispose of stolen goods through pawnshops.
Although reliance on pawnbroker loans is an expensive method of personal finance, individuals may
confront situations where this is the best source available to them. What our observations suggest is
that although research by those who look only at the most prolific pawners may inflate the role of
pawnbrokers in disposal of hot goods, evidence still indicates that more of these goods flow through
shops than scholarly research has acknowledged. The amount may not be as high as 25 percent of
total value, but it is certainly greater than the fractions of 1 percent noted earlier.

The Dallas Police Department, for example, reports recovery of $2.4 million in stolen
property from pawnshops in 1997. This represents about 3.5 percent of the average annual value of
transactions during 1991-96. 9 Given that police recover only goods with unambiguous markings
that determine true ownership, it seems likely that the share of stolen goods is much greater, greater
even than the 5 percent dismissed by Oeltjen (1996).

What that higher share might be is uncertain. Caskey and Zikmund (1990) report pledge
forfeit rates in three states ranging from 14 to 22 percent of loans, and 10 to 20 percent of loan value.
Our interviews with brokers suggest 20 to 25 percent of loans. Johnson and Johnson (1998) indicate
that 29 percent of pawners forfeited at least once during the year. Of more interest, they also report
that while 23 percent of individuals who pawned only once lost their pledges, 34 percent of those
pawning four times or more forfeited at least once, 20 percent at least twice. In other words, as might
be expected if one suspects that the frequent pawner population contains many thieves, prolific
pawners are much more likely to walk away from their goods than infrequent pawners.

Given the many legitimate reasons that people might have to forfeit, it seems unlikely that hot
goods constitute as much as 20 percent of all things pawned. By the same token, high reported rates
of forfeiture make it equally unlikely that they represent less than 5 percent - especially when 15 to
20 percent of transactions are straightforward sales of goods, not loans.10 The actual proportion
seems likely to lie somewhere in between. If it is in the vicinity of, say, 10 percent, then the annual
value of stolen goods at secondhand market prices might average about $64,000 per pawnshop.
Extrapolated to the universe of 15,000 brokers, the total approaches $1 billion.

This is substantial in relation to pawnshop turnover because brokers make nearly as much
money selling forfeited goods as they do collecting interest. 11 Income from interest and fees
averaged $145,100 per pawnshop in 1997 or, extrapolating to 15,000 shops, a total of about $2.2
billion (U.S. Census Bureau, 2001b). Sale of forfeit property produced an average of $253,400 per
shop, suggesting a net revenue of around $126,700 after deducting loss of capital. 12 For all shops,
these figures imply a gross income of $3.8 billion and a net of $1.9 billion from sale of merchandise.
That is, $1 billion may represent more than a quarter of the pawnbroking industry's gross, and more
than half of its net proceeds from sale of property.

At the same time, if the total value of pilfered merchandise is close to our estimate of $40 to
$45 billion per year, then this $1 billion also represents 2.0 to 2.5 percent of all stolen goods. And if,
say, half the items are retained by the people who steal them, it represents 4 to 5 percent of all goods
disposed. 13 This is a modest share of the whole market. But pawnshops are only one component of
used merchandise trade. Excluding repair and other shops selling secondhand items as sidelines, the
census counted an additional 105,000 used merchandise stores with gross sales of $8.3 billion in
1997 (U.S. Census Bureau, 2001e). If stolen goods comprised a quarter of these sales as well, then
one could account for another $2.1 billion and thus another 8 to 10 percent of the market for stolen goods. The entire used merchandise sector, including pawnshops, might then be handling $3.1 billion
per year, or 12 to 16 percent of the market for hot goods.

