oversight

Action Needed to Improve Case Processing Time at National Labor Relations Board Headquarters

Published by the Government Accountability Office on 1990-10-03.

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

                     United States General Accounting Of&e              /4/23      )cB;?:
                     Testimony


                                                                        I Ill
                                                                           142362

For Release          Action    Needed     to    Improve        Case
on Delivery          Processing      Time      at   National
Expected    at
9:JO a.m.     EST    Labor    Relations        Board    Headquarters
Wednesday
October    3, 1990




                     Statement  of
                     Franklin  Frazier,    Director               of
                     Education  and Employment                 Issues
                     Human Resources    Division

                     Before    the
                     Subcommittee     on Employment     and Housing
                     Committee     on Government    Operations
                     House of Representatives




GAO/T-HRD-91-1                                                                  GAO Form 160 W/87)
                 SUMMARYOF TESTIMONY BY FRANKLIN FRAZIER
            ON  ACTION NEEDED TO IMPROVE CASE PROCESSING TIWE
              AT NATIONAL LABOR RELATIONS BOARD HEADQUARTERS

The National      Labor Relations       Act of 1935, as amended, provides       the
basic framework qoverninq          labor-management        relations  in the
private   sector.      The act created        the National     Labor Relations
Board to administer        and enforce      the act.     The Board decides two
kinds of cases:       (1) allegations       of unfair    labor practices     by
employers     OK unions (unfair       labor     practice   cases) and (2)
disaqreements      about elections       to determine      whether employees wish
to be represented        by a union     (representation       cases).
All cases oriqinate           in the Board's      regional      offices,     and most
cases are resolved          there.     If a party      in the case contests            the
reqional       decision,    the case comes to the five-member                Board at
headquarters         for review and decision.            Because of concern about
the lenqth       of time the headquarters          Board was taking          to decide
some    cases,     we were asked to examine the decisionmaking                    process.
We focused primarily            on decisionmaking        in fiscal      years 1984-89.
Our   objectives        were to (1) assess how lonq the Board took to decide
cases and whether delays were excessive;                     (2) identify      factors     that
contributed        to delays;     and (3) identify         administrative        or
legislative        action   that miqht be warranted.
Increase    in median case processing    times after    regional  action.
The median times for the five-member        Board to decide cases during
the 1984-89 period,      after  regional action    was completed,  were
amonq the highest     in Board history.     Median times for unfair       labor
practice    cases in these years were two to three times higher           than
durinq   the 1970s.     Medians for representation     case were also higher
than during     the 1970s.
Some cases were at headquarters   over 2 years. About                      17 percent--
over 800 --of the cases appealed to the Board during                       the 6-year
period  took over 2 years to decide.
Three factors     primarily     contributed     to headquarters    delays.
First,    the Board lacks standards          and procedures    to prevent
excessive    delays.      Second, lack of timely       decisions   on lead cases
have delayed     related    cases.     Third,   Board member turnover      and
vacancies    also caused delays.
Board corrective       action    needed.      We recommend that the Board
establish    standards      for (1) the total       lenqth of time a case should
be at the Board and (2) a time            for each decision     stage that,   when
exceeded,    requires     corrective     action.     We also KeCOmmend that it
specify   the actions       to be taken when those standards         are exceeded.
Legislative      options  available.       Congress could amend the act to
allow Board members whose terms are ending to either             (1) remain
until*Keplacements       are confirmed      OK  (2) continue for a limited
period     while a replacement       is being sought.
     Mr.   Chairman   and Members of      the      Subcommittee:
 I am pleased to be here today to discuss               the lenqth of time the
five-member     headquarters    National       Labor Relations       Board (NLRB)
takes to decide some cases         appealed      to it from the regions.          As you
requested,    I will    (1) describe       how long the Board has taken to
decide cases, particularly         since 1984, and whether there were
excessive    delays;    (2) discuss      factors    that contributed       to such
delays;    and (3) suqgest administrative             OK legislative      action  that
miqht be warranted.
Cases generally      reach the headquarters         Board when parties           contest
decisions     made by NLRB reqional      offices      or administrative          law
judqes    (ALJs).    Cases involve    either      (1) allegations         of unfair
labor   practices    by employers OK unions          (unfair      labor   practice
cases) OK (2) disagreements        about     elections       to determine       whether
employees wish to be represented           by a union         (representation       cases).

