oversight

Forest Service: Factors Affecting Bids on Timber Sales

Published by the Government Accountability Office on 1997-06-17.

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

      United States
GAO   General Accounting  Office
      Washington, D.C. 20548

      Resources, Commuuily,   and
      Economic Development    Division


      B-277087


      June 17, 1997


      The Honorable John R. Kasich
      Chairman, Committee on the Budget
      House of Representatives

      Subject:    Forest Service: Factors Affecting Bids on Timber Sales

      Dear Mr. Chairman:

      As agreed with your office, we are providing you with information on the
      factors that affect bid premiums for the Forest Service’s timber sales. Our
      objectives were to (1) identify what are the key factors related to bid
      premiums in these sales, (2) determine what effects the key factors have on
      bid premiums, (3) determine if there are any regional differences in the key
      factors, and (4) determine what other factors could affect bid premiums that
      might not be evident in the Forest Service’s timber sales reports. These issues
      are discussed below and in enclosure I. Because a centralized database was
      not available, we sampled the agency’s timber sales reports for all Forest
      Service regions except Alaska. We performed statistical analyses on the data
      contained in the reports to determine the factors related to bid premiums and
      their relative importance.

      According to the Forest Service, its regions, excluding Alaska, sold
      approximately $2.6 billion worth of timber in reportable sales in fiscal years
      (FY) 1992-96. We analyzed reports from this period that we estimate totaled
      about $2.47 billion in sales. For those sales, we found that the appraised
      prices before bidding began totaled approximately $1.8 billion. The overall bid
      premium expressed as a bid ratio (the ratio of winning bid price to appraised
      price) was 1.45. Approximately 59 percent of the sales analyzed sold for less
      than $100,000. About 8 percent of the sales sold for $700,000 or more.




                                                   GAO/RCED-97-175R   Bids on Timber   Sales
B-277087

SUMMARY

We found three factors that were significantly and positively related to higher
bid premiums (the amount that the winning bidder paid over the appraised
price).’ One of these factors, the use of sealed bids, could have increased
timber sales receipts by $56 million if it had occurred in place of some oral
auctions.

-       The number of bidders per sale has a significant relationship with bid
        premiums. Regions 5 (Pacific Southwest) and 6 (Pacific Northwest) had
        the highest number of bidders per sale and had the highest bid premiums.
        Region 5 averaged about 2.97 bidders per sale and had a bid ratio of about
        1.70. Region 6 averaged about 3.05 bidders per sale and had a bid ratio of
        about 1.68. Regions 3 (Southwestern) and 4 (Intermountain) were among
        the three regions that averaged less than 2 bidders per sale and had the
        lowest bid premiums. Region 3 averaged about 1.59 bidders per sale and a
        bid ratio of about 1.17. Region 4 averaged about 1.78 bidders per sale and
        a bid ratio of about 1.17. Since the number of bidders was a key factor
        associated with higher bid premiums, we examined the factors that
        increased the number of bidders. One factor, the amount by which the
         estimated price of the timber being offered for sale is reduced (known as
         the rollback), was significantly related to the number of bidders. Regions 5
         and 6 had relatively large rollbacks and the greatest estimated number of
         bidders per sale. Regions 2 (Rocky Mountain), 3, and 4 had relatively small
         rollbacks and the lowest estimated number of bidders per sale. Forest
         Service officials noted, however, that at some point, if there are not many
         bidders, the Service begins reducing the rollbacks, since they are not
         serving the intended purpose. They believe this practice may account for
         the results in Regions 2 and 3.

-       The sealed bid auction method was significantly related to higher bid
        premiums. In sealed bid sales, individual bidders do not know how many
        bidders they are bidding against. Consequently, the bidders must assess the
        likelihood of competing bidders and base their bid accordingly. On the
        other hand, in oral auction sales, the bidders know if anyone else is
        competing against them. As a result, a single bidder at an oral auction can



    ‘See Federal Timber Sales: Process for Annraising Timber Offered for Sale
    Needs to Be Improved (GAOKRCED-90-135,May 2, 1990) for a description of
    our model, which determined statistical relationships by applying our sample
    data to a model of overbid percentages.
    2                                              GAO/WED-97-175R   Bids on Timber   Sales
B-277087
    win the sale by bidding the Forest Service’s appraised price. The appraised
    price is determined by rolling back the estimated value of the timber by
    some factor to encourage more bidders and adjust for possible
    inadequacies in the estimate, including timber market conditions. In effect,
    the bidder gets the timber at a discount from the Forest Service’s estimated
    value of the timber. The advantage gamed at an oral auction by a single
    bidder over bidders at any other sale is shown by the difference in the bid
    ratio. In sales in which there is only one bidder (29.4 percent of the sales
    analyzed), the bid ratio is 1.01 for oral auction sales compared with 1.14 for
    sealed bid sales. If the oral auction, single-bidder sales had been sealed bid
    sales, we estimate that revenues from these sales could have been $56
    million higher.