Conclusions

Although our estimates may eventually prove inaccurate, there seems no way to circumvent a
few essential facts. One is that pawnbrokers, as omnipresent today as McDonald’s restaurants, offer
thieves a potentially convenient method of disposing of merchandise, especially items with no
obvious markings. Another fact, found in our data and by journalists in Fort Lauderdale and police in
Portland, is that the population of prolific pawners contains a large segment of people with robust
arrest records. Combined with findings from burglar interviews, this strongly intimates that the
population contains a substantial corps of habitual thieves who actually do rely on pawnbrokers for
their recurrent service needs. A third fact, as we have tried to show, is that a modest percentage of the
total value of pawnbroker transactions is sufficient to constitute a noteworthy share of our estimated
$40 to $45 billion per year in stolen property. Even if our numbers are very wrong, there is enough
circumstantial evidence here to warrant much more scholarly research, the quantitative sort
especially, on connections between pawnbroking (and other components of the used merchandise
retail sector) and hot goods. This is our first conclusion.

Our second conclusion is that the idea of deliberately disrupting markets for stolen goods
does not seem well founded. In the case of pawnshops, Sutton’s (1998) notion that businesses can
avoid buying such goods by use of closed-circuit televisions, photographs and/or signs is not
convincing because these actions are the same or similar (e.g., fingerprints) to measures that most
pawnbrokers undertake routinely. Such tactics and police oversight may have reduced the flow of
stolen goods through pawnshops at different times, but there is no evidence to confirm this. More
likely, given the criminal records of prolific pawners, is that they have not dissuaded thieves from
availing themselves of pawnbrokers. One reason, proposed by Hall (1935), may be that enforcement
of pawnshop regulations is too perfunctory to interfere with receipt and disposal of stolen goods.
Another is that enforcement has been effective, but only to the extent of displacing part of the trade to
other, less regulated enterprises, such as secondhand, precious metal and antique dealers or, where
these are also under perpetual scrutiny (e.g., Illinois, Washington), flea markets and the like. A third
possibility, the most plausible, is that most stolen goods are not identifiable as such.

There are several dimensions to this issue. As discussed by Clark (1999) and Sutton (1998),
one is technical. Most stolen items are not unique, do not have serial numbers, engraved codes or
other property identifiers, or else have markings that are easy to remove. Another dimension is social.
Most households neither record serial numbers of what they buy nor engrave them and, together with
firms, do not report their loss to police.

The third aspect, the most important, is economic. Because the annual volume of hot goods is
large, perhaps measuring in the hundreds of millions of items, the societal outlay required to create a
record for each reported item and, at the same time, a reference base for the billions of things that
businesses buy legitimately, is larger. It is for the moment prohibitive. In other words, stolen goods
are not identifiable largely because it costs too much to identify them. They are, as a result, invisible
in the daily exchange of millions of secondhand items between sellers and buyers.

Viewed in this context, the claim that markets for stolen goods act as underlying incentives to
steal makes sense only if one subscribes to the notion that these markets are clearly separate or
separable from retail trade in general and used merchandise trade in particular. Because they are
neither clearly separate or nor separable, because it is usually impossible to know what is or is not stolen, a recommendation to deliberately disrupt demand for hot goods is a recommendation to
deliberately disrupt demand for secondhand goods in general. This is unwise counsel.

From this, our third and final conclusion is that wiser would be support of actions to render
more efficient the monitoring of people and things circulating through pawnshops, secondhand stores
and similar establishments. The premise here is that identification and apprehension of thieves need
to remain the focus of police and criminologist attention. More efficient monitoring of suspicious
pawners and goods, achieved through strengthened pawn details, speedier transfer of transaction
records from pawnshops to police computers, accelerated analysis of the data, and similar means can
help in this. To the extent that pawnbrokers cooperate with police in improvement of monitoring,
these actions may also protect the interests of firms that prefer to shun hot goods.

The problem, now and in the past, is that pawn units (called details or squads sometimes) are
relatively understaffed, partly because police departments are asked to concentrate on crimes against
persons, partly because policy makers do not see gain from spending on data collection, and partly
because most pawnbrokers object to the extra cost and intrusion into their affairs. The units, as
consequence, are usually behind in data entry. Fort Lauderdale's pawn unit recorded less than 50
percent of the pawn information it received during 1995 (Larrubia and Glover, 1996). Dallas police
recorded 100 percent of all information for only one continuous 12-month span during 1991-96,
managing an average completion rate of 70 to 80 percent through the period. Consequently, items
were often identified as hot after pawnbrokers disposed of them, and transaction trails that could
justify surveillance of suspicious pawners were often identified after they ran cold.

