oversight

Foreign Investment: Difficulties in Addressing Public Policy Concerns

Published by the Government Accountability Office on 1990-07-19.

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

For  Release on       FOREIGN INVSSTME3T:      Difficulties         in    Addressing
Dellvery   Expected   Public i'ol Icy Concerns
at ;O:Xl a.m. EL'7
Thursday
July   i?, 1392




                                                                                       p


                      Ztaiement     cf
                      Allan     I. Mendelowitz
                      Iirector,     International      Trade,   Energy,
                      and Tinancz       Issues
                      Zefcre     the
                      Zcnr,ittee     i=~: Csmmerco,   Sciexe,    and
                      Yri
                      ..  r.~aortaiian
                      0.5 * ienate




GAO/T-NSIAD-90-54
Mr. Chairman                   and      Members           of the Committee:


I    am pleased                to testify             today     before           this     Committee             on the         subject
of    foreign            direct             investment         in the United                   States,


AS you know, recent                            increases        in       foreign         investment             have raised
concerns            about            the     consequences            of     foreign            ownership         of U.S.             assets
and about               the        adequacy          of government               statistics          on forelqn
investment.                    It      is    understandable               that      the public            wants       to know
more       about         foreign             investment,          since          Eoreign         ownership          inherently
means that               key business                 decisions           can be made overseas                      on such
matters            as    research,              investment,              employment,             and location.


In    our     work            on     this      subject,        we have tried                   to identify          specific
concerns            about            foreign         investment           and the types              of    data      available
or needed to analyze                            these       concerns.              In general,            we believe
improvements                   in the          available        foreign           investment             data     would         benefit
i.-' lit      understanding                     of    an    economic         trend        likely         to continue                 over
the next            few years.                  We also        believe           that     if     GAO had         access         to

certain            types           of foreign             investment         data        that      are now closely
restricted,                this         access        would     allow        us to better                respond          to
congressional                       interest         in    improving         public            understanding              of    foreign
investment.


We have found that                           some questions               cannot         be answered             because             of the
limitations                of the government's                       foreign            investment         data.           The

                                                                     1
general       problems         with      the data        system         of   the     Bureau         of     Economic

Analysis       include         timeliness,             consolidated           reporting,             classification
methodoloqy,           and access          to data.               We have also             found      some       questions
whose answers           require         either         policy        judgments        about         the      future
behavior       of     foreign         investors         or       different         kinds        of data        than
foreign       investment         data.


PURPOSES AND LIMfTATfONS                     OF GOVERNMENTDATA


Let    me first        take     a moment          to    summarize the purposes                       and
limitations           of the      government's               databases         on foreign             direct
investment          in the      United       States.l


Cumulative          foreign      direct          investment            in the United               States         is
measured        in two different                 ways by the Commerce Department's                                     Bureau
of     Economic Analysis                (BEA),      which         has primary          responsibility                   for
collecting          such data.            The first              method measures                net foreign
capital       flows     into      the     United        States         and shows the               total
accumulation           of     foreign      direct            investment        over        time.          This     is the
Foreign       Direct        Investment           Position           and Balance            of    Payments Flows
data      series.       The second           method measures the total                             value         of a U.S.
affiliate's           assets      in     which     a foreign            investor           has an equity                share




1For a more complete discussion,     see Foreign                                     Investment:    Federal
Data Collection   on Foreign  Investment  in the                                     United States,
                                                                                                 -
{GAO/NSIAD-90-25BR,   October 3, 1989).

                                                             2
of at      least          10 percent.           This      is the Operations                of U.S.         Affiliates
of Foreign           Companies          data      series.


In addition,               another      BEA data          set,     U.S.      Business           Enterprises
Acquired         or Established                by Foreign          Direct      Investors.              measures

annual      transaction              values       for     new acquisitions                of existing               U.S.
companies          and for          start-up       investments.               Finally,           the    Benchmark
Survey      of      Foreign         Direct      Investment          in the United                States,

comprising           all     three      data      series,         provides      more detailed
information               every     5 years.


Usefulness           and Limitations


Each of these               data      series      is useful         in capturing                a different              aspect
of the      total          investment          picture:          however,      each also              has limitations
inherent          in its          methodology.


The following               hypothetical           case      is a simplified               example            of    how the
three      data      series         capture       a different          aspect        of    foreign         direct
investment           transactions.                Let     us suppose          that     ABC Company of the
United      Kingdan          makes an initial                investment          in the XYZ Company in
the     United       States,           ABC Company pays $100 million                             to    buy a SO-
percent      share          of the      XYZ Company, which                  has a total               worth        of    $200
million,            Its     transaction           is    structured           so that       it     raises           $50
million       in the         United       Kingdom,          and another          $50 million               is borrowed
in the United               States      through         a U.S. affiliate.                  The balance                  of

                                                             3
payments        data       series            would      report           the           $50 million          capital             inflow
from     the    United           Kingdom;         the       enterprises                   acquired          or established
data     series         would         report      the $100 million                            investment           outlay;             and
the affiliates              data         series        would            report           the $200 million                      company as
a foreign         affiliate.


