Supporting Statement, SLT(1505-0235)(May 2021)

SUPPORTING STATEMENT

Request for revision of Treasury International Capital (TIC)

Form SLT (OMB No. 1505-0235)

A. Justification

1. Circumstances necessitating the collection of information

Treasury International Capital (TIC) Form SLT, “Aggregate Holdings, Purchases and Sales, and Fair Value Changes of Long-Term Securities by U.S. and Foreign Residents”, is filed by U.S.-resident custodians, U.S.-resident issuers of long-term securities, and U.S.-resident end-investors (including endowments, foundations, pension funds, mutual funds, and other investment managers/advisors/sponsors) in long-term foreign securities. U.S.-resident holders, both banking and non-banking institutions, of foreign portfolio equity and/or debt may be required to report.

Form SLT is part of a series of reports to fulfill E.O. Number 10033 of February 8, 1949 and implementing Treasury Regulations (31 C.F.R. 128); the International Investment and Trade in Services Survey Act (22. U.S.C. 3103); and the Bretton Woods Agreements Act (Sec. 8(a) 59 Stat. 515; 22 U.S.C. 286f). They improve the preparation of the U.S. balance of payments accounts and the U.S. international investment position, as well as the formulation of U.S. international financial and monetary policies.

The global financial crisis of 2008-09 highlighted the importance of enhanced surveillance of the world and national economies, including more frequent and accurate data on external claims and liabilities. Securities constitute a large portion of U.S. external claims and liabilities. Before the crisis, the TIC data reporting system collected monthly aggregate data on purchases and sales of long-term securities, monthly aggregate data on holdings of short-term securities, and annual survey data on holdings of long-term securities. In response to the need the crisis showed for more frequent and accurate data, in 2011 the Department of the Treasury began collecting data on monthly holdings of long-term securities through Form SLT. It measures aggregate holdings of long-term securities by country.

To keep up with the evolution of financial markets and to further improve the accuracy of Form SLT data, the Department of the Treasury proposes to create, for the first time, a collection of data on “changes in fair value.” The Treasury also proposes to obtain the three main data types (holdings, purchases and sales, and change in fair value) from the same source. To eliminate duplication, the Treasury proposes to discontinue TIC Form S, “Purchases and Sales of Long-Term Securities by Foreign-Residents”, after a brief overlap period ensuring that the new data mesh appropriately with the old data. These changes should improve data quality, contributing to better preparation of the U.S. balance of payments accounts and the U.S. international investment position, as well as to the formulation of U.S. international financial and monetary policies.

2. Use of data

Data collected on TIC Form SLT constitute the most complete and readily available information on cross-border holdings of long-term domestic and foreign securities by U.S. and foreign residents. With the proposed changes, they will also constitute the most complete and readily available information on transactions and changes in valuation. The data are necessary for compiling the U.S. international financial transactions accounts, for calculating the U.S. international investment position, and in formulating U.S. international financial and monetary policies. Treasury, the Bureau of Economic Analysis of the Department of Commerce, and the Board of Governors of the Federal Reserve System are the most significant users of the data for these purposes. There is public interest in the aggregate data and the news media report on the monthly Treasury press release of monthly TIC data.

Respondents covered by the reporting requirements of Form SLT report directly to the district Federal Reserve Banks, which act as fiscal agents of the Treasury. The data are centrally aggregated and processed by the Federal Reserve Bank of New York (FRBNY) and then transmitted to the Treasury to be used for the purposes outlined above. Further, the data are published regularly in aggregate statistical formats, for the benefit of private users, in the Board of Governors’ Federal Reserve Bulletin and in the Commerce Department’s Survey of Current Business. Historical time series data, both aggregate and by country, reported on Form SLT also are posted to the Treasury TIC website at http://www.treasury.gov/resource-center/data-chart-center/tic/Pages/index.aspx.