Benefits of improved monitoring have already shown themselves through increased recovery
and apprehension. Murray (1996), for example, reports that police in Atlanta, Georgia, entered only
25 percent of transaction data in 1996. A year later, after installation of a computer system with
electronic data transfer from brokers, police entered 100 percent of pawn information, reduced
processing time from several weeks to 24 hours and increased recovery rates from 12 to 42 items per
month (Murray, 1997). In Florida, similarly, Perez (2000) reports a rise in recovery after the Broward
County Sheriff established an automated pawn tracking database. The new system also helped catch
175 parole violators and 110 felons pawning firearms in 1998. By 2000, some 50 state and local
agencies were using similar tracking systems, including the Florida Department of Law Enforcement
(FDLE) which initiated a project to install a statewide database in that year.

In addition to property recovery and apprehension of criminals, there is also the prospect of
using transaction data to map suspicious behavior in “real” time. People pawning twenty diamond
rings or watches or electric tools or city street directories or anything else of value within a relatively
short period, especially if encumbered with interesting criminal histories, earn immediate suspicion
of stealing or of receiving hot goods. If the items are not reported stolen or if they lack markings,
then arrest is not possible. But it is possible to conduct surveillance to determine whether initial
suspicions are justified, whether there are networks of accomplices warranting police attention, or
whether there are other ways to identify and maybe apprehend these or other thieves.

The prime obstacle to improved tracking is pawnbroker resistance. Rarely conceding the
utility of anything but “article only” tracking to help identify stolen goods, brokers habitually oppose
collection of personal information that might reduce the flow of patrons. In 2001, for instance, the
Florida Pawnbrokers Association responded to the introduction of automated tracking by threatening
to initiate legislation that would delete all customer data from pawn unit computer systems. 14 The
association did not achieve this goal, but it convinced the legislature to reduce funding for FDLE’s
statewide database project from $1 million in 2000 to $275,000 in 2001, and then to zero in 2002. 15 Pawnbrokers are nowhere strong enough to eliminate police scrutiny. But as the industry
organizes for common cause, strength such as that shown in Florida may spread. Helping in this is
the ability of brokers to propagate image-enhancing media stories containing references to academic
studies that point to minuscule quantities of stolen goods in their merchandise. Scarcity of proper
research on the role of pawnshops and other secondhand stores in hot goods trade thus makes it
easier for brokers to thwart efforts at improved tracking. Accordingly, the first conclusion put
forward above, about the need for more scholarly research, is in our opinion the most important.

References

Auster, B. (1997, Dec. 8), “The Pawn of a New Era is Upon Us,” US News and World Report 123
(22). p. 64.

Berg, E. (1991, Nov. 13), “Pawnshops Shed Their Dingy Image.” New York Times, pp. D4, D10.

Blakey, G. R. and Goldsmith, M. (1976), “Criminal Redistribution of Stolen Property: The Need for
Law Reform,” Michigan Law Review 74, 1512.

Breyer, R. (1995, Oct. 21), “Buying into a New Image: Pawnshops Jump into Respectability- Big
Business with Millions Using Them for Short-term Loans,”Austin American Statesman, pp. Al, A12.

Calkins, L. B. (1987, Aug. 10), “Wealthy Flock to Hock in Posh Pawnshop.” Houston Business
Journal, Vol. 17, No. 8. p. 1A.

Cash America (2002), Cash America International Inc., 2001 Annual Report [Online].
<http://www.cashamerica.com/pdf/CashAm01_rpt.pdf>[Dec. 2].

