oversight

District of Columbia: Information on Court-Ordered Tenant Evictions

Published by the Government Accountability Office on 1990-12-28.

Below is a raw (and likely hideous) rendition of the original report. (PDF)

GAO                             Report to the Vice-Chairman,
                                Committee on the District of Columbia,
                                House of Representatives

-.~
l)~~c~~ul)~~~*I’*J’IJO
                                 DISTRICT OF
                                 COLUMBIA
                                Information on Court-
                                Ordered Tenant
                                Evictions
                                                                      ,   .’




                                                                               llllllll
                                                                                   11111
                                                                                143093
                                                                                      ll


                                                      RELEASED
                         RESTRICTED--Not       to be released outside the
                         General Accounting oface unless spedfically
                         approved by the OfPice of Congressional
                         Relations.
             united
               states
GAO          General Accountiug Office
             Washington, D.C. 20648

             General Goverument Division

             B-242125

             December 28,199O

             The Honorable Stan Parris
             Vice Chairman, Committee on the
               District of Columbia
             House of Representatives

             Dear Mr. Vice Chairman:

             This report responds to your request for information on court ordered
             tenant evictions in the District of Columbia. Specifically, you asked us to
             determine (1) the time that elapsed between the landlord’s filing of an
             eviction request, the court issuance of the eviction order, and the car-
             rying out of the eviction by U.S. deputy marshals for evictions carried
             out in fiscal year 1990; (2) reasons for the elapsed times; and
             (3) whether using contractors to carry out evictions, instead of deputy
             marshals, would be legal and appropriate.


             To evict tenants, landlords must first have a writ of restitution (referred
Background   to as an eviction order) issued by the Landlord and Tenant Branch of
             the District of Columbia Superior Court. The landlord files a complaint
             for possession of real estate with the court, which is to set a hearing
             date for not less than 3 weeks from the date the complaint was filed. In
             the interim, the landlord is responsible for having the tenant served
             with the complaint and notified of the scheduled hearing. If the landlord
             prevails at the hearing or at any trial that might occur, judgment for
             eviction occurs. If not stayed (deferred) pending tenant compliance with
             an agreement to pay the rent due, the eviction order is to be issued and
             sent for execution to the U.S. Marshal for the District of Columbia. The
             marshal is under the supervision of the Attorney General and the
             United States Marshal Service (USMS), an agency within the Department
             of Justice. In performing evictions and other duties for the court, the
             marshal serves in a role typically filled by a sheriff in other
             jurisdictions.

             Upon receipt of the eviction order, the marshal is to stamp on the order
             the period of time (3 to 35 calendar days) during which the eviction
             could be executed, notify the tenant of this, and file the order by the
             applicable District geographic quadrant (e.g., northwest). Evictions are
             scheduled by quadrant, with priority given to the oldest orders. They
             are carried out by deputy marshals, who supervise the movement of
             property from the premises to street curb by laborers provided by the
             landlord. If an eviction is not done within the 35-day time limit and is


             Page 1                                           GAO/GGD91-29   Tenant Evictions
                   B-242125




                   still wanted, the landlord must have the order reissued by the court. A
                   reissued order is referred to as an alias writ.

                   The eviction process is described in more detail in appendix I and illus-
                   trated in appendix II. Overall statistics on eviction orders issued and
                   their disposition during fiscal year 1990 are included and explained in
                   appendix III.


                   The marshal performed 2,966 evictions in the District of Columbia
Results in Brief   during fiscal year 1990. An average of 114 calendar days (about
                   4 months) elapsed from the date landlords requested evictions and when
                   they were done. On average, 62 calendar days elapsed from the land-
                   lord’s complaint to the court issuance of the eviction. There are no
                   overall criteria on how long the court’s part of the eviction process
                   should take. The marshal’s portion of the eviction process averaged
                   52 calendar days. The marshal completed about 54 percent of the evic-
                   tions under the initial order, which has a 35-day time limit; the process
                   took an average of 25 days from issuance of the order to the actual evic-
                   tion. The remaining evictions involved reissued orders, which averaged
                   83 days from the date of the initial order to the actual eviction.

