US Attorneys > USAM > Title 4 > Civil Resource Manual
prev | next


Probation Letter for Convictions Where the Count of Conviction Occurred Prior to November 1, 2001

This is a letter analyzing loss under U.S.S.G. 2F1.1. This applies to offenses completed before November 1, 2001. For offenses that continued or were completed after November 1, 2001, use the U.S.S.G. 2B1.1 letter at Civil Resource Manual 182.

Note: this letter assumes that you have or can obtain a declaration from the case agent which provides:

  • average high and low mileage on vehicles whose odometers were rolled back;

  • the average mileage rollback;

  • total number of vehicles whose odometers were rolled back (Section III C below discusses how to estimate this number); and

  • retail price information for the vehicles clocked (or at least information which can be used to estimate retail price).

Often these numbers are derived from only a portion of the vehicles involved in a case, since "perfect" information is seldom available. References to the declaration should be incorporated in the letter where these numbers (average rollback, number of vehicles, price information) are used.

Early in the case, try to have your agent know he or she needs to obtain this type of information.


Probation Officer
United States Probation Office
[  ] District of [state]
[city, state  zip]

     Re:  United States v. [name], et al., 
          Criminal No. [docket number]

Dear [     ]:

     This letter will provide background on the defendant[s] in this
case and [his/her/their] crimes.  It will also discuss pertinent
considerations under the Sentencing Guidelines, and outline the United
States's views on a Guidelines' analysis.  I trust the information will be of
value to you.  


     On [date] a federal grand jury in [city] returned a
[#] count indictment against [#] defendants.  All charges
stemmed from an odometer rollback scheme.  All defendants were charged in each
count [except . . . .].  

     The offenses charged in the indictment were as follows:

[Describe the charges in the indictment.]


     The defendants pled guilty to the following counts:

     [List the counts to which the defendants pled guilty.] 

     Enclosed are copies of the plea agreements which each defendant entered. 
As set out more fully in each of the agreements, the United States has agreed
to recommend [describe any provisions in the plea agreement relating
specifically to recommendations on terms of sentences, departures, substantial
assistance, sentencing at the bottom or top of the range].  


     A.   The Odometer Fraud Scheme

     All odometer-tampering enterprises, of necessity, involve several
interrelated activities.  First, a late-model, high-mileage vehicle is
purchased at a low price that reflects the significant wear and tear on the
vehicle.  This vehicle will then be "reconditioned" to remove many outward
appearances of long use, and the odometer will be reset, often more than
40,000 miles below its true reading.  

     In addition to the cosmetic "reconditioning" of the car, the odometer
tamperer must also "recondition" the paperwork accompanying the vehicle.  Most
automobile titles include a declaration of the vehicle's mileage at the time
ownership was transferred to the individual involved in odometer fraud.  To
hide the actual mileage that is declared on the title when the car was sold to
the odometer tamperer, the individual must take one of several measures.  

     First, the odometer tamperer may simply alter the mileage figures
appearing on the vehicle's title documents so as to cause the documents to
reflect the false, lower mileage figure after the rollback.  In so doing, the
odometer tamperer may also create one or more transfers of the vehicle to
fictitious dealerships, thereby seeking to make it unclear which entity was
responsible for the odometer rollback and title alteration.  

     Second, the odometer tamperer may destroy the original certificate of
title documents indicating the vehicle's true high mileage, and obtain a
duplicate certificate of title, upon which the false, lower mileage figure is

     Third, the odometer tamperer may engage in a practice known as "washing"
the automobile titles.  Here, the odometer tamperer transfers ownership of the
vehicle on paper to a "straw" dealership, which may exist or be fictitious. 
This "straw" dealership uses the altered title documents to obtain a new
replacement title listing the false low mileage reading.  

     Under each of these scenarios, the result is the same.  The odometer
tamperer now possesses an altered, forged, or replacement title document(s)
(which are securities under federal law) containing a false low mileage
reading.  He uses these documents to sell the car (for several thousand
dollars above its actual value) to a purchaser who is deceived regarding the
vehicle's remaining useful life by the altered odometer, by the vehicle's
outward appearance, and by the counterfeit, low-mileage paperwork (title and
odometer statement).

     B.   The Defendants' Scheme

     The charges in this case focus upon [lead defendant] and his

          1.   [name of lead defendant]—the Ringleader

     Defendant has been in the used car business in [place] for many
years.  He has conducted business under a variety of names, including
_____________________.  For the most part, defendant has not dealt directly
with the public, but has purchased and sold vehicles at wholesale.  He has
acquired used motor vehicles from numerous sources, both in [name of state
where he did business] and out of state.  Similarly, he has sold vehicles
to firms as near to home as [place], and as far away as [place]. 

     Defendant gave direction to his co-conspirators, deciding what cars to
purchase, how much to pay for them, and how far to roll back their odometers. 
He also arranged and paid for the rollbacks, as well as the falsification of
title documents to show false low mileage on the titles.  He personally sold
many of the vehicles.  In short, he was in control of the conspiracy.  

     [Describe lead defendant's role with any details we have showing the
number of vehicles involved; evidence of knowledge of wrongdoing; evidence of
profits; and examples of controlling the conspiracy, such as giving directions
to others.]