Caskey, J.P. (1994), Fringe Banking: Check-Cashing Outlets, Pawnshops, and the Poor. New York:
Russell Sage Foundation.

Caskey, J.P. (1995), “Explaining the Boom in Check-Cashing Outlets and Pawnshops.” Consumer
Finance Law Quarterly Report, Vol. 49, No. 4. Winter.

Caskey, J.P., and Zikmond B. (1990), “Pawnshops: the Consumers Lender of Last Resort.” Economic
Review: Federal Reserve Bank of Kansas City. 75.2, 5-18.

Chuang, T. (1998, Jan.1), “Easy Money: Pawnshops Losing Old Reputation, Gaining New
Customers.” Dallas Morning News, Arlington Morning News ed. [Online]
<http://www.purdue.edu/uns/clips/clips.01.98.980118.dallasmorn> [June 14].

Clarke, R.V. (1999), Hot Products: Understanding Anticipating and Reducing Demand for Stolen
Goods, Police Research Series Paper 112, London: Home Office Policing and Reducing Crime Unit.

Cromwell, P. F. (1991), Breaking and Entering, An Ethnographic Analysis of Burglary, Beverly
Hills: Sage Publications.

Drysdale, L. and Keest, K.E. (2000), “The Two-Tiered Consumer Financial Services Marketplace:
The Fringe Banking System and its Challenge to Current Thinking About the Role of Usury Laws in
Today's Society,” South Carolina Law Review 51, pp. 589-669

Florida Committee on Criminal Justice. (2000), “A Review of Florida’s Pawnbroking Law,” The
Florida Senate, Committee on Criminal Justice. Interim Project Report 2000-26.

Freiberg, A. (1997), “Regulating Markets for Stolen Property” Australian and New Zealand Journal
of Criminology, 30, 237-258

Glover, S. and Larrubia E. (1996, Nov. 24), “A System in Hock: Quick Cash with Few Questions,”
South Florida Sun-Sentinel, p. 22A

Gryta, M. (1998, Jun. 3), “Pawnshop Owner Pleads Innocent.” Buffalo News: p.B5

Hall, J. (1935), Theft, Law and Society, Boston, MA: Little, Brown, and Co.

Hammond, B.(1997, Dec.3), “Pawnshops Link Addicts and Crime.” The Oregonian, p. 5A

Illinois Office of Banks and Real Estate. (2001, July 14), “Annual Statewide Disclosure Results,”
Pawnbroker Regulation Section, [Online]
http://www.obre.state.il.us/CBT/PAWNBROK/pawnrept.htm [July, 14, 2001]

Ingrassia, R. (1995, Apr. 22), “A New Brand of Danger- Ex-police Officer Shot, Injured in Robberies
Separated by 8 Years at Pawnshops.” Dallas Morning News, p. A37.

Johnson, R.W. and Johnson, D.P. (1998), Pawnbroking in the U.S.: A Profile of Customers, Credit
Research Center, Georgetown University, Monograph #34, July.

Kleinfield, N. (1989, Aug 13), “Running the Little Man's Bank,” The New York Times, p. F4.

Kock, E., Kemp, T., and Rix, B. (1996), Disrupting the Distribution of Stolen Electrical Goods,
Crime Detection and Prevention Series Paper 69, London: Home Office Police Research Group.

Krause, K. (1998, Jun. 13), “Pawnshop Crackdown on Table; Customers Would be Taped, Looking
Out for Stolen Goods,” Fort Lauderdale Sun-Sentinel: p. 1B

Larrubia, E. and Glover, S. (1996, Nov 25), “Pawn Law Tips Scales Against Crime Victims,” South
Florida Sun-Sentinel. Digital Edition [Online].<http://www.sunsentinel.com/news/1663.htm> [1997,
Apr. 21].

Levine, S.W. (1913), The Business of Pawnbroking. New York: D. Halpern Company.

Lundy, A. (1994, Oct. 25), “150 Protest Pawnshop in South Oak Cliff: We Help the Community,
Owners Say,” Dallas Morning News, p. A26.