3. Use of information technology

Electronic filing of the TIC Form SLT report has been mandatory since reports as of June 30, 2014. Section II.H of the instructions, Submission of Reports, specifies that the TIC Form SLT report must be submitted electronically by using the Federal Reserve System’s “Reporting Central” submission system. That online system is easy to use, secure, provides confirmation of the receipt of the data, and performs several validity checks of the file format of the report being filed.

3. Efforts to identify duplication

There will be duplication of the Purchase and Sales data with the TIC Form S data for three months after the revised Form SLT begins. This period of overlap will allow TIC and the data filers to make any adjustments necessary so that the Purchases and Sales component of the new form is consistent with the Form S that it will replace.

After the overlap period, there will be no duplication in the collection of these data. Similar information does not exist. The Treasury is the sole U.S. Government collection authority for these international portfolio capital positions and transactions data.

3. Impact on small entities

The reporting threshold of $1 billion, which is applied to total portfolio holdings of long-term securities, effectively excludes small custodians, issuers, and end-investors from the reporting population. Data reporters will file one consolidated report for all U.S.-resident parts of their organizations, and the total will be the aggregate of holdings as issuer, end-investor, and custodian.

3. Consequences of less frequent collection and obstacles to burden reduction

The data are collected on an ongoing basis. The monthly submission date for filing Form SLT is twenty-three calendar days following the month to which the report applies. If the data were collected less frequently, valuable information on trends in international capital movements would not be observed and made available to the U.S. Government. The lack of timely information could seriously impair the formulation of U.S. financial and monetary policies. The primary legal obstacle to reducing burden is the statutory mandate to collect the information. (See Item 1 above.)

3. Circumstances requiring special information collection

Frequency of reporting: The circumstances that require Form SLT to be collected more often than quarterly are the unpredictable qualities and magnitudes of the data due to the rapidly changing factors influencing the securities industry, and the U.S. government's need for timely and reliable information on trends in international holdings of securities. The likely consequences of collecting this information less frequently are outlined in Item 6 above.

Apart from the foregoing case, there are no special circumstances regarding Form SLT that are inconsistent with the conditions outlined in Item A.7 of the Specific Instructions for Supporting Statements for Paperwork Reduction Act Submissions.

3. Solicitation of comments on information collection

Treasury’s notice in the Federal Register soliciting public comment on Form SLT was published Thursday, January 28, 2021, vol. 86, pages 7459-7461. The deadline for receiving comments was Monday, March 29, 2021. To bring attention to the Federal Register notices, the Federal Reserve Bank of New York (FRBNY) contacted all respondents reporting data on securities to FRBNY, and other district Federal Reserve Banks that process TIC data.

The Federal Register notice of January 28 invited comments on the following proposed changes in the instructions for Form SLT:

(1) Justification: One important aim of this revision of the SLT data collection is to create, for the first time, a data collection of “changes in fair value” for the TIC securities data. Users of TIC data often compare the change in the holdings of long-term securities reported on the Form SLT, with the net purchases (purchases less sales) of long-term securities reported on the Form S. There is general agreement that the difference between the change in holding and the net purchases is due largely to the change in fair value of the securities, with less important factors making up the remainder of the difference. In mathematical terms, “Change in holdings” equals “purchases less sales” plus “change in fair value” plus “other factors”. Different assessments between TIC data users often arise because each one has to create their own estimates of the “change in fair value” despite lacking detailed information on the holdings of, and transactions in, the many securities in the TIC system. Another aim of this revision of the SLT data collection is to obtain the three main data types (holdings, purchases and sales, and change in fair value) from the same source. The result should greatly improve the connections between the holdings data and the purchases and sales data and the “change in fair value” data. Lastly, while there is an increase in the reporting burden on custodians from the revision of the SLT, after 2022 it is expected that this increase in burden will be significantly offset by the decrease in burden when the Form S is discontinued.

(2) No changes are made in the collection of holdings data; i.e., no changes are made in the columns and rows of the Form SLT or in the instructions regarding the holdings of long-term securities. In both the current and revised Form SLT there are eleven such columns covering three types of foreign securities and four types of U.S. securities, where for each type of U.S. security there is a column for foreign-official-held and a separate column for other-foreign-held.

