PRECEDENTIAL

         UNITED STATES COURT OF APPEALS
              FOR THE THIRD CIRCUIT
                   _____________

                       No. 15-2862
                      _____________

            UNITED STATES OF AMERICA

                             v.

              MAXIMO MATEO-MEDINA
                a/k/a David Contreras
                   a/k/a Luis Nieves
                 a/k/a Joseph Robles,

                                 Appellant
                      _____________


      On Appeal from the United States District Court
          for the Eastern District of Pennsylvania
             District Court No. 2-15-cr-00055-1
      District Judge: The Honorable Gerald J. Pappert

     Submitted Pursuant to Third Circuit L.A.R. 34.1(a)

                      April 18, 2016

    Before: McKEE, Chief Judge,1 FUENTES, and ROTH,
                     Circuit Judges

                  (Filed: January 9, 2017)



1
 Judge Theodore McKee concluded his term as Chief of the
United States Court of Appeals for the Third Circuit on
September 30, 2016. Judge Brooks Smith became Chief
Judge on October 1, 2016.
Brett G. Sweitzer, Esq.
Federal Community Defender
Office for the Eastern District of Pennsylvania
601 Walnut Street
The Curtis Center, Suite 540 West
Philadelphia, PA 19106
       Counsel for Appellant


Bernadette A. McKeon, Esq.
Clare P. Pozos, Esq.
Office of United States Attorney
615 Chestnut Street
Suite 1250
Philadelphia, PA 19106
       Counsel for Appellee
               _________________________

                 OPINION OF THE COURT
               _________________________

McKEE, Chief Judge

       Maximo Mateo-Medina appeals his sentence of twelve
months plus one day imprisonment for illegally reentering the
United States, in violation of 8 U.S.C. § 1326(a) and (b)(1).
Although Mateo-Medina pled guilty to the offense, he now
appeals the sentence, arguing that the sentencing court
violated his Due Process Clause rights by impermissibly
considering, among other things, arrests that did not result in
convictions. The Presentence Investigation Report (PSR) that
disclosed those arrests did not contain any of the underlying
conduct. For the reasons set forth below, we agree and we
will therefore vacate the sentence that was imposed and
remand for resentencing.




                              2
I. Factual & Procedural Background

       Mateo-Medina, a citizen of the Dominican Republic,
was initially deported from the United States in December
2012 after being convicted of unlawfully obtaining a U.S.
passport and serving a five-month sentence for that offense.
Shortly after he was deported, his common law wife,
Milagros Rasuk, a U.S. citizen with whom Mateo-Medina had
been residing for fifteen years prior to his deportation, was
diagnosed with terminal colon cancer. Rasuk had two adult
children from a prior marriage, both of whom had become
drug addicts, and one of whom, Miguel, resided with Mateo-
Medina and Rasuk. Rasuk’s oldest son, Risdael, who suffers
from mental health issues, had his own child, Angel. No
doubt because of Risdael’s drug addiction, he abandoned
Angel for all practical purposes, and Angel was raised by
Mateo-Medina and Rasuk.

       When Mateo-Medina received word that Rasuk had
been diagnosed with terminal cancer, he returned to the
United States to care for her during her final months of life.
She died in February 2014.2 Angel was no older than eleven
when his grandmother, Rasuk, died.             Mateo-Medina
continued to care for Angel and became his sole caretaker
following Rasuk’s death.3

        According to Mateo-Medina, Miguel’s continued
presence in the household became increasingly disruptive and
problematic following Rasuk’s death because of Miguel’s
involvement with drugs and alcohol. Mateo-Medina claims
that when he (Mateo-Medina) attempted to intercede and
confront Miguel about his behavior, Miguel reported Mateo-
Medina to the immigration authorities, informed them of his
illegal reentry, and kicked Mateo-Medina out of the home.


2
  Mateo-Medina’s mother, who was also a United States
citizen, died of lung cancer 10 months later.
3
  Angel’s therapist credits Mateo-Medina with providing a
“stable, reliable, loving environment,” for Angel following
Rasuk’s death. Appellant’s Br. at 6 (citing J.A. at 68).
                              3
       Miguel’s strategy apparently worked because Mateo-
Medina was subsequently arrested and charged with illegal
reentry. He thereafter pled guilty to one count of reentry after
removal in violation of 8 U.S.C. § 1326(a) and (b)(2).