-   Salvage sales, which generally contain dead or dying trees, appear to have
    a significant relationship with bid premiums. However, that occurs, in part,
    because some regions reduce the estimated value of salvage timber by a
    greater amount than normal when dete r-mining the appraised price in order
    to spur sales. This action can result in higher bid premiums when a
    salvage sale tree sells at the same price as a green sale tree. About 60
    percent of salvage sales were less than $50,000, while about 36 percent of
    green sales were less than $50,000.

The factors that related to more bidders and higher bid premiums varied
across regions. In Regions 5 and 6, the factors related to more bidders and
higher bid premiums were higher, on average, than the national estimates.
Regions 5 and 6 had the highest average number of bidders and bid premiums
of the regions. In Regions 3 and 4, the factors related to the number of
bidders and bid premiums were lower, on average, than the national estimates.
Regions 3 and 4 had the fewest average number of bidders and the lowest
average bid premiums. Tables II.1 and II.2 in enclosure II show all estimates
and sampling errors for all regions on the factors related to the number of
bidders and the amount of bid premium.

We spoke with Forest Service headquarters, regional, and forest-level
personnel; industry sources; and timber experts to obtain their views on other
possible factors not included in timber sale reports that might affect bid
premiums and the number of bidders. They identified three major areas:

-   Species nrotection and changes in forest management nhilosonhv. These
    factors have reduced the allowable size of harvested trees and increased
    the use of more costly harvesting practices, such as the use of helicopters
    to reduce environmental damage. Also, management philosophies requiring

                                               GAO/WED-97-175R   Bids on Timber   Sales
B-277087
        more intensive timber management activities by loggers, such as replanting
        understocked stands or thinning to reduce fire danger, have increased
        logging costs, resulting in lowered value to the bidder.

-       Concentration in the timber industrv. In some areas, there are fewer mills
        operating than previously, giving those closest to the sales a cost advantage
        over more distant mills. Unless the value of the trees is high enough to
        offset the disadvantage of distance, this factor could lead to fewer bidders.
        There are also fewer mills to buy logs from the independent loggers who
        might be interested in bidding on the sale.

-       Use of the transaction evidence annraisal (TEA) method. TEA uses past
        sales as the starting point for timber appraisals. Because prior sales are
        used as the basis for the appraisal, TEA may be slow to reflect price
        decreases in declining timber markets. If the appraisal overvalues the
        proposed sale, there is a likelihood that no one will bid on the sale. Forest
        Service officials believe this is true for any appraisal system that relies on
        existing data, not just TEA.

AGENCY COMMENTS

We provided the Forest Service with a draft of this report for review and
comment. We met with the Deputy Director, Timber Management Staff, and
members of his staff at Forest Service headquarters. They agreed with the
information presented in our report and provided us with clarifications, which
we have incorporated where applicable.



    We performed our work from January 1997 through May 1997 in accordance
    with generally accepted government auditing standards. We reviewed journal
    articles, industry reports, reports by the U.S. Department of Agriculture’s
    Office of Inspector General and prior GAO reports to discover what factors
    were previously found to be important in timber sales. Because data tapes
    recording timber sale data were not available, we sampled the Forest Service’s
    Reports of Timber Sales from f&al years 1992 through 1996 for all Forest
    Service regions except Alaska. The Forest Service Eles reports only for sales
    that are for at least 2 million board feet or more than $2,000; however, such
    sales account for most of the timber sold. We briefed your office on the




    4                                               GAO/RCED-97-175R   Bids on Timber   Sales
B-277087

results of our work on May 21, 1997. Please call me on (206) 2874810 if you
or your staff have any questions. Major contributors to this report were
Edward M. Zadjura, Jonathan Bachman, and SaraAnn Moessbauer.