(3) To accomplish the aims in (1) above, both the Form SLT and the instructions are expanded to add the collection of data on the total change in the fair (market) value over the month for all securities held at the end of the month for each type of long-term security. In the Form SLT, one column is added for each of the 11 columns of holdings mentioned in (2) above; in the instructions, sections II.F.4 and III.G are added.

(4) To accomplish the aims in (1) above, both the Form SLT and the instructions are expanded to add the collection of data on U.S. purchases and U.S. sales of long-term securities by U.S.-residents with foreign-residents; in the Form SLT, two columns are added for each of the 11 columns of holdings mentioned in (2) above; in the instructions, sections II.F.3, II.F.5, III.E and III.F are added.

(5) Note that while purchases and sales in the revised SLT data collection appear to be generally the same as in the Form S data collection, there are three important differences: (i) Purchases and sales in the Form SLT are reported by the custodian or issuer or end-investor that is also reporting the holdings, while in the Form S purchases and sales are reported by a trader (e.g., broker-dealer, prime broker, principal trading firm); (ii) the Form SLT data are recorded from the U.S. point of view, while the Form S data are recorded from the foreign point-of-view (e.g., Form SLT “purchases” are made by U.S. residents from foreign-residents, whereas Form S “purchases” are made by foreign-residents from U.S.-residents); and (iii) purchases and sales of foreign securities in Form SLT are recorded opposite the foreign country that issued the security, whereas the Form S data are recorded opposite the country that purchased or sold the security. The Form SLT data are much more informative about U.S. claims on individual foreign countries.

(6) The revised Form SLT no longer has Parts A and B, where previously a custodian reported data in part A and an issuer and/or end-investor reported data in part B. In the revised Form SLT the reporting firm must check one or both of the two boxes in the top-center section of the cover page to specify whether the data is from a custodian or from an issuer and/or end- investor or from both; see II.A in the instructions. So, a firm that reports data for both a custodian and an issuer/end-user can combine both types of data into one report, and no longer needs to report them separately in part A and part B.

(7) To allow time for respondents to revise their reporting systems, the revised form and instructions are scheduled to become effective for reports as of February 2022.

(8) Until the revised form becomes effective in 2022, the currently approved Form SLT and instructions will continue to be in effect.

(9) The name of the revised Form SLT on the cover page and elsewhere is expanded to “Aggregate Holdings, Purchases and Sales, and Fair Value Changes of Long-Term Securities by U.S. and Foreign Residents.” Added on the cover page under the name, is the phrase “Effective for reports beginning as of February 2022”.

(10) After the revised Form SLT becomes effective in February 2022, there will be a duplication of the Purchase and Sales data with the Form S for roughly six months. This period of overlap for comparison of the two sources of data will allow the agencies to make any necessary adjustments to the revised Form SLT and/or instructions. After the six-month overlap period ends, and if the purchases and sales data from the revised Form SLT are acceptable, then the Form S will be discontinued.

(11) Some other clarifications and format changes may be made to improve the instructions.

Letters from three entities were received in response to the Federal Register notice.

(A) A letter from the Commerce Department’s Bureau of Economic Analysis (BEA), dated February 10, 2021, strongly supported the continued collection of data by the Form SLT, including the proposed changes to the form and instructions. BEA especially supported the changes described in sections 3 and 4 under current actions in the Federal Register Notice (FRN). BEA agreed with the statement in the FRN that collecting holdings, transactions, and changes in fair value in the SLT will “greatly improve the connections between the holdings data and the purchases and sales data and the ‘change in fair value’ data.”

(B) The second response was an email from KeyBank, inquiring about the definition of New Issuances. Staff of the Federal Reserve Bank of New York (FRBNY) responded to the inquiry over the phone and explained the reporting instructions related to the issuance of securities from the current TIC Form S instructions in Section II H.1. Those instructions are now included also in section II.F.3.d of the SLT instructions.