                   Delays in the eviction process occurred for a variety of reasons. In the
                   court’s part of the process, delays occurred because issuance of the evic-
                   tion order was stayed for a period of time because the tenant agreed but
                   subsequently failed to pay the rent due, or the case went through a trial.
                   In the marshal’s part of the process, delays occurred because evictions
                   were not carried out on bad weather days, holidays, and weekends; few
                   deputy marshals were assigned to evictions; or the landlord failed to
                   comply with arrangements agreed to with the marshal (e.g., failed to
                   have agreed-upon number of movers present when deputy marshals
                   arrived at the eviction site).

                   A legal barrier exists to contracting out the marshal’s part of the evic-
                   tion process. The Department of Justice’s 1991 appropriation act pro-
                   hibits any contracting out in fiscal years 1991 and 1992 for law
                   enforcement or litigating purposes without specific congressional
                   approval. If not for this law, USMS officials believe that contracting out
                   for evictions would probably be legal. However, they would not want to
                   do so without the approval of the court and the Department of Justice.




                   Page 2                                          GAO/GGD91-29   Tenant Evictions
                        B-242125




                        USMSofficials maintain that contracting out is not needed given the addi-
                        tional staff they will be able to assign to evictions. On the basis of evic-
                        tion and other workload problems, Congress authorized 24 additional
                        deputy marshals in October 1990 and another 17 are expected to come
                        on board in January, 1991. Consequently, the marshal plans to assign
                        sufficient deputies so that the eviction backlog will be generally limited
                        to eviction orders under 3 days old and orders that have not been car-
                        ried out for uncontrollable reasons, such as bad weather. Officials in
                        USMSand the marshal’s office believe they will be able, with the addi-
                        tional staff, to execute all orders within the initial 35day period except
                        those for which nonservice was the fault of the landlord or other uncon-
                        trollable reasons.


                        To obtain elapsed times, we randomly sampled 277 of the 2,966 evic-
Objectives, Scope,and   tions we determined were done during fiscal year 1990.’We arrived at
Methodology             the total number of evictions by reviewing USMS'daily eviction lists and
                        accounting for all days on which evictions were performed during the
                        year. For each eviction sampled, we reviewed the individual case files
                        maintained by the court and recorded, among other things, the dates the
                        eviction was requested, ordered, and carried out.

                        To determine reasons for the elapsed times, we interviewed court and
                        IJSMSofficials and reviewed court and USMSdocumentation on how the
                        eviction process is supposed to work and how it was carried out during
                        fiscal year 1990. We also determined if the evictions in our sampled
                        cases had involved a trial, a stay because of a tenant’s promise to pay
                        the rent that was due, or the issuance of an alias order. We compared
                        elapsed times for these evictions with evictions not involving these fac-
                        tors. When the information was available, we also recorded why reis-
                        sued orders had not been previously carried out.

                        The decision to limit the file review to evictions performed was based on
                        both availability of data and time constraints, Because cases ending in
                        eviction may be quite different from the larger universe of all com-
                        plaints filed or all eviction orders issued, our sample allows us to gener-
                        alize only about cases resulting in actual evictions. Unless otherwise
                        noted, all sampling errors are less than 9 percent of the estimate, at the
                        95 percent confidence level. In other words, except where noted, the
                        chances are 19 out of 20 that if we had analyzed all 2,966 evictions, the

                        IThe original sample consisted of 301 evictions, but 24 files could not be located, resulting in a sample
                        size of 2’77.



                        Page 3                                                             GAO/GGD-91-29      Tenant Evictions
                                        5242126




                                        results would differ from our sample results by less than the lo-percent
                                        sampling error. (See app. IV for sampling error figures.)