          2.   [name]—the "Spinner"  [if the spinner
               was charged]

     A "spinner" is a person who does the physical work of altering an
odometer.  Defendant was a professional spinner.  [Describe how we know the
number of vehicles he rolled, and how he was paid (e.g., in cash).  Outline
any cooperation provided, and admissions in debriefing, unless this last item
is precluded by the plea agreement.]  

          3.   [name]—Document Alterations  [if the
               title alterations person was charged]

     This defendant performed the work of altering the titles to reflect each
vehicle's new, lower mileage, after the odometers were rolled back.  This work
is essential to a rollback scheme because mileage is recorded on the title or
title reassignment on sale of a car, and when an odometer is rolled back,
title documents relating to the car have to be altered to reflect the false
low mileage.  

     [Describe how long this defendant was involved in alterations;
estimate the number of titles altered, and provide the basis for this
estimate.  Describe who paid this person, and how (e.g., paid in cash). 
Describe any cooperation during investigation, and admissions in debriefing,
unless this last item is precluded by the plea agreement.]  

          4.   [name]—[role]

     [Provide similar information for any other defendant, describing
their role, the number of vehicles with which they were involved, their
cooperation, if any, and admissions made in debriefing.]  

     C.   Total Cars Purchased and Sold with Rolled-back Odometers

     [Describe how the government knows the total number of clocked cars. 
It may be that we checked for rollbacks on 300 cars, and found 200 rollbacks. 
This may be all we use for sentencing.

     But if there is reason to believe this understates the defendants'
criminal activity, explain why that is so and provide a reliable estimate of
the total number of cars clocked.  For example, provide the number of vehicles
the government knows the defendants purchased and on which we checked for
rollbacks, and state the number of rollbacks found among those vehicles.  Such
as, "Defendants purchased 300 vehicles for which the government checked for
rollback activity, discovering 200 rollbacks.  This is 67% rollback rate for
vehicles the defendants purchased."

      Then explain how many vehicles we believe the defendants actually
purchased and sold during the period they were clocking cars.  For example,
"Defendants produced no records to the United States showing their total
volume.  Moreover, they did business in numerous places and dealt with
multiple banks and other institutions, making it difficult to collect records
showing a complete picture of their operation over the entire period of
illegal activity.  Nevertheless, bank records, auction records, dealer
records, and witness interviews establish that defendants purchased and sold
at least 900 cars during the conspiracy period."  Explain the basis for this

     United States v. Berndt, 86 F.3d 803, 811 (8th Cir. 1996),
supports this type of estimation.  In Berndt, although the government
could identify only 67 cars involved in the fraud, the court held that the
estimate of 90 vehicles was not unreasonable.  In support of this holding, the
court noted:  "Considering that fraud involves the element of deceit and
secrecy, it is likely that there are more automobiles involved in this
odometer-tampering scheme than the government can track down."  Id.

     Provide sufficient information to draw this sort of conclusion:  "From
the above, we know that defendants handled 900 vehicles during the period they
were clocking cars, and that their rollback rate was 67%.  Thus, the best
estimate of the number of cars defendants clocked is 67% of 900 vehicles, for
a total of 600 rollbacks."]

     D.   Victim Impact

     The primary victims of the defendant[s]' odometer fraud scheme
were the consumers who ultimately purchased the cars with altered odometers. 
The harms suffered by odometer-tampering activities generally fall upon those
persons least able to afford them.  Buyers of used cars include elderly people
on fixed incomes, younger people who have not yet earned enough to afford new
cars, and others who, for many reasons, are simply unable to buy new cars.
Additionally, rolling back a car's odometer directly affects the safety of the
car.  Indeed, in enacting the federal odometer laws, Congress expressly found
that "an accurate indication of the mileage assists a buyer in deciding on the
safety and reliability of the vehicle[.]"  49 U.S.C. § 32701(a)(3).

     As explained further below, each purchaser was harmed in several ways. 
First, the consumers were provided an inaccurate indication of the mileage
previously traveled by the vehicle, which impeded their ability to determine
the vehicle's value, safety, and reliability.  Accordingly, the consumer paid
a higher price for the vehicle than its fair market value.  Second, cars with
altered odometers have substantially reduced resale value once the fact that
the odometer has been altered has been made known (as must occur when the
consumer attempts to resell the car).  Thus, consumers will lose significant
money when they sell their cars.  Third, cars with higher mileage are more
costly to maintain, and more likely to require extensive repairs.  Thus,
consumers will lose significant money if they do not sell their cars.  Fourth,
consumers incur increased expenses (such as higher insurance premiums, lost
time) than they would have if they had known the cars' true mileage. 


     In this letter, the United States discusses elements of analysis under
the United States Sentencing Guidelines (U.S.S.G.) that are common to all
defendants.  We discuss elements that pertain only to one defendant, and
application of general principles to individual defendants, in separate
attachments applicable to each defendant.

