E-Authentication Guidance for Federal Agencies

EXECUTIVE OFFICE OF THE PRESIDENT
OFFICE OF MANAGEMENT AND BUDGET
WASHINGTON, D.C. 20503
The Director

December 16, 2003
M-04-04
MEMORANDUM TO THE HEADS OF ALL DEPARTMENTS AND AGENCIES
FROM:

Joshua B. Bolten
Director

SUBJECT:

E-Authentication Guidance for Federal Agencies

The Administration is committed to reducing the paperwork burden on citizens and
businesses, and improving government response time to citizens – from weeks down to minutes.
To achieve these goals, citizens need to be able to access government services quickly and easily
by using the Internet. This guidance document addresses those Federal government services
accomplished using the Internet online, instead of on paper. To make sure that online
government services are secure and protect privacy, some type of identity verification or
authentication is needed.
The attached guidance updates guidance issued by OMB under the Government
Paperwork Elimination Act of 1998, 44 U.S.C. § 3504 and implements section 203 of the EGovernment Act, 44 U.S.C. ch. 36. This guidance also reflects activities as a result of the EAuthentication E-Government Initiative and recent standards issued by the National Institute of
Standards and Technology (NIST). In preparing this guidance, we have worked closely with and
incorporated comments from agency Chief Information Officers.
This guidance takes in account current practices in the area of authentication (or eauthentication) for access to certain electronic transactions and a need for government-wide
standards and will assist agencies in determining their authentication needs for electronic
transactions. This guidance directs agencies to conduct “e-authentication risk assessments” on
electronic transactions to ensure that there is a consistent approach across government. (see
Attachment A). It also provides the public with clearly understood criteria for access to Federal
government services online. Attachment B summarizes the public comments received on an
earlier version of this guidance.
For any questions about this guidance, contact Jeanette Thornton, Policy Analyst,
Information Policy and Technology Branch, Office of Management and Budget, phone (202)
395-3562, fax (202) 395-5167, e-mail: [email protected].

Attachments
Attachment A – E-Authentication Guidance for Federal Agencies
Attachment B – Summary of Public Comments and Responses

Attachment A

E-Authentication Guidance for Federal Agencies
Section 1:
Section 2:
Section 3:
Section 4:
Section 5:

Introduction
Assurance Levels and Risk Assessments
Assessing Confidence in Credential Service Providers
Implementing an Authentication Process
Effective Dates of Guidance

1. Introduction
1.1. Summary
This guidance requires agencies to review new and existing electronic transactions to ensure that
authentication processes provide the appropriate level of assurance. It establishes and describes
four levels of identity assurance for electronic transactions requiring authentication. Assurance
levels also provide a basis for assessing Credential Service Providers (CSPs) on behalf of
Federal agencies. This document will assist agencies in determining their e-government
authentication needs. Agency business-process owners bear the primary responsibility to
identify assurance levels and strategies for providing them. This responsibility extends to
electronic authentication systems.
Agencies should determine assurance levels using the following steps, described in Section 2.3:
1. Conduct a risk assessment of the e-government system.
2. Map identified risks to the applicable assurance level.
3. Select technology based on e-authentication technical guidance.
4. Validate that the implemented system has achieved the required assurance level.
5. Periodically reassess the system to determine technology refresh requirements.
1.2. Scope
•

This guidance applies to remote authentication of human users of Federal agency IT
systems for the purposes of conducting government business electronically (or egovernment). Though that authentication typically involves a computer or other
electronic device, this guidance does not apply to the authentication of servers, or other
machines and network components.

•

This guidance is intended to help agencies identify and analyze the risks associated with
each step of the authentication process. The process includes (but is not limited to)
identity proofing, credentialing, technical and administrative management, record
keeping, auditing, and use of the credential. Each step of the process influences the
technology’s overall conformance to the desired assurance level.

•

This guidance supplements OMB Circular A-130, Management of Federal Information
Resources, Appendix II, Implementation of the Government Paperwork Elimination Act
(GPEA).

•

This guidance does not directly apply to authorization. Authorization focuses on the
actions permitted of an identity after authentication has taken place. Decisions
concerning authorization are and should remain the purview of the business process
owner.