(C) The Bank Policy Institute (BPI) submitted a seven-part letter, dated March 29, 2021, and also sent two supplementary letters later in response to Treasury’s first two replies. The BPI’s comments requested changes and clarifications in the instructions. Each of Treasury’s three replies addressed different matters raised by the BPI.

(C.1.) Treasury’s first response (called Part 1) (April 13, 2021) to the first three matters in the BPI letter (March 29, 2021) was as follows. NOTE: BPI’s concerns are listed using Roman numerals and bold type; the word “you” refers to the BPI, and the word “I” refers to the Treasury Department. End of NOTE.

Thank you for your helpful comments and suggestions. I believe we understand your concerns and will try to accommodate what we can. In that spirit, I would like to respond in three parts, this email being the first part dealing with your first three recommendations. The second part will cover your last four items and propose new language for the instructions that would provide the clarifications you request. After I get your responses to the first two parts and make any appropriate changes, then I would send you a third part which would be a draft final response to your letter that I hope would constitute the finalization of the form and instructions that you request in your first two recommendations.

I. The proposed revisions to the TIC Form SLT should not be effective until at least 18 months after the revisions are finalized and the final forms and instructions are released.

We appreciate your concerns and we agree to the 18-month development period for the revised-SLT form. Assuming that the form and instructions can be finalized this month, then the first reports would begin as of end-October 2022 and be filed in November 2022 with the Federal Reserve Bank of New York, acting on behalf of the Treasury Department.

II. The proposed six-month period of overlap in TIC Form SLT and TIC Form S reporting should be eliminated, with TIC Form S reporting ceasing immediately upon implementation of the proposed TIC Form SLT revisions.

We understand your opposition to the proposed overlap period when filing would be required of both the S reports and the revised-SLT reports. However, there will be series breaks and we need an overlap period to compare and measure the breaks between the transactions data in the S form, which has been around since the mid-1930s, with the new measure of transactions in the revised-SLT form. While there may be theoretical arguments both pro and con about why the measurement of transactions by custodians on the revised-SLT form should be the same as the measurement by brokers and dealers on the S form, we believe almost all users of TIC data would strongly prefer that we provided a comparison of the S data and revised-SLT data to better understand the series breaks.

For such reasons, (a) we have a suggestion that might reduce the burden of two codes mentioned in your second and fourth paragraphs; and (b) we agree to the suggestion in your next-to-last paragraph to reduce the overlap period to 3 months.

Regarding (a), we are unclear about the problem because the S form code is in the databases of broker/dealers while the revised-SLT code is in the databases of custodians. So, I present this suggestion in brief terms to see if you are interested in this idea to eliminate the need for two codes during the overlap period: At the beginning of the overlap period, data providers would change the code used for transactions on the S form to be the same as the one used on the revised-SLT form. As a result, the S-data would be similar to the revised-SLT transactions data; and for foreign securities, that would change the location of the foreign country in the S-data from the country-of-transaction to the country-of-issuer. Please let me know your view, as some feel that you would not want to make such a change in the S-form for only 3 months. Regarding a different matter you mentioned, I am not sure why changing the “point of view” creates an additional burden. Noting that “the foreign purchase of a US security” is identical to “the US sale of a US security”, there would seem to be little difference in the burden for reporting either one since the difference seems to be simply changing the data label from “foreign purchase” to “U.S. sale”. However, I understand there may be larger differences for data providers.

III. The submission date of the revised TIC Form SLT should be extended to at least the 25th calendar day of the month following the report as-of date.

We acknowledge the concerns of having the due date as the 20^th of the month. The 20^th was in-between the current due date of the 15^th for the TIC S and the 23^rd for the TIC SLT. We will agree to move the date back to the 23^rd for the revised TIC SLT.

(C.2.) Treasury’s second response (Part 2) (April 28, 2021) to the last four matters in the BPI letter of March 28, 2021 and to the BPI supplementary letter of April 21, 2021, was as follows.