Our major        points   are   the   following:
--         During fiscal        years 1984-89,  the five-member    Board's   annual
           median times to decide cases, after           regional  action   was
           completed,      were among the highest      in Board history.     Annual
           unfair    labor    practice   case medians ranged from 273 to 395
           days --from     two to three times higher      than during    the 1970s.
           Representation        case  medians ranged from 190 to 256 days--
           also higher       than medians during    the 1970s.
--         About 17 percent   (820) of the            cases appealed    to the
           headquarters  Board during  the            6-year period    took   more   than   2
           years to decide.
WV
           Factors    contributing     to the delays      in deciding     cases were (1)
           lack    of standards     and procedures     to prevent     excessive     case
           delays;      (2) lack of timely      decisions    on lead cases, which
           delay related       cases; and (3) Board member turnover             and
           vacancies.
--         Although    timeliness    improved in 1989, we believe   additional
           Board actions      are needed to further   improve  case processing
           timeliness.       Also, the Congress may wish to consider      amending
           the National      Labor Relations  Act (NLRA) to provide    for greater
           continuity     of Board members.
Before       discussing   these points in detail,              I would like to provide
some       background   on our study and describe              our methodology.


The National     Labor Relations     Act of 1935, as amended, provides        the
basic framework governing        labor-management      relations    in the
privatie sector.      The act created     the NLRB to administer       and
enforce   the act.     Its two statutory      missions    are (1) to prevent
and remedy unfair      labor practices      by employers     or unions and (2)
to conduct elections       to determine     whether   employees wish to be
represented    by a union.    The agency's  functions     are divided
between its General Counsel,      which has responsibility       for
activities    of the 34 reglonal   offices,   and a five-member      Board.
The President    appoints  Board members for S-year terms with the
consent of the Senate.
When a Board member's term ends, that position            is vacant until      a
replacement      has been nominated    and confirmed.     In contrast,     some
other federal       agencies operate   under laws that allow a member
whose term is ending to either          (1) remain for a fixed period       of
time or (2) continue        until a replacement     has been confirmed.       Such
agencies    include    the Equal Employment Opportunity        Commission,   the
Federal   Communications      Commission,   the Federal   Trade Commission,
the Interstate       Commerce Commission,     and the Securities    and Exchange
Commission.
All cases originate         in the regions,      and most cases are resolved
there within        1 year or less.     In fiscal      year 1988, most unfair
labor practice        cases were resolved      in the regions       without  formal
litigation,       and half of those were disposed           of informally    in 50
days OK less.         The median time to obtain         a decision     when unfair
labor practice        cases were litigated       before an ALJ was about a
year.       Representation    cases were generally         resolved    more quickly
than unfair       labor practice    cases.
If the regional       decision      is contested,       the case goes for review to
the headquarters        five-member       Board, which affirms,       modifies,  or
reverses      the regional      decision.   1 In 1989, fewer than 5 percent         of
all cases were decided at the headquarters                   Board; about 75 percent
of these were unfair          labor practice       cases.      The number of cases
decided     by the Board has declined           steadily     since 1980, roughly
paralleling      a decline      in cases originating         in the regions.    In
fiscal     year 1989, 874 cases were assigned              to Board members,
compared with 1,875 in 1980.
Most cases are decided at the Board by three-member                panels rather
than the full    five-member    Board.    After   assignment     to a Board
member, cases proceed through        three stages --analysis        and research
 (Stage 11, drafting     (Stage II),    and circulation      (Stage III)--
before   the Board issues its decisions.           The Board has established
expected    time targets   for progress    of cases through these stages
and procedures     for tracking   case progress.
Board decisions      on unfair    labor practice       cases, but generally    not
representation      cases, can be appealed        to   circuit   courts  and may,
in turn,     be appealed to the Supreme Court.              The number of cases
appealed     to the circuit    courts   in fiscal      year 1989 was about 13
percent    of all Board decisions       issued in      that year.