•

This guidance does not address issues associated with “intent to sign,” or agency use of
authentication credentials as electronic signatures. For more information on electronic
signatures, see the OMB guidance on implementing GPEA1 and the Electronic
Signatures in Global and National Commerce Act, Pub. L. No. 106-2292.

•

This guidance does not identify which technologies should be implemented. The
Department of Commerce’s National Institute for Standards and Technology (NIST) is
developing complementary e-authentication technical guidance that agencies will use to
identify appropriate technologies, based on the analysis process described here.

•

This document does not confer, and may not be used to support, any right on behalf of
any person or entity against the United States or its agencies or officials.

1.3. Overview
This document provides agencies with guidance on electronic authentication (eauthentication). The National Research Council report, “Who Goes There? Authentication
Through the Lens of Privacy”3 defines e-authentication as the process of establishing
confidence in user identities electronically presented to an information system. It defines
individual authentication as the process of establishing an understood level of confidence that
an identifier refers to a specific individual.
Authentication focuses on confirming a person’s identity, based on the reliability of his or
her credential. Authorization focuses on identifying the person’s user permissions.
To successfully implement a government service electronically (or e-government), Federal
agencies must determine the required level of assurance in the authentication for each
transaction. This is accomplished through a risk assessment for each transaction. The
assessment identifies:
a) risks, and
b) their likelihood of occurrence.
OMB Circular A-130, Management of Federal Information Resources, states that agencies
must prepare and update a strategy that identifies and mitigates risks associated with each
information system. This guidance will help agencies map identified risks to corresponding
assurance levels.

1 OMB Memorandum M-00-10, 4/25/00, http://www.whitehouse.gov/omb/memoranda/m00-10.html
2 OMB Memorandum M-00-15, 9/25/00, http://www.whitehouse.gov/omb/memoranda/m00-15.html
3 March 31, 2003, http://www.nap.edu/books/0309088968/html/

Section 5 of the GPEA guidance details the risk factors agencies should consider in planning
and implementing e-government transactions and systems. This document expands on
Section 5 by instructing agencies how to implement e-authentication processes by
•

outlining a process for assessing risk,

•

describing four levels of identity assurance, and

•

explaining how to determine the appropriate level of identity assurance.

1.4. Applicability
Not all Federal electronic transactions4 require authentication; however, this guidance
applies to all such transactions for which authentication is required, regardless of the
constituency (e.g. individual user, business, or government entity).
Transactions not covered by this guidance include those that are associated with national
security systems as defined in 44 U.S.C. § 3542(b)(2). Private-sector organizations and state,
local, and tribal governments whose electronic processes require varying levels of assurance
may consider the use of these standards where appropriate.
There are two types of individual authentication:
a) Identity authentication—confirming a person’s unique identity.
b) Attribute authentication—confirming that the person belongs to a particular group
(such as military veterans or U.S. citizens).
Attribute authentication is the process of establishing an understood level of confidence that
an individual possesses a specific attribute. If the attribute does not provide ties to the user’s
identity; it would be considered an anonymous credential (discussed further in Section 4.2).
Attribute authentication is not specifically addressed in this document, however agencies
may accept ‘anonymous credentials’ in certain contexts.

2. Assurance Levels and Risk Assessments
2.1. Description of Assurance Levels
This guidance describes four identity authentication assurance levels for e-government
transactions. Each assurance level describes the agency’s degree of certainty that the user has
presented an identifier (a credential5 in this context) that refers to his or her identity. In this
context, assurance is defined as 1) the degree of confidence in the vetting process used to
establish the identity of the individual to whom the credential was issued, and 2) the degree
of confidence that the individual who uses the credential is the individual to whom the
credential was issued.
The four assurance levels are:
4 For the purposes of this document, a transaction is defined as: a discrete event between user and systems that supports a business or
programmatic purpose.
5 A credential is defined as: an object that is verified when presented to the verifier in an authentication transaction.

•

Level 1: Little or no confidence in the asserted identity’s validity.

•

Level 2: Some confidence in the asserted identity’s validity.