This is the second part of our response to your letter dated March 29, 2021, that concerns “Revisions to Treasury International Capital Form SLT, ‘Aggregate Holdings, Purchases and Sales, and Fair Value Changes of Long-Term Securities by U.S. and Foreign Residents’ (OMB Control Number: 1505–0235)”.

First, regarding your supplementary letter of April 21, we agree that the effective date should be 18 months following the release of the agreed-upon form and instructions. For example, if the final proposed form and instructions were published on the TIC website in early May, and the Federal Register Notice with a 30-day comment period were published a little later in May, then the effective date would begin with the report for November 2022. In my experience, OMB [the Office of Management and Budget] always accepts the proposals agreed upon by data reporters and Treasury; OMB procedures will take roughly three more months to complete. We also agree with your rejection of the coding-change suggestion.

This Part 2 responds to the last four items in the BPI letter and includes new language for the instructions that would provide the clarifications you request.

Please note that there are two changes in the revised-SLT form. The expiration date on the cover page is changed to November 2022. The title for columns 4, 8, 12, 16, 20, 24, 28, 32, 36, 40, 44 on the form is changed from “Fair Value” to “Valuation”.

IV. Clarification is needed with respect to the breaks in data created by the exclusion of certain non-reportable events from the transactional data in the TIC SLT.

The BPI comment # IV seeks clarifications on a broad range of matters. We first note that there are no changes from the current SLT in what is reported in the columns for end-month holdings (in columns 1, 5, 9, 13, 17, 21, 25, 29, 33, 37, and 41) in the revised-SLT form i.e., the figures on the revised SLT are the same as would have been reported on the SLT.

Next, here is what is included, and not included, in the two new sections of the revised-SLT form. The Federal Register Notice of January provided a framework in mathematical terms — the “Change in holdings” equals “purchases less sales” plus “change in fair value” plus “other factors”.

a) The new Purchases/Sales columns. These two columns cover financial transactions relating to purchases (including issues) or sales (including redemptions) of securities (U.S. securities for columns 1-32; foreign securities for columns 33-44).

b) For this discussion, we modify the second term in the equation to read “valuation”, meaning that this column covers only the portion of the total change in the fair value of holdings that results from changes in the prices of the securities held (U.S. securities for columns 1-32; foreign securities for columns 33-44).

c) The “other factors” have no column in the revised-SLT form. Section II.F.5 lists the kinds of items that affect the fair value of holdings, but that are not to be included in either of the two new sections (purchases & sales, valuation) of the form. These factors often include transfers of new long-term securities into or out of a custodian’s database as a result of addition or departure of clients, interest and earnings added to holdings of long-term securities, interest and earnings paid out of holdings of long-term securities, and fees, penalties, other charges deducted from holdings of long-term securities. They may also include unusual events such as (but not limited to): debt cancellation and write-off, change from a portfolio holding to a direct investment holding, and change in the country-of-residence of a holder or an issuer.

So, we agree with the point in this BPI comment that client onboarding and offboarding “would therefore result in shifts in the positional holdings data of long-term securities but would not be reflected in the transactional data.” Regarding another point in this BPI comment — “As a result, there is the potential for substantial breaks in the data when comparing holdings balances on sequential reports” — we believe there is the same potential as there is with the current SLT, and our process will not change: when we see large moves, we will contact reporters and ask about those moves. Reporters are welcome to provide background information about substantial breaks in the data with their data submissions to the Federal Reserve Bank of New York.

V. The instructions should be clarified to allow issuers and end-investors the same flexibility afforded to custodians with respect to reporting on the TIC SLT when data on valuation changes are not available.

We agree with the BPI letter’s recommendation that issuers and end-investors should be allowed the same option provided to custodians for reporting revisions to their estimates of monthly valuation changes in subsequent submissions. Please note that all reporting entities are required to report on time their best estimates of the valuation changes at that time. We are changing the effective date to November 2022 on both the cover page of the form and the cover page of the instructions. In addition, we are preparing several FAQs to be published together later.