1Two kinds of cases also come from the regions      for Board review
without  first having regional    hearings: Motions for Summary
Judgment and Stipulations    to the Board.
                                          2
SCOPE AND F'IETHODOLOGY
In conducting     our review,   we COnCentKated      on activities    that take
place at Board headquarters        from the time the Executive        Secretary
assigns     a case to a Board member to the time the Board issues its
decision.      However, for perspective,       we also obtained    selected
information     about NLRB regional    activity,     such as the number of
cases that are resolved       in the regions.
We collected     information      from   the   following    sources:
--    prior    studies    and published        literature,      including     NLRB's
      annual appropriation         justifications;
--    current     and former Board members and staff                attorneys    (using   a
      combination      of interviews        and a questionnaire);          and
--    NLRB files,      including     computerized          data from its management
      information      system.
In describing     delays,    we focused on 1984-1989 because the Board
said its computer data base was incomplete            for earlier   years.     We
included     in our analyses     only those cases that came to the
headquarters     Board after     one of the parties    appealed the decision
of an NLRB regional       office    OK administrative    law judge (that   is,
we excluded     summary judgments and stipulations          to the Board).
To help identify     factors    contributing       to delays,     we analyzed   (1)
20 cases selected      judgmentally      to illustrate       delays in
decisionmaking    and (2) 90 cases randomly selected               from those with
the longest    and the shortest       processing       times during    the most
recent   complete  fiscal    year (1989).
STUDY RESULTS
Excessive  Delays     in
Case Processinp
In the 6-year period,      1984 through  1989, the headquarte,rs   Board
decided   about two-thirds    of the 5,000 cases appealed to it within
1 year from the date the case was assigned        to a Board member, but
10 percent    of the cases took over 3 years to decide.        A few cases
took more than 7 years to decide.
We found 1984-89 headquarters         case processing    times to be
excessive     when compared with two criteria        for excessive  delays:
 (1) median processing       times prior  to 1984 and (2) a "more-than-2-
years"    timeframe     that former Board Chairmen identified      as an
unreasonable      length of time to decide any case.
Median case processing
tire8 baoe increased
In   1984-89,     the median times from the date regional           action  was
completed     to the date the Board decision       was issued were among the
highest     in Board history,      as shown in the first      figure.2     (By
regional     action     we mean either  an ALJ decision     or a regional
hearing.)       The medians for unfair      labor practice      cases ranged from
a low of 273 days to a high of 395 days --between               two and three
times higher        than medians during   the 1970s.     Only one previous       year
had a median as high (324 days in 1983).             For representation        cases,
the medians ranged from 190 to 256 days --also             higher     than medians
during    the 1970s.



GAO Median Time to Decide
    Contested Cases (1960-89)
           400
           350
           300
           250
           200
            150
           100




                  Fiscal Years
                  -           Unfair Labor Practice Cases
                  - - -   l   Representation Cases



2This Includes      up to 60 days for parties    to file    documents
regarding   their    appeal of the regional   action.      The Board does
not publish    historical   data about median processing        trmes from
 (1) the date all documents are received       and the case is assigned
to a Board member to (2) the date the decision           is issued.
                                                 4
Sow cases took over
2 years to decide
According     to the previous      Board Chairmen we interviewed,             all cases
should be decided within         2 years.       However, during the 1984-1989
period,   the Board took more than 2 years to decide about 17 percent
of its cases.       Of its unfair      labor    practice     cases, 752 (19 percent)
took over 2 years to decide;           of its representation          cases, 71 (7
percent)    took over 2 years.         The percent       of unfair    labor   practice
cases taking     longer than 2 years ranged from 8 percent                  in fiscal
1984 to 30 percent       in fiscal     1988, as the figure         below shows.
Representation      cases taking     over 2 years ranged from 2 percent                in
fiscal   year 1984 to 13 percent           in fiscal     year 1987.