While        the BEA data              series         help     to piece                  together          the     foreign
investment         picture,              there        are     important                  gaps and disparities                          in
available         data.            For       instance,         the         balance             of payments               data      provide
information             about         the     amount of            foreiqn               funds      that       come into               the
United        States,           but    they      are not designed                         either         (1)     to capture                 the
amount of the              total            investment         transaction                     if   funds        from U.S.
sources        are used or (2)                   to     identify             the         ultimate          beneficial              owner.
These data         also          are based on               book         value,           the value             of the
investment         when it             was initially                    made.


The data        on enterprises                   acquire5               or established                   provide          information
about        the total           investment             outlay           or transaction                    cost,         but     only            for
new investments                  in that         given        year,               If     an existing             investment                 is
expanded         in the          following            year,        it      would          not be captured.                       This
data     series      also         classifies             investments                     under the          primary             line        of
business        of the           new enterprise,                   unlike               the balance             of payments                  or
affiliates         data          series,         which        classify                  investments             under the              line
of     business         of the         foreign          investor.                      This    classif         ication
difference         means that                 the data         on enterprises                       acquired             or



                                                                   4
established             cannot        easily        be compared              to the      other       data series              on a
sector       basis.


The most          detailed           of    the data       series,            the affiliates                data,     provides
information             about      the      value     of the         total      asset         in which the
foreigner             has invested.              Because        these         data      include        all      firms        that
have more than                 lo-percent           foreign       interest,            the question                arises      as
to what portion                  is held       by relatively                 small      foreign        equity
interests             in U.S.      firms.           EIEA data        showing         the affiliates'                 sources

of     funds,         however,        indicate        that      foreign          firms        hold     about 80
percent         of the equity               in such      firms.          Therefore,              using        the    IO-
percent          equity        interest        rule     to define             foreign       direct           investment
does not          particularly              distort      the      investment             data.         One exception
is     in the         reporting           of Japanese          interests             in the      nonbank           finance

sector,         where      Japanese          firms     have purchased                  minority            interests          in
U.S.      securities            brokerage         or investment                firms,      which           have very
large      assets         relative          to equity.


unlike      the        balance        of payments            data,      the affiliates                data do
include         the     U.S.- financed            portion         of the investment                   and      attempt         to
report      the       country        of the ultimate                 beneficial           owner.             Finally,        due
to the detailed                information            collected,             which       includes          balance          sheet
items      and statements                 of income,          these      data take            longer          to be
reported,         checked,           and compiled:                preliminary             data       are      2 years        old
by the time             they      are publicly           reported,             and revised            data are 3
years     old.
General             Problems


This         brlnqs      US to       the     subject         of the general                 problems           in the         BEA
data.


1.     Timeliness             - BEA publishes                the preliminary                  balance          of payments
and new enterprises                     data     about        6 months after                  the end of the                  report
year.          It     should     be recognized                that        BEA does publish                   these        data   on
a timely             basis.      However,         because               of late          reports     and corrections,
the     revised         data     published             in    the        subsequent          year have often
reflected             significant            changes.              Timeliness             is an issue              with     regard
to     the     affiliates            data,     which         is,        however,         the most detailed                    data
series.


2.     Classification                methodology             - The data             are limited              by the
industrial             classification              system              used.       This     is     roughly          based on
the     three-digit             Standard         Industrial               Classification                  (SIC)      level,
which         BEA     feels     is    the     lowest         level        to reasonably              classify
enterprise             data.         BEA does not             require            greater         detail        in
identifying              industrial           classification                    in company reporting.                         This
means that             many important             industry               sectors         are not          listed
separately,              including           numerous          sectors           vital      to national              security,
such as aerospace                    and semiconductors.


3. Consolidated                 reporting         - Data are reported                        on an enterprise
rather         than     an establishment                    basis.             In other      words,         the      U.S.


                                                                   6
affiliate            corporation               files             a single           report          for dll     of     its        assets,
including            its     own subsidiaries.                                For example,            a U-S.        petroleum
corporation                with     foreign              ownership               interests           may own many
subsidiaries                in varying               industry                 sectors,        each one of uhich                    may
have many establishments                                 (which           can be individual                   factories            or
properties).                 But      it     will         file           one report           for     the whole enterprise,
all      of which           will      be classified                       under the petroleum                   sector            by BEA.
This      can skew the data                         in    all       sectors.              Although        establishment-
level        data      are needed for                     adequate               analysis           of important             public
pal icy       questions,              BEA collects                       only      limited       data on sales                and
employment             below the             enterprise                   level.          According           to BEA,
collection            of more industry                          detail           below that           level     is constrained
by     the    requirements                  of the Paperwork                        Reduction          Act     to minimize               the
burden        on respondents.