Sincerely yours,




Associate Direc    Energy, Resources,
 and Science

Enclosures-2




5                                           GAO/RCED-97-175R   Bids on Timber   Sales
ENCLOSURE I                                   ENCLOSURE I




GKJ Resources, Community, and Economic
    Development Division

        REVIEW OF FACTORS AFFECTING
        BIDS ON FOREST SERVICE
        TIMBER SALES




                                                                     *’         i’,,,
                                                                          .,.




 6                         GAO/RCED-1’75R   Bids on ‘Jhber   Sales
ENCLOSURE I                                            ENCLOSURE I




w         Timber Sales Competition -
          Objectives
    l   What are the key factors related to bid
        premiums in Forest Service timber sales?
    l   What effects do the key factors have on bid
        premiums?
    l   Are there regional differences in the key
        factors?
    l   What other factors could affect bid
        premiums that might not be evident in sales                          n



        reports?




7                                   GAO/RCED-175R   Bids on Timber   Sales
ENCLOSURE I                                        ENCLOiUFtE I




w         Timber Sales Competition -
          Scope and Methodology
      l   Reviewed journal articles, industry
          reports, USDA OIG and GAO reports
      l   Discussed our work with Forest Service,
          CRS, and CBO personnel
      l   Sampled reports of timber sales from
          FYs 1992-96 for all regions except
          Alaska because Forest Service did not
          have a computerized database available
          with sales data




                                  GAO/RCED-175R   Bids on Timber   S&es
ENCLOSURE I                                        ENCLOSURE I




GLQ~ Timber Sales Competition -
       Scope and Methodology
      8 Performed statistical analyses on data in
        the reports to determine factors related
        to increased bid premiums




 9                               GAOiRCED-175R   Bids on Timber   Sales
ENCLOSURE I                                             ENCLOSURE I




GQ        Timber Sales Competition -
          Scope and Methodology
      l       Regression analyses
          l    Performed on a national level
          l    Examined impact of each factor in
               determining bid premiums and number
               of bidders
          l    Controlled for the impact of other
               factors when examining relationships




 10                                   GAO/RCED-175R   Bids on Timber   Saks
ENCLOSURE I                                              ENCLOSURE I




m        Timber Sales Competition -
         Background
     l       During FYs 1992-96, Forest Service
             Regions 1-9 sold about $2.6 billion in
             timber
     l       GAO analyzed a sample of the large
             sales that occurred during this period
         l    The estimated total sales represented
              by the sample was about $2.47 billion




11                                    GAOIRCED-175R   Bids on Timber   Sales
ENCLOSURE I                                                 ENCLOSURE I




G+Q Timber Sales Competition -
    Background
        l       Of the sales GAO analyzed:
            l       The appraised prices before bidding
                    began totaled about $1.8 billion
            l       The overall bid ratio (the ratio of bid to
                    appraised prices) was about 1.45
                l   The bid ratio is based on data from
                    sales in which both winning bid price
                    and appraised price were reported




                                             GAO/WED-175R   Bids on Timber   Sales
ENCLOSTJREI                                                                     ENCLOSURE I




m      Timber Sales Competition -
       Background
                   Distribution    of Timber   Sales Selling   Prices

              Less than $100,000   59.4%




                                                                    No data    4.9%



                                                                $700,000 or more      8.4%


                                                          $500,000-699,999     3.5%




13                                                        GAOLRCED-175R      Bids on Tuber   Sales
ENCLOSURE I                                                                                         ENCLOSURE I




w          Timber Sales Competition -
           Factors Related to Higher Premiums
       l       National-level regression results
           l       Three factors had significant and positive
                   impact on higher bid premiums
               l    Use of sealed bids--about 75% of all
                    sales used sealed bids
               l    Number of bidders--the average number
                    of bidders was about 2.66
                l   Salvage sales--about 42% of all sale
                    Note: Table 11.1lists significant factors affecting bid premiums, by region, with their sampling errors.