VI. The flowchart for reporting responsibilities scenario 2 should be clarified so that U.S. investments made through a foreign master fund are not required to be reported by the U.S. investment manager and are instead reported by the U.S. custodian or the U.S. issuer.

We cannot change this requirement. Investment Managers (IMs) could have the responsibility to report in certain scenarios when the IMs manage the ownership of U.S. securities for foreign clients. For example: If a US custodian manages the safekeeping of reportable securities and the custodian knows the identity of the foreign owner, then the custodian would have the responsibility to report. If the securities were issued directly into a foreign market, then the U.S. issuer would have the responsibility to report. However, in a case where a U.S. custodian manages the safekeeping of the securities but only sees the IM as their client and cannot look through the IM to see its foreign clients, then neither the custodian nor the issuer has the means to report, so the IM has the responsibility to report the foreign ownership of U.S. securities. So we see no need for a change in the instructions.

VII. Clarification is needed with respect to the reporting treatment of principal payments for prepaying bonds on the revised TIC SLT.

We agree that these monthly principal payments on asset-backed-securities (ABS) are significant. As illustrated in the attached table (attachment 2), taken from the TIC website for February 2021, they change our estimate of total net-foreign-purchases-of-US-securities from negative $7,329 million, based solely on S-data, to negative $49,531.^1/ These principal payments on ABS are redemption-like transactions, and in the revised-SLT form are reportable in the new transactions section, by country, in the columns for US-Purchases-of-US-Securities (e.g. in columns 10, 14, 18 and 22). End-of-month holdings data, just as in the current SLT form, will reflect the lower value of holdings after accounting for the repayments.

Attachment 1 below is our revised instructions. All the revisions are in section II.E (for item V; the fourth sentence begins with “firms” instead of “custodians”), section II.F, and the cover page (effective date).

Thank you for your reply to our Part 1 response. We await your additional comments to this response.

[footnote 1] These figures are from rows 105-112 in column B of the attached file “snetus.csv”. This file is updated every month, at line A.1.a on the Securities (A) webpage at: https://home.treasury.gov/data/treasury-international-capital-tic-system-home-page/tic-forms-instructions/securities-a-us-transactions-with-foreign-residents-in-long-term-securities.

Attachment 1: Proposed revised instructions re items IV, V, and VII above, and supplemental BPI letter.

Attachment 2: file “snetus (released 4-15-2021).csv”

(C.3.) Treasury’s Part 3 response of May 14, 2021 to BPI’s supplementary letter of May 5, 2021 was as follows. BPI’s concerns are listed using Arabic numerals and bold type.

This is Part 2b of our response to your letter dated March 29, 2021, that concerns “Revisions to Treasury International Capital Form SLT, ‘‘Aggregate Holdings, Purchases and Sales, and Fair Value Changes of Long-Term Securities by U.S. and Foreign Residents” (OMB Control Number: 1505–0235)”. In particular, this Part 2b responds to your May 5, 2021 supplemental letter requesting further clarification of several items in the instructions.

First, we thank you for the heads-up about the “code-freezes” that are employed during December by most reporting entities. We expect that this Part 2b reply will complete the current revision process, and for that reason I will soon post the finalized form and instructions on the TIC website. We will then move to publish a Federal Register Notice to formally inform all reporting entities, including those that are not members of the Bank Policy Institute (BPI), about the final form and instructions.

Here are our responses to the items you mentioned as needing further clarification.

1. The BPI understands, from a previous Treasury response regarding the reporting of valuation, that if there is any revision to the estimated valuation that was reported, firms would merely be expected to make an adjustment to subsequent submissions, as opposed to continually amending prior submissions. BPI requests that Treasury confirm that this understanding is correct, and that the expectation of respondent firms is to revise such items prospectively and not to revise previous submissions to accommodate for such adjustments.