•

Level 3: High confidence in the asserted identity’s validity.

•

Level 4: Very high confidence in the asserted identity’s validity.

2.2. Risks, Potential Impacts, and Assurance Levels
While, this guidance addresses only those risks associated with authentication errors, NIST
Special Publication 800-30, “Risk Management Guide for Information Technology
Systems,” recommends a general methodology for managing risk in Federal information
systems. In addition, other means of risk management, (e.g., network access restrictions,
intrusion detection, and event monitoring) may help reduce the need for higher levels of
authentication assurance.
Potential Impact Categories: To determine the appropriate level of assurance in the user’s
asserted identity, agencies must assess the potential risks, and identify measures to minimize
their impact. Authentication errors with potentially worse consequences require higher levels
of assurance. Business process, policy, and technology may help reduce risk. The risk from
an authentication error is a function of two factors:
a) potential harm or impact, and
b) the likelihood of such harm or impact.
Categories of harm and impact include:
•

Inconvenience, distress, or damage to standing or reputation

•

Financial loss or agency liability

•

Harm to agency programs or public interests

•

Unauthorized release of sensitive information

•

Personal safety

•

Civil or criminal violations.

Required assurance levels for electronic transactions are determined by assessing the
potential impact of each of the above categories using the potential impact values described
in Federal Information Processing Standard (FIPS) 199, “Standards for Security
Categorization of Federal Information and Information Systems.” The three potential impact
values are:6
•

Low impact

•

Moderate impact

•

High impact.

6 For the purposes of this document, the impact value not applicable may apply to the categories of harm.

The next section defines the potential impacts for each category. Note: If authentication
errors cause no measurable consequences for a category, there is “no” impact.
Determining Potential Impact of Authentication Errors:
Potential impact of inconvenience, distress, or damage to standing or reputation:
• Low—at worst, limited, short-term inconvenience, distress or embarrassment to any
party.
• Moderate—at worst, serious short term or limited long-term inconvenience, distress or
damage to the standing or reputation of any party.
• High—severe or serious long-term inconvenience, distress or damage to the standing or
reputation of any party (ordinarily reserved for situations with particularly severe effects
or which affect many individuals).
Potential impact of financial loss:
• Low—at worst, an insignificant or inconsequential unrecoverable financial loss to any
party, or at worst, an insignificant or inconsequential agency liability.
• Moderate—at worst, a serious unrecoverable financial loss to any party, or a serious
agency liability.
• High—severe or catastrophic unrecoverable financial loss to any party; or severe or
catastrophic agency liability.
Potential impact of harm to agency programs or public interests:
• Low—at worst, a limited adverse effect on organizational operations or assets, or public
interests. Examples of limited adverse effects are: (i) mission capability degradation to
the extent and duration that the organization is able to perform its primary functions with
noticeably reduced effectiveness, or (ii) minor damage to organizational assets or public
interests.
• Moderate—at worst, a serious adverse effect on organizational operations or assets, or
public interests. Examples of serious adverse effects are: (i) significant mission capability
degradation to the extent and duration that the organization is able to perform its primary
functions with significantly reduced effectiveness; or (ii) significant damage to
organizational assets or public interests.
• High—a severe or catastrophic adverse effect on organizational operations or assets, or
public interests. Examples of severe or catastrophic effects are: (i) severe mission
capability degradation or loss of to the extent and duration that the organization is unable
to perform one or more of its primary functions; or (ii) major damage to organizational
assets or public interests.
Potential impact of unauthorized release of sensitive information:
• Low—at worst, a limited release of personal, U.S. government sensitive, or
commercially sensitive information to unauthorized parties resulting in a loss of
confidentiality with a low impact as defined in FIPS PUB 199.
• Moderate—at worst, a release of personal, U.S. government sensitive, or commercially
sensitive information to unauthorized parties resulting in loss of confidentiality with a
moderate impact as defined in FIPS PUB 199.

•

High—a release of personal, U.S. government sensitive, or commercially sensitive
information to unauthorized parties resulting in loss of confidentiality with a high impact
as defined in FIPS PUB 199.