     A.   The Fraud Guideline, U.S.S.G. &§ 2F1.1, Applies

     The Sentencing Guidelines that will be in effect on the date the
defendant is expected to be sentenced (and are in effect as of the date of
this letter) include an amendment to the fraud guideline which consolidates
the theft, property destruction, and fraud guidelines into a new guideline,
U.S.S.G. &§ 2B1.1.  However, because the amendment took effect
November 1, 2001, after the defendant's offenses were committed, but before
sentencing, they should not be applied because doing so would increase the
sentence and would violate the Ex Post Facto Clause of the Constitution. 
See United States v. Young, 932 F.2d 1035, 1038 n.3 (2d Cir.
1991).  Therefore, the 2000 edition of the Sentencing Guidelines Manual, in
effect on the date the defendant committed his offenses should be applied. 
See U.S.S.G. &§ 1B1.11(b).

     Each defendant pled guilty to violations of odometer tampering and
related statutes (failure to keep odometer records, providing false odometer
statements), in violation of [cite applicable statutes].  Guideline
&§ 2N3.1, "Odometer Laws and Regulations," cites to these statutory
provisions.  The background for this guideline explains that it applies only
where "a single vehicle" was involved.  It goes on to provide: "If more than
one vehicle was involved, the guideline for fraud and deception,
&§ 2F1.1, is to be applied because it is designed to deal with a
pattern or scheme."

     Finally, all counts of conviction should be grouped together.  This is
because all counts of conviction were part of a common scheme or plan
(U.S.S.G. &§ 3D1.2(b)), and because the fraud guideline applies to
each count and determines an offense level largely on the basis of total loss
(U.S.S.G. &§ 3D1.2(d)).

     Thus, U.S.S.G. &§ 2F1.1 applies, with a base offense level of
six.  U.S.S.G. &§ 2F1.1(a).

     B.   Amount of Loss Increase

     Guideline 2F1.1(b)(1) provides for an increase in offense level based on
the amount of loss incurred by a scheme or course of conduct such as that
here.[FN1]  Further, the amount of "loss need not be determined with
precision."  Application Note 9.  Rather, a "reasonable estimate" is
sufficient, given "available information."  Id.

     The guidelines specifically contemplate an estimate based on the
"approximate number of victims and an estimate of the average loss to each
victim."  Application Note 9.  That Note also states that an offender's gain
from the fraud "ordinarily will understate the loss."

     In this case, the amount of loss is estimated by multiplying a figure
representing the average loss per victim by the number of cars with rolled-
back odometers.  [Describe any sentencing stipulations regarding amount of
loss here.  Even when amount of loss is stipulated, it is valuable to include
a rationale for the agreed upon loss amount, so some or all of the following
analysis remains relevant.  Similarly, it is important that the Court be
provided with a rationale for stipulated loss amounts, either in the PSR or
through a sentencing brief from the government.]

          1.   Average Loss Per Victim - $4,000  [The loss may be
               higher, depending on retail prices.  $4,000 is based on 40%
               of retail.]

     The loss to the ultimate purchaser of each car the defendant[s]
sold with a rolled odometer can conservatively be estimated as $4,000.  This
is based on case law and other factors that establish the validity of a loss
per vehicle based on 40% to 50% of the retail price paid by consumer victims. 
There are a number of approaches that reach this same conclusion.

               a.   Courts Consistently Find Loss to be $4,000 to
                    $6,000 per Vehicle

     Several Federal Courts of Appeals have affirmed sentences in odometer
fraud cases where the trial court found consumer loss of $4,000 per car or
more.  Several of these cases are "unpublished," but most are available on
Westlaw.  Some are also published in the Federal Reporter series, as indicated
below.  These decisions provide guidance in this area based on other courts
which have examined these issues.  Moreover, the guidelines specifically
permit estimating loss "based on . . . the nature and duration of the fraud
and the revenues generated by similar operations."  Guideline 2F1.1,
Application Note 9.

     United States v. Whitlow, 979 F.2d 1008, 1012 (5th Cir. 1992),
explicitly affirms a loss finding of $4,000 per vehicle based on 40% of retail
price consumers paid.  In Whitlow, the district court had noted that
the National Automobile Dealers Association guide for used car values stated
that "High Mileage" deductions should not reduce the value of a vehicle by
more than forty percent.  Accordingly, the district court calculated the loss
per vehicle as forty percent of the average $10,000 retail purchase price of
the cars.  This resulted in a loss per car of $4,000, which the Fifth Circuit
said was plausible in light of the record as a whole.  979 F.2d at 1012.  The
analysis of the Fifth Circuit in Whitlow is directly applicable.  Here,
the cars the defendant sold had an average retail sale price of approximately
$_______________.  If the loss per vehicle is calculated at forty percent of
the average retail purchase price, the loss per vehicle here will be  
$_____________ [40% of avg. retail sale price].

     United States v. Berndt, 86 F.3d 803, 811 (8th Cir. 1996).  The
defendant rolled back the odometer on  80 - 100 cars.  The loss finding of
$4,000/car was not contested, but is mentioned in the opinion.

     United States v. Jarrahi, et al., Nos. 97-4289, 4311 (4th Cir.,
May 11, 1998), 1998 WL 230825.  Defendants appealed sentencing guideline loss
findings.  The Fourth Circuit held that the district court did not commit
clear error by holding the defendants responsible for rollbacks committed by
others even though the defendants were in a rather "loose-knit" association in
which they purchased vehicles for each other.  The loss finding was just over
$4,800 per car on a total of 364 cars.

