2014 Rule 17g-7 Supporting Statement v2

SUPPORTING STATEMENT
for the Paperwork Reduction Act Information Collection Submission for
“Rule 17g-7”
A.

Justification
1.

Necessity of Information Collection

Rule 17g-7 of the Securities Exchange Act of 1934 (“Exchange Act”) requires nationally
recognized statistical rating organizations (“NRSROs”) to include in any report accompanying a
credit rating with respect to an asset-backed security (“ABS”) (as that term is defined in Section
3(a)(77) of the Exchange Act) a description of the representations, warranties, and enforcement
mechanisms available to investors and a description of how they differ from the representations,
warranties, and enforcement mechanisms in issuances of similar securities.
Rule 17g-7 was adopted by the Securities and Exchange Commission (the
“Commission”) in 2011 as part of a set of rules designed to implement Section 943 of the DoddFrank Wall Street Reform and Consumer Protection Act (“Dodd-Frank Act”) relating to assetbacked securities. Specifically, Rule 17g-7 is designed to implement the requirements of Section
943 of the Dodd-Frank Act by providing investors with information regarding the use of
representations and warranties in the ABS markets.
2.

Purpose and use of the Information Collection

As noted in the response to Item 1, the Commission adopted Rule 17g-7 to implement
Section 943 of the Dodd-Frank Act which, among other things, requires NRSROs to provide
investors with certain information regarding the representations, warranties, and enforcement
mechanisms available to investors in an ABS offering. Such disclosure would not be made
directly with, or used by, the Commission. Rather, the information regarding the representations,
warranties, and enforcement mechanisms with respect to an ABS transaction would be provided
in any report accompanying a credit rating by an NRSRO in connection with an ABS transaction.
The information is intended to help ensure that investors in ABS transactions are provided with
important information about such transactions prior to the point at which they make an
investment decision.
3.

Consideration Given to Information Technology

Not applicable. Rule 17g-7 disclosure is not required to be filed with the Commission.
4.

Duplication

Commission staff is not aware of any rules that conflict with or substantially duplicate
Rule 17g-7.
5.

Effect on Small Entities

Rule 17g-7 would not impact a significant number of small entities. Rules in the

2
Securities Act of 1933 (“Securities Act”) and Exchange Act define an issuer, other than an
investment company, to be a “small business” or a “small organization” if it had total assets of $5
million or less on the last day of its most recent fiscal year. Currently, there are 10 NRSROs and,
based on their most recently filed annual reports pursuant to Rule 17g-3 of the Exchange Act,
only two NRSROs are small entities under the above definition. In addition, Rule 17g-7 does not
prescribe how an NRSRO must fulfill its responsibility to compare the terms of a deal to those of
similar securities, thereby providing additional flexibility in the case of an NRSRO that may be
subject to a more limited application of the rule due to a smaller volume of credit ratings issued
in connection with ABS transactions.
6.

Consequences of Not Conducting Collection

The objectives of offering disclosure requirements under the Securities Act and the
ongoing disclosure requirements under the Exchange Act could not be met with less frequent
collection of this information for asset-backed securities.
7.

Inconsistencies with Guidelines in 5 CFR § 1320.5(d)(2)

There are no special circumstances. This collection is consistent with the guidelines in 5
CFR § 1320.5(d)(2).
8.

Consultations outside the Agency

The required Federal Register notice with a 60-day comment period soliciting comments
on the collection of information was published. Three comment letters were received in response
to this submission.
One commenter focused on the overall utility of Rule 17g-7, noting that the burdens of
complying with the rule are “substantial” and not justified by the rule’s benefits. 1 In particular,
this commenter provided the Commission with quantitative information demonstrating what it
perceived to be evidence of a lack of demand for the information required to be provided under
Rule 17g-7 among the rule’s intended beneficiaries (i.e., investors) – namely statistics reflecting
the “hit rate” for users of its website as it relates to accessing Rule 17g-7 information. This
commenter did not offer suggestions for reducing the paperwork burdens under the existing
collection of information, other than to submit that “OMB should not extend Rule 17g-7.”
Further, this commenter did not provide a detailed analysis of the burden estimates provided in
the Federal Register notice, other than to suggest that: (i) the burden should contain estimates of
ancillary compliance costs, such as the creation and maintenance of policies and procedures to
ensure compliance with Rule 17g-7 and the time, effort, and financial resources to train
personnel on the applicable requirements, and (ii) the number of estimated respondents should be
lowered from ten to seven to reflect the actual number of NRSROs currently registered with
respect to the rating of ABS transactions.
The second commenter submitted a report analyzing the reporting burden created by the
1

Comment letter of DBRS (Apr. 15, 2014) (“DBRS Comment Letter”).

