GAO-20-135, Federal Real Property: GSA Should Improve Accuracy, Completeness, and Usefulness of Public Data

FEDERAL REAL

PROPERTY

GSA Should Improve

Accuracy,

Completeness, and

Usefulness of Public

Data

Report to the Committee on

Transportation and Infrastructure, House

of Representatives

February 2020

GAO-20-135

United States Government Accountability Office

Highlights of GAO-20-135, a report to the

Committee on Transportation and

Infrastructure, House of Representatives

February 2020

FEDERAL REAL PROPERTY

GSA Should Improve Accuracy, Completeness, and

Usefulness of Public Data

What GAO Found

The General Services Administration (GSA) has worked in recent years to

improve reliability of the Federal Real Property Profile (FRPP), which tracks

federal real property assets. However, numerous errors in the database were

carried into the public version. GSA extracted data from the FRPP’s 398,000

civilian federal assets to create a public database to be used, for example, by

researchers and real estate developers. However, GSA’s data verification

process did not address key errors. GAO found that 67 percent of the street

addresses in the public database were incomplete or incorrectly formatted. For

example, the database lists “Greenbelt Road” as the address for over 200

buildings at NASA’s Goddard Space Flight Center, but the road stretches over

6.3 miles, thereby reducing a user’s ability to locate specific buildings.

Challenges Mapping Incomplete “Greenbelt Road” Street Address

The public database is not complete because GSA and selected agencies

decided not to provide certain useful information. Specifically, GSA withheld

assets’ information without consulting those agencies managing the assets and

allowed agencies to withhold information that is already publicly available. For

example, GSA withheld the name “Goddard Space Flight Center” from the public

database, but NASA’s website lists this name and the Center’s location.

Unnecessarily withholding information limits the database’s utility and

undermines analysis.

The public database’s usefulness is further limited by how GSA presents the

information. Because the database does not identify if an asset is part of a

secure installation, the public does not know if assets, such as the unnamed

buildings at Goddard, are accessible to the public. Unless GSA improves the

public database’s accuracy, completeness, and usefulness, its benefits may not

be realized.

View GAO-20-135. For more information,

contact

Lori Rectanus at (202) 512-9847 or

rectanus[email protected]

Why GAO Did This Study

The lack of reliable data on federal

assets is one of the main reasons

Federal Real Property Management

remains on GAO’s high risk list. In

2016, legislation required GSA to

publish a single, comprehensive, and

descriptive database of federal real

property that would be available to the

public. The database could be used for

research and other potential

applications. GAO was asked to study

the public database. This report

assesses (1) GSA’s efforts to improve

the reliability of FRPP’s data and the

public database, (2) the public

database’s completeness, and (3) the

presentation of the data in the public

database.

GAO reviewed federal laws,

documents, and data, including GSA’s

fiscal years 2017 and 2018 FRPP and

public databases. GAO interviewed

officials at GSA and from six federal

agencies selected in locations with

enough questionable data in the public

database to analyze, among other

things, and studied assets in

Washington, D.C., Illinois, and New

Mexico. GAO also interviewed selected

stakeholders involved in federal real

property management, such as real

estate brokers.

What GAO Recommends

GAO is making six recommendations

to GSA, including improving the

accuracy of the database, consulting

with agencies on assets’ information

withheld from the database, and

improving the public database’s

presentation.

GSA agreed with five of

the recommendations. GAO clarified

the recommendation on withholding

information on agencies’ assets, to

address GSA’s comments.

Page i GAO-20-135 Federal Real Property

Letter 1

Background 5

GSA’s Efforts Have Not Effectively Addressed FRPP’s Reliability

Issues, Which Affect the Public Database 7

GSA and Agencies Withheld Information That Reduces the

Completeness of the Public Database 15

Data Presentation and Limited Stakeholder Awareness Hinder

Usefulness of the Public Database 22

Conclusions 27

Recommendations 28

Agency Comments and Our Evaluation 28

Appendix I Categories of Data Withheld from the Public Database 31

Appendix II Comments from the U.S. General Services Administration 32

Appendix III GAO Contact and Staff Acknowledgments 34

Tables

Table 1: Example of Information Included and Not Included in the

Public Data for a National Aeronautics and Space

Administration (NASA) Building 17

Table 2: Summary of Information Selected Agencies Withheld

from Public Database for Fiscal Year (FY) 2018 20

Table 3: Summary of Selected Information All Agencies Withheld

from the Public Database for Fiscal Year (FY) 2018

across All Reporting Agencies 21

Table 4: The 15 Categories of Data in the Federal Real Property

Profile That the General Services Administration (GSA)

Withholds from the Public Database 31

Figures

Figure 1: Extent to Which Street Information Is Included in the

General Services Administration’s (GSA) Federal Real

Property Profile (FRPP) and Public Database, for Fiscal

Year 2018 8

Contents

Page ii GAO-20-135 Federal Real Property

Figure 2: Incomplete “Greenbelt Road” Street Address Provides

the Public with Inaccurate Locations for Buildings at the

National Aeronautics and Space Administration’s (NASA)

Goddard Space Flight Center 9

Figure 3: Examples of Data Identified as Anomalous but Validated

as Correct by the Department of Energy 12

Figure 4: Examples of Federal Assets Identified as Anomalous by

the General Services Administration’s Validation and

Verification Process and Incorrectly Validated 13

Figure 5: Validation and Verification (V&V) Results for All

Agencies as of October 2019 14

Figure 6: Steps for a User to find National Aeronautics and Space

Administration’s (NASA) Goddard Space Flight Center’s

Buildings in the Public Database 18

Figure 7: National Aeronautics and Space Administration’s

(NASA) Goddard Space Flight Center Buildings: Current

Reporting as Individual Assets Compared to Reporting as

a Campus 23

Figure 8: Frequency of Accessing the General Services

Administration’s (GSA) Inventory of Owned and Leased

Property Database and Public Real Property Database

from December 2017 through July 2019 27

Page iii GAO-20-135 Federal Real Property

Abbreviations

DHS Department of Homeland Security

DOD Department of Defense

DOE Department of Energy

DOI Department of Interior

FASTA Federal Assets Sale and Transfer Act of 2016

FCC Federal Communications Commission

FRPP Federal Real Property Profile

GIS geographic information system

GSA General Services Administration

ICE Immigration and Customs Enforcement

ISC Interagency Security Committee

NASA National Aeronautics and Space Administration

NPS National Park Service

OGP Office of Government-wide Policy

OMB Office of Management and Budget

PBS Public Buildings Service

V&V validation and verification (process)

This is a work of the U.S. government and is not subject to copyright protection in the

United States. The published product may be reproduced and distributed in its entirety

without further permission from GAO. However, because this work may contain

copyrighted images or other material, permission from the copyright holder may be

necessary if you wish to reproduce this material separately.

Page 1 GAO-20-135 Federal Real Property

441 G St. N.W.

Washington, DC 20548

February 6, 2020

The Honorable Peter A. DeFazio

Chairman

The Honorable Sam Graves

Ranking Member

Committee on Transportation and Infrastructure,

House of Representatives

The General Services Administration (GSA) reported that in fiscal year

2018 civilian federal agencies spent billions of dollars to operate about

398,000 civilian real property assets (buildings, structures, and land)

across every state, including nearly 127,000 buildings covering 1.1-billion

square feet. GSA tracks the federal government’s real property assets

using a government wide database known as the Federal Real Property

Profile (FRPP) Management System, which contains data submitted

annually by agencies. We have previously reported on problems with the

reliability of the data in the FRPP, and it is one of the main reasons that

managing federal real property remains on our high-risk list.