Potential impact to personal safety:
• Low—at worst, minor injury not requiring medical treatment.
• Moderate—at worst, moderate risk of minor injury or limited risk of injury requiring
medical treatment.
• High—a risk of serious injury or death.
The potential impact of civil or criminal violations is:
• Low—at worst, a risk of civil or criminal violations of a nature that would not ordinarily
be subject to enforcement efforts.
• Moderate—at worst, a risk of civil or criminal violations that may be subject to
enforcement efforts.
• High—a risk of civil or criminal violations that are of special importance to enforcement
programs.
Determining Assurance Level:
Compare the impact profile from the risk assessment to the impact profiles associated with
each assurance level, as shown in Table 1 below. To determine the required assurance level,
find the lowest level whose impact profile meets or exceeds the potential impact for every
category analyzed in the risk assessment (as noted in step 2 below).
Table 1 – Maximum Potential Impacts for Each Assurance Level

Assurance Level Impact
Profiles

Potential Impact Categories for Authentication Errors
Inconvenience, distress or damage to standing or
reputation
Financial loss or agency liability
Harm to agency programs or public interests
Unauthorized release of sensitive information
Personal Safety

1
Low

2
Mod

3
Mod

4
High

Low
N/A
N/A
N/A

Mod
Low
Low
N/A

Mod
Mod
Mod
Low

Civil or criminal violations

N/A

Low

Mod

High
High
High
Mod
High
High

In analyzing potential risks, the agency must consider all of the potential direct and indirect
results of an authentication failure, including the possibility that there will be more than one
failure, or harms to more than one person. The definitions of potential impacts contain some
relative terms, like "serious" or "minor," whose meaning will depend on context. The agency
should consider the context and the nature of the persons or entities affected to decide the
relative significance of these harms. Over time, the meaning of these terms will become
more definite as agencies gain practical experience with these issues. The analysis of harms

to agency programs or other public interests depends strongly on the context; the agency
should consider these issues with care.
In some cases (as shown in Table 1), impact may correspond to multiple assurance levels.
For example, Table 1 shows that a moderate risk of financial loss corresponds to assurance
levels 2 and 3. In such cases, agencies should use the context to determine the appropriate
assurance level.
2.3. Determining Assurance Levels and selecting authentication solutions using Risk
Assessment
Agencies shall use the following steps to determine the appropriate assurance level:
Step 1: Conduct a risk assessment of the e-government system. Guidance for agencies in
conducting risk assessments is available in A-130, Section 5 of OMB’s GPEA guidance and
existing NIST guidance. The risk assessment will measure the relative severity of the
potential harm and likelihood of occurrence of a wide range of impacts (to any party)
associated with the e-government system in the event of an identity authentication error.
Note: An E-government system may have multiple categories or types of transactions, which
may require separate analysis within the overall risk assessment. An E-government system
may also span multiple agencies whose activities may require separate consideration.
Risk analysis is to some extent a subjective process, in which agencies must consider harms
that might result from, among other causes, technical failures, malevolent third parties,
public misunderstandings, and human error. Agencies should consider a wide range of
possible scenarios in seeking to determine what potential harms are associated with their
business process. It is better to be over-inclusive than under-inclusive in conducting this
analysis. Once risks have been identified, there may also be ways to adjust the business
process to mitigate particular risks by reducing the likelihood that they will occur (see Step
4).
Step 2: Map identified risks to the required assurance level. The risk assessment should
be summarized in terms of the potential impact categories in Section 2.2.
To determine the required assurance level, agencies should initially identify risks inherent in
the transaction process, regardless of its authentication technology. Agencies should then tie
the potential impact category outcomes to the authentication level, choosing the lowest level
of authentication that will cover all of potential impacts identified. Thus, if five categories
of potential impact are appropriate for Level 1, and one category of potential impact is
appropriate for Level 2, the transaction would require a Level 2 authentication. For example,
if the misuse of a user’s electronic identity/credentials during a medical procedure presents a
risk of serious injury or death, map to the risk profile identified under Level 4, even if other
consequences are minimal.
Step 3: Select technology based on the NIST e-authentication technical guidance. After
determining the assurance level, the agency should refer to the NIST e-authentication