3
collection of information requirements contained in Rule 17g-7. 2 This analysis focused on the
time and costs (both internal and external) the commenter incurred in the implementation of the
rule, based on figures calculated during the first full year of the rule’s effectiveness. The report
also contained an analysis intended to demonstrate whether the information required to be
disclosed under Rule 17g-7 is “necessary” and has “practical utility,” which, similar to the other
commenter, involved a calculation of the “hit rate” for users of its website. The report
concluded that the cost of implementing Rule 17g-7 greatly exceeded the estimates proffered at
the time the rule was adopted and that the information disclosed pursuant to the rule appears to
be of limited use to investors. 3
The third commenter questioned the accuracy of the burden estimates contained in the
Federal Register notice, asserting that such estimates understated the amount of time required to
comply with the rule on an ongoing basis. 4 Specifically, the commenter believes that the
Commission staff’s estimate of the time it would take to review and update benchmarks (i.e., 100
hours per NRSRO per year, for an industry-wide burden of 1,000 hours per year) was
significantly lower than the actual burden incurred by this particular commenter (estimated by the
commenter to be 568 hours per NRSRO per year, for an industry-wide burden of 5,680 hours per
year). In addition, this commenter – based on its anecdotal observation of the industry – believes
that the Commission staff’s estimate of the average number of NRSROs preparing reports for
each ABS transaction is low (2.5 NRSROs per ABS transaction versus a Commission staff
estimate of two NRSROs per ABS transaction), which difference would be magnified when
incorporated into the overall industry-wide estimate. The commenter also believes that
Commission staff failed to account for certain audit activities that flow from compliance with
Rule 17g-7. Finally, the commenter expressed concern as to the necessity of the rule and the role
that it serves in the capital markets, and, based on its own readership data, questioned the utility
of the rule as it relates to its usefulness to investors. Like the other two comment letters,
however, the commenter did not offer suggestions for reducing the paperwork burdens under the
existing collection of information.
While Commission staff appreciates the information received from all three commenters,
we ultimately believe that the estimates contained in the Federal Register notice remain valid,
particularly since they reflect the average burden incurred by all respondents, as opposed to the
particular experience of individual entities. 5 Specifically, NRSROs may vary in terms of their
2

Comment letter of Kroll Bond Rating Agency, Inc. (Apr. 17, 2014). Notably, the analysis contained in
this report was based entirely on the PRA estimates contained in the 2011 Commission release adopting
Rule 17g-7. The report did not refer to any of the burden estimates provided in the Federal Register
notice, nor did it provide any suggestions for reducing the current burden.
3

The report also requests that “SEC staff engage in a rigorous analysis of the benefit and cost of pending
rules – which are proposed to be finalized in fiscal 2014 – before final rules are adopted.” Commission
staff notes that this comment, which focuses on a pending Commission rulemaking to modify Rule 17g-7,
is outside the scope of the submission to extend OMB approval of the existing collection of information
requirements. See infra footnote 6.
4

Comment letter of Moody’s Investor Services (Apr. 28, 2014) (“Moody’s Comment Letter”).

5

The Commission also notes that points raised by one commenter regarding lowering the number of

4
overall activity in the ABS markets – both the number or different types of asset classes rated and
the number of actual transactions rated per year – and in terms of their level of experience in and
exposure to the creation of benchmarks. Moreover, Commission staff would expect there to be
some variation from NRSRO to NRSRO in terms of the amount of time it takes to create new
benchmarks, as compared to the amount of time devoted to reviewing and updating existing
benchmarks. Furthermore, Commission staff believes that the figure used for the number of
respondents (i.e., ten) is the appropriate number to be used for these purposes because it is
intended to represent an estimate as to the number of parties that could potentially be subject to
the rule over the course of the requested approval period (i.e., the next three years), as opposed to
the actual number of respondents subject to the rule at a moment in time. Finally, Commission
staff notes that any suggested changes to the rule itself (including any requests for the rule to be
eliminated in its entirely) would need to be effected pursuant to a Commission rulemaking. 6
9.

Payment or Gift

Not applicable.
10.

Confidentiality

Rule 17g-7 disclosure is not filed with the Commission; such disclosure would be
provided in any report accompanying a credit rating by an NRSRO in connection with an ABS
transaction.
11.

Sensitive Questions

Not applicable. No questions of a sensitive nature are asked. The information collection
does not collect any personally identifiable information.
12.