Marino, J. (1997, Aug. 14-20), “Easy Money in Hard Times,” Memphis Flyer [Online]
<http://www.memphisflyer.com/backissues/issue443/cvr443.htm> [1998, May 22].

McGeevy, P. (1997, Oct.18), “Pawnshop Computing; LAPD Wants All Dealings Put On Disk to
Help Hunt Criminals.” Los Angeles Daily News: p. N4

Murray, B. (1997, Dec. 8), “New Software Helps Police Finger Hot Pawned Goods.” Atlanta
Business Chronicle [Online]. <http://www.amcity.com/atlanta/stories/120897/story6.html> [1998,
June 12].

Murray, B. (1996, Sep. 2) “City's pawn policing system broken.” Atlanta Business Chronicle,
[Online]. <http://www.amcity.com/atlanta/stories/090296/story1.html> [1997, April 21].

Nickles, S.H. and Adams, E.S. (1994), “Pawnbrokers, Police, and Property Rights - A Proposed
Constitutional Balance,” Arkansas Law Review, Vol. 47, 793-815.

Oeltjen, J. C. (1996), “Florida Pawnbroking: An Industry in Transition,” Florida State University
Law Review, Vol. 23, 995-1041.

Pastore, A.L. and Maguire, K. eds. (2000), Sourcebook of Criminal Justice Statistics [Online].
<http://www.albany.edu/sourcebook/> [2001, July 20].

Patterson, W.R. (1899), “ Pawnbroking in Europe and the United States,” Bulletin of the Department
of Labor, 21. March. pp. 173-310.

Perez, J.C. (2000, Dec. 11), “Sheriff's Office Uses Technology to Catch the Crooks,” InfoWorld, Vol.
22, Issue 50, p. 47.

“Quick Cash.” (1997), narr. Mike Wallace. Sixty Minutes. WCBS: New York, 21 December

Richard, W. and Decker, S.H. (1993), “Exploring the House Burglar's Perspective: Observing and
Interviewing Offenders in St. Louis, 1989-1990" [Computer file]. St. Louis, MO: Richard Wright and
Scott H. Decker [producers], 1992. Ann Arbor, MI: Inter-university Consortium for Political and
Social Research [distributor].

Ruisseaux, R, (1995, Oct. 23), “Personal Finance: Pawnshops Clean up Image, Enter New Markets.”
Detroit News, n. pag.

Scott, D. (1992, June 26), “Pawnshops Multiply Despite Seedy Image: Big Chains Gobble Mom-andpops,
Spawn Industry Transformation.” Dallas Business Journal, pp. 1, 54-55.

Sheley, J.F. and Bailey, K.D. (1985), “New Directions for Anti-Theft Policy: Reductions in Stolen
Goods Buyers,” Journal of Criminal Justice 13, (5) pp. 399-415.

Stroer, J. (1997, Dec. 28), “The Pawning Game.” Athens Daily News Online Athens [Online].
<http://www.onlineathens.com/1997/122897/1208.o1pespective.html> [1998, June 14].

Sutton, M. (1995), “Supply by Theft: Does the Market for Second-Hand Goods Play a Role in
Keeping Crime Figures High?” British Journal of Criminology, 35, (3), pp. 400-416

Sutton, M. (1998), Handling Stolen Goods and Theft: A Market Reduction Approach, Home Office
Research Study 178, London: Home Office.

U.S. Bureau of Justice Statistics. (2001, July 26), Criminal Victimization in United States, 1999:
Statistical Tables, NCJ 184938. Office of Justice Programs, Department of Justice. [Online].
<http://www.ojp.usdoj.gov/bjs/abstract/cvusst.htm [July 26].

U.S. Bureau of Labor Statistics. (1999, June 25), “Consumer Price Index CPI-U.” [Online].
<http://146.142.4.24/cgi-bin/dsrv [June 25].

U.S. Bureau of the Census. (1953), Statistical Abstract of the United States: 1953, Washington D.C.:
U.S. Government Printing Office.

