Managing Risk at International Ports of Entry

1
Managing Risk at International Ports of Entry
Patrick L. Gurian, Department of Civil,
Architectural, and Environmental Engineering,
Drexel University Nicholas Dudley Ward,
Department of Statistics, The Wharton School,
University of Pennsylvania Josiah McC. Heyman and
Cheryl Howard, Department of Sociology and
Anthropology, University of Texas at El Paso
A Simple Normative Model
Public Perceptions Results from a survey of 704
border crossers in the summer of 2004.
Ethnographic Background How do border inspectors
make decisions? Summarized results from previous
studies
Illegal crossers may be recruited from any
societal subgroup but the total number is
constrained. The total number of detailed
inspections performed is constrained. Recruitment
of illegal crossers is designed to minimize
interceptions with the knowledge that inspectors
will select group with the highest proportion of
illegal crossers in order to maximize
interceptions. As a result, both recruitment and
inspection will be uniform across societal
groups. Duration of inspections should be
independent and identically distributed over time
- no serial correlation or systematic patterns as
such patterns can be exploited by illegal
crossers.
Perceptions of Inspectors Survey Question Now,
Ill read you some choices, please tell me which
comes closer to your views about the inspectors
at customs a. would you say they are fair
or unfair or it depends on the
inspector, b. would you say their work is
professional or unprofessional or it
depends on the inspector.
Additional Inspections Survey Question In
general, how often do you have to submit to
additional searches at customs? Frequently,
Occasionally, Rarely, or Never.

• Major constraint trade-off between intensity of
  inspection and
• traffic build-up. Port management receives
  pressure from local
• influentials to keep traffic moving. Port
  management pressures
• inspectors to achieve a satisfactory level of
  interdicted drugs while
• at the same time maintaining an average
  through-put of one vehicle
• per 30 seconds (El Paso) and not developing
  abnormally long queues.
• Secondary constraints (A) do not harass
  apparently influential
• crossers who will call lawyers, congressional
  offices, or embassies.
• (B) incomplete information--access to car and
  drivers license
• records, visas or immigration documents is
  available, but much other
• information is missing or cannot be
  cross-checked. Only a few
• crossers receive automated fingerprint checks.

Reported Frequency of Additional Inspections
Fairness of Inspectors
Inspections Observations
Normal Inspections 1,228 inspections of
northbound, non-commercial traffic observed at
Santa Fe Bridge in El Paso, TX (1) Average
duration 34 seconds compared to 30 second
goal (see ethnographic background section) (2)
45 last 20 seconds or less (3) 2 referred to
secondary inspections (4) 79 do not involve
opening a vehicle compartment Based on Villegas,
Gurian, Heyman, et al. 2006. Tradeoffs between
Security and Inspection Capacity Policy
Options for Land Border Ports of Entry,
Proceedings of the Annual Meeting of the
Transportation Research Board, Washington, DC
(accepted)
Response English Language Respondents Spanish Language Respondents
NEVER 37 18
RARELY 37 48
OCCASIONALLY 17 21
FREQUENTLY 9 13
Response English Language Respondents Spanish Language Respondents
FAIR 65 28
UNFAIR 3 9
DEPENDS ON INSPECTOR 32 63

• Inspectors detection strategies an overall
  gestalt created of many factors at once
  inspectors often call this gestalt a sixth
  sense or intuition.
• Elements of the gestalt
• (1) Direct documentary and detection system
  evidence if available, such as computer database
  records (intelligence) of vehicle plates or
  narcotic sniffing dogs.
• (2) Visible condition of the vehicle (if
  applicable).
• (3) Nerves/non-verbal cues.
• (4) Travelers story does it make sense
  (various socially based rubrics, such as apparent
  social class).
• (5) Kind of traveler
• (a) nation of origin immigration or visa
  status.
• (b) age, gender.
• (c) apparent ethnicity (body phenotype).
• Goals traffic moves steadily, officers get
  satisfactory number of busts, and avoid criticism
  by congress, lawyers, embassies, etc.

Professionalism of Inspectors
Response English Language Respondents Spanish Language Respondents
PROFESSIONAL 63 41
UNPROFESSIONAL 7 8
DEPENDS ON INSPECTOR 30 51
Different demographic groups perceive the process
differently with Spanish-language survey
respondents reporting a higher frequency of more
detailed inspections and less favorable attitudes
towards the inspectors.

• What is the role of profiling in these
  strategies?
• It does occur the need for rapid through-put of
  traffic results in tactics for quick
  decision-making, including profiling by apparent
  ethnicity, age, gender, apparent social class,
  type of vehicle, etc.
• The profiles are not static or simple. Officers
  report that they change profiles as smuggling
  strategies change but how much lag or
  over-simplification occurs is not
  well-documented.
• A characteristic use of profiling to manage
  traffic/interdiction trade-offs was a remark by a
  port manager who said they did not want time
  wasted by inspectors on grandmas. In fact,
  smugglers use many kinds of crossers but
  simplifications help officers and supervisors
  satisfy constraints.
• (3) The application of profiles differs between
  officers, but this is complex and not well
  understood.

Duration of inspections Half-hour observation
periods have been concatenated into a single time
series shown above. (1) Inspection durations are
not independent and identically distributed. (2)
Time-series analysis reveals a weakly
mean-reverting process consistent with pressure
to limit average inspection time and maintain
traffic flow.
Conclusions
Ethnographic research, survey results, and
inspection observations support the conclusion
that inspections are not uniform over time or
across societal groups as would be predicted by a
simple normative model. This has implications
both for social justice and for the effectiveness
of the inspection process as a risk management
tool. Time-series analysis of inspections shows
clusters of longer duration inspections and
shorter duration inspections. However, these
clusters do not follow a predictable pattern and
tend to persist for relatively brief periods of
time, making it difficult for prospective illegal
crossers to take advantage of them. Because the
simple normative model does not fully explain
this system, we are investigating more complex
and realistic models.
Sources Heyman, Josiah McC. 2004 Ports of
Entry as Nodes in the World System, Identities
Global Studies in Culture and Power 11
303-327. 2001 "United States Ports of Entry on
the Mexican Border, Journal of the Southwest,
43 681-700 1999 Why Interdiction? Immigration
Law Enforcement at the United States-Mexico
Border, Regional Studies 33 619-30. Gilboy,
Janet 1992 Penetrability of Administrative
Systems Political Casework and Immigration
Inspectors, Law and Society Review 26
273-314. 1991 Deciding Who Gets In Decision
making by Immigration Inspectors, Law and
Society Review 25 571-599. Nature of the data
ethnographic, based on direct observation at land
and air ports, interviews at the time of decision
of officers, and in-depth interviews of officers.
Multiple independent researchers report similar
results. Data is not quantified. Data was
collected in late 1980s and early 1990s, prior
to the current project (2003-2005). Changes
since this time period include better detection
technologies and stronger access to databases,
but limited interviews in 2004 indicate broad
applicability to the present. The current
project did not have direct access to officers.
When 30-minute observation periods are divided
into thirds (left), the mean durations of
inspections typically differ by ANOVA (right). A
single example analysis is shown above.
Acknowledgement This project was supported by
NSF award number 0332001