The BEA and the                    Census Bureau are currently                                      working     on establishing
a data        link         between          their         databases.                 Census data contain                      detailed
individual             establishment-level                               information            but do not highlight
foreign        ownership,                  while         BEA tracks                foreign          ownership         but     collects
only      enterprise-level                     data.              A pilot           project          is underway             to    use      the
firms'        Internal             Revenue Service                        Employer           Identification                 Numbers to
match        the BEA enterprise                          data       to the Census establishment                               data.
Commerce could                 then         publish             these           new data       annually         at an
aggregated             level.              Even in aggregated                         form,      the data            would        provide
greater        sectoral             breakdowns                   (such          as for       aerospace         or
semiconductors)                    and at           least         partially              address       the classification

                                                                          7
and      consolidation               problems            discussed              above.            The objective            of this
new      data     set      is to have more                  reliable                and detailed             information           on
foreign          direct        investment               without             placing           a greater          burden on the
private          sector.            Achieving            this      will            require        overcominq         technical
problems          and obtaining                  additional                 resources.


4.     Access to data                - The individual                        firms'           reports      are confidential,
Access          to this        information               is statutorily                       limited      to officials
specifically               designated             by the         President               to perform           analytical           and
statistical               functions          under         the International                        Investment        and Trade
in     Services           Survey      Act        of 1976.               Only aggregated                   data     are released
outside          of     BEA.        This     limits         others            with           a legitimate          need to know
                                                                                                                                             !   j
from      examining            the    data        to     further             track           foreign      direct     investment
in     specific           industry          sectors,              It        also      results           in substantial         data
                                                                                                                                             i   i
suppression               in the      state-level                data         in order            to protect         the
confidentiality                    of respondents,                     who could              be identified          when the
number of             affiliates            is    small.


GAO ACCESS TO DATA


The legislative                    branch        needs to              be    fully           informed      about     the nature,
extent,          and effects           of        foreign        direct              investment            in the     United
States.           Proposed           legislation                providing               GAO access           to foreign
direct          investment           data        that      is otherwise                      restricted          would be
consistent              with    GAO's mission                   and appropriate                     and beneficial            to
legislative              branch       oversight             in this             area,            Providing         GAO access           to

                                                                   8
-




    these       data    would allow                    it     to perform                such      evaluations           for        the

    Congress        and to assess                  the          adequacy              of federal        government                 data        on
    foreign        investment.2



    The issue          of     safeguarding                    the     confidentiality                  of data          should            be

    separate        fran       the        issue         of who should                    have access.              Access           should
    be     determined          on a need-to-know                            basis.             Confidentiality                of data           is
    safeguarded             by the         procedures                 for        gaining          access     and handling,                     and
    by regulations                 prohibiting                  unauthorized                   disclosure,          not       by

    unnecessarily              restricting                    access.


    We believe          that        GAO should                  have access                to foreign          direct
    investment          data        for     the             reasons         I    have      just     stated.          We also              nc%e
    that      GAO has         in place            strict            and rigorous                  programs         to maintain                 the
    security        and confidentiality                             of      information             consistent            with       the
    requirements              of    the     originating                     agency.



    GAO conducts            many reviews                      of agency programs                      that     necessitate
    access       to confidential,                           proprietary,                sensitive          and,     in many cases,
    highly       classified               information.                     We routinely               have access              to
    national        security              information                 of        the     utmost      sensitivity.                   We also
    routinely          have        access         to         proprietary                information          more similar                  in




    2This issue is discussed      in greater detail  in our June 13, 1990
    testimony,  National   Security   Review of Two Foreiqn Acquisitions  in
    the Semiconductor    Sector (GAO/T-NSIAD-90-47).
nattire     to      the foreign              direct                investment        data,      such as that          acquired
by federal              bank      authorities                 in     requlating         bank activities.                In
addition,           GAO has             access          to    Internal            Revenue Service           tax     data,
including           files         that      show names and social                        security          numbers.          In
all      these      cases,             procedures             are      in place       that      are designed          to
prevent          unauthorized               disclosure.


In     summary,          GAO has been entrusted                              with    access      to much of the most
sensitive           data         held      by the            federal         government         and has an excellent
record      in safeguarding                       such        data.