 14                                                                          GAO/RCED-175R       Bids on Timber     Sales
ENCLOSURE I                                                                                      ENCLOSURE I




G&Q Timber Sales Competition -
    Factors Related to Higher Premiums
     Average impact of factors on winning bid price
       l   increase in number of oral bidders                                    increase in amount bid
           l   From 2 to 3 bidders                                                        + 8.5 %
           l   From 3 to 4 bidders                                                        + 3.9 %
           l   From 4 to 5 bidders                                                        + 2.3 %
               Note: Sales with only 1 bidder not included in our analysis of oral bid sales because single bidders
               have no reason to bid above the minimum advertised price since they know there are no other
               bidders.

       l   Increase in number of sealed bidders                                 Increase in amount bid
           l   From 1 to 2 bidders                                                        + 38.2 %
           l   From 2 to 3 bidders                                                        + 9.2 %
           0 From 3 to 4 bidders                                                          + 4.2 %
           . From 4 to 5 bidders                                                          + 2.4 %
       l   Change from oral to sealed bid sale                                            + 8.6 %
       l   Change from green to salvage sale                                              +17.1 %




15                                                                        GAOIRCED-175R       Bids on ‘lhnber   Sales
ENCLOSURE I                                          ENCLOSURE I




GAD Timber Sales Competition -
    Factors Related to Higher Premiums
  l       Regions 5 and 6 averaged the highest
          number of bidders per sale and the
          highest bid premiums
      l    Region 5 averaged about 2.97 bidders
           per sale and had a bid ratio of about
           1.70
      l    Region 6 averaged about 3.05 bidders
           per sale and had a bid ratio of about
           1.68




 16                                 GAO/RCED-175R   Bids on Timber Sales
ENCLOSURE I                                          ENCLOSURE I




GAI Timber Sales Competition -
    Factors Related to Higher Premiums
           Regions 3 and 4 averaged among the
           lowest number of bidders per sale and
           had the lowest bid premiums
       l    Region 3 averaged about 1.59 bidders
            per sale and had a bid ratio of about
            1.17
       l    Region 4 averaged about 1.78 bidders
            per sale and had a bid ratio of about
            1.17




17                                GAOIRCED-175B   Bids on Timber   Sales
ENCLOSURE I                                          ENCLOSURE I




GALI Timber Sales Competition -
     Factors Related to Higher Premiums
      l   In sealed bid sales, individual bidders do
          not know how many other bidders they
          are competing against
      l   In cases with a single bidder (29.4% of
          all sales), sealed bid sales had an
          overall bid ratio of 1.14, while oral
          auction sales had an overall bid ratio of
          1 .Ol




 18                                 GAO/RCED-175R   Bids on Timber   Sales
ENCLOSURE I                                           ENCLOSURE I




MO     Timber Sales Competition -
       Factors Related to Higher Premiums
       l   If these same single-bidder oral auctions
           had been single-bidder sealed bid sales,
           we estimate that sales revenue could
           have been about $56 million more




19                                 GAO/RCED-175R   Bids on Timber   Sales
ENCLOSURE I                                          ENCLOSTJREI




GM Timber Sales Competition -
   Factors Related to Higher Premiums
      l   While salvage sales had a significant
          impact on bid premiums, that occurs, in
          part, because some regions reduce the
          estimated value of the timber by a
          greater amount than normal when
          determining the appraised price for
          salvage sales to spur bidding
      l   Numerically, salvage sales represented
          about 42% of all sales and 32% of the
          total dollar value




 20                                GAOLRCED-175R   Bids on l’hber   Sales
ENCLOSURE I                                           ENCLOSURE I




w        Timber Sales Competition -
         Factors Related to More Bidders

     l   Since the number of bidders was a key
         factor that cut across all other factors,.
         such as the type of sale (green vs
         salvage) or the method of sale (oral vs
         sealed bid), we examined the factors that
         increased the number of bidders
         0 There were 6 factors that increased the
           number of bidders at the national level




21                                 GAOLRCED-175R   Bids on Thnber   Sales
ENCLOSURE I                                                                                       ENCLOSURE I




GLQ~ITimber Sales Competition -
     Factors Related to More Bidders
          l   These factors relate to either sale
              administration or sale contents




              Note: Table 11.2lists significant factors, by region, with their sampling errors.




 22                                                                     GAO/WED-175R         Bids on ‘Ember Sales
ENCLOSURE I                                                                                        ENCLOSURE I




G$Q Timber Sales Competition -
    Factors Related to More Bidders


     l       Administrative factors
         l    Amount of rollback
         l    Sealed bids
         l    Purchaser road credits


              Note: A rollback is a reduction in the estimated value of timber to calculate the appraised price for
              the sale. The rollback is intended to encourage more bidders to participate in the sale.