     United States v. Carroll, et al., Nos. 97-4022, 4259 (4th Cir.,
November 19, 1998), 1998 WL 801880.  Richard Carroll and Charles Granata were
convicted after trial and sentenced to prison.  The court found that the
ultimate consumer purchaser was the real victim of the crime, and that a loss
estimate of $6,000 per car was reasonable.  

     United States v. Alami, 1997 WL 570867 (4th Cir., September  16,
1997).  The court upheld a loss finding under the Sentencing Guidelines of
$6,000 per car.  That was the average difference between the price defendants
paid for the cars, and the ultimate consumer purchase prices.

     See United States v. Sprague, 35 F.3d 559 (5th Cir. 1994)
(table) (affirming $4,000 loss per vehicle estimate); United States v.
David Allen Hatley, Crim. No. SA-96-CR-230 (W.D. Tex. December 13, 1996)
(applying $6,000 los per vehicle); United States v. Hampton, No. SA CR
96-40-GLT (C.D. Cal. Oct. 16, 1996) (rejecting PSR recommendation of $3,200
per vehicle loss estimate and holding that loss estimate should be $4,000 per
vehicle); United States v. Sadeghi, No:95CR00267-001 (M.D.N.C. May 7,
1996) (loss of $4,000 per victim); United States v. Rossi, Cr. No. 94-
506 (E.D. Pa. Aug. 1, 1995) (loss finding $4,000/car loss on 1,600 cars; $6.4
million total loss); see also United States v. Welch,
Crim. No. 93-30004-F (D. Mass. October 19, 1993) (adopting PSR loss estimate
using $4,000/car figure); United States v. Cooper, Cr. No. HCR 92-0149
(N.D. Ind. March 3, 1993) (same); United States v. Coker, Cr. No. H-92-
00089 (S.D. Tex. Aug. 7, 1992) (same).

               b.   Loss of Value at Resale

     As indicated above, the guidelines contemplate determining the amount of
loss on the basis of a reasonable estimate of loss per victim.  Guideline
2F1.1, Application Note 9; see, e.g., United States v.
Tardiff, 969 F.2d 1283, 1288 (1st Cir. 1992).  The guidelines specify how
this estimate is to be accomplished in a consumer fraud case such as this. 
Application Note 8(a) provides:  "In a case involving a misrepresentation
concerning the quality of a consumer product, the loss is the difference
between the amount paid by the victim for the product and the amount for which
the victim could resell the product received."

     In cases involving odometer rollbacks, this application note requires
estimating the value of cars with rolled odometers, and essentially giving the
defendant a "credit" for this amount when compared to the consumer price. 
(The consumer is the victim because, even when the defendant initially sold
vehicles to other car dealers, those dealers merely passed the vehicles along
as low-mileage vehicles.)  

     When the odometer on a motor vehicle is altered, and the vehicle resold,
it becomes impossible to know the vehicle's true mileage.  As a result, such
vehicles must be sold with notice to the buyer of an odometer discrepancy. 
See 49  U.S.C. § 32705(a) and 49 C.F.R. Part
580.[FN2]  Not surprisingly, consumers resist purchasing cars branded as
having an odometer discrepancy, which diminishes their value dramatically.

     A study of odometer fraud conducted by the Pennsylvania Attorney
General's Bureau of Consumer Protection (copy attached) explains:  "Many
consumers and dealers have told the Pennsylvania Bureau of Consumer Protection
and NHTSA odometer fraud investigators that they simply would not knowingly
purchase a rolled back vehicle regardless of price."  (Section D of the study,
"Monetary Damage to Consumers," page 13.)  The United States Department of
Transportation contracted for this study.

     Richard Diklich, an Instructor of Automotive Technology at Longview
Community College in Lee's Summit, Missouri, has quantified the effect of this
consumer resistance.  Mr. Diklich has extensive experience in the automobile
industry, and has developed expertise in evaluating the diminution of value
that arises when a vehicle is sold with notice that its odometer has been
rolled back.  (Diklich Dec., attached, ¶¶ 1 - 8.)  Mr. Diklich

     A vehicle with a Not Actual Mileage title does not have the same
     value in the marketplace as does a vehicle with the same odometer
     reading and a clean title.  The diminution in value in most cases
     will be 40% to 50% of fair market value.  This is true whether the
     vehicle is being sold at wholesale or retail.

Id., ¶ 8.  Mr. Diklich's Declaration goes on to explain the
reasons for this decline in value.  Id., ¶¶ 9 - 11.

     Thus, consumers have lost 40 to 50% of what they paid for the vehicle as
a result of the diminution in value caused by the necessity of disclosing the
rollback on the title.  This is precisely what Guideline 2F1.1, Application
Note 8(a) provides should be the measure of loss, and the measure approved in

               c.   Actual Expenses While Operating Clocked Vehicle

     Another approach to showing consumer loss is to look at the types of
actual expenses a consumer pays as a direct result of this type of fraud.  A
typical rollback in this scheme was [avg. rollback] miles.  See
Declaration of case agent, attached.  In addition to paying a fraudulent
purchase price, a consumer who purchases a vehicle with a rolled-back odometer
incurs a series of costs which stem directly from the fraud perpetrated by the
defendant.  These increased costs include the following:

     Maintenance Costs:increased maintenance costs are not always
reflected in purchase price considerations, as some owners have vehicles that
are virtually worthless as the result of a rollback.