The FRPP

itself is not available to the public but does not contain any classified

national security information. The lack of publicly available data and data

quality issues have posed problems for people wanting to use the federal

real property data for various purposes, such as leasing or purchasing

space that the federal government no longer needs.

The Federal Assets Sale and Transfer Act of 2016 (FASTA) directed GSA

to release to the public a single, comprehensive, descriptive database of

federal real property.

In April 2018, the then-chair of the House

Subcommittee on Economic Development, Public Buildings, and

Emergency Management stated that a key goal for making the data public

was to provide transparency and help hold federal agencies accountable

for reporting accurate information. In response to this act, GSA created

and released a publicly available version of the FRPP in December 2017.

High-risk issues with managing federal real property include: costly leasing; data

reliability; excess and underused property; and physical security. We have stated that

among other things, OMB and GSA should continue working with federal agencies to

improve the reliability of their real property data. See GAO, High-Risk Series: Substantial

Efforts Needed to Achieve Greater Progress on High Risk Areas, GAO-19-157SP

(Washington, D. C.: March 6, 2019).

Pub. L. No. 114-287, § 21(a). (c)(2) (2016).

Letter

Page 2 GAO-20-135 Federal Real Property

You asked that we review GSA’s new public real property database. This

report examines:

• GSA’s efforts to improve the reliability of FRPP data and the public

database;

• the completeness of the public database; and

• how the data are presented in the public database.

For all objectives, we reviewed applicable laws and the White House’s

and the Office of Management and Budget’s (OMB) directives.

Additionally, we reviewed GSA guidance to agencies on FRPP data

submissions for fiscal years 2016 through 2019.

Further, we reviewed

prior GAO reports on federal real property. We downloaded and analyzed

data in the public database for fiscal years 2017 and 2018 (the most

recent data available since FASTA was enacted) and obtained GSA’s

original FRPP data on civilian properties for this time period.

determine the reliability of this data, we evaluated the FRPP and public

data to identify what agency data were missing per GSA’s annual

guidance to agencies. We found the data sufficiently reliable for the

purpose of our reporting objectives.

We interviewed GSA officials, as well as a non-generalizable selection of

officials from six selected agencies that are required to submit data to the

FRPP: the Department of Homeland Security (DHS); Department of

Energy (DOE); the Department of Interior (DOI); the Federal

Communications Commission (FCC); the National Aeronautics and

Space Administration (NASA); and GSA’s office that submits its own

FRPP data. We selected these agencies because they were located in

areas with enough questionable FRPP data as identified by GSA to

analyze. We also interviewed a non-generalizable sample of 14

stakeholders from groups likely to use the public database, identified by

their frequent interaction with federal real property issues, among other

GSA, Federal Real Property Council: 2016 Guidance for Real Property Reporting,

Version 3 (Washington, D. C.: Oct. 31, 2016); 2017 Guidance for Real Property Reporting,

Version 2 (Washington, D. C.: Sept. 11, 2017); 2018 Guidance for Real Property

Reporting, Version 2 (Washington, D. C.: Aug. 27, 2018); 2019 Guidance for Real

Property Reporting, Version 1 (Washington, D. C.: Jun. 7, 2019).

FASTA allows the exclusion from reporting federal real property for reasons of (1)

national security and (2) those exempt from disclosure pursuant to section 552(b) of title 5,

United States Code. Pub. L. No. 114-287, § 21(a), (e) (3) (2016).

Page 3 GAO-20-135 Federal Real Property

things, and included real estate brokers, lessors, and developers.

Specifically, we identified groups to contact from the list of participants at

a congressional roundtable discussion on the uses of the public

database

and by asking officials at GSA and private-sector professionals

who specialize in federal real estate through a snowball-sampling

technique. We then used that list of potential users and search terms

related to real property data to conduct a web search for other groups that

indicated they used federal real property data and confirmed their use of

real property data.

To assess GSA’s efforts to improve the reliability of FRPP data and the

public database, we analyzed data from fiscal years 2017 and 2018 from

GSA’s Validation and Verification effort, which GSA implemented to

improve the quality of FRPP. We determined how the results of this effort

carried over to the public data. We also tested the location information in

the fiscal year 2018 public data to determine the extent to which this

information was consistent with guidance for reporting inventory GSA

provided to agencies and whether the location information was user-

friendly. Specifically, we determined if the data were machine-readable

(directly usable by a computer) and could be displayed in a map using

commercial off-the-shelf, geographic-information system (GIS) software.

We used the data to select and visit a non-generalizable sample of

properties from the six agencies mentioned above that were located in

proximity to our headquarters and Chicago, Illinois, field offices. We also

selected agency properties in the Los Alamos and Sandia, New Mexico,

areas because they had enough questionable data in those locations to

provide a basis for analysis. We determined if the properties’

characteristics matched reported data and assessed the extent to which

these properties are accessed by the public, through direct observation or

review of agency photographs.

The 14 stakeholders we interviewed were Bismarck Realty, Carpenter Robbins, Colliers

International, Equus Capital Partners, Federal Real Property Association, Holland and

Knight, Jones Lang, Lasalle, PublicAssets, Public Properties LLC, Reis, RSM US LLP,

Savillis Studley, Seven Properties, and Tower Properties.

Committee on Transportation and Infrastructure, U. S. House of Representatives,

Roundtable Policy Discussion on “Saving Taxpayer Dollars and the Federal Real Property

Database” (Washington, D. C., Apr. 11, 2018).

A Geographic Information System (GIS) is a computer system that analyzes and displays

geographically referenced information. It uses data that is attached to a unique location,

identified by longitude and latitude, referred to hereafter as “geo-coordinates.”

Page 4 GAO-20-135 Federal Real Property

To assess the completeness of the public database, we reviewed GSA

instructions to agencies and a memorandum from DHS’s Interagency

Security Committee (ISC) recommending processes and criteria for

agencies to follow when determining what, if any, information to withhold

from the public database.

We also reviewed selected agencies’ guidance

and processes, if any, for identifying assets to withhold from the public

database. We compared the FRPP to the public database to identify data

withheld from the public database for fiscal years 2017 through 2018 to

assess how this withholding affected the completeness and usefulness of

the database and to analyze trends and consistency in selected agencies’

withholding decisions. We also interviewed GSA, DHS-ISC, and selected

agency officials on internal processes for identifying what types of data

categories and specific assets they withheld from the public database.

To assess factors that affect the usefulness of the public database, we

reviewed GSA’s instructions on presenting data, specifically with regard

to how to report individual assets on secure installations that are not open

to the public. We analyzed the public database and FRPP database to

determine how agencies report assets on secure installations and how

this reporting affects the usefulness of the database. We also observed

secure federal installations in the Washington, D.C., area and in New

Mexico, to assess how the presentation of these assets in the public

database affects the usefulness of the data. These sites were among

those selected to assess GSA’s efforts to improve the reliability of the

database, as described above. We interviewed GSA officials to determine

how they communicate the availability of the public database. We also

assessed GSA’s website to determine how it communicates the

availability of the public database as well as three other real-property

databases as they related to the relevant provision of the Open

Government Data Act.

We analyzed GSA data on the frequency with

which the public accessed the public database from December 2017

through July 2019. We also interviewed selected agency officials to

determine their views on GSA’s organization of the database, such as

reporting by individual assets rather than installations. Lastly, we

The Interagency Security Committee (ISC) develops physical security policies and

standards, promotes key management practices, and facilitates mitigation of threats to

employees and the visiting public and is chaired by the DHS Assistant Director of the

Infrastructure Security Division within the Cybersecurity and Infrastructure Security

Agency. Members include chief security officers and other senior executives from 60

federal agencies and departments.