technical guidance to identify and implement the appropriate technical requirements.
Step 4: After implementation, validate that the information system has operationally
achieved the required assurance level. Because some implementations may create or
compound particular risks, conduct a final validation to confirm that the system achieves the
required assurance level for the user-to-agency process. The agency should validate that the
authentication process satisfies the systems’s authentication requirements as part of required
security procedures (e.g., certification and accreditation).
Step 5: Periodically reassess the information system to determine technology refresh
requirements. The agency must periodically reassess the information system to ensure that
the identity authentication requirements continue to be valid as a result of technology
changes or changes to the agency’s business processes. Annual information security
assessment requirements provide an excellent opportunity for this. Agencies may adjust the
identity credential’s level of assurance using additional risk mitigation measures. Easing
identity credential assurance level requirements may increase the size of the enabled
customer pool, but agencies must ensure that this does not corrupt the system’s choice of the
appropriate assurance level.
2.4. Assurance Levels and Risk Profiles: Descriptions and Examples
Level 1—Little or no confidence exists in the asserted identity. For example, Level 1
credentials allow people to bookmark items on a web page for future reference.
Examples:
•

In some instances, the submission of forms by individuals in an electronic
transaction will be a Level 1 transaction: (i) when all information is flowing to the
Federal organization from the individual, (ii) there is no release of information in
return, and (iii) the criteria for higher assurance levels are not triggered. For example,
if an individual applies to a Federal agency for an annual park visitor's permit (and
the financial aspects of the transaction are handled by a separate contractor and thus
analyzed as a separate transaction, the transaction with the Federal agency would
otherwise present minimal risks and could be treated as Level 1.

•

A user presents a self-registered user ID or password to the U.S. Department of
Education web page, which allows the user to create a customized “My.ED.gov”
page. A third party gaining unauthorized access to the ID or password might infer
personal or business information about the individual based upon the customization,
but absent a high degree of customization however, these risks are probably very
minimal.

•

A user participates in an online discussion on the whitehouse.gov website, which
does not request identifying information beyond name and location. Assuming the
forum does not address sensitive or private information, there are no obvious inherent
risks.

Level 2—On balance, confidence exists that the asserted identity is accurate. Level 2
credentials are appropriate for a wide range of business with the public where
agencies require an initial identity assertion (the details of which are verified
independently prior to any Federal action).

Examples:
•

A user subscribes to the Gov Online Learning Center (www.golearn.gov). The
site’s training service must authenticate the person to present the appropriate course
material, assign grades, or demonstrate that the user has satisfied compensation-or
promotion-related training requirements. The only risk associated with this
transaction is a third party gaining access to grading information, thereby harming the
student’s privacy or reputation. If the agency determines that such harm is minor, the
transaction is Level 2.

•

A beneficiary changes her address of record through the Social Security web site.
The site needs authentication to ensure that the entitled person’s address is changed.
This transaction involves a low risk of inconvenience. Since official notices regarding
payment amounts, account status, and records of changes are sent to the beneficiary’s
address of record, it entails moderate risk of unauthorized release of personally
sensitive data. The agency determines that the risk of unauthorized release merits
Assurance Level 2 authentication.

•

An agency program client updates bank account, program eligibility, or payment
information. Loss or delay would significantly impact him or her. Errors of this sort
might delay payment to the user, but would not normally result in permanent loss.
The potential individual financial impact to the agency is low, but the possible
aggregate is moderate.

•

An agency employee has access to potentially sensitive personal client information.
She authenticates individually to the system at Level 2, but technical controls (such as
a virtual private network) limit system access to the system to the agency premises.
Access to the premises is controlled, and the system logs her access instances. In a
less constrained environment, her access to personal sensitive information would
create moderate potential impact for unauthorized release, but the system’s security
measures reduce the overall risk to low.

Level 3—Level 3 is appropriate for transactions needing high confidence in the asserted
identity’s accuracy. People may use Level 3 credentials to access restricted web services
without the need for additional identity assertion controls.

Examples:
•

A patent attorney electronically submits confidential patent information to the US
Patent and Trademark Office. Improper disclosure would give competitors a
competitive advantage.