23                                                                         GAO/WED-175R        Bids on ‘Jhnber Sales
ENCLOSURE I                                            ENCLOSURE I




w         Timber Sales Competition -
          Factors Related to More Bidders

      l       Content factors
          l    Sale volume
          l    Salvage sale
          l    Quality of the trees (measured by the
               appraised price per thousand board
               feet)




 24                                   GAO/RCED-175R   Bids on Timber   Sales
ENCLOSURE1                                                                                       ENCLOSURE1




MO Timber Sales Competition -
   Factors Related to More Bidders
     l   Rollback had a significant impact on the
         number of bidders
         Note: Rollbacks could not be calculated for Region 1 sales because of the methodology used to establish
         rollbacks in that region. This precluded the use of Region 1 in any analysis involving rollbacks or appraised
         prices.


     l   Regions 5 and 6 had relatively large
         rollbacks and the highest number of
         bidders per sale
     l   Regions 2, 3, and 4 had relatively small
         rollbacks and the lowest number of bidders
         per sale




25                                                                       GAOIRCED-175R       Bids on Timber Sales
ENCLOSURE I                                                 ENCLOSURE I




GBI Timber Sales Competition -
    Variations Among Regions
      l       There were regional variations in the 6
              factors that related to more bidders
          l        Regions 5 and 6 almost always had
                   higher average values for the factors
                   positively related to higher bid
                   premiums and more bidders than the
                   national averages
               l    Regions 5 and 6 had the highest bid
                    ratios and the most bidders




 26                                        GAO/RCED-175R   Bids on Timber   Sales
ENCLOSURE I                                             ENCLOSURE I




w        Timber Sales Competition -
         Variations Among Regions
     l       Regions 3 and 4 usually had lower
             average values for the factors positively
             related to higher bid premiums and more
             bidders than the national scores
         l    Regions 3 and 4 had the lowest bid
              ratios and among the fewest bidders




                                     GAO/RCED-175R   Bids on Timber Sales
ENCLOSURE I                                               ENCLOSURE I




w              Timber Sales Competition -
               Other Factors Affecting Bid Premiums
      l       Forest Service and other knowledgeable
              parties have cited other factors as
              potentially affecting bid premiums and the
              number of bidders
          l    Species protection and changes in forest
               management philosophy
          l    Concentration in the timber industry
          l    Use of transaction evidence appraisal
               (TEA) method in declining markets




 28                                      GAO/RCED-175R   Bids on l’lmber   Sales
ENCLOSURE I                                             ENCLOSURE I




MI       Timber Sales Competition -
         Other Factors Affecting Bid Premiums
     l       Species protection and changes in forest
             management philosophy
         l    Have reduced size of trees being
              harvested
         l    Have changed logging practices to
              include more costly harvesting methods
         l    Have reduced value of federal timber to
              potential bidders




29                                   GAO/RCED-175R   Bids on Thnber Sales
ENCLOSURE I                                                 ENCLOSURE I




MO        Timber Sales Competition -
          Other Factors Affecting Bid Premiums
      l       Concentration in the industry
          l       With fewer mills, those that are closest
                  to sales have distinct cost advantages
                  over those farther away, which may
                  reduce the number of bidders
              l   Bidders planning on selling to mills
                  have fewer options, leaving less room
                  for differences in bids




 30                                        GAO/RCED-175R   Bids on lhber   Sales
ENCLOSURE I                                                  ENCLOSURE I




w              Timber Sales Competition -
               Other Factors Affecting Bid Premiums
l        TEA method
     l       Because the TEA method of appraising
             timber sales is based on past sales, its
             results may be slow to reflect price
             decreases in a declining market
     l       Consequently, the amount of rollback may
             be a significant factor in bringing in bidders
         l    Only areas with higher rollbacks may
              have prices low enough to attract bidders




31                                        GAO/RCED-175R   Bids on ‘IEmber Sales
ENCLOSURE II                                                                                        ENCLOSURE II

                    SIGNIFICANT FACTORS RELATED TO HIGHER BID
                         PREMIUMS AND NUMBER OF BIDDERS
Table 11.1: Sianificant Factors Related to Hiaher Bid Premiums

                                      .’
                                                 dl                  Significant factors