                                        To determine any barriers to having a contractor perform the eviction
                                        functions now performed by the marshal, we requested the views of
                                        USMSand the Superior Court on any legal issues the idea would raise and
                                        on the appropriateness of such an endeavor. We also asked the Office of
                                        Management and Budget (OMB) whether such contracting would be con-
                                        trary to the guidance in OMBCircular A-76, “Performance of Commercial
                                        Activities.” We reviewed the law establishing USMSand proscribing its
                                        duties and responsibilities; USMS'regulations; the Justice Appropriation
                                        Act for 1991 and its legislative history; the Superior Court Rules of Pro-
                                        cedure for the Landlord and Tenant Branch; and the District of
                                        Columbia statutes relating to evictions. We also reviewed the request for
                                        eviction contract proposals USMSissued in 1986 and obtained the views
                                        of USMSofficials on why a contract was never let.

                                        We also interviewed USMSand court officials regarding any plans they
                                        had for shortening the time needed to do evictions, obtained statistics on
                                        the eviction process, and observed deputy marshals carry out two evic-
                                        tions and cancel a third because movers were not present.

                                        We did the audit work from September to December 1990, in accordance
                                        with generally accepted government auditing standards.


                                        We estimate that the 2,966 evictions performed by USMSduring fiscal
Average of About 4                      year 1990 took, on average, 114 calendar days (about 4 months) from
Months Elapsed                          the date the landlord filed a complaint to the day the eviction was done.
Between Landlord                        Table 1 .l shows the mean, median, and range of elapsed times (1) from
                                        complaint to eviction, (2) from complaint to issuance of the eviction
Complaint and                           order (the court’s part of the process), and (3) from eviction order to
Eviction                                eviction (the marshal’s part of the process).

Table 1.1: Elapsed Calendar Days for
Evictions Performed October 1, 1989 -                                              Complaint to            Complaint to              Order to
September 30,199O                       Time
                                        -_--~___                                      eviction                   order               eviction
                                        Mean                                                  114                       6Za                 52
                                        Median ---_~
                                        ._-_.._-              ---                              99                       43                  35
                                        Shortest______--_____                                  37              -._-     21 --__              6
                     v                  Lonaest                                               945                      924                 276
                                        aThe sampling error is + 9 days (14.5 percent of the mean) at the 95-percent confidence level.




                                        Page 4                                                            GAO/GGD-91-29      Tenant Evictions
                      B-242126




                      We did not assessthe reasonableness of the elapsed times for the
                      1990 evictions, given the lack of overall criteria on how long it should
                      take from complaint to eviction. Some perspective on the court’s part of
                      the process, however, is possible through an analysis of the hearing
                      dates. Court procedures provide that the initial hearing is to be sched-
                      uled for a date not less than 3 weeks from the date the landlord files the
                      complaint. We were able to determine the initial hearing date for all but
                      seven of the evictions we sampled. We found that 237 of the 270 evic-
                      tions (88 percent) had the initial hearing after 3 weeks. The average for
                      all 270 evictions was 4 weeks.

                      Concerning the marshal’s part of the process, some perspective on the
                      reasonableness of the elapsed time is possible through an analysis of
                      evictions based on initial and reissued eviction orders. As earlier noted,
                      an eviction order has a time limit of 35 calendar days and must be reis-
                      sued if that period expires and the landlord still wants an eviction. We
                      estimated that 54 percent of the 1990 evictions were carried out under
                      the initial eviction order. These evictions, on average, took 25 days from
                      the date of the eviction order. Those involving eviction orders that had
                      to be reissued took about 83 days, which means the order, on average,
                      had been reissued twice.


                      Our analysis of the steps required in the eviction process, discussions
Factors Lengthening   with court officials and deputy marshals, and review of 277 randomly
Time Neededto         sampled 1990 evictions revealed several reasons why some evictions
Complete Eviction     take longer than others. As far as the court’s part of the eviction process
                      is concerned, stays and trials are major factors affecting elapsed time.
Process               We estimate that stays were involved in about 33 percent of the fiscal
                      year 1990 evictions. For our sample, these cases took, on average,
                      97 days from complaint to the issuance of the eviction order versus an
                      elapsed time of 46 days for the cases not involving a stay.