       Mateo-Medina’s PSR calculated his offense level at
ten, and his criminal history category was II. This resulted in
a recommended sentence of eight to fourteen months’
imprisonment. The PSR noted that Mateo-Medina had two
previous convictions, one for driving under the influence in
2000, and one for fraudulently applying for a United States
passport in 2012. Mateo-Medina was arrested under a
different alias each time. The PSR also noted that Mateo-
Medina had “numerous” arrests that did not lead to
conviction.4 Aside from the arrests leading to his two
convictions, Mateo-Medina had been arrested six other times.
However, each of the charges involved in his arrests had been
withdrawn or dismissed, except for one which lacked a
recorded disposition. As we noted earlier, the PSR did not
describe any of the underlying conduct purportedly leading to
those arrests.

        Mateo-Medina argued for a downward departure from
the suggested eight to fourteen-month guideline range. The
District Court did depart downward one level. The court
adjusted Mateo-Medina’s Guidelines range downward to six
to twelve months’ imprisonment based on an offense level of
nine and a criminal history category of II. At the sentencing
hearing, both the prosecutor and the defense argued for a
sentence of time served, which would have been equivalent to
roughly six months, or the lower end of the Guidelines range.
In spite of this, the District Court sentenced Mateo-Medina to
twelve months plus one day, followed by two years of
supervised release.5

4
  PSR at ¶ 6.
5
  While this sentence was technically an upward variance
from the Guidelines range of six to 12 months, the sentencing
court explained that imposing a sentence greater than one
year would make Mateo-Medina eligible for “good time
credit” which could reduce his term of imprisonment by 54
days (15%). 18 U.S.C. § 3624(b)(1).
                               4
        In calculating Mateo-Medina’s sentence, the District
Court relied on the relevant factors in 18 U.S.C. § 3553(a)
and information contained in the PSR. Significantly, the
District Court also relied in part on Mateo-Medina’s record of
arrests that did not lead to conviction. The court explained:

        I also cannot overlook the defendant’s rather
        extensive and I think we all have our own
        barometer of what is extensive versus what is
        not extensive interaction with the criminal
        justice system. But there were as I counted, I
        believe seven arrests, two convictions in
        three states since 1988. So, the defendant
        who was in this country initially illegally
        since at least the 80s has engaged in conduct
        which to the Court’s view belied and made
        ring hollow a little bit his desire to merely
        come to America to seek a better life.6

II. Standard of Review7

         As a threshold matter, the parties disagree on the
applicable standard of review. The Government argues that,
because Mateo-Medina’s objection to the District Court’s
statement regarding the defendant’s prior arrest record was
not preserved at sentencing, it is reviewed for plain error.8
Under that standard, Mateo-Medina would bear the burden of
establishing the District Court committed plain error.9 This,
the Government urges, Mateo-Medina has failed to do
because, even if he could show plain error, he cannot show
that it affected the outcome of the proceedings.




6
  J.A. at 115 (emphasis added).
7
  We have jurisdiction under 28 U.S.C. § 1291 and 18 U.S.C.
§ 3742(a).
8
  Gov’t Br. at 12; see also United States v. Flores-Mejia, 

, 256 (3d Cir. 2014) (en banc); United States v.
Couch, 

, 252-54 (3d Cir. 2002).
9
  Johnson v. United States, 

, 467 (1997).
                              5
        Mateo-Medina counters that he did indeed preserve the
issue, and review should therefore be plenary.10 Mateo-
Medina points out that counsel objected to the inclusion of
the arrest record in the PSR and at the sentencing hearing. He
claims that the District Court understood the objection as an
attempt to exclude the arrests as a sentencing consideration,
and overruled it based on the court’s (erroneous) view that
arrests are “appropriate for the Court to consider [] under the
statutory [sentencing] factors.”11 Thus, in response to a
defense objection, the District Court expressly ruled on the
exact issue Mateo-Medina raises on appeal. Mateo-Medina
argues that this amounts to preservation, not forfeiture. We
need not address whether Mateo-Medina preserved his
objection at sentencing because our precedent clearly
demonstrates that a district court’s consideration, even in part,
of a bare arrest record is plain error.12

III. Discussion

A. Error

        Our review of a criminal sentence “proceeds in two
stages.”13 First, we review for procedural error, “such as
failing to calculate (or improperly calculating) the Guidelines
range, treating the Guidelines as mandatory, failing to
consider the § 3553(a) factors, selecting a sentence based on
clearly erroneous facts, or failing to adequately explain the
chosen sentence—including an explanation for any deviation
from the Guidelines range.”14 Under the plain error standard,
a defendant must show: (1) error, (2) that is plain or obvious,
and (3) that affects a defendant’s substantial rights.15 “If all
three conditions are met, an appellate court may then exercise