                                                        Average             Percent            Percent
                   Number of      Overall bid         number of              sealed            salvage
                       sales            ratio            bidders               bids              sales
  Total U.S.   10,423 (+197)     1.45(~06)            2.7 (~1)        75.1 (+1.2)          42.2 (k2.3)
  Region 1        1145 (+78)     1.58 (-r-.21)        2.5 (k-2)        47.5 (k6.2)         58.4 (dfi.9)
  Region 2         432 (k18)     1.29 (k.07)           1.7 (k.1)       90.9 (k3.3)         40.1 (-1-5.3)
  Region 3’         180 (&O)       1.17 (+O)           1.6 (k0)          31.9 (k0)           27.2 (k0)
  Region 4         632 (a37)     1.17 (k.07)           1.8 (-t-.2)     35.3 (k5.7)         60.1 (k5.6)
  Region 5        1162 (k84)     1.70 (k.24)           3.0 (2.3)       85.1 (k4.5)         70.4 (k5.7)
  Region 6        1472 (-1-65)   1.68 (k.18)           3.0 (k.3)       15.4 (k3.9)          57.1 (k5.5)
  Region 8       3546 (k117)     1.31 (k.05)           2.7 (k.2)        100.0 (k0)         30.2 (-1-4.8)
  Region 9        1854 (k79)     1.34 (~~06)           2.8 (k.2)        100.0 (+O)         14.5 (k4.5)

Note: Sampling errors are in parentheses.

‘All sales for Region 3 were sampled.




 32                                                                             GAO/RCED-175R        Bids on Timber   Sales
     ENCLOSURE II                                                                                         ENCLOSURE II

Table 11.2: Sianificant Factors Related to Higher Number of Bidders

                  ;    :.

                Average
                number            Average
                                                                  Significant factors

                                                                                              -1
                                                                                              Average quality         Percent
                      Oi           percent b                                                       (appraised including road
                bidders           rollback                                                         valuelmbf)          credits
Total U.S.    2.7 (e-1)       15.3 (kO.2)
 Region 1’    2.5 (t.2)     Not available
 Region 2     1.7 (-c.l)       8.9 (kO.2)
 Region 3’     1.6 (+O)          6.8 t+Ol
 Region 4     1.8 (k-2)
 Region 5     3.0 (+.3)         20.0 (%O)
*Region 6     3.0 (k.3) ~ 31.1 (kO.5)
 Region 8    1 2.7 (k.2)       9.4 (kO.2)      901.8 (4105.2) 1   100.0 (&O) 130.2 (k4.8)
 Region 9     2.8 (+.2)         15.0 (+O) 1265.9 (~120.4) [

Note: Sampling errors are in parentheses.

‘Rollbacks could not be calculated for Region 1 because of the methodology used in that region to establish
rollbacks. This precluded the use of Region 1 in any analysis involving rollbacks or appraised prices.
‘All sales for Region 3 were sampled.




(141011)


33                                                                                      GAO/RCED-175R    Bids on Timber    Sales
J
Ordering    Information

The first copy.of each GAO report and testimony is free.
Additional   copies are $2 each. Orders should be sent to the
following address, accompanied by a check or money order
made out to the Superintendent     of Documents, when
necessary. VISA and MasterCard      credit cards are accepted, also.
Orders for 100 or more copies to be mailed to a single address
are discounted 25 percent.

Orders by mail:

U.S. General Accounting Office
P.O. Box 6015
Gaithersburg, MD 20884-6015

or visit:

Room 1100
700 4th St. NW (corner of 4th and G Sts. NW)
U.S. General Accounting  Office
Washington, DC

Orders may also be placed by calling (202) 512-6000
or by using fax number (301) 258-4066, or TDD (301) 413-0006.

Each day, GAO issues a list of newly available reports and
testimony.   To receive facsimile copies of the daily list or any
list from the past 30 days, please call (202) 512-6000 using a
touchtone phone. A recorded menu wili provide information         on
how to obtain these lists.

For information on how to access GAO reports on the INTERNET,
send an e-mail message with “info” in the body to:

info@www.gao.gov

or visit GAO’s World Wide Web Home Page at:

http:Bwww.gao.gov
United States
General Accounting  Office
Washington, D.C. 20548-0001

Official   Business
Penalty    for Private   Use $300

Address    Correction    Requested