                      While trials do not occur often, they do significantly affect the elapsed
                      time. Trials were held in 10 of the 270 sampled evictions (about 4 per-
                      cent) where we could determine whether a trial occurred. These
                      10 cases took, on average, 112 days from complaint to the issuance of
                      the eviction order versus an elapsed time of 61 days for the orders not
                      involving a trial.

                      According to a court official, another factor affecting elapsed time in the
                      court’s part of the process is the failure of landlords to take required
                      actions, such as assuring that a complaint is properly served on the


                      Page 5                                          GAO/GGDQl-29   Tenant Evictions
                                        B242126




__.--   ..-- ..__.-
                -----. ----
                                        tenant prior to the initial hearing. However, information on the fre-
                                        quency and impact of this factor was not available.

                                        Concerning the marshal’s part of the eviction process, we identified
                                        three major reasons for delays in executing evictions. First, the number
                                        of evictions depends on the number of days on which evictions are done.
                                        We determined that evictions were performed on 173 days during fiscal
                                        year 1990. Table 1.2 shows the number of days on which evictions were
                                        not done and why.

Table 1.2: Noneviction Days During
Period October 1,1989 - September 30,                                           Number of days
1990                                                                   Evictions not Evictions scheduled          Total nonevict.
                                        Reason                           scheduled           but not done       ____.   ~- days
                                        Weekend                                 105                         .                        105
                                        Holiday
                                        __-- -.-. (non-weekend)
                                                          __-.                   10                         .
                                                                                           -~.___--_----.~--~.-.-              -..-..- 10
                                                                                                                            ..~~
                                        Bad weather                               .                       58                           58
                                        Operational (e.g., training)              2           --~___        4                           6
                                        Unknown
                                        -~-. .- ___-___                          12                         1                          13
                                        Totals                                  129                       63                         192


                                        As shown, evictions are not scheduled for weekends and holidays, and
                                        bad weather is the primary reason for canceling scheduled eviction
                                        days. IJSMSdefines bad weather as a 50-percent chance of rain or an
                                        expected temperature below 25 degrees Fahrenheit. The weather fore-
                                        cast is obtained from the National Weather Service early on the sched-
                                        uled eviction day.

                                        Of the 13 days shown in the table as unknown reasons, 9 occurred in
                                        November and December 1989; this may reflect the marshal’s policy of
                                        not doing evictions around the Thanksgiving and Christmas holidays.
                                        These 2 months had 108 and 98 evictions respectively, compared to the
                                        next lowest month (181) and a monthly average of 247 for all of fiscal
                                        year 1990.

                                        A second major factor affecting the marshal’s part of the eviction pro-
                                        cess is the number of deputy marshals assigned to evictions. Officials
                                        told us that they generally had four deputies (two teams) doing evic-
                                        tions out of an authorized staff of 37 deputies during fiscal year 1990.
                                        They believe that two deputies should be present at each eviction so
                                        that one is inside and the other outside the premises to fully observe all
                                        aspects of the eviction. As to the overall number of deputies assigned to
                                        evictions, we were told that the remaining deputies were needed on


                                        Page 6                                                 GAO/GGD-91-29        Tenant Evictions
                                                                                  __

1

    B-242126




    other duties , such as providing courtroom security and transporting
    prisoners.

    W e cannot determine how much time the deputy marshals spent on evic-
    tions in fiscal year 1990. This information is not separately recorded on
    overall reports or on indiv idual deputies ’ time and attendance records.

    O ffic ials in the marshal’s office told us that on various days during
    fiscal year 1990 there were more than two teams working on evic tions ,
    but they could not tell us how often this occurred. Residential evic tions
    are scheduled to take 1 hour and are generally scheduled to s tart from
    9 a.m. to 3 p.m. Busines s evic tions could be scheduled for longer than
    1 hour. Some evic tions may take longer than the scheduled time, causing
    the deputies to run behind schedule and to work a longer day than
    planned. These fac tors suggest that a team can accomplish up to 7 evic-
    tions a day; 2 teams can do 14. During fiscal year 1990, there were
    59 days when 14 or fewer evic tions were done and 114 days when over
    14 evic tions were done. The average number of evic tions carried out per
    day was 17.