Pub. L. No. 115-435, § 202 (2019).

Page 5 GAO-20-135 Federal Real Property

interviewed the 14 third party stakeholders identified earlier to determine

their familiarity with the database and its usefulness to their work.

We conducted this performance audit from September 2018 to February

2020 in accordance with generally accepted government auditing

standards. Those standards require that we plan and perform the audit to

obtain sufficient, appropriate evidence to provide a reasonable basis for

our findings and conclusions based on our audit objectives. We believe

that the evidence obtained provides a reasonable basis for our findings

and conclusions based on our audit objectives.

Two GSA offices have roles in managing data related to federal real

property. The Public Buildings Service (PBS) acts as a landlord for the

federal government by acquiring new space for government agencies and

tracking data on the property it acquires. PBS manages and publishes

three databases that provide information to public stakeholders and

researchers on federally owned and leased properties, and on properties

eligible for disposal. Another office, the Office of Government-wide Policy

(OGP), collects, manages, and reports on all federal real-property data

through the FRPP database. OGP has managed the FRPP since its

inception in fiscal year 2005 by collecting data from federal agencies on

their real property assets.

OGP is also responsible for compiling and

managing the public database required by FASTA.

FRPP is the most comprehensive database of federal real property

holdings, containing details for about 398,000 assets (buildings,

structures, and land). It is not public, but it also does not contain any

classified national security information. FRPP data show the range of

agency assets, including single buildings in a given location or multiple

buildings located on installations, like a national park or research center.

The FRPP identifies whether buildings are on installations, but does not

identify whether buildings are public-facing or secure (and thus

inaccessible by the public).

Before FASTA was enacted, GSA was directed to establish a single, comprehensive,

and descriptive database, which resulted in the FRPP. Exec. Order No. 13327, 69 Fed.

Reg. 5897 (Feb. 6, 2004).

Background

Page 6 GAO-20-135 Federal Real Property

We have repeatedly identified reliability issues with the FRPP,

and GSA

has taken actions to improve the reliability of FRPP data. Specifically, in

2016, GSA established its validation and verification (V&V) process. After

agencies submit their data annually to FRPP, GSA identifies questionable

entries (called anomalies) from 20 separate categories. Through these

categories, GSA flags assets that are very small in size, changed from

the previous year, or have unusual financial statistics, among other

things. GSA then provides an annual list of anomalies to the agencies

that entered the data. Agencies have 10 months to research each

anomaly and correct errors or validate that the data are correct. GSA has

provided instructions to agencies on how to respond to the V&V process.

GSA also requires agencies to certify accuracy of the data and

established database rules that require agencies to submit complete

information on assets. GSA officials said that it must ultimately rely on

agencies to submit correct data.

FASTA required GSA to publish a single, comprehensive, descriptive

database of all federal real property by December 16, 2017, while

allowing it to exclude assets for reasons of national security, such as

those that are secure installations.

FASTA also required the database to

be made public to the extent its release is consistent with national

security and procurement laws.

GSA officials said that GSA used the

FRPP as the basis for developing the database it released to the public at

the end of 2017.

GSA presents the data in two ways: as a downloadable

spreadsheet or in a searchable mapping application.

GAO, High Risk Series: Progress on Many High Risk Areas, While Substantial Efforts

Needed on Others, GAO-17-317 (Washington, D.C.: Feb. 15, 2017) and GAO: High Risk

Series: Substantial Efforts Needed to Achieve Greater Progress in High Risk Areas,

GAO-19-157SP (Washington, D.C.: Mar. 6, 2019).

GSA-OGP creates database rules for the FRPP Management System, the software

system agencies use to submit FRPP data. The database rules establish what values the

system will accept for each data category. For example, the database rules will not allow

an agency to submit a blank value for a required data category, such as the asset type

(building, land, or structure).

Pub. L. 114-287, § 21.

For example, cost information could be withheld so they don’t affect procurement of

assets.

Although the public database is a subset of the FRPP, they are two separate, distinct

databases.

Page 7 GAO-20-135 Federal Real Property

FASTA requires that the public database be machine-readable and permit

searching or sorting of data to the extent practicable.

Further, GSA

guidance also calls for agencies to provide accurate and complete data.

Specifically, GSA requires agencies to include either a complete street

address or geo-coordinates for all 398,000 assets in the FRPP; for

example, GSA’s FRPP data dictionary establishes the format agencies

are to use when inputting asset addresses—number, street, city, zip

code. This requirement carries over to the 305,000 assets included in the

public database.

We found that almost 214,000 of the assets in the public database

included some street address information, but most of the addresses

were incomplete or incorrectly formatted. Specifically, only approximately

70,000 (33 percent) fully met the standards. Since another 91,000 assets

did not include a street address, a computer would only be able to locate

about 23 percent of the 305,000 civilian federal assets using street

addresses in the public database (See fig. 1.) GSA officials who manage

the FRPP said that they were aware that many street addresses were not

readable and have asked agency officials to review the accuracy of

address information and correct it in future submissions. They

acknowledged, however, that their efforts were not fully successful. As

discussed later, GSA is currently taking steps to ensure that agencies

provide more complete geo-coordinates when they submit data to the

FRPP.

Pub. L. No. 114-287, § 21. The term ‘machine-readable,’ when used with respect to

data, means data in a format that can be easily processed by a computer without human

intervention while ensuring no semantic meaning is lost. See 44 U.S.C. § 3502(18).

GSA guidance pertaining to annual FRPP data submittal also applies to the public

database.

GSA’s Efforts Have

Not Effectively

Addressed FRPP’s

Reliability Issues,

Which Affect the

Public Database

Most Street Addresses in

Public Data Are

Incomplete or Otherwise

Unusable

Page 8 GAO-20-135 Federal Real Property

Figure 1: Extent to Which Street Information Is Included in the General Services

Administration’s (GSA) Federal Real Property Profile (FRPP) and Public Database,

for Fiscal Year 2018

Agencies are not required to include street addresses for all locations. Agencies can submit street

addresses, geo coordinates, or both.

For the remaining 67 percent of the assets (144,000) with some street

address information that did not fully meet the standards, we found two

types of problems—incomplete addresses and addresses that were not

formatted correctly. First, more than 28,000 assets had street addresses

that were incomplete. For example, instead of having individual address

listings, we found that all 215 buildings at the Goddard Space Flight

Center had a single listing of “Greenbelt Road.” This road actually

stretches over 6 miles and many other buildings are located along the

road. The front gate’s complete address is “8800 Greenbelt Road.” In

these instances, GSA officials said that its public-mapping program

selects the mid-point of the street, which in this case is over a mile from

the public entrance to the installation. (See fig. 2.) As a result, someone

using the database would not be able to determine exactly where

Goddard is.

Page 9 GAO-20-135 Federal Real Property

Figure 2: Incomplete “Greenbelt Road” Street Address Provides the Public with

Inaccurate Locations for Buildings at the National Aeronautics and Space

Administration’s (NASA) Goddard Space Flight Center

Second, we found about 115,000 assets had street address information

that was incorrectly formatted based on FRPP instructions. While we did

not conduct a complete analysis of all these assets, we found examples

of some of the address issues, such as:

• Extra descriptive information about the property in the address field.

For example, “N220 AG Science Bldg North U of Kentucky” and

“Beltsville AG Research Center, 10300 Baltimore Avenue.” The data

in the address field for these two assets—which belong to the

Department of Agriculture—could not be directly read by a computer

or displayed on a map.