GAO Percent of Contested Cases Taking
    Over Two Years to Decide (1984-89)
           35 Percent
           30
           25
           20




                                                        u
                    1984       1985   1988       1987    1988      1989
                Fiscal Years
                0     Unfair Labor Practice Cases
                      Representation Cases




                                             5
c



    Timeliness     improved in     1989
    but further     improvem ent    is needed
    The tim eliness    of Board decisionm aking       improved in 1989.       Both the
    m edian num ber of days to decide cases and the num ber of cases at
    the Board for over 2 years were lower in fiscal                year 1989 than in
    1988.    In addition,    the num ber of cases undecided          at the end of
    1989 (437) was lower than in any previous             year of the decade.        The
    Board Chairm an attributed      this    improvem ent,     at least  in part,   to
    actions    taken at the Board to expedite         certain    kinds of cases and
    focus on deciding     the oldest     cases.
    Nevertheless,      in fiscal    year 1989, the m edian time   to decide
    unfair    labor practice     cases (300 days) was still     substantially
    hiqher   than at the start       of the decade (133 days in 1980), and               21
    percent     of the unfair    labor practice   cases decided had been at
    headquarters      over 2 years.
    Three Factors   Prim arily
    Contributed   to Delays
    Our analysis    of the Board's    system for deciding     cases identified
    three factors     that, we believe,    prim arily explain   why there have
    been decisionm akinq    delays.
    Lack of standards     and procedures
    to prevent  excessive    delays
    The Board has no standard            for (1) the total       length of tim e it
    considers      acceptable      for a contested     case to be at the Board or
     (2) the lenqth       of tim e a case can rem ain in each decision          stage
    before    corrective      action   is required.        In the absence of such
    standards,      its m onitorinq      procedures    do not require    Board m embers
    or their     staffs    to focus proactively        on cases most likely     to show
    excessive      delays unless corrective         action    is taken.
    The Board does m onitor       the proqress     of cases through     each stage,
    but it prim arily     uses as targets      the estim ated   tim es in which
    cases are likely      to m ove through     each stage.     For typical    cases,
    the Board's     tim e targets    are 3 weeks in stage I, 3 weeks in stage
    11, and 2 weeks in stage III.            In com bination,   these targets       total
    about 2 m onths --a length       of tim e with lim ited    value in focusing
    attention     on cases m ost in need of corrective         action.     In fact,       in
    fiscal    year 1989, about 90 percent         of all cases were at the Board
    longer    than 2 m onths before     a decision     was issued.
    Lack of tim ely    decisions    on “lead”
    cases delays    related    cases
    When several    undecided    cases deal with the sam e issue,        the Board
    selects  one case to serve as the principal           or "lead"   case and
    suspends  further    processing    on all related     cases until    the lead
    case is decided.       The lack of tim ely   decisions     on lead cases
    delays all related      cases.
                                               6
Our analysis     of     a random sample of 45 of the lengthiest         cases
decided during        fiscal   year 1989 disclosed      that delays attributable
to waitinq   for      lead case decisions     was a major factor     in at least
13 (29 percent)         of those cases.    In contrast,      none of the 45
shortest   cases      had been delayed for a lead case.
Our review of 20 other judgmentally-selected              lengthy   cases also
determined      that lead cases had caused delays in 13 of the 20
cases.     Some cases were delayed more than once; four cases had
been delayed       three different     times during their     processing  by
different     lead cases.       Almost all of the lead cases that delayed
these 13 cases either         took more than 2 years to decide or had been
pendinq    for more than 2 years at the end of fiscal             year 1989.
Board member turnover
and vacancies
Board member turnover           and Board vacancies     during  the 1980-84
period    contributed     to    a backlog of pending cases at the start      of
the 1984-89 period.3            In addition,  Board member turnover      and
vacancies     durinq  the      6-year period  continued     to affect case
processinq      even after      the backlog was no longer a problem and
fewer new cases were           coming to the Board.
Board member turnover         during    the 1980-84 period was the highest            in
aqency history:    the Board had as many new members (six) during
that time as it had during           the entire     1970-79 decade, and also
more than durinq     the 1960-69 decade.            Five very experienced       Board
members --with  over 60 years cumulative             experience   as members--
were replaced   during      fiscal     years 1980-83.       One newly-appointed
member served less than 17 months, another                served less than 3
months.    TUKnOVeK continued         during    the 1985-89 period when six new
members replaced     Others      who were appointed       during  the 1980-84
period.
Turnover     contributes    to delayed decisions   in several    ways.    The
Board must add departing        members' undecided   cases to remaining
members' caseloads,       and new members require    time to hire senior
staff  and become familiar       with the issues in cases they inherit.
In addition,       some cases get sent back to earlier     decision    stages
because new Board members disaqree         with the previous    decision.
DUKinq both 5-year periods    (1980-84 and 1985-89),     the Board had
fewer than five members for a total     of 3 out of 5 years.
Likewise,  each S-year period   had a total   of 8 months in which two
of the five positions  were vacant.     Vacancies  inctease   the workload
for other members and cause some cases --major     cases the Board
believes  should have all five members voting--to      be delayed.