     Lost Time:  buyers of high mileage vehicles frequently spend
considerable time dealing with unexpected maintenance--often time is lost from
work when a car is the person's method of transport to his or her employment.

     Taxes:  the sales tax imposed on the amount of a purchase price
attributable to the rollback, and annual property taxes imposed based on a
falsely inflated purchase price.

     Finance Charges:  the cost of financing the portion of a
vehicle's cost that stems from the rollback.

     Insurance Costs:  unnecessary insurance carried due to a false
belief that a vehicle is low mileage.

     The Pennsylvania Attorney General's study attempts to quantify losses
that consumers face.  The study reaches a figure of $6,653 as the total loss,
not including amounts for various losses that are difficult to quantify
(Pennsylvania Attorney General study, page 17).  Consideration of all of these
factors[FN3]ūsome of which (e.g., property taxes and insurance costs)
are recurringūdemonstrates that the $4,000 per vehicle the government suggests
is rather conservative, particularly so when one considers that the
Pennsylvania Attorney General's study was released in 1993, and has not been
adjusted for inflation.

               d.   Loss per Mile Rolled Back Alternative

     Another approach to viewing loss is through the use of "book value." 
The Pennsylvania Attorney General's study cites automotive industry
publications that provide deductions from "book" value that are associated
with mileage.  The figures cited are six cents and ten cents per mile.[FN4] 
For a [avg. rollback] mile rollback, this deduction would convert to a
$___________ (at six cents per mile) or a $____________ (at ten cents per
mile) fraud.

     However, for reasons discussed above, and in the Pennsylvania Attorney
General's study itself, such "book" value deductions underestimate the impact
of this type of fraud.  "Book" values compare two vehicles with honest
odometers.  At most, this reflects only loss associated with purchase price as

     Nevertheless, support for conservatively evaluating loss at ten cents
per mile removed can be found in a February 1993 bulletin published by the
National Association of Fleet Resale Dealers (NAFRD).  The bulletin reported a
1992 study by Associates Leasing which showed that resale value of cars fall
off markedly after 60,000 miles (copy attached, page 2).  The NAFRD bulletin
reported an 18% drop in value at about 60,000 miles, and another 10% reduction
at about 70,000 miles.  In other words, this report indicates a drop in value
of 28% at these two marks alone.  

     The NAFRD bulletin also included a table (on page 2) showing that
variable costs per mile for a car increase slowly to about ten cents per mile
at about 45,000 miles, but reach twenty cents per mile after about 80,000
miles.  Using the averages involved in this case, rolling an odometer from
[avg. pre-rollback mileage] to [avg. post-rollback mileage]
(see Declaration of case agent) would thereby push a car from being a
high cost per mile vehicle to falsely appearing to be a low cost per mile
vehicle.  The costs of operating such a vehicle will thus be considerably
higher than expected as a result of the fraud.  If these excessive operating
expenses are ten cents per mile to repeat the [avg. rollback] miles
rolled off the odometer, the consumer would be paying another $________ in
costs attributable to the fraud.[FN5]

     United States v. Fraaza, No. 97-3863 (7th Cir., March 12, 1998),
1998 WL 122159, involved low-value vehicles.  In that case, the Seventh
Circuit agreed with an approach to loss based on six cents per mile rolled
back.  Such old vehicles can give rise to particularly difficult valuation
issues because consumers pay relatively small amounts for the cars.  Fraaza
pled guilty to one count of making false odometer disclosure statements, and
received a 10-month jail term.  He had altered odometers on nine cars which
were 9-13 years old, and sold at retail for a total of less than $30,000,
which means each car cost the consumer around $3,000.  The Seventh Circuit
upheld the district court's loss finding, which was based on 50% of consumer
price or six cents per mile removed, which led to a loss of $10,000 to
$20,000.  (This case thus also supports 50% of retail as being a good measure
of loss.)

     Since the defendant here was not selling very old cars as in
Fraaza, a loss of ten cents per mile removed from the odometer,
consistent with the Pennsylvania Attorney General's study and the NAFRD
bulletin discussed above, would be more appropriate.  This would lead to an
average loss of more than $4,000 per vehicle.

               e.   Total Loss Alternative

     Application Note 8, U.S.S.G. &§ 2F1.1, states that the
valuation of loss discussion in the Commentary to U.S.S.G. &§ 2B1.1
applies.  See United States v. Wilson, 900 F.2d 1350, 1356 (9th
Cir. 1990).  Guideline 2B1.1, Application Note 2, notes that ordinarily fair
market value of property taken is used to evaluate a loss.  However, where the
"market value is difficult to ascertain or inadequate to measure harm to the
victim, the court may measure loss in some other way, such as reasonable
replacement cost to the victim."  Thus, the Probation Office is not obligated
to calculate a precise fair market value for the TMU cars, and may consider
replacement cost if it determines that market values are too difficult to

     The Pennsylvania Attorney General's study of odometer fraud states that
one measure of damages to consumers is the entire price paid for the car:

     Many consumers and dealers have told the Pennsylvania Bureau of Consumer
     Protection and NHTSA odometer fraud investigators that they simply would
     not knowingly purchase a rolled back vehicle regardless of price. 
     Therefore, the proper measure of damages would often be the entire price
     the consumer paid for the car.

