GDT -- Support Statement Part A 12-20-2017 v4

SUPPORTING STATEMENT

UNITED STATES INTERNATIONAL TRADE COMMISSION QUESTIONNAIRE

Inv. Nos. 332-562 and 332-563

Global Digital Trade 2: The Business-to-Business Market, Key Foreign Trade Restrictions, and U.S. Competitiveness; and Global Digital Trade 3: The Business-to-Consumer Market, Key Foreign Trade Restrictions, and U.S. Competitiveness

Part A-Justification

1. Request for regular action

The U.S. International Trade Commission (USITC or Commission) is seeking approval for use of a questionnaire in connection with two reports it is preparing for the U.S. Trade Representative (USTR). The Commission’s reports, Global Digital Trade 2: The Business-to-Business Market, Key Foreign Trade Restrictions, and U.S. Competitiveness (Investigation No. 332-562) and Global Digital Trade 3: The Business-to-Consumer Market, Key Foreign Trade Restrictions, and U.S. Competitiveness (Investigation No. 332-563), were requested by the USTR on January 13, 2017. The Commission will deliver the results of its investigation of the business-to-business market to the USTR by October 29, 2018 and its investigation of the business-to-consumer market to the USTR by March 29, 2019.

Pursuant to section 332(g) of the Tariff Act of 1930 (19 U.S.C. 1332(g)) and the authority delegated by the President of the United States to the U.S. Trade Representative under Executive Order 12661, the Commission, whenever requested, “shall put at the disposal of the President […] all information at its command, and shall make such investigations and reports as may be requested by the President […].” Section 333(a) of the Tariff Act of 1930 authorizes the Commission, in carrying out its functions and duties in connection with any investigation authorized by law, to obtain information, including by subpoena or other order to furnish information. Copies of section 332(g) and section 333(a) of the Tariff Act of 1930 are attached as supplementary documents.

Questionnaires are one method through which the Commission will collect information for this request. The Commission is using one questionnaire for both investigations to obtain relevant information for each of these investigations. On November 9, 2017 the Commission posted its draft questionnaire on its website (https://www.usitc.gov/globaldigitaltrade), and published a request for public comments in the Federal Register on the draft. Public comments will be accepted through January 15, 2018.

As indicated above, the Global Digital Trade 2: report is due to the USTR by October 29, 2018. This deadline necessitates prompt issuance of questionnaires. Requests to complete the online questionnaire are tentatively scheduled to be mailed to respondents on or before February 6, 2018 in order to allow sufficient time for each recipient to complete the questionnaire, which is due back to the Commission by March 8, 2018.

2. Purpose

In its request letter, the USTR directed the Commission to produce two reports that analyze the impact of regulatory and policy measures in key foreign markets on (1) the ability of U.S. firms to develop and/or supply digital products and services abroad, and (2) the competitiveness of U.S. firms engaged in the sale of digital products and services, as well as on international trade and investment flows associated with digital products and services.

USTR specified that the Commission’s reports be based on a review of available data, including a survey of U.S. firms in selected industries particularly involved in digital trade. The use of a questionnaire to collect relevant information is thus an integral part of the USTR’s request and will aid the Commission in fulfilling its statutory duty to provide the requested information. The information to be collected is critical to the Commission's task in addressing the elements of the USTR’s request, since such data are not publicly available. The information collected from questionnaires will be aggregated by the Commission to preserve the confidentiality of information received from the respondent firms.

As drafted, the questionnaire mandates a response from recipients. Pursuant to section 333(a) of the Tariff Act of 1930, the Commission is authorized to compel persons, firms, a partnership, a corporation, or an association to furnish in writing, in such detail and in such form as the Commission may prescribe, information in their possession pertaining to such investigation. Failure to require mandatory responses would likely significantly depress response rates. Without a robust response, the Commission will be unable to furnish the information requested by the USTR and will be unable to satisfactorily discharge its responsibility under section 332(g) of the Tariff Act of 1930.

3. Use of technology

The Commission will use survey software to allow respondents to complete an online questionnaire. The online version will incorporate skip-logic and piping to ensure respondents only see questions that are relevant to them or that include options based on responses to previous questions. The online form incorporates quality control functions, submission buttons, and summation functions and will require an individual ID for each respondent to access. A fillable form version of the questionnaire will also be available. This version will be available for download from the USITC website.

4. Non-duplication of available data

To the extent possible, the Commission’s investigations rely on existing publicly available data. Commission staff has reached out to government, academic, and industry leaders in this sector, and have confirmed that there is no publically available data needed for the analysis required in the reports and that sending a questionnaire is the only way to obtain primary data. Further, after a thorough background search of data sources for these investigations, it has been determined that no other industry, government, or academic organizations collect or publish data that are duplicative of the data requested in the questionnaire.

5. Impact on small businesses

Minimum firm sizes (by employment and revenue) were used to exclude firms that were unlikely to take part in cross-border digital trade. To minimize the reporting burden, the questionnaire was designed to be as brief as possible, consistent with information requirements. Skip-logic ensures that firms answer only relevant sections. Check-the-box and select-from-a-list response options are used to simplify the questionnaire where appropriate. In addition, the questionnaire indicates that carefully prepared estimates are acceptable, which should further reduce the potential burden on smaller firms that may not have sufficient administrative resources or automated record-keeping systems.

6. Consequences of non-collection

Due to the lack of suitable data from other sources, without this information collection, the Commission will be unable to fully respond to the USTR’s request and therefore will be unable to satisfactorily discharge its responsibility under section 332(g) of the Tariff Act of 1930 (19 U.S.C. 1332(g)).

7. Frequency of data collection

This is a one-time, nonrecurring data collection.