    Various fac tors could cause a team to do more or fewer evic tions than
    seven a day. Sometimes two evic tions are carried out at about the same
    time when, for example, two apartments in the same building are to be
    evic ted. Also, some evic tions take les s than an hour, s ince the tenant has
    already vacated the premises, requiring only a “walk-through” by the
    deputy marshals , or s ince the evic tion involves the removal of a small
    amount of personal property. On the other hand, ins tead of being
    assigned to an actual evic tion, the deputies sometimes perform pre-
    evic tion surveys to determine special arrangements that may be needed
    before scheduling the evic tion. This generally occurs when a busines s is
    involved.

    A third major fac tor affec ting elapsed time in the marshal’s area is the
    landlord’s failure to follow through on agreed-upon arrangements for
    the evic tions . O ffic ials in the marshal’s office told us that landlords
    often are a cause of delay s in executing evic tions . For example, the land-
    lord may not have the required number of movers on hand when the
    deputies arrive at the evic tion scene, despite having previous ly agreed
    on the number of movers to be present. If there are not enough movers,
    the evic tion is not done, and the evic tion order is canceled and returned
    to the court, If the landlord s till wants the evic tion done, the order must
    be reissued.



    Page 7                                            GAO/GGDQl-29   Tenant Evictions
                          B-242126




                          There are other factors that affect how long the marshal takes to per-
                          form evictions. For example, we were told that an increasing number of
                          evictions involve suspected “crack houses.” These evictions entail
                          working out cooperative arrangements with the District’s police
                          department.

                          Although there is no indication that it affected elapsed time in fiscal
                          year 1990, the marshal does have an agreement with the District’s
                          Department of Human Services to do no more than 60 evictions a day.
                          According to USMS, the Department believes that any higher number
                          would overburden its ability to provide emergency shelter for those
                          evicted tenants who need it.


                          The marshal is currently prohibited from contracting out for evictions
Contracting for           without congressional approval. The Department of Justice Appropria-
Evictions Is Prohibited   tion Act for 1991 prohibits any contracting out in fiscal years 1991 and
Without Specific          1992 for law enforcement and litigating activities of USMS as well as
                          other components of the Department of Justice unless the contracting
Congressional             out proposals are specifically approved by an Act of Congress. (Pub. L.
Approval                  No. 101-515, section 212). The conference committee said that:

                          “The enforcement of Federal law is an inherently governmental function which
                          should not be contracted out to private industry. Furthermore, there are concerns
                          that, after a decade in existence, in many instances, contracting out programs have
                          failed to produce both desired management improvements and projected savings.
                          Under the conference agreement, no positions, workyears or associated funding for
                          the law enforcement and litigating components of the Department of Justice associ-
                          ated with contracting out initiatives can be reduced unless specifically approved by
                          the Congress. The conferees agree that it is incumbent upon the Administration to
                          prove that future privatization proposals will produce savings and management
                          improvements before such proposals are submitted to and approved by the Con-
                          gress.” H.R. Rep. No. 909, 1Olst Cong., 2nd Sess. 62, 53 (1990).

                          Besides the aforementioned prohibition, USMS officials told us that con-
                          tracting out would probably be legal but it would be subject to challenge
                          for a number of reasons. Thus, they would not agree to contracting out
                          without first obtaining the approval of the Superior Court and the
                          Department of Justice.

                          Because of the short time frame for doing our work, we have not been
                          able to develop and analyze all the information necessary to determine
                          if, except for the prohibition in the appropriation act, contracting out




                          Page 8                                                GAO/GGDOl-29   Tenant Evictions
    1


        B-242126
I
/




        would be legal or appropriate. OMB’S preliminary v iews are that con-
        tracting out for evic tions along the lines described in USMS’ 1986 request
        for contract proposals would be inappropriate. OMBis concerned about
        (1) the possibility of private persons operating under color of law and
        having to deal with v iolent persons on the premises and (2) the extent to
        which the judgment of these persons may limit or eliminate appropriate
        government discretion. The Superior Court informed us that it would be
        inappropriate for the court to take a position on the contracting out
        issue.