U.S. Bureau of the Census. (1978), Census of Retail Trade, 1977: Vol. 11, Geographic Area
Statistics, U.S. Summary, Washington D.C.: U.S. Government Printing Office.

U.S. Bureau of the Census. (1996), “Retail Trade Establishments-Number, Sales, Payroll, and
Employees, by Kind of Business: 1987 and 1992, (Table 1291),” Statistical Abstract of the United
States:1995, Washington D.C.: U.S. Government Printing Office.

U.S. Census Bureau. (2001a, May 20), “Retail Trade Sales- Summary: 1980 to 1999, (Tables 1274
and 1275), Statistical Abstract of the United States: 2000,” [Online]
<http://www.census.gov/prod/2001pubs/statab/sec27.pdf>[May 20]

U.S. Census Bureau. (2001b, June 16), “Sources of Revenue1997,” 1997 Economic Census, Finance
and Insurance Subject Series, [Online]. <http://www.census.gov/prod/ec97/97f52-ls.pdf> [June 16].

U.S. Census Bureau. (2001c, July 10), “Robbery and Property Crimes by Type and Selected
Characteristic: 1980 to 1993, (Table 316), Statistical Abstract of the United States:1995," [Online].

<http://www.census.gov/prod/1/gen/95statab/law.pdf> [July 10].

U.S. Census Bureau. (2001d, July 12), “Robbery and Property Crimes, by Type and Selected
Characteristic: 1990 to 1997, (Table 350), Statistical “bstract of the United States: 1999," [Online].

<http://www.census.gov/prod/99pubs/99statab/sec05.pdf> [July 12].

U.S. Census Bureau. (2001e, July 16), “Nonemployer Statistics, Retail Trade,@ 1997 Economic
Census, [Online] <http://www.census.gov/epcd/nonemployer/1997/us/US000_44.htm>[July 16].

U.S. Census Bureau. (2001f, July 20), “Robbery and Property Crimes, by Type and Selected
Characteristic:1990 to 1998, (Table 337), Statistical Abstract of the United States: 2000," [Online]

<http://www.census.gov/prod/2001pubs/statab/sec05.pdf >

U.S. Census Bureau. (2001g, Aug. 30), “Estimated Gross Margin as Percent of Sales by Kind of
Business: 1993 to 1999, (Table 6),@ [Online]
<http://www.census.gov/svsd/retlann/view/table6.html>[August 30].

U.S. Census Bureau. (2002a, Sept.14), “Unadjusted and Adjusted Estimates of Monthly Retail and
Food Services Sales by Kinds of Business:1999,@ [Online]
http://www.census.gov/mrts/www/data/html/nsal99.html [September 14]

U.S. Census Bureau. (2002b, Sept. 14), “Gross Domestic Product in Current and Real (1996) Dollars
by Industry: 1990 to 1999 (Table 641),@ Statistical Abstract of the United States:2002 [Online]
http://www.census.gov/prod/2002pubs/01statab/income.pdf [September 14]

U.S. Census Bureau. (2002c, Nov. 28), “Manufacturing and Trade Inventories and Sales, 1992-
2002,@ [Online] http://www.census.gov/mtis/www/data/text/mtis-sales.txt [November 28].

U.S. Federal Bureau of Investigation (1995-2000), Crime in the United States: Uniform Crime
Reports (annual), Department of Justice. Washington D.C.: U.S. Government Printing Office.

University of Florida. (2001, July 18), “2000 National Retail Security Survey, 9th Edition,@
Gainesville, FL: University of Florida. [Online]. <http://www.soc.ufl.edu/srp.htm> [July, 18]

Wheat, W.J. (1998), “A Study on the Status of the Pawnbrokers Industry in the State of Oklahoma,@
Consumer Finance Law Quarterly Report, 52, pp. 87-92.