• Unrecognizable text. For example, “2881 F;B Road” and “1-15, Exit

172, 1 Mile East.” The data for these assets, which belong to the

Department of Agriculture, could not be directly read by a computer or

displayed on a map.

GSA officials said that users may be able to interpret the individual asset

addresses in the database but that GSA’s automated computer system

could not map unreadable addresses. Similarly, a private-sector user who

tried to use the public data to map federal facilities for clients said that he

was unable to map many of the assets because addresses were not

readable by his computer. As a result, he said that he excluded

incomplete or unreadable addresses from the database he created. He

noted that incomplete data would reduce clients’ interest.

Page 10 GAO-20-135 Federal Real Property

We also found problems with assets for which agencies provided geo-

coordinates (latitude and longitude). Specifically, GSA guidance states

that geo-coordinates must include a minimum of four decimal places. Of

the 305,000 assets included in the public database, almost 220,000

included geo-coordinates but more than half—about 141,000—did not

meet FRPP standards because they were not precise enough to map the

location of the assets.

GSA officials noted agencies are required to

enter some type of information in the field for address or geo-coordinates,

but an “open data” format did not prevent agencies from reporting

information that was not strictly a street and address number.

Consequently, some agencies may have entered incorrect values for the

geo-coordinates just to complete the field. Our analysis supports this

view; few (550 of about 131,000) of the assets with both sufficiently

detailed geo-coordinates and street addresses pointed to the same

location. In addition to the open data issue described above, officials also

explained that GSA did not have a “business validation rule” in place that

prevented agencies from inputting coordinates with less than four decimal

places.

GSA has taken a number of actions to correct the issues with geo-

coordinates that they say should help address this problem for the next

release of the public data in 2020. For example, GSA added V&V

anomaly categories for fiscal year 2018 data that identified GPS

coordinates pointing to unlikely locations, such as a location in the water,

which identified about 80,000 potential anomalies. Agencies are currently

checking these. Additionally, GSA added a feature to the fiscal year 2019

FRPP submission form that will force agencies to provide geo-

coordinates that are detailed enough for their data to be accepted. GSA

officials said that they would consider taking additional steps once they

have analyzed the results of the GPS coordinate anomaly categories.

GSA has asked agencies to review addresses for accuracy, and officials

indicated that they have discussed plans to improve this data. However,

GSA’s requirement that geo-coordinates be a minimum of four decimal places allows

precision to plus or minus 11.1 meters.

GSA’s guidance states that street addresses must be in a “geo-codable” format (i.e., an

address that can be mapped by Geographic Information System (GIS) software or used

by an overnight delivery service to deliver packages). An example of a geo-codable

address is “123 Main Street.” Further, the street address data can contain any

combination of letters and numerals up to 100 characters total, which is also called an

“open-data” format.

Page 11 GAO-20-135 Federal Real Property

GSA has not taken specific steps to work with agencies to ensure they

input correct street addresses in the public database in light of the “open

data” format. The lack of correct street addresses can affect users who

may be interested in acquiring or leasing assets or who may be interested

in installing telecommunications devices on an asset,

from knowing

exactly where those assets are located. As a result, until the street

address information is complete and correctly formatted, the public may

unknowingly pursue assets that are not available or suited to their needs.

We found that while GSA has identified close to 30,000 potential errors in

the FRPP database over the first 2 years of the V&V process, agencies

confirmed only 5 percent as errors (1,291 of 28,572). Agencies validated

the remaining 27,281 anomalies as correct or left them unresolved.

The

low number of errors being identified indicates that GSA’s V&V process is

not efficiently identifying errors in the data, either in terms of the anomaly

categories themselves or the thresholds at which GSA flags data as an

anomaly. This situation could ultimately mean that agencies are spending

time researching correct information that was flagged as potentially

erroneous or not fully actually researching anomalies and allowing

mistakes to remain uncorrected.

Agencies identified no anomalies as errors for five of GSA’s 16 anomaly

categories in 2017, raising questions about the anomaly categories GSA

has identified. OMB guidance suggests that agencies only do extra tasks

that are justified by their cost.

GSA officials who manage the V&V

process said that the high number of anomaly categories for which

agencies found no errors could reflect that the anomaly categories are

flagging correct data as anomalies or that agencies are validating data as

correct without actively checking it.

We found examples of both. For example, we examined a selected

sample of 14 V&V data anomalies at DOE sites in New Mexico. GSA

The MOBILE NOW Act amended FASTA to require the federal real property database to

include information on the ability of the asset to support private telecommunications

infrastructure. See Pub. L. No. 115-141, div. P. § 608(a) (2018).

As mentioned previously, GSA identified questionable entries (called anomalies) and

flags assets that are very small in size, changed from the previous year, or have unusual

financial statistics, among other things.

OMB, Circular A-11: Preparation, Submission, and Execution of the Budget

(Washington, D.C.: June 2019).

GSA’s V&V Process Does

Not Efficiently Identify

Erroneous Data

Page 12 GAO-20-135 Federal Real Property

flagged the buildings for being very small—office buildings less than 400

square feet and warehouses less than 64 square feet—and found that the

information in the public database was correct. Figure 3 illustrates how

such information flagged as being questionable, is actually correct

according to GSA’s reporting rules for agencies, which specify data

categories, such as the types of buildings GSA considers to be

warehouses. Specifically, GSA flagged assets at DOE’s Los Alamos and

Sandia National Laboratories because their square footage fell below

certain amounts. But, in reality, these assets met GSA’s criteria for offices

and warehouses despite being small.

Figure 3: Examples of Data Identified as Anomalous but Validated as Correct by the

Department of Energy

We also found instances where an agency verified information as correct

that was incorrect. Figure 4 illustrates examples data validated as correct

that was actually erroneous. Specifically, an agency erroneously reported

water towers and antenna arrays as office buildings. Staff responsible for

managing the V&V process for their agency’s assets said that they did not

always consult the personnel with the best knowledge of the assets in

resolving anomalies. Instead, they relied on their own judgment when

determining whether to forward the anomalies to asset managers to

ultimately check the data and correct any errors. This resulted in some

errors going uncorrected.

Page 13 GAO-20-135 Federal Real Property

Figure 4: Examples of Federal Assets Identified as Anomalous by the General Services Administration’s Validation and

Verification Process and Incorrectly Validated

Thresholds—the points at which GSA flags data as anomalies—lead to a

large number of data elements flagged, which can challenge the

resources of affected agencies. Officials at two of our selected agencies

said that the number of anomalies that the V&V process produces

Page 14 GAO-20-135 Federal Real Property

annually overwhelms their ability to validate the data. The large number of

unresolved V&V anomalies appears to support this conclusion. GSA’s

guidance allows agencies 10 months to validate the anomalous data, but

the number of anomalies that remain unresolved after 10 months has

risen sharply. Figure 5 shows that while agencies addressed all

anomalies in the first year, they have since struggled to keep up. As of

October 2019, 106,231 anomalies, or approximately 71 percent,

remained unresolved after 10 months.

Figure 5: Validation and Verification (V&V) Results for All Agencies as of October

2019

Data validated represent anomalies for which agencies affirmed that the data were correct.

Unresolved anomalies represent anomalies which the agencies neither determined were errors nor

affirmed were correct.

Errors identified represent anomalies that agencies determined were errors to be corrected.