        W hile at one time they favored the idea of having contractors do evic-
        tions , USMSoffic ials now believe that there is no need to contract out. On
        the basis of evic tion and other workload problems, additional deputy
        marshals are being assigned to the Dis tric t. In O c tober 1990, 24 new
        deputies completed training and became operational; another 17 are
        expected to become operational in January 199 1.

        Consequently, the marshal plans to assign sufficient deputies so that the
        evic tion backlog will be generally limited to evic tion orders under 3 days
        old and orders that have not been carried out for uncontrollable reasons,
        such as bad weather. O ffic ials in USMSand the marshal’s office believe
        they will be able, with the additional s taff, to execute all orders within
        the initial 35-day period except those for which nonservice was the fault
        of the landlord or for other uncontrollable reasons. They believe this I
        will be achieved in January when they expect to have 8 to 10 teams
        carrying out evic tions .


        W e discussed the information in this report with the head of the Court’s
        Landlord and Tenant Branch, the Marshal and the Chief Deputy Mar-
        shal for the Dis tric t of Columbia, and USMS'general counsel, who gener-
        ally agreed with the fac ts presented.

        As arranged with the committee, we plan no further dis tribution of this
        report until 30 days after the date of this letter, unles s you or your suc-
        cessor as Vice Chairman public ly announce its contents earlier. As that
        time, we will send copies to the Superior Court, the Dis tric t Marshal,
        IJSMS,and other interes ted parties .




        Page 9                                           GAO/GGDOl-29   Tenant Evictions
---~..-__..
              B-242126




              Major contributors to this report are listed in appendix V. If you have
              any questions about this report, please call me on 275-8389.

              Sincerely yours,



              L-u Dbyfz-
              Lowell Dodge
              Director, Administration
                of Justice Issues




              Page 10                                         GAO/GGD-91-29 Tenant Evictions
Page 11   GAO/GGD-91-29   Tenant Evictions
                                                                                                 .
Contents


Letter
Appendix I
The Eviction Process
Appendix II                                                                                          16
The Eviction Process
Illustrated
Appendix III                                                                                         18
Eviction Orders
Subject to Service and
Their Disposition
During Fiscal Year
1990
Appendix IV                                                                                          19
Sampling Errors for
Estimates Used in This
Report
Appendix V                                                                                           20
Major Contributors to
This Report
Tables                   Table 1.1: Elapsed Calendar Days for Evictions                                   4
                             Performed October 1,1989 - September 30,199O
                         Table 1.2: Noneviction Days During Period October 1,                             6
                              1989 - September 30,199O
                         Table IV. 1: Confidence Intervals for Duration Estimates                    19
                             (in Days)
                         Table IV.2: Confidence Intervals for Population Estimates                   19
                             for Number of Cases With Conditions Met
                             (Population=2,966)



                         Page 12                                       GAO/GGD-91-29   Tenant Evictions
          Contents




Figures   Figure II. 1: The Court’s Part of the Eviction Process                     16
          Figure 11.2:The Marshal’s Part of the Eviction Process                     17




          Abbreviations

          OMB        Office of Management and Budget
          USMS       United States Marshal Service


          Page 13                                       GAO/GGDOl-29   Tenant Evictions
Annendix I

The Eviction Process


              In general, the eviction process begins with the landlord or the land-
              lord’s representative filing a complaint for possession of real estate.
              (According to a court official, about 69,000 complaints were filed during
              fiscal year 1990.) Court officials will assign a case number to the com-
              plaint and set a hearing date for not less than 3 weeks from the com-
              plaint date. In the interim, the landlord is responsible for having the
              tenant served with the complaint at least 7 full days before the hearing
              date (excluding Sundays and legal holidays). An affidavit of service
              must be filed with the court at least 4 days before the hearing.