Pennsylvania Attorney General Study, page 13.  This method to calculate loss
is applied by some state courts.  See, e.g., People v.
Ross, 25 Cal. App. 3d 190, 195, 100 Cal. Rptr. 703 (1972) (treating
odometer fraud as theft and holding that proper determination of theft amount
is full purchase price paid by the victim).

     Given that many defrauded purchasers of cars with rolled-back odometers
will be unable to sell their cars at all, this translates into a consumer loss
figure that equals the consumer's full purchase price of the used vehicle.  In
comparison to this amount, a loss figure of only $4,000 as suggested by the
government is an extremely conservative means of calculating loss under
Application Note 8(a).  

               f.   Summary

     In sum, the government's loss figures are well-founded yet conservative. 
They are based on methodology embraced by the courts.  The loss estimate
reflects the loss of value when a car is resold as well as expenses that are
incurred when a vehicle is kept and maintained.  The loss per vehicle is
conservative, when compared to the Pennsylvania Attorney General's study and
the total loss analysis set forth above.

     The substantial loss consumers face results from the reality that
clocked vehicles have but a fraction of the value of cars with known low
mileage, which is what the consumers believed they had purchased.  A consumer
with a clocked car is in a poor position to discover the true mileage of the
vehicle, and faces the fact that dealers simply do not want to purchase
clocked cars at all.

     In light of all these factors, an estimate of $4,000 per vehicle is
truly a highly conservative estimate of the losses suffered by a consumer who
purchases a rolled-back car.

[Note:  Defendants frequently ask that the court use "Blue Book" values to
determine loss.  To some extent "book values" can be used to support our
evaluations, as is done above, and in another respect book values
underestimate loss, for reasons discussed above.  In addition, OCL has
obtained declarations from the publishers of Kelley's Blue Book and the
N.A.D.A. Official Used Car Guide which state that their publications are not
appropriate to use in attempting to estimate the loss in value caused by an
odometer rollback.

Nevertheless, DOT's National Highway Traffic Safety Administration in 2002
released a study that uses high and low-mileage figures from the N.A.D.A.
Official Used Car Guide ("book value analysis") to conclude that consumers pay
about $2,336 more for a clocked car than they should.  The study makes it
clear that this figure is merely an estimate of increased purchase price, and
does not include other losses consumers suffer from odometer fraud.  The
Executive Summary of the study states:

     That sum [$2,336] does not include inflated financing, insurance
     and tax costs; additional amounts consumers pay for vehicle
     repairs; other consequential damages; the decreased resale value
     due to the vehicle having an altered odometer; or the many
     indirect or intangible costs of odometer fraud:  time spent
     waiting for vehicle repairs and road service, consumers' anger and
     frustration at being cheated and getting a car they wouldn't have
     wanted, and costs of government programs to detect and deter
     odometer fraud.

Study, p. vii.  See also id. at p. 35.  

     Thus, the most significant items beyond purchase price that contribute
to consumer loss are not considered by the NHTSA report:  repair costs and
loss of resale value.  These items are properly counted in measuring loss
under U.S.S.G. &§ 2F1.1.  Accordingly, the figure in the NHTSA study
only measures a portion of what constitutes loss under U.S.S.G.
&§ 2F1.1, making the study essentially irrelevant to loss analysis
under the Guidelines.

     As mentioned above, OCL has declarations from the authors of the
N.A.D.A. Official Used Car Guide and Kelley's Blue Book which state that using
their publications to try to establish losses caused by odometer fraud is not
appropriate.  The Diklich declaration (¶ 9) makes the same point.  If you
need a copy of these declarations or the NHTSA study to use at a sentencing,
contact OCL.]

          2.   The Total Number of Vehicles Involved in the Fraud

     As discussed above, the evidence indicates that defendant should be held
accountable for ___ vehicles.

          3.   Total Loss Caused by Defendant's Fraud

     Thus, there were ___ vehicles involved in the fraud, at $4,000 loss per
vehicle.  The total fraud is $___________, an extremely conservative figure. 
An additional ____ offense levels should be added to a defendant's total
offense level under 2F1.1(b)(1)(_).

     C.   Minimal Planning - Multiple Victim Increase

     The offenses here required extensive planning and created multiple
victims.  Accordingly, the upward adjustment of 2 offense levels provided by
U.S.S.G. &§ 2F1.1(b)(2) applies.  

     D.   Acceptance of Responsibility

     In each plea agreement, the United States agreed to a [2 or 3]
level downward adjustment for acceptance of responsibility under U.S.S.G.
&§ 3E1.1.  These agreements stemmed from both the
defendant[s]' agreements to enter pleas of guilty [well in advance
of trial], and from the indication of acceptance provided by cooperation
given to the government by discussing criminal activity in this case and
related odometer fraud investigations.  