•

A supplier maintains an account with a General Services Administration
Contracting Officer for a large government procurement. The potential financial loss
is significant, but not severe or catastrophic, so Level 4 is not appropriate.

•

A First Responder accesses a disaster management reporting website to report an
incident, share operational information, and coordinate response activities.

•

An agency employee or contractor uses a remote system giving him access to
potentially sensitive personal client information. He works in a restricted-access
Federal office building. This limits physical access to his computer, but system
transactions occur over the Internet. The sensitive personal information available to
him creates a moderate potential impact for unauthorized release.

Level 4—Level 4 is appropriate for transactions needing very high confidence in the
asserted identity’s accuracy. Users may present Level 4 credentials to assert identity
and gain access to highly restricted web resources, without the need for further
identity assertion controls.
Examples:
•

A law enforcement official accesses a law enforcement database containing
criminal records. Unauthorized access could raise privacy issues and/or compromise
investigations.

•

A Department of Veteran’s Affairs pharmacist dispenses a controlled drug. She
would need full assurance that a qualified doctor prescribed it. She is criminally liable
for any failure to validate the prescription and dispense the correct drug in the
prescribed amount.

•

An agency investigator uses a remote system giving her access to potentially
sensitive personal client information. Using her laptop at client worksites, personal
residences, and businesses, she accesses information over the Internet via various
connections. The sensitive personal information she can access creates only a
moderate potential impact for unauthorized release, but her laptop’s vulnerability and
her non-secure Internet access raise the overall risk.

2.5. Scope and Elements of Risk
When determining assurance levels, one element of the necessary risk assessment is the risk
of denial (or repudiation) of electronically transmitted information. Section 9c of OMB’s
GPEA guidance states agencies should plan how to minimize this risk by ensuring user
approval of such information. Section 8c of the OMB Procedures and Guidance on
Implementing GPEA includes guidance on minimizing the likelihood of repudiation.
OMB’s GPEA guidance states that properly implemented technologies can offer degrees of
confidence in authenticating identity that are greater than a handwritten signature can offer.
Conversely, electronic transactions may increase the risk and harm (and complicate redress)
associated with criminal and civil violations. The Department of Justice's “Guide for Federal
Agencies on Implementing Electronic Processes”7 discusses the legal issues surrounding
electronic government. Legal and enforcement needs may affect the design of an eauthentication system and may also entail generation and maintenance of certain system
management documentation.
Legal issues can present significant policy challenges for agencies. Agencies should consider
these issues when assigning transactions to assurance levels. Risk assessments should include
the potential effects of illegal activities and process failures with respect to:
•

agency enforcement priorities

•

agency programmatic interests

•

broader public interests such as national security, the environment, and economic
markets.

Some of these harms (e.g., financial loss or release of personal information) are described in
each assurance level; others depend on the agency's programmatic interests. The risk analysis
process is necessarily highly contextual, and agencies should consider whether their systems
present any distinctive risks.
The risk analysis incorporates this by discussing the risks associated with criminal and civil
violations, and harm to agency programs or the public interest. Agencies should remember to
consult appropriately with their counsel's office in their determination of this impact. When
assessing this risk and designing a process, agencies should consider single acts and patterns
of action that could affect agency programs. For example, if sensitive information is
available from an agency website, the agency should consider the effects of single acts and
possible patterns of such activity when assessing risk levels. (18 U.S.C. 1029, 1030)
Agencies may also decrease reliance on identity credentials through increased risk-mitigation
controls. For example, an agency business process rated for Level 3 identity assertion
assurance may lower its profile to accept Level 2 credentials by increasing system controls or
‘second level authentication’ activities. (See Section 2.3, Step 5)
Agencies are expected to follow all relevant guidance issued by the National Achieves and
Records Administration (NARA) regarding the handling of electronic records. This guidance
7 http://www.usdoj.gov/criminal/cybercrime/eprocess.pdf

addresses implementation issues further in section 4.1.