              While some hearings result in the judge granting a landlord or tenant
              request for a trial, most result in either an agreement reached between
              the landlord and the tenant or a default judgment. On the day of the
              hearing, the landlord and tenant may reach a settlement through negoti-
              ation. In a settlement, the tenant often agrees to pay rent due. When
              such an agreement is reached, the court enters judgment for the landlord
              but defers (stays) issuing the eviction order as long as the tenant com-
              plies with the agreement. If the tenant fails to comply, the landlord may
              request the court to issue the eviction order. The tenant is to be notified
              of the landlord’s intent to do so.

              A default judgment is entered when the tenant fails to appear for the
              hearing. The eviction order can be issued 2 days after the judgment,
              giving the tenant time to appeal.

              The U.S. Marshal for the District of Columbia is responsible for sched-
              uling and executing all eviction orders issued by the District’s Superior
              Court, Essentially, the marshal serves in the role typically served by a
              sheriff in other jurisdictions, The marshal is under the general supervi-
              sion of the Attorney General and LJSMS.

              As the system is designed, the marshal receives eviction orders for the
              court twice daily, stamps the period of time for service on the top and
              bottom parts of the order, and mails the bottom part to the tenant. Evic-
              tion orders can be served from 3 days (72 hours) up to 35 calendar days
              from the date of the order. This time span includes weekends and holi-
              days, with one exception: The final day cannot be on a weekend or hol-
              iday. If the order is not served or otherwise canceled during that period,
              it expires and the landlord must have it reissued if the eviction is still
              wanted. The reissued orders are also good for 35 calendar days. The
              tenant does not have to be advised of the landlord’s intent to have the
              order reissued unless 90 days have elapsed since the date of the judg-
              ment or the date any stay was lifted.


              Page 14                                         GAO/GGD-91-29 Tenant Evictions


                                                I   .   .
Appendix 1
The Evktlon Procew




The eviction orders are scheduled for service by geographic quadrant
(e.g., northwest portion of the District), with the oldest orders being
done first. Each eviction involves two deputy federal marshals who
oversee movers arranged and paid for by the landlord. The landlord is to
arrange for a sufficient number of movers so that the eviction can be
completed in 1 hour. The marshal and the landlord determine the
number of movers needed.




Page 15                                       GAO/GGD-91-29   Tenant Evictions
Appendix II

The Eviction ProcessIllustrated


Figure 11.1:The Court’s Part of the Eviction Process




                             Landlord and Tenant
                             Branch Hearing                Landlord and Tenant
                                                       D      Branch of D.C.
                                          a
                                     Landlord Files
                                                              Superb Court


                                     Complalnt

                                                                     t
                                                                    @
                                 L
                                                           Special Process
                              Landlord                     Server Files Affidavit    Tenant
                                                           of Service With
                                                           Landlord and Tenant
                                                           Branch


                                           0
                                     Landlord Hires
                                     Special Process
                                     Server
                                                       D ,&*l.     Server           Special Process
                                                                                    Server Serves
                                                                                    Summons and
                                                                                    Complaint on
                                                                                    Tenant




                                               Page 16                                        GAO/GGD91-29   Tenant Evictions
                                            Appendix II
                                            The Eviction Process Illuetrahd




Flgure 11.2:The Marshal’s Part of the Eviction Process




               Landlord and
              Tenant Branch

                                      Marshals Service
                                      Contacts Landlord
                                      and Schedules
                                      Evlctlon Date and
     Landlord and                     TkW
     Tenant Branch                                                                                            @
     Issues EZvlctlon                                                                                   Landlord Hires
     Order                                                                                              Movers


                                      Malls 72-Hour
                                    % Evictlon Notlce
                United States         to Tenant                I                          I
               Marshals Service                                       Tenant




                                                         Deputy Marshals, Landlord, and
                                                         Movers Meet at Scheduled Time
                                                         and Date to Evict Tenant




                                             Page 17                                          GAO/GGD91-29   Tenant Evictions
Appendix III

Ekiction OrdersSubjectto Serviceand Their ’
DispositionDuring l?iscaIYew 1990’

                                                                                                        .-.