10
   Appellant’s Br. at 3, 12 n.7.
11
   J.A. at 100.
12
   See United States v. Berry, 

, 281-84 (3d Cir.
2009).
13
   United States v. Tomko, 

, 567 (3d Cir. 2009)
(en banc).
14
   Gall v. United States, 

, 51 (2007).
15
   United States v. Goodson, 

, 539 (3d Cir. 2008)
(citations omitted).
                               6
its discretion to notice a forfeited error, but only if (4) the
error seriously affect[s] the fairness, integrity, or public
reputation of judicial proceedings.”16 If we find procedural
error “our preferred course is to remand the case for re-
sentencing, without going any further.”17 In the absence of
procedural error, we will then determine whether the sentence
imposed was substantively reasonable. When reviewing for
substantive reasonableness, “we will affirm [the sentence]
unless no reasonable sentencing court would have imposed
the same sentence on that particular defendant for the reasons
the district court provided.”18

       Mateo-Medina relies on our opinion in United States v.
     19
Berry to argue that the District Court plainly erred in
considering his bare record of arrests not leading to
conviction when imposing his sentence.           There, the
sentencing judge considered the relevant factors under 18
U.S.C. § 3553(a) but also speculated about the reasons why
the defendants’ robbery charges had been nol prossed. In the
process, the prosecutor misread the PSR regarding Berry’s
bald arrest record.20

      The sentencing judge also inflated the defendant’s
propensity for crime by speculating that it was “rather
obvious that the reason he doesn’t have any actual adult
convictions is because of the breakdowns in the court—in the

16
   

v. United States, 

, 467
(1997) (citation omitted)).
17
   United States v. Merced, 

, 214 (3d Cir. 2010)
(citation omitted). However, on rare occasions, we have
chosen to proceed to the second step of the analysis,
substantive reasonableness, despite finding procedural error.
See United States v. Lychock, 

, 219-20 (3d Cir.
2009). We find Lychock to be the exception rather than the
rule, as that case concerned a defendant who was found guilty
of possessing over 150 images of child pornography, but who
received no jail term at sentencing, and where the district
court provided no reasonable rationale for its actions.
18
   

.
19
   

(3d Cir. 2009).
20
   

.
                              7
state court system—and not because of innocence.”21 The
court also considered appropriate factors under Section
3553(a) such as the seriousness of the crimes of conviction.
However, in imposing the sentence on Berry and his
codefendant, the court explained:

        Taking all those factors in to account, given
        the fact that their criminal points . . . I don’t
        think reflect quite adequately, the seriousness
        of their criminal exposure in the past. The
        fact that they were charged with crimes and
        then, the prosecution was dropped because
        nobody showed up to prosecute or something
        like that, means that their criminal history
        points were probably understated.22


        Berry appealed, arguing that the district court should
not have considered, even in part, his bald arrest record, when
those arrests did not lead to a conviction. He also argued that
the sentencing court erred in speculating about why some
prior charges were nol prossed and assuming he was guilty of
offenses that were dismissed. Because neither Berry nor his
codefendant objected during sentencing, we reviewed for
plain error.23 We specifically noted as a threshold matter
“that resentencing would be required here even without the
district court’s speculation about the reasons for prior charges
being nol prossed because of the misstatement of the
defendant’s arrest record and the district court’s misreading
of the PSRs.”24 Further, we explained:

        A defendant cannot be deprived of liberty
        based upon mere speculation. We therefore
        follow the reasoning of the majority of our
        sister appellate courts and hold that a bare
        arrest record—without more—does not
        justify an assumption that a defendant has

21
   

at 279.