Officials who are responsible for resolving anomalies at two selected

agencies said that more realistic anomaly categories or thresholds could

reduce the number of anomalies and better target actual errors, an

approach that could help agencies better prioritize their resources when

researching anomalies. GSA staff who manage the FRPP said that they

brainstormed internally and used industry standards and policy initiatives

to develop anomaly categories. They also explained that they adjust

thresholds within each category. However, GSA officials said they had not

reviewed the anomaly categories or their thresholds to see if they

consistently capture incorrect data. This approach puts the stated goals of

the V&V process—which are to improve data accuracy, promote data

Page 15 GAO-20-135 Federal Real Property

consistency among the agencies, and enable OMB to measure data

quality improvement—at risk.

In the absence of better information about

the validity of categories and thresholds, the current process for V&V is

taking up limited agency resources without efficiently correcting errors in

the data.

GSA and reporting agencies decided not to provide certain useful

information from the public database in two ways, thereby reducing the

data’s completeness and ultimately its utility. First, GSA withheld data

from the public database without consulting agencies about their

sensitivity. Second, selected agencies withheld information that was

already publicly available or withheld similar types of information

inconsistently within their agencies.

GSA chose to withhold 15 categories of data from the public database for

all agencies. FASTA authorized the withholding of information from the

public database for national security or procurement-related issues.

GSA officials who manage the FRPP said that GSA does not have the

security or intelligence expertise to issue guidance on national security

issues. As a result, they sought input from the ISC on what information to

withhold. ISC reviewed the security risks of FRPP data and provided

written recommendations in a memo to GSA in November 2017.

Specifically, ISC recommended that certain categories of data on assets

be withheld from the public database because of the security risk that

they could pose individually or in combination. ISC also recommended

that agencies use internal guidance on restricting the public release of

real property information and ISC’s mission criticality criteria

determine any individual real property assets to withhold entirely from the

public database.

OMB, Management Procedures Memorandum M-2016-01: Improving Federal Real

Property Data Quality – Required Data Validation and Verification Procedures

(Washington, D. C.; Jan. 28, 2016).

Pub. L. No. 114-287, § 21(c)(2).

ISC’s “mission criticality criteria” are based on the relative risk involved with the

mission(s) carried out by a facility’s federal tenants. The mission criticality criteria are one

set of criteria derived from ISC guidance. For more information, see ISC, The Risk

Management Process for Federal Facilities: An Interagency Security Committee Standard

(Washington, D. C.: Nov. 2016).

GSA and Agencies

Withheld Information

That Reduces the

Completeness of the

Public Database

GSA Withheld Data from

the Public Database

Page 16 GAO-20-135 Federal Real Property

GSA implemented ISC’s first recommendation by withholding 15 FRPP

data categories for all assets from the public database without consulting

the relevant agencies on this decision, considering the specific sensitivity

of these categories for all assets, or assessing the effect withholding them

would have on the database.

ISC officials acknowledged that the memo

that they prepared for GSA could have been clearer as to ISC’s intent that

departments and agencies should consider the recommendations in

making a final determination. According to ISC officials, they believed that

implementation would involve GSA communicating these

recommendations and leaving decisions on what to withhold to officials

within individual departments and agencies who control real property

assets.

Specifically, the following five categories of data were among the 15

withheld by GSA:

• property’s/installation’s name,

• replacement value of an asset,

• annual operating and maintenance costs for owned assets,

• annual-operating and maintenance costs for leased assets, and

• breakdown of annual operating and maintenance costs (e.g., utilities

costs, janitorial costs, sewage costs, etc.).

Because GSA did not consult with agencies on this decision, the

agencies did not have an opportunity to consider whether or not the 15

data categories GSA withheld included information that is sensitive or

already publicly available. As a result, the public database is incomplete

in ways that adversely affect users and limits agencies’ public

accountability for reporting accurate information. For example, identifying

assets in the public database is difficult without the property’s name—one

of the data categories GSA withheld—especially given the insufficient

location data in the database discussed earlier. Returning to the

incomplete address example discussed earlier (NASA Goddard Space

Flight Center), the public data also do not include the property’s name,

“Goddard Space Flight Center,” leaving users with limited information to

identify the buildings. As a result, someone using the public database

cannot identify assets on NASA’s Goddard Space Flight Center campus

without using outside sources for additional information. (See table 1.)

See Appendix I for a list of the 15 withheld data categories.

Page 17 GAO-20-135 Federal Real Property

Table 1: Example of Information Included and Not Included in the Public Data for a National Aeronautics and Space

Administration (NASA) Building

Source: GAO analysis of General Services Administration data. | GAO-20-135

Note: This table shows a selection of data categories included and not in the Public Database. A list

of all categories withheld from the public database is in appendix I.

As discussed in the next section, we found that some of the information

from these 15 excluded data categories, such as property names, is often

already in the public sphere. For example, “Goddard Space Flight Center”

and its address are clearly disclosed on NASA’s public website, but GSA

withheld the name for 215 NASA buildings at this address, including

Goddard’s public visitors’ center. Using the public database alone, a

member of the public would need to go through numerous steps to

determine if assets are part of Goddard Space Flight Center and still have

no way of being sure. (See fig. 6.)

Information

Included in Public Database

Not Included in Public Database

Agency

NASA

Street Address

Greenbelt Road, Greenbelt, MD 20771

Real Property Type

Building

Real Property Use

Warehouse

Square Feet

Legal Interest

Owned

Status

Current Mission Need

Utilization

Utilized

Year of Asset Construction

2015

Installation Name

Goddard Space Flight Center

On Secure Installation Not Open to Public

Yes

Number of Buildings at Same Installation

215

Operating and Maintenance Costs (actual

dollars)

$87

Replacement Value

(actual dollars)

$4,019

Location of front gate

8800 Greenbelt Road

Page 18 GAO-20-135 Federal Real Property

Figure 6: Steps for a User to find National Aeronautics and Space Administration’s (NASA) Goddard Space Flight Center’s

Buildings in the Public Database

Moreover, third-party, private sector stakeholders we spoke with such as

brokers, lessors, consultants, and a non-profit organization that work in

federal real-property markets, noted that some of the data categories

GSA withheld would be among the most useful to their work. For

example, 10 of 14 stakeholders we spoke to said that financial data, such

as operating costs and annual rent, were among the most useful FRPP

data categories to their analyses of real property markets and business

opportunities. Additionally, four stakeholders cited the property’s name as

among the most important data categories for their work in analyzing

federal real property.

While GSA withheld the 15 categories of data across all agencies, it

allowed each agency to determine if any specific assets should be

withheld entirely from the public database, in accordance with ISC’s

second recommendation. ISC officials told us that this was appropriate

because individual departments and agencies that control real property

assets should determine what information to withhold.

GSA provided agencies with guidance that explained its decision to

withhold the 15 data categories and instructed agencies to consult ISC’s

mission criticality criteria and any additional internal agency criteria in

determining what information to withhold from public release. ISC’s

mission criticality criteria provide a page-long list of uses of real property

assets that warrant consideration for national security exclusion, but do

not provide other instructions for agencies to consult while making

decisions on what information to withhold. Further, OMB Circular—

Management of Reporting and Data Integrity Risk also instructs agencies

to integrate a risk-based approach towards meeting reporting objectives,

Agencies Withheld

Publicly Available

Information and Withheld

Similar Assets

Inconsistently, Making

Analysis Difficult

Page 19 GAO-20-135 Federal Real Property

an approach that requires “management practices that identify, assess,

respond, and report on risks.”

However, we found that our selected

agencies did not consistently identify internal guidance to supplement

GSA’s instructions within their agencies.

In September 2018, ISC recommended that GSA not withhold from the

public database newly added data categories that provide information

already in the public sphere. Additionally, the OPEN Government Data

Act requires OMB to foster greater sharing, dissemination, and access to

public information and issue guidance that, among other things, takes into

account the requirement that data must be disclosed if it would otherwise

be made available under a Freedom of Information Act request.