MORE THAN IYPROVED FOREIGN INVESTMENT DATA IS NEEDED TO                                                            ANSWER      SOME
IMPORTANT QUESTIONS


I would          like       to    poir.t     out         that        there        are also      some important
qu?stians           about         foreign          in*Jestment               that    cannot      be resolved           by
improving           foreign             investment              data.3



In the      banking              sector,          for        example,         Federal        Reserve       Board data         are
comprehensive                and publicly                    available.             Specific      transactions              and
foreign      investment                  trends          can be identified,                    showing      that     foreign
investors           hold         over      25 percent                of    U.S.     banking      assets.           What is not
clear,      however,              is     whether             there        is some level          of foreign          ownership


3For    a detailed  discussion                             of these   issues, see Foreign   Investment:
Concerns in the Banking,                                Petroleum,    Chemicals, and Biotechnology
Sectors    (GAO/NSIAD-90-129,                                May 30, 1990).
beyond which                 foreign          control       would be undesirable.                       For example,
could         lending         decisions           affecting           the growth          and direction              of tJ.S.
industry             be subordinated                to     foreign       interests?             The type       of data
needed         for     this        analysis        would       be information              relating,          for
example,             to bank lending                patterns          and competitive              behavior.


In the petrole*xn                    sector,        the     Departments          of Commerce and Energy
have detailed                 data      showing           the extent          and nature          of foreign
investment,                 and it      is     known that            members of the Organization                          of
Petroleum             Exporting            Countries          have invested              in U-S.         refining         and
marketing             facilities.                The difficult            question         is     whether       such a
link     enhances             security           of petroleum            supplies         or raises          the risk           of
greater         U.S.         vulnerability                to events        in the      Middle       East.           Analysis
of     this     question             needs to focus              both      on finding           ways to reduce                 the
likelihood             that         OPEC member countries                     would choose          to     exploit         any
vulnerability                 and on reducing                 that      vulnerability.                  For petroleum,
of course,             it     should          be recognized            that     the    basic       issue      is U.S.
dependence              on oil         imports          and that        the potential             for     disruption
exists         regardless             of      foreign        investment         levels.


In the biotechnology                          sector.        formal      government         data         cannot      be
disaggregated                 to     show foreign             investments,            but private            sector        data
can identify                 trends          indicating         increased         .%reign         participation                in
the U.S.             biotechnology               industry,           particularl>-         oy tapanese              firms.
The central                 question          in this        sector      is whether             the short-term
benefits             from     such investment                 will     match the          consequences              of

                                                                11
transfers            of      core      technologies,            includinq        the benefits            from
commercializing                  them.        Answers to this               question        would have to            come
from     a closer             government            focus     on the       direction          and types       of
technology            transfers            that      are occurring.              Indeed,        the     key question
is    the     basic          public       policy       question       of    whether       the U.S.         government
should        try     to      ensure       that      commercially           strategic          technologies
originating                in the       united       States      are developed              by U.S.      firms      into
commercial            products.


With        regard        to national              security      concerns4          about      foreign
acquisitions                 of U.S.       firms,       we found       that      the most difficult
questions            posed by individual                      investments        required         (1)     identifying
the     U.S.        firm's       link      to national           security        and (2) assessing
foreign         investor              intentions        regarding          technology          transfer,         assured
supply        to the          Defense        Department          or its       contractors,            and the
security            implications             of any commercial                advantages         gained         through
the     investment.                  Improved        foreign        investment         data     cannot      answer
these        questions,              which    are frequently               raised       in the        interagency
review        process          for      national        security-related               investments,           as
provided            in the          1988 Exon-Florio             Amendment to the               Defense
Production             Act.5

4For a detailed    discussion    of these concerns,   see Foreign
Investment:   Analyzing    National   Security Concerns (GAO/NSIAD-90-94,
March 29, 1990).

5These issues were also covered in our June 13, 1990 testimony,
National  Security    Review of Two Foreign Acquisitions in the
Semiconductor    Sector (GAO/T-NSIAD-90-47).

                                                               12
From a macroeconomic                          point         of view,              we should         recognize        that      the
United         States      is experiencing                           increased           foreign        participation           in       *,

the     U.S.     economy in part                       because             U.S.     budget         and trade        deficits
require         foreign           capital.        to        finance            them.       As long as U.S. domestic
savings         fail      to match             federal               budget       deficits          and private           sector
investment             needs,         the      United          States           will      be dependent           on foreign
financing.              Therefore,               foreign              direct           investment        will    remain        one
element         of such foreign                   participation                     in the       U.S.     economy.



Overall,         we     believe              improvements                  can be made in the government's
foreign         investment              data,         but      we also            believe        the     nation's       highest
policymakers              need to             address          some of the                more difficult             questions
raised       by increased                    foreign         investment                 in the      United      States.


                                                             -m-w_




This     concludes          my statement,                     Mr.          Chairman.          I will         be happy to           try
to     answer      any questions                  the       Committee              may have.




                                                                      13
                                                                                                                                              t