Appendix: The Value of Stolen Goods

We estimate the net market value of stolen merchandise at $40 to $45 billion annually in
recent years, based on three sets of data and, for better or worse, several assumptions. The data sets
are: reported offenses in the Uniform Crime Reporting System (UCRS); reported and unreported
offenses experienced by households in the National Crime and Victimization Survey (NCVS); and
reported and unreported losses sustained by firms in the National Retail Security Survey (NRSS).

UCRS data indicate that reported offenses involving loss of property declined during the
1990s: theft-larcenies (excluding motor vehicles) by 12 percent, burglaries by 32 percent, and
robberies by 36 percent (Pastore and Maguire, 2000). The net value of these losses, however, stayed
constant at about $6.6 billion in current dollar terms. Households reported $3.7 billion of this sum
and businesses and others $2.9 billion.16 Jewelry and precious metals comprised the largest class of
merchandise, 16 percent of value, followed by electronics (14 percent), office equipment and
supplies (8 percent), clothing (4 percent), household goods (3 percent), and firearms (1.5 percent).

But property crimes are usually not reported. For households, NCVS data show that 73
percent of thefts, 48 percent of burglaries and 34 percent of robberies went unreported in 1999 (U.S.
Bureau of Justice Statistics, 2001). Our NCVS-based estimate of the net value of property taken from
households, accordingly, is much greater than that based on the UCRS, $7.3 vs $3.7 billion.17 The
composition of goods taken is nonetheless similar. Jewelry and clothing were stolen most frequently
(19 percent of all cases), followed by electronics and photographic gear (9 percent), vehicle parts (7
percent), tools and machinery (5 percent), household furnishings (4 percent) and firearms (1 percent).


Turning to business, NRSS data intimate that gross losses from theft and fraud in the retail
sectors it sampled, valued at prices sellers paid for them, came to $18.3 billion in 1999, or about 1.1
percent of the $1.74 trillion in total sales of these sectors (University of Florida, 2001). 18 Employee
theft was responsible for $9.2 billion, shoplifters $6.8 billion, vendor theft $1.2 billion, and check
and credit card fraud $1.1 billion.19 Main items taken were clothing and shoes, compact disks,
cassette tapes, video games, movies, over-the-counter remedies, health and beauty aids, automobile
accessories, jewelry, hand tools, cigarettes, and batteries. Although the loss rate of 1.1 percent of
sales held steady, the total value of goods stolen increased substantially in step with a 6.3 percent
average annual increase in retail sales through the decade (U.S. Census Bureau, 2001a).

There are no reliable data on loss recovery to help estimate net losses in the retail sector, and
no reliable figures on the proportion of incidents reported which could help determine the share of
losses already included in the UCRS. We therefore make assumptions. First we assume that the
recovery rate for retail stores is 7 percent, the same as for theft in the UCRS and NCVS. This lowers
merchandise loss from a gross of $18.3 billion to a net of $17 billion. Second, we assume that the
$2.9 billion in losses reported by businesses and others in the UCRS are mainly reports by retail
firms, and that they are already included in the $17 billion. Our estimate of the net value of goods
stolen from households and from firms of kinds covered by the NRSS is then $24.3 billion, of which
$7.3 billion in household losses and $17 billion in retail business losses. 20

To this we must add losses in retail sectors not covered by the NRSS with sales of $1.1
trillion in 1999, losses in manufacturing and wholesale trade with combined sales of $6.5 trillion, and
losses in other sectors, especially from employee pilfering, in services and government.21 With no
better basis to guess, we assume that theft of goods is (relatively) negligible in government, services
and other sectors that are merchandise poor and thus less exposed to removal of objects than retail
firms. We assume also that net losses in manufacturing, wholesale trade and the balance of retail
amount to 0.25 percent of their total sales of $7.6 trillion in 1999, or $19 billion. Adding this to the
$24.3 billion lost by households and retail firms in the NRSS yields $43.3 billion, and the estimate
range of $40 to $45 billion.


Table 1: Pawnbrokers in the United States in 2002, by state a