Burden of Information Collection

Commission staff estimates that the 10 currently-registered NRSROs would each spend
an average of approximately 100 hours per year reviewing and updating benchmarks for various
types of securities for purposes of comparing representations, warranties, and enforcement
mechanisms, resulting in an annual industry-wide reporting burden of 1,000 hours (10
respondents X 100 hours/respondent). On a deal-by-deal basis, Commission staff estimates that
NRSROs (from ten to seven) would have the effect of lowing the overall PRA burdens (see DBRS
Comment Letter at note 5) and that the points raised by another commenter to increase the number of
burden hours per NRSRO would have the opposite effect (see Moody’s comment Letter at 2-3). While
the Commission believes the estimates it provided for each of these factors are appropriate for the
reasons cited above, the Commission notes that the net effect of these comments, taken together, would
not result in a significant change in the overall PRA burdens estimate.
6

The Commission is considering rules that it proposed in 2011 that would, among other things, amend
Rule 17g-7. See Proposed Rules for Nationally Recognized Statistical Rating Organizations, Exchange
Act Release No. 64514 (May 18, 2011), 76 FR 33420 (Jun. 8, 2011). All three comment letters are
being placed in the comment file for this pending rulemaking.

5
it would take each NRSRO an average of approximately: (i) one hour to review each ABS
transaction to review the relevant disclosures prepared by an issuer, which an NRSRO would
review as part of the rating process, and convert those disclosures into a format suitable for
inclusion in any report to be issued by an NRSRO, and (ii) 10 hours per ABS transaction to
compare the terms of the current deal to those of similar securities. When the Commission
adopted Rule 17g-7, it estimated the average annual number of ABS offerings to be 2,067 and
the average number of credit ratings per issuance of ABS to be four, resulting in 8,268 annual
responses. 7 Commission staff believes that these estimates continue to be valid and, accordingly,
estimates that the total industry-wide annual reporting burden of complying with the disclosure
requirements under Rule 17g-7 is 90,948 hours (8,268 responses X 11 hours/response). As a
result, Commission staff estimates a total aggregate burden of 91,948 hours per year for
complying with the rule (1,000 hours for reviewing and updating benchmarks + 90,948 hours for
complying with disclosure requirements).
13.

Costs to Respondents

Not applicable. It is not anticipated that respondents will have to incur any capital and
start-up costs, nor any additional operational or maintenance costs (other than as provided in Item
12), to comply with the collection of information.
14.

Costs to Federal Government

Not applicable. Rule 17g-7 would not result in any costs to the federal government
beyond normal full-time employee labor costs, nor does the rule require the Commission to hire
any new employees or reallocate existing employees to ensure compliance with the rule.
15.

Explanation of Changes in Burden

When the Commission first adopted rules under the Credit Rating Agency Reform Act of
2006, it estimated that approximately 30 credit rating agencies ultimately would be registered as
NRSROs. 8 Accordingly, the Commission used 30 respondents for purposes of calculating its
PRA burden estimates when it adopted Rule 17g-7. 9 Since that time, 10 credit rating agencies
have registered with the Commission as NRSROs. This number has remained constant for
several years. Consequently, when the Commission last proposed rules regarding the oversight of
NRSROs, it stated that it believed it to be more appropriate to use the actual number of NRSROs
for purposes of the PRA. 10
7

See Disclosure for Asset-Backed Securities Required by Section 943 of the Dodd-Frank Wall Street
Reform and Consumer Protection Act, Release No. 33-9175; 34-63741 (Jan. 20, 2011), 76 FR 4489,
4508 (Jan. 26, 2011) (“Rule 17g-7 Adopting Release”).
8

See Oversight of Credit Rating Agencies Registered as Nationally Recognized Statistical Rating
Organizations, Release No. 34-5587 (Jun. 5, 2007), 72 FR 33564, 33607 (Jun. 18, 2007).
9

See Rule 17g-7 Adopting Release, 76 FR at 4506.

10

See Proposed Rules for Nationally Recognized Statistical Rating Organizations, Release No. 34-64514
(May 18, 2011), 76 FR 33420, 33499 (Jun. 8, 2011) (stating that “while the Commission expects several

6
The decrease in the total aggregate annual burden from 96,948 hours to 91,948 hours is
attributable to: (i) the change in calculating the number of respondents as described above (which
resulted in a reduction of 2,000 hours annually) and (ii) the elimination of 3,000 initial hours
(100 hours per NRSRO) to conduct an initial review and setup.
16.

Information Collections Planned for Statistical Purposes

Not applicable because the information is not used for statistical purposes.
17.

Approval to Omit OMB Expiration Date

Not applicable. The Commission is not seeking approval to omit the expiration date.
18.

Exceptions to Certification

Not applicable. This collection complies with the requirements in 5 CFR 1320.9.
B.

Collections of Information Employing Statistical Methods
Not applicable because the collection of information does not employ statistical methods.

more credit rating agencies may become registered as NRSROs over the next few years, the Commission
preliminarily believes that the actual number of NRSROs should be used for purposes of the PRA.”). As
previously noted, however, only seven NRSROs are currently registered with respect to the rating of
ABS transactions. Nevertheless, the Commission believes that the figure used for the number of
respondents (i.e., ten) is the appropriate number to be used for these purposes because it is intended to
represent an estimate as to the number of parties that could potentially be subject to the rule over the
course of the requested approval period (i.e., the next three years).