For

purposes of this report, we refer to this requirement as “assuming

openness.”

However, GSA’s instructions to agencies lacked specifics to help

agencies apply a consistent, risk-based approach in determining which, if

any, assets or asset-specific information should be withheld from public

release. As a result, we found that some of the selected agencies

withheld asset-related information from the public database that is

available on their own public websites or from other official sources.

Withholding information that is already publicly available unnecessarily

reduces the completeness and utility of the public database that FASTA

indicated should be comprehensive. For example:

• DHS’s Immigration and Customs Enforcement (ICE) withheld

buildings at five of its publicly-accessible service-processing centers

that are shown on a detention facility locator mapping system on its

own website. ICE officials told us that they did not consider what

information is already publicly available when deciding what

information to withhold from the public database.

• FCC withheld all of its real property assets. FCC’s own website and

regulations, however, list the locations and functions of FCC offices.

• The U.S. Coast Guard withheld information on its public-recruiting

offices and lighthouses that it advertises on its public website. All

OMB, “Appendix A,” Management of Reporting and Data Integrity Risk, OMB Circular

No. A-123 (Washington, D.C.: June 6, 2018).

Pub. L. No. 115-435, § 202(b) (2019).

47 C.F.R § 0.121.

Page 20 GAO-20-135 Federal Real Property

buildings and structures that were not specifically used for the

purpose of aids to navigation were withheld from the public data set.

As a result public users can look up information on the Coast Guard’s

aids to navigation, but cannot look up some of its publicly accessible

locations, such as recruiting offices and lighthouses.

In contrast, DOE decided to withhold none of its 20,378 assets from the

public database. According to a DOE official responsible for submitting

data to FRPP, DOE does not have a specific process for assessing what

properties to make public. However, it is aware that much of the

information in the public database is also publicly available through other

sources. Table 2 shows how selected agencies took different approaches

to withholding information from the public database.

Table 2: Summary of Information Selected Agencies Withheld from Public Database for Fiscal Year (FY) 2018

Source: GAO analysis of GSA data. | GAO-20-135

Notes: We did not include our other selected agencies, the Department of Energy and General

Services Administration, because they did not withhold any assets.

Under risk-based criteria assuming openness (as mentioned earlier),

agencies may consider whether information made public in one instance

should be withheld in another instance. However, neither ISC’s mission

criticality criteria nor GSA’s instructions addressed the issue of

consistency within specific agencies. Specifically, we found that selected

agencies withheld

• the same assets differently over time, and

• similar assets inconsistently.

U.S. Coast Guard officials said that this decision caused information on lighthouses to

be released, but not information on other assets at lighthouse sites.

Withheld in FY 2018

Number of Assets

Withheld

Total Number of Assets

(overall and by agency)

Percent

Withheld

Total Overall

93,246

397,993

23%

By Selected Agency

Department of Homeland Security

23,611

52,192

45%

National Aeronautics and Space

Administration

3,696

5,370

69%

Federal Communications Commission

100%

Department of the Interior

2,226

151,429

Page 21 GAO-20-135 Federal Real Property

Table 3 shows how reporting agencies made different decisions on

whether to withhold the same types of assets. At times, some agencies

withheld certain asset types that ISC’s mission criticality criteria did not

identify as warranting withholding, resulting in almost 7,000 assets such

as parking structures and disposed assets being withheld.

This led to

inconsistencies as to whether these agency assets were included or not

in the public database, limited transparency about these assets, and

prevented users from fully analyzing federal real property assets in these

categories.

Table 3: Summary of Selected Information All Agencies Withheld from the Public Database for Fiscal Year (FY) 2018 across

All Reporting Agencies

Source: GAO analysis of General Services Administration (GSA) Data. | GAO-20-135

In other cases, selected agencies withheld similar assets inconsistently,

did not always follow written procedures and withheld similar assets. For

example:

• DOI headquarters provided its bureaus with GSA’s instructions on

withholding assets, but individual bureaus applied the instructions

differently. For example:

• The Fish and Wildlife Service reports that it has 369 publicly

accessible national wildlife refuges, but it withheld selected real

property assets at 11 of them. However, the withheld assets are

the same types as the assets the Service disclosed at other

refuges. For example, it reported all but two of 447 restrooms and

10 of 2,066 recreational structures on its national wildlife refuges.

The Fish and Wildlife Service told us it will re-evaluate its

withholding for the fiscal year 2019 FRPP database.

• The National Park Service (NPS) reported that it has 374 publicly

accessible national parks, monuments, memorials, historic sites,

Disposed assets are no longer in an agency’s inventory.

Withheld in FY 2018

Number of Assets

Withheld

Total Number of Assets

in Category

Percent of Assets in

Category Withheld

Disposed Assets removed from agency inventory

3,272

6,452

51%

Parking Structures

1,824

19,003

10%

Recreational Structures

1,538

26,392

Monuments/memorials/museums

204

1,899

11%

Page 22 GAO-20-135 Federal Real Property

and recreation areas. NPS withheld some real property assets

from 15 of those sites. For example, it reported all but 2 of 1,045

service buildings at its sites. These withheld assets are the same

types as those disclosed at other sites.

• NASA withheld assets at a centralized level, but headquarters officials

told us that they have not established instructions or policies for these

decisions. NASA officials told us that they withhold real property

assets shared with agencies working in defense and/or national-

security, which led NASA to withhold 1,517 assets in fiscal year 2017.

In fiscal year 2018, however, we found that NASA withheld all assets

at certain field centers, causing the number to more than double from

1,517 in fiscal year 2017 to 3,696 in fiscal year 2018.

Finally, our comparison of the fiscal year 2018 FRPP and public

databases found that seven agencies did not identify whether data on

3,845 assets should be withheld despite GSA guidance to do so for every

asset. GSA included these assets in the public database without

consulting agencies on the assets’ sensitivity or risks in releasing

information on them. GSA officials said that these data should not have

been accepted and that they had implemented controls to ensure that

agencies identify whether data should be withheld.

It is difficult for a user of the public database to determine when assets

are located on a secure installation that the public cannot access. For

example, returning to the NASA Goddard Space Flight Center illustration

from earlier in the report, assets located at the Space Flight Center are

listed individually, with no indication that the assets are all located on a

secure installation. The public database lists all 215 assets at the same

location—Greenbelt Road in Greenbelt, MD, but provides no further

indication that the assets are part of a larger, secure facility. (See fig. 7.)

Data Presentation

and Limited

Stakeholder

Awareness Hinder

Usefulness of the

Public Database

Data Presentation Issues

Limit the Usefulness of the

Public Database

Page 23 GAO-20-135 Federal Real Property

Figure 7: National Aeronautics and Space Administration’s (NASA) Goddard Space

Flight Center Buildings: Current Reporting as Individual Assets Compared to

Reporting as a Campus

Currently, GSA requires civilian agencies to report individual assets,

including those on secure installations. Detailed, asset-specific

information could be useful for government decision makers, and GSA

applied this approach to the public database. However, asset-level

information can cause challenges for users when they are located on

secure installations because GSA withheld the installation names from

the public database.

Listing assets individually could prompt fruitless public interest in

inaccessible secure facilities. One expected use of the public database is

for the private sector to identify possible locations for installing

commercial telecommunications infrastructure, such as cell towers and

antennas.

However, as this infrastructure cannot be installed on secure

installations, the public database would be more useful to such

companies if they could readily determine whether a potential location

was on a secure installation or not. For example, officials on a secure

installation we visited told us that reporting individual buildings does not

make sense because there are few, if any, legitimate reasons for public

interest in the individual assets on a secure installation.