     E.  Restitution

     [The "mandatory restitution" provisions of 18
 U.S.C. § 3663A-3664 have applied to Title 18 fraud
offenses since 1996.  Title 49 odometer offenses themselves are not subject to
mandatory restitution.  See 18
 U.S.C. § 3663A(c)(1).  However, U.S.S.G.
&§ 5E1.1(a)(2) directs that a court impose a restitution order as a
term of supervised release for offenses [such as odometer tampering] that are
not restitution-authorized under 18  U.S.C. § 3663(a)(1). 
Under U.S.S.G. &§ 5E1.1(b)(2)(B), that provision does not apply
where "determining complex issues of fact related to the cause or amount of
the victim's losses would complicate or prolong the sentencing process to a
degree that the need to provide restitution to any victim is outweighed by the
burden on the sentencing process."

     For mandatory restitution offenses, section 3664(f)(1)(A) states that
the defendant's economic circumstances may not be considered in determining
the amount of restitution, though they can be considered in ordering nominal
periodic payments in lieu of full restitution at once.  18
 U.S.C. § 3664(f)(3)(B).  Nevertheless, section
3663A(c)(3) provides that restitution need not be ordered if (A) "the number
of identifiable victims is so large as to make restitution impracticable" or
(B) if "determining complex issues of fact related to the cause or amount of
the victim's losses would complicate or prolong the sentencing process to a
degree that the need to provide restitution to any victim is outweighed by the
burden on the sentencing process."

     The victims are generally the first consumer purchaser of the
vehicles.  However, if that information is unavailable and current owners are
known, the current owners can properly be viewed as victims, as they cannot
readily resell the vehicles once notified of the rollback.

     With large numbers of vehicles and the vagaries of consumer car repair
bills, determining restitution amounts consumer by consumer is virtually
impossible, suggesting that the average loss per vehicle amount used to
calculate loss under &§ 2F1.1 should be used as the restitution
amount where restitution is practicable.  In those cases, probation should be
provided as complete a list of victims as possible, and urged to grant as
restitution the figure that is used to calculate "loss" under 2F1.1.

     Some jurisdictions view use of an "average" loss figure as
inappropriate.  In those jurisdictions, the Probation Office contacts victims
seeking individualized loss information.  If OCL or the USAO is sending victim
notification letters (informing consumers that they are victims of the
offense) in the period of time before sentencing, it is frequently efficient
to send in the same mailing to victims the notice that the Probation Office
provides to victims, in which the Probation Office solicits information about
consumer losses.

     In cases where restitution is not practicable, the following, reflecting
the language of the statute, may apply:]

     It is difficult to identify all the victims of the fraud scheme.  Not
only did the wholesale and retail car dealers, to whom the clocked cars were
sold, suffer a monetary loss, but each dealer who thereafter bought and sold
the vehicle, as well as the consumers who purchased the vehicles for their own
use, was defrauded.  By this point in time, some of the consumers may have
already resold the clocked cars to second- or third-generation purchasers
without suspecting the fraud.  Tracing chains of ownership for the many
vehicles involved here would be labor-intensive and time-consuming.

     Moreover, the number of victims in the case, the difficulty in setting a
fair market value for each vehicle at the time of its fraudulent sale, and the
difficulty of calculating the value of extra charges and expenses (such as the
increased finance charges, insurance costs, and repairs) makes it almost
impossible to set an exact restitution amount that would be appropriate for
each consumer who owned one of defendant's rollbacks for a period of time.

     The government will provide victim information related to the specific
vehicles identified as having been rolled back in the Indictment to which
defendant pled guilty.  However, the number of victims is so large that
fashioning a broader restitution order is impracticable.  The difficulty of
identifying victims, and other considerations discussed above, leads the
government to believe that determining complex factual issues related to the
cause and amount of victim losses would complicate and prolong the sentencing
process to a degree that the need to provide restitution on a broader scale
(related to all the vehicles involved in this case) is outweighed by the
burden on the sentencing process.  Accordingly, a broader restitution order is
not appropriate.  18  U.S.C. § 3663A(c)(3); U.S.S.G.
&§ 5E1.1(b).

     F.   Special Condition of Supervised Release

     Finally, the United States urges that an occupational restriction be
placed on the defendant as a condition of supervised release.  The defendant
engaged in a massive consumer fraud, causing large economic harm to numerous
consumers.  In light of this, a condition of supervised release prohibiting
the defendant both from operating a used car business and from being employed
in automobile sales is necessary to protect the public.

     Imposing this type of employment restriction as a special condition of
supervised release is specifically provided for by U.S.S.G.
§&§ 5D1.3(b), and 5F1.5.  The sentencing guidelines provide for
the imposition of special conditions of supervised release that are reasonably
related to (1) the nature and circumstances of the offense, (2) the need for
adequate deterrence of further criminal conduct, and (3) the need to protect
the public.  U.S.S.G. &§ 5D1.3(b).  Pursuant to U.S.S.G.
&§ 5F1.5(a), an employment restriction is permissible as a condition
of supervised release or probation if:

     (1)  a reasonably direct relationship existed between the defendant's
     occupation, business, or profession and the conduct relevant to the
     offense of conviction; and 

     (2)  imposition of such a restriction is reasonably necessary to protect
     the public because there is reason to believe that, absent such
     restriction, the defendant will continue to engage in unlawful conduct
     similar to that for which the defendant was convicted.

     Both of these factors exist in this case.  The criminal conduct is a
direct result of activities in the used car business.  Moreover, in light of
the defendant's multi-year involvement in odometer fraud there is substantial
reason to believe that the defendant will continue to engage in such offenses
absent such a court-imposed restriction.  