.
24
   

                               8
        committed other crimes and it therefore
        cannot support increasing his/her sentence in
        the absence of adequate proof of criminal
        activity.25

       Here, the District Court also not only considered but
misstated Mateo-Medina’s prior arrests.26 While the District
Court considered a host of relevant factors under
Section 3553(a), it also erroneously and puzzlingly relied on
his misstated bald arrest record, stating:


        I also cannot overlook the defendant’s rather
        extensive and I think we all have our own
        barometer of what is extensive versus what is
        not extensive interaction with the criminal
        justice system. But there were as I counted, I
        believe seven arrests [actually six], two
        convictions in three states since 1988.27

       The Government argues that these statements indicate
only the District Court’s doubt as to Mateo-Medina’s
credibility in stating his reasons to return to the United States,
not his criminal nature, a point that the government terms the
“crucial” distinguishing factor between Berry and this case.28
However, Mateo-Medina had only two convictions in the
United States since 1988; one was a fifteen-year-old DUI, and
25
   

26
   Although 18 U.S.C. § 3661 provides that "[n]o limitation
shall be placed on the information concerning the
background, character, and conduct of a person convicted for
an offense which a court in the United States may receive and
consider for the purpose of imposing an appropriate
sentence," such information must be reliable. See 

(explaining that facts considered at sentencing
must be proved by preponderance of the evidence to satisfy
requirements of Due Process Clause). As explained more
fully herein, the bare arrest record relied on by the District
Court in this case does not meet this requirement of
reliability.
27
   J.A. at 115.
28
   Gov’t Br. at 17.
                                9
the other was for the passport violation for which he was
deported in 2012. It strains credulity to argue, as the
Government does, that the sentencing court was referring
only to these two convictions as an extensive interaction with
the criminal justice system.

       Accordingly, we conclude that, given our holding in
Berry, the District Court’s consideration of Mateo-Medina’s
record of prior arrests that did not lead to conviction was
plain error under the circumstances here.

         The Government next argues that:

           It was certainly not unreasonable for the
           district court to consider that persons
           genuinely occupied with the care of a
           terminally ill relative and a child typically do
           not have numerous interactions with the
           criminal justice system.29

       This argument is both irrelevant and illogical. It
assumes that one in Mateo-Medina’s circumstance who is
caring for a terminally ill relative does not venture outside the
confines of the home—a nonsensical proposition. It also
ignores the rationale that we clearly explained in Berry:

           [R]eliance on arrest records may also
           exacerbate sentencing disparities arising
           from economic, social and/or racial factors.
           For    example,     officers   in    affluent
           neighborhoods may be very reluctant to
           arrest someone for behavior that would
           readily cause an officer in the proverbial
           “high crime” neighborhood to make an
           arrest. A record of a prior arrest may,
           therefore, be as suggestive of a defendant’s




29
     Gov’t Br. at 19.
                                 10
        demographics as his/her potential           for
        recidivism or his/her past criminality.30

        Since we wrote Berry, substantial research and
commentary has only reinforced the regrettable circumstances
that we emphasized in disallowing consideration of bare
arrest records at sentencing. In 2013, The Sentencing Project
released a shadow report to the United Nations Human Rights
Committee, Regarding Racial Disparities in the United States
Criminal Justice System (Sentencing Project Report).31 The
Sentencing Project Report pointed to a wide body of
scholarship indicating that socioeconomic factors influenced
disparities in arrest rates.32

        The Sentencing Project Report also remarked on recent
research indicating that police are more likely to stop, and
arrest, people of color due to implicit bias. Implicit bias, or
stereotyping, consists of the unconscious assumptions that
humans make about individuals, particularly in situations that

30
   

(citing Barbara Bennett Woodhouse,
Youthful Indiscretions: Culture, Class Status, and the
Passage to Adulthood, 51 DEPAUL L. REV. 743 (2002); Jane
W. Gibson–Carpenter & James E. Carpenter, Race, Poverty,
and Justice: Looking Where the Streetlight Shines, 3–SPG
KAN. J.L. & PUB. POL'Y 99, 101 (1994) (“Police officers who
have worked in many types of neighborhoods acknowledge
that they call home to middle-class parents more readily.
Between suburban and urban departments, the difference can
be even more striking. A department of college-educated
officers in a suburb of Minneapolis in the 1970s went so far
as to invite parents and children into the station to discuss
their problems confidentially, with virtual immunity from
formal handling.”)).
31
   The Sentencing Project, Report of The Sentencing Project
to the United Nations Human Rights Committee Regarding
Racial Disparities in the United States Criminal Justice
System (August 2013), available at
http://sentencingproject.org/wp-
content/uploads/2015/12/Race-and-Justice-Shadow-Report-
ICCPR.pdf (hereinafter Sentencing Project Report).
32
   Sentencing Project Report at 3.
                              11
require rapid decision-making, such as police encounters.33
“Extensive research has shown that in such situations the vast
majority of Americans of all races implicitly associate black
Americans with adjectives such as ‘dangerous,’ ‘aggressive,’
‘violent,’ and ‘criminal.’”34