The Mobile Now Act requires that real property database include information on the

ability of the asset to support telecommunications infrastructure. Pub. L. No.115-141, §

608 (a).

Page 24 GAO-20-135 Federal Real Property

FASTA required GSA to develop a comprehensive database and provide

the public with database access, but recognized the importance of

protecting national security. In that respect, a key organizational issue

faced by GSA and agencies is how to present data for reporting assets on

campuses that are not accessible to the public. While non-disclosure is

permitted, such actions to withhold this information may reduce the

usefulness of the public database as a whole.

The Department of Defense (DOD) takes a different approach for its

secure military bases in the public database. According to GSA officials,

DOD submits a separate summary-level report for public release. This

summary-level information shields sensitive information and alerts users

that those assets are not accessible or of use to private-sector interests.

Civilian agencies’ assets located on closed federal installations are similar

to those on DOD bases in that the public may have less interest in or

reason for knowing about assets that are not available to the public.

Officials from NASA and two DHS bureaus said that the installation-level

approach to reporting would be more appropriate for their circumstances

than the asset-level reporting currently applied to civilian agencies and

would likely allow them to release more information to the public. Officials

from DHS added that they already release some information to the public

on the web site. We found that other selected agencies also release

information about secure installations on their public websites, including

NASA and its Goddard Space Flight Center.

In our interviews with 14 private sector stakeholders, we found varying

levels of awareness and understanding of GSA’s publicly available real-

property datasets. Of the 14 private sector stakeholders we interviewed,

eight told us that they were aware of the public database. Of these, five

told us they tried to use it. Several selected stakeholders—regardless of

whether or not they had used the database—cited concerns about the

usefulness of the data, specifically with its reliability, completeness,

formatting, and organization. For example, officials from one brokerage

firm told us that, while the information could theoretically be useful for

agency consolidation efforts, the database was too cumbersome to

analyze for that purpose. Similarly, officials with a federal real-estate-

consulting firm told us that they do not refer customers to the public

database because they believe that the data are not complete, correct, or

intuitive. Moreover, one member of a federal real-property trade

association noted serious limitations in the database’s completeness and

organization. In addition, one user said that he hoped the public release

would allow better access to real property data but that the poor quality,

Stakeholders’ Lack of

Awareness of the Public

Database and Confusion

with Other Databases

Limits Usefulness

Page 25 GAO-20-135 Federal Real Property

completeness, and organization of the data means access to data is no

better than it was before the release.

Further, six of the private sector stakeholders we interviewed were not

aware of the public database, including a stakeholder who confused it

with GSA’s Lease Inventory database. The lack of a single location on

GSA’s website that contains information about all of GSA’s real property

databases may contribute to the awareness, confusion, and usefulness

issues expressed by these stakeholders. Specifically, public access to the

FRPP public database, the GSA’s Lease Inventory database and two

other publicly available real-property databases is found in different

places on GSA’s website:

• Public FRPP http://publicfrppdata.realpropertyprofile.gov (managed by

GSA’s Office of Government-wide Policy)

• GSA lease inventory https://www.gsa.gov/real-estate/real-estate-

services/leasing-policy-procedures/lease-inventory (managed by

GSA’s Office of Leasing)

• GSA inventory of owned & leased properties

https://www.gsa.gov/tools/buildings-real-estate-etools/inventory-of-

owned-and-leased-properties (managed by GSA’s Public Building

Services)

• GSA disposal inventory https://disposal.gsa.gov/s/ (managed by

GSA’s Office of Property Disposal)

The Open Government Data Act requires the Administrator of GSA to

maintain a single public interface online as a point of entry dedicated to

sharing an agency’s data assets with the public.

While the databases

serve different purposes, some asset-level data are similar, such as

location or size. According to a GSA official, these databases are

operated by different offices within GSA. This situation poses challenges

to listing the database on a consolidated webpage. Nevertheless, GSA

officials agreed that there could be clearer links and said that they plan to

add them based on our findings. Without a consolidated webpage or clear

links showing how the databases relate to each other and how to access

each database, users of the various databases may not be aware of what

databases do exist to search for assets that could be available to the

public.

Pub. L. No. 115-435, § 202.

Page 26 GAO-20-135 Federal Real Property

The public database’s presentation issues, combined with stakeholder

confusion and lack of awareness, could contribute to low numbers of

people who accessed the database compared to another GSA-managed

real property database. GSA data indicate that users accessed civilian

agency data from the public database 147 times per month on average

from December 2017 through July 2019 and some months fewer than 10

times.

However, according to a GSA official, the number of times users

access the public database through the GSA website doesn’t necessarily

reflect the extent to which people use the data. The official explained that,

since GSA only issues the data once a year, users only need to access

and save it once for use in a given year and that GSA usually sees a

peak in users accessing the data when GSA publishes its annual update

to the database. As indicated in figure 8, there was a peak in users

accessing the database when GSA first issued the 2016 data in

December 2017, and again in March and April 2018 when GSA published

2017 data (28 and 162 times, respectively), and in June 2019 when GSA

published the 2018 data (170 times). In comparison, users access

another real property database, GSA’s Inventory of Owned and Leased

Property database—which is updated weekly—more often than they

access the public database. Users access the Inventory of Owned and

Leased Property database to search for properties controlled by GSA.

Specifically, since the public database was released in December 2017,

the public has continued to access GSA’s Inventory of Owned and

Leased Property almost 10 times more per month than the public

database on average (see fig. 8).

The FRPP Public Database includes data for both civilian and military agencies. Our

report focuses on the civilian database and users.

Data’s Presentation Issues

May Affect the Level of

Use

Page 27 GAO-20-135 Federal Real Property

Figure 8: Frequency of Accessing the General Services Administration’s (GSA) Inventory of Owned and Leased Property

Database and Public Real Property Database from December 2017 through July 2019

Federal agencies spend billions of dollars annually to operate and

maintain hundreds of thousands of real property assets. GSA’s public

database, extracted from FRPP data, is a comprehensive, descriptive,

database of federal real property. Through the database, the public

should be able to learn about federal assets, whether people are

conducting research or interested in potential uses such as leasing or

purchasing. Issues with the data, however, undermine these uses. GSA

has taken a number of actions to improve the accuracy of the data, such

as implementing the V&V process for identifying and correcting possible

errors. But until GSA has better processes to ensure accuracy of street

address information and identify anomalies, the public data will continue

to lack the type of database most useful to the public. Moreover, the

absence of a risk-based, consistent approach for withholding assets from

the public database or reporting assets to it further erodes its utility.

Finally, utilization of the data base is low; GSA’s choices on how the

Conclusions

Page 28 GAO-20-135 Federal Real Property

database information is presented and how users find out about and

access the public database and other real-property databases may

contribute to this lack of use. Unless GSA improves the accuracy,

completeness, and usefulness of the public database, its intended

benefits—to the public and the federal government—will remain

unrealized.

We are making the following six recommendations to GSA:

The Administrator of GSA should coordinate with agencies to ensure that

street address information in the public database is complete and

correctly formatted. (Recommendation 1)

The Administrator of GSA should coordinate with agencies to review V&V

anomaly categories to better target incorrect data. (Recommendation 2)

The Administrator of GSA should work in consultation with agencies to

determine which, if any, data should be withheld from public release.

(Recommendation 3)

The Administrator of GSA should instruct each agency to apply a

consistent, risk-based approach in determining which, if any, assets or

asset-specific information should be withheld from public release.