3. Assessing Confidence in Credential Service Providers
Since identity credentials are used to represent one’s identity in electronic transactions, it is
important to assess the level of confidence in the credential. Credential Service Providers
(CSPs) are governmental and non-governmental organizations that issue and sometimes
maintain electronic credentials. These organizations must have completed a formal
assessment against the assurance levels described in this guidance.
The CSP’s issuance and maintenance policy influences its e-authentication process
trustworthiness. The E-Authentication Initiative will therefore develop an assessment process
for the government to determine the maximum assurance level merited by the CSP. For
example, if a CSP follows all process/technology requirements for assurance Level 3, a user
may use a credential provided by the CSP to authenticate himself for a transaction requiring
assurance Levels 1, 2, or 3.

4. Implementing an Authentication Process
4.1. The E-Authentication Process
Each step of the authentication process influences the assurance level chosen. From identity
proofing, to issuing credentials, to using the credential in a well-managed secure application,
to record keeping and auditing—the step providing the lowest assurance level may
compromise the others. Each step in the process should be as strong and robust as the others.
Agencies will achieve the highest level of identity assurance through strong identity
proofing, a strong credential, and robust management (including a strong archive and audit
process). However, the best authentication systems result from well-engineered and tested
user and agency software applications. A process currently being developed for enabling
authentication across Federal agencies will be published for implementation when complete.
To determine the level of credential required to validate a user’s identity, an agency must
understand how its business applications process the credential. The agency must identify the
requirements for each step in the e-authentication (and authorization) process. This includes
the following steps:
•

Initial enrollment

•

Subsequent visits to agency application

•

Verification of identity credential

•

Transaction management

•

Long term records management

•

Periodic system tests

•

Suspension, revocation, re-issuance

•

Audit.

Each step is explained in the e-authentication technical guidance. Responsibility for these
steps lies with the business process owner, designated agency, or cross-agency authority.
4.2. Using Anonymous Credentials
Unlike identity authentication, anonymous credentials may be appropriate to use to evaluate
an attribute when authentication need not be associated with a known personal identity. To
protect privacy, it is important to balance the need to know who is communicating with the
Government against the user’s right to privacy. This includes using information only in the
manner in which individuals have been assured it will be used. It may be desirable to
preserve anonymity in some cases, and it may be sufficient to authenticate that:
•

The user is a member of a group; and/or

•

The user is the same person who supplied or created information in the first place;
and/or

•

A user is entitled to use a particular pseudonym.

These anonymous credentials have limited application and are to be implemented on a caseby-case basis. Some people may have anonymous and identity credentials. As general matter,
anonymous credentials are appropriate for Levels 1 and 2 only.
4.3. Information Sharing and the Privacy Act
When developing authentication processes, agencies must satisfy the requirements for
managing security in the collection and storage of information associated with validating
user identities. The E-Government Act of 2002, section 208 requires agencies to conduct
privacy impact assessments for electronic information systems and collections. This includes
performing an assessment when authentication technology is added to an electronic
information system accessed by members of the public. For additional information on
privacy impact assessments, consult OMB guidance.8
Most e-authentication processes capture the following information:
•

Information regarding the individuals/ businesses/governments using the E-Gov
service

•

Electronic user credentials (i.e., some combination of public key certificates, user
identifiers, passwords, and Personal Identification Numbers)

•

Transaction information associated with user authentication, including credential
validation method

•

Audit Log/Security information.

To the extent that the authentication process captures information that is protected by the
8 OMB Memorandum M-03-22, http://www.whitehouse.gov/omb/memoranda/m03-22.html

Privacy Act (because it is information about an individual that the agency retrieves by an
individual's name or other identifier and thus is maintained in an agency Privacy Act system
of records), the agency needs to comply with the Privacy Act with respect to such
information.
Authentication data must be protected from unauthorized disclosure or modification. The
Privacy Act generally requires that registered users be allowed to have access to and request
amendment to information about them maintained in a system of records. Information from
the system of records should not be shared, except in accordance with the Privacy Act and
other applicable laws.
4.4. Cost/Benefit Considerations
Like any capital purchase, implementing e-authentication requires consideration of the
benefit and costs, and thus a cost-benefit analysis is required by the Capital Programming
Guide.9 It is also important to match the required level of assurance against the cost and
burden of the business, policy, and technical requirements of the chosen solution.
Benefits typically include:
•

increased speed of the transaction

•

increased partner participation and customer satisfaction

•

improved record keeping efficiency and data analysis opportunities

•

increased employee productivity and improved quality of the final product

•

greater information benefits to the public

•

improved security

•

extensive security for highly sensitive information

Costs typically include:
•

initial capitalization for application design

•

technology acquisition

•

testing

•

deployment of the functional implementation

•

long-term maintenance.