For a couple of reasons, we cannot agree to that understanding. First, we maintain Section II.H.7, which states, “the respondent may be asked by the Federal Reserve Bank’s staff to provide supplemental information, including reasons for significant data changes between reporting periods, or submit revisions as necessary.” For example, there may be cases with a significant impact where we would need to request a reporting entity to revise prior months. Second, if a change in valuation significantly affected the holdings data, then we would expect a revised report just as we do now for the current SLT reporting. The proposed new form in no way affects the priority we place on obtaining accurate and reliable monthly SLT holdings data.

2. Recommend that specific examples of how to report changes in fair value due to price be incorporated into the instructions or added to the FAQs that are currently in process. Specifically, confirmation that firms should apply the change in price to the prior period par value, as opposed to the current period par value. Further, for current month purchases, confirmation that the change would be applied to the par value as of the purchase date would be helpful.

We agree with all those points. We have included additional examples and have expanded the discussion in section II.F.4 concerning the changes in fair value due to price. The revised instructions are attached.

We confirm the specifics: that firms should apply the change in price to the prior period par value, as opposed to the current period par value; and for current month purchases, that the change should be applied to the par value as of the purchase date.

Thank you for all your comments and suggestions. They have improved the form and instructions.

END of section (C) regarding BPI comments on the Federal Register notice of January 28, 2021.

3. Provision of payments to recordkeepers

In respect of Form SLT or any other TIC form, there has been no provision of payments or gifts to respondents for any purpose.

3. Assurance of confidentiality

As is the case for all TIC forms, individual respondent data are considered confidential, and access to that information is strictly limited to selected staff of the Treasury, the Federal Reserve Board of Governors, and the district Federal Reserve Banks. Compliance with the Privacy Act is assured.

3. Justification of sensitive questions

There are no questions of a sensitive nature.

This collection does not gather personally identifiable information.

12. Total annual hour burden, beginning November 2022 when the proposed form goes into effect

Number of respondents ─ 438

Frequency of Responses ─ 12 per year

Annual Burden ─ 61,722 hours

Burden estimates are calculated for three different classes of data reporters.

• The first group is comprised of U.S. resident custodians, who provide safekeeping services for their own firm as well as for others. This type of organization will provide most of the data collected on Form SLT, and thus will have the greatest burden.

• The second class of respondents is U.S. resident end investors who either keep securities in custody at their own site or who arrange for safekeeping abroad. In either case, no U.S.-resident custodian is employed to hold these foreign securities in safekeeping. These investors would report on all these holdings of foreign securities.

• The last class of reporters consists of U.S. resident issuers of securities, who are not custodians, that have issued securities directly to foreigners and, in addition, those foreigners do not hold the securities with U.S. custodians. These issuers would report the foreign holdings of those securities.

Beginning November 2022 when the proposed form goes into effect, the respondent population subject to Form SLT is approximately 438 banks, brokers, and nonbanking business enterprises that fall into the three classes of respondents described above. We expect about 5,256 responses per year. We estimate there will be about 87 custodians that report relatively large amounts of data. With an estimated average reporting burden of 21.6 hours per filing by custodians and 9.3 hours per filing by 351 other respondents, the estimated total reporting burden on the public will be 61,722 hours per year.

# Respondents	# Responses per Respondent	Total Annual Responses	Hours Per Response	Total Burden
438	12	5,256	11.743	61,722

Estimates of annualized cost to respondents: Generally, completion and review of the forms involves two persons. It is estimated that the average wage of persons completing the form is $36.68 per hour (corresponding to an annual salary of $76,292), while that of supervisory or other more senior staff reviewing the forms is $55.31 per hour (corresponding to an annual salary of $115,055). For 351 respondents filing 12 times per year, and using an average of 8.3 hours per form for completion plus 1.0 hour per review, and for 87 custodians filing 12 times per year, and using an average of 19.6 hours per form for completion plus 2.0 hours per review, the total annualized cost to respondents for the burden hours is estimated to be $2,381,332 (this amount equals custodian costs plus costs of others; = 87x12x(19.6x$36.68+2x$55.31) + 351x12x(8.3x$36.68+1x$55.31).