               On hand at beginning of fiscal year                                                                  1,392
               Received durina the year                                                                            20,289
               Available for disposition                                                                           21,681
               Returned to the court during the year
                 Evictions oerformed                                                            2,966
                 Cancelled after beina scheduled for execution by USMS’                         3,116
                 Cancelled by landlords before being scheduled for executionb                   5,514
                 Expired-not executed within 35 days of issuanceC                               8,650              20,246
               On hand at end of fiscal vear                                                                        1,435

               *Reasons for cancellation include incorrect names or addresses on the eviction orders, payment of rent
               by tenants, and failure of landlords and/or movers to be at eviction sites at scheduled times.
               bNormally results when tenants pay rent
               %easons include bad weather limiting the number of days on which evictions can be performed, depu-
               ties normally working evictions being assigned to higher priority duties, too few deputies being
               assigned to evictions, and difficulty in scheduling dates and times with landlords for performrng
               evictions.




               ‘Numbers include both initial and reissued (alias) eviction orders.



               Page 18                                                               GAO/GGD-91-29 Tenant Evictions
           .
  Ppe

&&&ng Errors for EstimatesUsed in
This Report

Table IV.l: Confidence Intervals for
Duration Estimates (In Days)a                                                                        Lower                              Upper
                                                                                                     bound          Estimate            bound
                                       -.Full Sample (277 cases)
                                           Complaint to first eviction order                              53              62                71
                                          First order to eviction                                         47              52                57
                                          Complaint to eviction
                                       -________                                                         104             114               124
                                          Complaint to hearing
                                       _...~I__                                                           27              28            -- 29
                                       Cases where first order was executed (146
                                          cases)                                                                                   --.__
                                          Order to eviction
                                       --__---                                                            23              25                27
                                                                                                                                            -~
                                       Cases with alias orders
                                          (126 cases)
                                                 -.---.-       --
                                          Order
                                       -~-.__--. to  eviction              ____--                         76              83
                                                                                                                          __-__.            90

                                       Cases with stays (88 cases)
                                         Complaint to order                                               70              97             --124

                                       Cases with no stays
                                       --             ___- (182 cases)                                                             --
                                          Complaint to order                                              42              46                50

                                       Cases  with trials (10 cases)
                                       -._______-
                                          Complaint to order
                                                           ----. - ___..                                  84             112 ___ __..- 140

                                       Cases
                                       .---- with no trials (260 cases)
                                          Complaint to order                                              51              61                71
                                       BTheconfidence interval is the estimate plus or minus the sampling error. Some numbers are based on
                                       less than the complete number of cases cited due to missing data.

Table IV.2: Confidence Intervals for
Population Estimates for Number of                                                                   Lower                              Upper
Cases With Conditions Met                                                                            bound          Estimate            bound
(Population=2,966)@                    Initial order executed
                                       __..__                                                          1,398            1,563 -          1,728
                                       Trials                                                             45              107              169
                                       Stays                                                             788              942            1,096
                                       Initial hearing more than 3 weeks after
                                          comolaint                                                    2.430           2.538             2.646
                                       aThe confidence interval is the estimate plus or minus the sampling error.




                                       Page 19                                                            GAO/GGDBl-29     Tenant Evictions
Appendix V

Major Contributorsto This Report


General Government        Issues
Division, Washington,   Carl T. Trisler, Evaluator-in-Charge
DC.                     Mary B. Hall, Evaluator
                        Nelson S. Payne, Jr., Evaluator
                        Barry Jay Seltser, Social Science Analyst


                        Nancy Finley, Senior Attorney
Office of General
Counsel




(Isless)                Page 20                                     GAO/GGD91-29   Tenant Evictions
                             *

!’   ..-“..   _.I “.“l._   ..” .“,..-_-.~-   ..-._l _-___   --.-~_---__
                                                                                               ,I
                                                                                               ’‘f
   --    ..-_           --..-   -..._   - .-.-   -..--   .-.....-.--.   -   ---..   -.-_-.--




-.I---          -.---