(Recommendation 4)

The Administrator of GSA should allow agencies to provide summary

data for secure installations. (Recommendation 5)

The Administrator of GSA should link all of GSA’s publicly available real-

property data sources. (Recommendation 6)

We provided a draft of this report to GSA, DHS, DOE, DOI, FCC and

NASA for comment. GSA provided written comments, which are reprinted

in appendix II and summarized below. We received, via email from DOI,

technical comments, which we incorporated as appropriate. DOI, in its

email comments, also suggested revisions to two recommendations,

which we clarified as appropriate. DHS and NASA provided, in email,

technical comments, which we incorporated as appropriate. DOE and

FCC told us they had no comments.

Recommendations

Agency Comments

and Our Evaluation

Page 29 GAO-20-135 Federal Real Property

GSA agreed with five of our six recommendations but disagreed with our

third recommendation. GSA wrote that allowing agencies to unilaterally

determine which categories of data to withhold from the public would not

be useful and would complicate comparisons among agencies.

We did not intend that our recommendation allow agencies to decide

without consulting with GSA, and we have clarified our recommendation

accordingly. We continue to believe this recommendation, as clarified, is

valid.

As we reported, GSA currently withholds 15 variables—categories of

data—for all federal assets, including the name of every federal building

and structure. While this approach is consistent for all assets, it reduces

the overall usefulness of the data by withholding information that federal

agencies already make public.

In addition, the ISC told us that the landholding agencies, not GSA, are in

the best position to know what data about their assets are sensitive. We

amended the recommendation by removing the reference to categories of

data and adding that GSA work in consultation with agencies to

determine what data to withhold. This change would create a consistent

way for agencies to release useful data while withholding sensitive data

for individual assets, a step they already take by withholding assets from

the public database. GSA plans to work with the ISC and federal

agencies to review related guidance and modify it as needed. We

support these plans.

In addition, DOI suggested in email comments that we revise our second

recommendation to include coordinating with agencies to review V&V

anomaly categories to better target incorrect data. Our original

recommendation did not preclude coordination, and since we agree that

such coordination would help improve the V&V process, we clarified the

recommendation accordingly.

We are sending copies of this report to the appropriate congressional

committees, the Administrator of the General Services Administration, the

Acting Secretary of Homeland Security, the Secretary of Energy, the

Secretary of the Interior, Chair of the Federal Communication

Commission, the Administrator of the National Aeronautics and Space

Administration, and other interested parties. In addition, the report is

available at no charge on the GAO website at http://www.gao.gov.

Page 30 GAO-20-135 Federal Real Property

If you or your staff have any questions about this report, please contact

me at (202) 512-2834 or r[email protected]. Contact points for our

Offices of Congressional Relations and Public Affairs may be found on

the last page of this report. GAO staff who made key contributions to this

report are listed in appendix III.

Lori Rectanus

Director, Physical Infrastructure

Appendix I: Categories of Data Withheld from

the Public Database

Page 31 GAO-20-135 Federal Real Property

Table 4: The 15 Categories of Data in the Federal Real Property Profile That the

General Services Administration (GSA) Withholds from the Public Database

GSA withholds these categories for all agencies’ real property assets

Data Category Description

Reporting bureau The bureau (within an agency) that reports

the asset

Using organization The entity that uses the asset

Replacement value The cost to replace an existing asset

Repair needs Non-recurring costs necessary to restore

asset to original state

Condition index Condition of the asset based the

replacement value and repair needs

Owned and otherwise managed annual

operating and maintenance costs

Costs related to the everyday functions of

an asset owned and/or managed

Leased annual rent to lessor Rent paid to lessor minus the annual

operating and maintenance costs

Leased annual operating and maintenance

costs

Costs related to the everyday functions of a

leased asset

Annual operating and maintenance costs Costs related to the everyday functions of

an asset

Installation identifier Code to identify an installation (i.e.

buildings, structures, land or any

combination of these)

Sub-installation identifier Code to identify a part of an installation (i.e.

buildings, structures, land or any

combination of these)

Installation name Building name or the name of an entire

installation (such as an agency campus)

Number of federal employees Total number of full and part time federal

employees

Number of federal contractors Total number of full and part time contract

employees

Field office Identifies whether an asset is part of a field

office (any location that is not the

headquarters location for the agency)

Source: GAO Analysis of GSA’s federal real property reporting requirements | GAO-20-135

Appendix I: Categories of Data Withheld

from the Public Database

Appendix II: Comments from the U.S. General

Services Administration

Page 32 GAO-20-135 Federal Real Property

Appendix II: Comments from the U.S.

General Services Administration

Appendix II: Comments from the U.S. General

Services Administration

Page 33 GAO-20-135 Federal Real Property

Appendix III: GAO Contact and Staff

Acknowledgments

Page 34 GAO-20-135 Federal Real Property

Lori Rectanus, (202) 512-2834 or rectanusl@gao.gov

In addition to the contact named above, Keith Cunningham (Assistant

Director), Lynn Filla-Clark (Analyst-in-Charge), Melissa Bodeau, George

Depaoli, James Duke, Rami Khalfani, Terence Lam, John Mingus,

Joshua Ormond, Crystal Wesco, and Elizabeth Wood made key

contributions to this report.

Appendix III: GAO Contact and Staff

Acknowledgments

GAO Contact

Staff

Acknowledgments

(103032)

The Government Accountability Office, the audit, evaluation, and investigative

arm of Congress, exists to support Congress in meeting its constitutional

responsibilities and to help improve the performance and accountability of the

federal government for the American people. GAO examines the use of public

funds; evaluates federal programs and policies; and provides analyses,

recommendations, and other assistance to help Congress make informed

oversight, policy, and funding decisions. GAO’s commitment to good government

is reflected in its core values of accountability, integrity, and reliability.

The fastest and easiest way to obtain copies of GAO documents at no cost is

through our website. Each weekday afternoon, GAO posts on its website newly

released reports, testimony, and correspondence. You can also subscribe to

GAO’s email updates to receive notification of newly posted products.

The price of each GAO publication reflects GAO’s actual cost of production and

distribution and depends on the number of pages in the publication and whether

the publication is printed in color or black and white. Pricing and ordering

information is posted on GAO’s website, https://www.gao.gov/ordering.htm.

Place orders by calling (202) 512-6000, toll free (866) 801-7077, or

TDD (202) 512-2537.

Orders may be paid for using American Express, Discover Card, MasterCard,

Visa, check, or money order. Call for additional information.

Connect with GAO on Facebook, Flickr, Twitter, and YouTube.

Subscribe to our RSS Feeds or Email Updates. Listen to our Podcasts.

Visit GAO on the web at https://www.gao.gov.

Contact FraudNet:

Website: https://www.gao.gov/fraudnet/fraudnet.htm

Automated answering system: (800) 424-5454 or (202) 512-7700

Orice Williams Brown, Managing Director, WilliamsO@gao.gov, (202) 512-4400,

U.S. Government Accountability Office, 441 G Street NW, Room 7125,

Washington, DC 20548

Chuck Young, Managing Director, youngc1@gao.gov, (202) 512-4800

U.S. Government Accountability Office, 441 G Street NW, Room 7149

Washington, DC 20548

James-Christian Blockwood, Managing Director, spel@gao.gov, (202) 512-4707

U.S. Government Accountability Office, 441 G Street NW, Room 7814,

Washington, DC 20548

GAO’s Mission

Obtaining Copies of

GAO Reports and

Testimony

Order by Phone

Connect with GAO

To Report Fraud,

Waste, and Abuse in

Federal Programs

Congressional

Relations

Public Affairs

Strategic Planning and

External Liaison

Please Print on Recycled Paper.