In some cases initial capital cost may be small, with higher long-term maintenance cost. It is
therefore important to assess the costs over the life cycle of the system. Authentication errors
can result in significant costs to agencies and the public. These costs will be identified by the
risk analysis set forth in section 2, and should be included in any cost/benefit analysis.

9 OMB Circular A-11, Supplement, http://www.whitehouse.gov/omb/circulars/a11/cpgtoc.html

Burden consists of the following two factors:
a) Costs imposed on non-federal entities
b) Any time demands not captured by the cost estimate, but imposed on the entities
by the technical solution.
Overly burdensome systems may affect non-federal entities’ use of the system diminishing
anticipated benefits and lowering return on investment. If a technical solution for an
assurance level is too costly or burdensome, agencies should consider reducing the required
assurance level and implementing management controls or business process adjustments to
achieve the same level of assurance. If the cost cannot be reduced to an acceptable level
through such risk mitigation approaches, then the agency may need to reconsider its vision
for the e-authentication system.
Higher assurance levels may require more costly credentials. Minimizing the number of
credentials can reduce costs. Refer to Section 3 of the GPEA guidance for additional
information on assessing risks, costs, and benefits. The e-authentication technical guidance
will provide alternatives for addressing assurance levels, which may help agencies to better
manage authentication costs.
5. Effective Dates of Guidance
Agencies must categorize all existing transactions/systems requiring user authentication into
one of the described assurance levels by September 15, 2005. Agencies should accomplish
this in the following order:
•

Systems classified as “major” must be completed by December 15, 2004.

•

New authentication systems should begin to be categorized, as part of the system
design, within 90 days of the completion of the final E-Authentication Technical
Guidance issued by NIST.

The chosen assurance level must be made publicly available through the agency website, the
Federal Register, or other means (e.g., upon request). Agency application assurance levels
will be posted at a central location for public access by the E-Authentication Initiative.
Beginning in 2004, agencies will be asked to report on their progress in implementing this
guidance in their annual E-Government Act Reports to OMB required by section 202(g) of
the E-Government Act.

Attachment B

Summary of Public Comments and Responses

On July 11, 2003, in a Federal Register notice [68 FR 41370], the General Services
Administration, in coordination with the Office of Management and Budget, published a Draft EAuthentication Policy for Federal Agencies.
We received 47 comments representing Federal agencies, technology vendors and other
organizations on the proposed guidance. We considered all comments in developing this final
OMB guidance. Comments on the guidance supported establishing the requirement for agencies
to follow government-wide e-authentication guidance when implementing electronic transactions
and to conduct risk assessments. The following paragraphs summarize the general groupings of
comments and our responses.
Comments and Responses
Comments fell into three categories: 1. General comments about the need for, cost of,
and philosophies of authentication guidance; 2. Comments specific to sections of the document;
and 3. Editorial and stylistic comments. Overall, approximately one third of the submitted
comments were adopted, resulting in significant revision and reorganization.
Over half of the comments pertained to the Section 2, Assurance Levels and Risk
Assessments, and focused on requests for clarification of the guidance. While the entire
guidance was extensively revised, the most extensive revision was in this section. The revisions
included improving the assurance level descriptions and eliminating the vague naming
convention, clarifying how to determine the appropriate assurance level, as well as the potential
impacts of authentication errors, and ensuring accuracy of the examples. The scope and
applicability sections were improved, confusing terminology was defined and the guidance was
reviewed for greater consistency. The guidance was also changed to more closely align with the
recently issued NIST standards (the Federal Information Processing Standard (FIPS) 199) and
applicable laws. The guidance was also revised for legal accuracy.