13. Estimated total annual cost burden to respondents, beginning November 2022 when the proposed form goes into effect

Total annual cost burden:

(a) Total annualized capital and start-up costs associated with Form SLT are estimated to be $0 (zero dollars). In general, reporting on the forms requires neither specialized capital equipment, nor fixed or variable costs that are not already associated with the customary and usual business practices of respondents.

(b) Total annualized operations, maintenance, and purchases of services costs are estimated to be $0 (zero dollars). Reporting on the forms does not in general impose operations, maintenance, or specialized services costs that are not already associated with the customary and usual practices of respondents.

The above cost estimates are not expected to vary widely among respondents.

Note: As required by OMB, the Federal Register notice of January 28, 2021 included an explicit request for public comments on the estimates of cost burdens that are not captured in the estimates of burden hours. No comments on cost estimates were received.

13. Estimated cost to the Federal government, beginning November 2022 when the proposed form goes into effect

Consistent with procedures for all TIC reports, Form SLT is printed, circulated, collected, and edited by the Federal Reserve Banks that process TIC data. All TIC forms, including Form SLT, are made available on the Treasury TIC website at http://www.treasury.gov/resource-center/data-chart-center/tic/Pages/forms-slt.aspx.

The total annualized cost to the Federal government, beginning November 2022 when the proposed form goes into effect, is estimated to be approximately $1,885,736. The figures are the best estimates by the staff of the Federal Reserve Bank of New York using their standard accounting and costing procedures and are based in part on experience gained by conducting other TIC surveys. Treasury Department staff has included additional expected costs for advisory services and dissemination of the information collected.

Estimated Annual Federal Costs for Form SLT

Salaries and benefits:	$ 687,032
Other direct expenses for operations, support and overhead:	$ 735,828
Computer systems and programming costs:	$ 327,390
Other Treasury and Board of Governors costs (mostly salary):	$ 135,486
Totals	$1,885,736

13. Reason for change

There is an increase in burden hours to 61,722 as compared to the estimate of 38,586 currently carried in OMB's Information Collection Inventory. The total overall increase of 23,136 hours is the result of an increase of 16,596 hours due to program changes (the program changes are detailed above, at the beginning of section 8, in items (3) and (4) of the Federal Register Notice) plus an increase of 6,540 hours from an increase in the number of respondents. Note: after the three-month overlap period with the S-form, it is expected that the S-form will be discontinued, resulting in a reduction in the total burden hours for all TIC forms.

13. Plans for tabulation, statistical analysis, and publication

Form SLT is needed to collect the data on an ongoing basis. Aggregate data in considerable detail by country, including breakdowns of holdings of domestic and foreign long-term securities, are published monthly in the Federal Reserve Bulletin and quarterly in the Survey of Current Business. Historical time series data, both aggregate and by country, reported on Form SLT are also posted to the Treasury TIC website (http://www.treasury.gov/resource-center/data-chart-center/tic/Pages/index.aspx). Data are published or otherwise made publicly available in aggregate form only, to avoid violating the confidentiality of any single respondent's submission. In addition, tabulations of these data frequently are made on an ad hoc basis for senior officials at the Treasury and the Board of Governors and other offices at Treasury and the Board.

13. Reasons why displaying the OMB expiration date is inappropriate

Approval to not display the expiration date for OMB approval on the TIC forms has previously been granted. The TIC forms on the website have expiration dates.

13. Exceptions to certification requirement of OMB Form 83-I

Regarding this request for OMB approval, there are no exceptions to the certification statement in item 19 of Form 83-I.

B. Collections of Information Employing Statistical Methods

The collection of this information does not employ statistical methods. Statistical methods are not appropriate for the type of information collected and would not reduce burden or improve accuracy of results.

May 2021