     In previous odometer fraud cases, district courts have not hesitated to
impose precisely this restriction.  See United States v.
Whitlow, 979 F.2d 1008, 1012 (5th Cir. 1992); United States v.
Mills, 959 F.2d 516, 519 (5th Cir. 1992).


     The above describes the conspiracy, summarizes the role of each
defendant in the conspiracy, and discusses guidelines applicable to all
defendants.  A copy of this letter will be provided to counsel for all
defendants.  In the following pages, we discuss how the guidelines apply to
each defendant.  A copy of the page(s) pertaining to each defendant will be
provided to that defendant's counsel.  Should you have any questions about any
of the discussion provided or need any additional material, please do not
hesitate to contact me.




1  Guidelines calculation for defendant

2  Declaration from case agent (providing average mileage rollback and retail
price information for the case, as well as information on the total number of
vehicles involved in the offenses.)

3  Pennsylvania Attorney General's Study

4  Declaration of Richard Diklich

5  Automotive Fleet, April 1993

6  National Association of Fleet Resale Dealers, February 1993 bulletin

[Copies of attachments 3 - 6 are available from the Office of Consumer
Litigation.]DEFENDANT A


2F1.1(a) BASE OFFENSE LEVEL                                  6

2F1.1(b)(1)(L) INCREASE ($800,000 - $1,499,999 LOSS)        11


3B1.1(a) ROLE IN THE OFFENSE                                 4

3C1.1 OBSTRUCTION                                            2

TOTAL OFFENSE LEVEL:                                        25


ADJUSTED OFFENSE LEVEL                                      22


     [Discuss any known criminal convictions and calculate criminal


     a.   Imprisonment

     An adjusted offense level of 23 combined with a Criminal History
Category of [x] results in a guideline range of [xx-xx] months

     b.   Supervised Release

     The crimes to which the defendant pleaded guilty are [Example] Class
E felonies.  See 18
 U.S.C. § 3559(a)(5).  Accordingly, because
[name]'s guideline sentence will be greater than one years'
imprisonment, a term of supervised release of one year is appropriate. 
U.S.S.G. §&§ 5D1.1(a), 5D1.2(b)(3).

     c.   Fine

     The sentencing guidelines require calculation of a minimum and a maximum
amount for a fine.  For an offense level of 23, under U.S.S.G.
&§ 5E1.2(c)(3), the fine guideline is $10,000 - $100,000. 

     d.   Special Assessment

     Pursuant to 18  U.S.C. § 3013, the defendant must
pay a special assessment of $100 for each count of conviction.  Thus,
[name] must pay special assessments in the amount of [$ xxx].

[Similar pages of analyses for each defendant in multiple defendant cases
should be added.]

FN 1. Application Note 7 provides: "The cumulative loss produced by a common scheme or course of conduct should be used in determining the offense level, regardless of the number of counts of conviction." Application Note 8 provides: "if an intended loss that the defendant was attempting to inflict can be determined, that figure will be used if it is greater than the actual loss." FN 2. The federal regulations implementing 49  U.S.C. § 32705 require that the odometer disclosure statement for a vehicle whose odometer does not display the vehicle's true mileage must state "that the odometer reading does not reflect the actual mileage, and should not be relied upon. This statement shall also include a warning notice to alert the transferee that a discrepancy exists between the odometer reading and the actual mileage." 49 C.F.R. &§ 580.5(e)(3). "True Mileage Unknown," or "TMU," is an industry colloquialism that refers to vehicles that must be sold with this disclosure. FN 3. Consideration of comparable factors affecting the victim's loss are frequently considered in determining total monetary loss attributable to a defendant's criminal conduct. See United States v. Curran, 967 F.2d 5, 6 (1st Cir. 1992) (victim's lost interest income considered as part of loss caused by defendant's embezzlement scheme); United States v. Pavao, 948 F.2d 74, 77 (1st Cir. 1991) (defendant's fraud contributed to victim's failure to keep up mortgage payments). FN 4. The six cents per mile figure was from Automotive Market Report (AMR), while the ten cents/mile estimate comes from Galves Auto Price List (Pennsylvania Attorney General study, p. 14). Both are among the more reliable mileage deduction indicators in the industry, according to an industry publication. "According to resale dealers, . . . [t]he used- car guides that did the best job of adding/subtracting for high miles, in order, were AMR, NADA, Galves, Kelley Blue Book, Black Book, and Gold Book." Automotive Fleet, April 1993, p. 68 (copy attached). Note that the graphic on page 1 of this article lists the "Top Factors in Determining Resale Price," and the first factor is "Mileage." FN 5. Losses suffered by victims cannot fairly be measured by gains to criminals. It is for reasons as those discussed in the text that the Guidelines provide that "[t]he offender's gain from committing the fraud is an alternative estimate [of loss under 2F1.1] that ordinarily will underestimate the loss." Guideline 2F1.1, Application Note 9. Consider by analogy the burglar who breaks into a home through a window, steals the stereo, and fences it for $100. The homeowner loses not only the stereo and suffers the cost of replacement, but must repair the broken window as well.

[updated May 2004] [cited in Civil Resource Manual 170; Civil Resource Manual 182]