3Cases pendinq at fiscal year end increased               from less than 500 in
1980 to over 1,300 in 1983 and 1984--almost               thtee times as many
as were pending 10 years earlier,  in 1973.
                                           7
CONCLUSIONS
During the 1984-89 period,             NLRB headquarters'        median case
processing    times    were    among the highest        in Board history,         and the
Board took more than 2 years to decide 17 percent                      (823 cases) of
the cases appealed to it from the regions.                     Factors that
contributed     to delays      included      (1) lack of standards         and
procedures    to prevent       excessive      delays,    (2) lack of timely
decisions   on lead cases, and (3) Board member turnover                      and
vacancies.      Such factors        indicate     the need for a more proactive
case management approach to systematically                   identify     cases most in
need of corrective        action.        Even though timeliness         improved in
1989, reduced delays were probably                 due not only to specific          actions
of the Board, such as focusing               on the oldest      cases, but also to
reduction   in the number of cases coming to the Board.                       We believe
additional    Board actions        are needed to further           improve case
processing    timeliness       and reduce inordinate          delays in deciding
cases at the five-member           Board.
RECOHMENDATIONS
We recommend that        the   Chairman     of the National       Labor     Relations
Board
--     establish    standards  for (1) the total     length of time a case
       should be at the Board and (2) a time for each decision         stage
       that,    when exceeded,  requires  corrective     action and
--    specify    the corrective        actions    that should be taken when
      those standards       are exceeded.         Such action     might include        more
      extensive     involvement      of Board members during           the first      two
      decision    stages or more frequent             use of the existing        policy
      option    of issuing     decisions      without     waiting   for untimely
      written    dissents    during     the final      decision   stage.
LEGISLATIVE     OPTION
To help reduce the problem of Board member turnover               and vacancies,
Congress may wish to provide        for more continuity     of members.
Congress could amend the NLRA to include         provisions       similar    to
those applicable    to other agencies,     such as the Equal Employment
Opportunity    Commission,    that would allow Board members whose terms
are ending to either       (1) stay at the Board until      their     replacement
has been confirmed     or (2) continue    for a limited     period while        a
replacement   is being sought.


Mr. Chairman, this concludes  my statement.                   I will      be glad   to
answer any questions  you may have.



                                             8