       In addition, a recent empirical study analyzed thirteen
years’ worth of data on race, socioeconomic factors, drug use,
and drug arrests.35 The study found that African-Americans,
Hispanics, and whites used drugs in roughly the same
percentages, and in roughly the same ways.36 The study
controlled for variables such as whether the participant lived
in high-crime, gang-controlled areas. Despite those controls,
the study concluded that “in early adulthood, race disparities
in drug arrest[s] grew substantially; as early as age 22,
African-Americans had 83% greater odds of a drug arrest
than whites and at age 27 this disparity was 235%.”37 With
respect to Hispanics, the study found that socioeconomic
factors such as residing in an inner-city neighborhood
accounted for much of the disparity in drug arrest rates.38

33
   

34
   

(internal citations omitted).
35
   Ojmarrh Mitchell & Michael S. Caudy, Examining Racial
Disparities in Drug Arrests, JUSTICE QUARTERLY (Jan. 2013),
available at
http://dx.doi.org/10.1080/07418825.2012.761721.
36
   

(“Contrary to popular explanations of racial
disparities in drug arrest[s], this research found that the racial
disparity in drug arrests between black and whites cannot be
explained by race differences in the extent of drug offending,
nor the nature of drug offending. In fact, in this sample,
African-Americans (and Hispanics) were no more, and often
less, likely to be involved in drug offending than whites.
Further, while minorities were more likely to live in the kinds
of neighborhoods with heavy police emphasis on drug control
and living in such neighborhoods had a strong relationship to
drug arrest; neighborhood context explained only a small
portion of racial disparity in drug arrests between African-
Americans and whites.”)
37
   

                               12
       Accordingly, we conclude here, as we did in Berry,
that the District Court plainly erred when it considered
Mateo-Medina’s bare arrest record when imposing sentence.

B. Substantial Rights

        Having concluded that the sentencing court committed
plain error in considering Mateo-Medina’s record of prior
arrests, we turn next to the question of whether the error
violated Mateo-Medina’s substantial rights. As explained in
United States v. Marcus,39 errors that violate substantial rights
“[i]n the ordinary case” must be “‘prejudicial,’ which means
that there must be a reasonable probability that the error
affected the outcome of the trial.”40

        Here, as we have explained, the sentencing court
erroneously considered Mateo-Medina’s bare arrest record
when determining the length of his sentence. It did so in spite
of the prosecution and defense counsel agreeing to a lighter
sentence and in spite of Mateo-Medina’s minimal record of
only two prior convictions for nonviolent offenses since the
1980s. We realize that the sentencing court also referenced
numerous other factors that were appropriate to consider in
deciding upon Mateo-Medina’s sentence. However, that is no
more palliative here than it was in Berry. The District
Court’s determination of an appropriate sentence for Mateo-
Medina was nevertheless influenced by the impermissible
consideration of Mateo-Medina’s arrest record. We think it
highly unlikely that the court was thereafter able to unring the
bell when considering the guidelines or the factors contained
in Section 3553(a). As we said in Berry, “The guidelines are,
after all, purely advisory, and unsupported speculation about
a defendant’s background is problematic whether it results in
an upward departure, denial of a downward departure, or
causes the sentencing court to evaluate the § 3553(a) factors
with a jaundiced eye.”41 Here, for example, the sentencing

39
   

(2010).
40
   

(citations omitted).

(citing United States v. Booker, 

(2005).
                               13
court stated that Mateo-Medina’s prior interactions with the
police made his statement that he came to the United States to
seek a better life “ring hollow.”42 We therefore conclude that
the court’s improper consideration of his bare arrest record
affected the entire sentencing hearing and resulted in
prejudicial error.

       Finally, calculating a person’s sentence based on
crimes for which he or she was not convicted undoubtedly
undermines the fairness, integrity, and public reputation of
judicial proceedings.

       Thus, all four plain error factors are met here, and
resentencing is required.

IV. Conclusion

       For the foregoing reasons, we will remand the matter
to the District Court for resentencing.43




42
   J.A. at 115.
43
   While Mateo-Medina has finished his term of incarceration,
he remains subject to the remainder of his two-year term of
supervised release, with all of the restrictions that supervised
release entails. We are confident that resentencing that
accurately reflects Mateo-Medina’s minor criminal history
will afford him some relief from those restrictions.
                              14