8a. Consistency with 5 CFR 1320.6 guidelines

No special circumstances exist that require the collection to be conducted in a manner inconsistent with the guidelines of 5 CFR 1320.6. If any respondents do not maintain information in the format requested by the questionnaire, they are requested to submit carefully prepared estimates based upon available information.

8b. Consultations with affected public

The Commission’s 60-day notice requesting public comment was published in the Federal Register on November 9, 2017. The notice and other information related to this study were published on the Commission’s internet site at https://www.usitc.gov/globaldigitaltrade.

The Commission’s 30-day notice of submission to OMB requesting clearance was published in the Federal Register on December 27, 2017 (82 FR 61325). The notice is also posted on the Commission’s internet site at http://www.usitc.gov/research_and_analysis/What_We_ Are_Working_On.htm. No public comments have been received to date.

Commission staff field tested the questionnaire with regard to the availability of data, reporting burden, product coverage and definitions, clarity of instructions, disclosure, and reporting format. All field testers received copies of the draft questionnaire. Commission staff requested comments from nine and received comments from eight firms that represent members of the manufacturing, software development, cloud computing, and e-commerce industries. Comments and responses are summarized in table 1.

TABLE 1 – Field testing comments and Commission response¹

Comment/Suggestion	Response
Q2 (Q1.3): Some larger firms may have difficulty choosing one aspect of their business as the primary industry, or fitting their industry in one of the options	The industry options are matched to the stratification being used for sampling, so the list cannot be aggregated further. The question suggests contacting the project team for clarification when needed.
Q3 (Q1.4): Suggest adding specific technical language to ask about the legal basis of the company in case the company acts like a holding company but is not technically a holding company. Firms may say ‘no’ to avoid having to coordinate/fill out several surveys.	These situations will only apply to a small number of firms and this nuanced issue would be better handled by contacting the project team. As the question already instructs the respondents to contact the project team in this circumstance, no change was made.
Q5 (Q1.6): Will firms understand how to interpret “U.S. employment (full figure)”	Changed “full figure” to “full-time equivalents” and added a definition for “full-time equivalents”
Q7 (removed question): Add a N/A or unsure column to round out options and clarify the overlap between categories	This question has been removed from the questionnaire as a result of the overlapping categories
Q8 (removed question): May not be binary, some firms with export and have foreign affiliates depending on the market	Removed this question.
Q9 (Q2.1): Concern with the use of the term “E-commerce”	Clarified in definitions, changed the term “E-commerce” to “Online commercial and retail services, and electronic payment services” throughout
Q9 (Q2.1): Move “Other services” to the bottom; add "Fintech" as an option	Reordered the products/services and put “other services” at the bottom; too few firms provide “fintech” services to allow for responses to be statistically significant, so a separate category will not be added
Q9 (Q2.1): Add ICT hardware in addition to software	Added language to include ICT hardware
Q11 (Q2.3): Suggest adding column for “Not applicable” in case they do not provide the service	Did not add. In the web version, firms will only see products/services they chose above.
Q11 (Q2.3): Define B2B and B2C; could be difficult to determine, suggest adding “Not possible to determine”	Added to definitions; this question is needed to filter between reports so cannot add “Not possible”
Q12 (Q3.1): Clarify that not all regulatory and privacy measures are negative	Added clarifying language to definitions
Q12 (Q3.1): Include “requirement to locate data centers and other physical IT infrastructure in-country as a condition of operation” in “In-country data storage processing requirements”	Added language into the definitions
Q12 (Q3.1): “Encountered” is too ambiguous. Indirectly vs. directly encountered? Time intensive to figure out.	Did not add. Specifying whether firms directly or indirectly encounter measures would introduce more confusion to the question.
Q12 (Q3.1): Consider adding ‘No’ column	Did not add. Based on standard survey design practice, fewer answer choices lead to fewer errors. Adding a “no” column would introduce greater opportunity for error (clicking incorrectly or skipping). Non-responses with two choices would also increase data checking and burden.
Q13 (Q5.2, Q5.3), 17 (Q5.6, Q5.7): Consider having fewer buckets for responses; more buckets may suggest the need for precision.	Because these questions are necessary for modeling purposes, cannot reduce the number of buckets. Added language to allow firms to roughly estimate.
Q15 (Q3.2): Suggest separating a and b into two separate questions or columns	Created an additional column, allowing firms to select either option for each market
Q18 (Q4.1.1): Should include “maintain internal operations”	Added to questions 4.1.1, 4.2.1, 4.3.1, 4.4.1, and 4.5.1
Q19 (Q4.1.2): Unsure how to interpret “market share”, “costs”; Suggest moving “Not applicable” to the far left.	Removed “Market share”; added “Operational and compliance” to “costs” in questions 4.1.2, 4.2.2, 4.3.2, 4.4.2, and 4.5.2; Did not move “Not applicable”, past experience suggests improved responses when this is the last option rather than the first
Q22 (Q4.1.6): Suggest adding positive column	Added a column for a positive effect response to market specific questions in each sub-section in section 4 (questions 4.1.6, 4.2.6, 4.3.6, 4.4.6, and 4.5.6). For questions 4.1.1, 4.2.1, 4.3.1, 4.4.1, and 4.5.1 a note is added that positive responses should select “Encountered but not restrictive” column. Added language to definitions and beginning of each sub-section in section 4 to “primarily consider” measures that negatively impact operations, rather than “only consider” the negatively impact; A question allowing firms to describe any positive impacts of certain regulatory and policy measures has been added (5.10).
Q23 (removed question): Question asks about removing measures, but removing data protection and privacy measures would be a bad thing	Removed this question from all measures sections
Q24 (Q4.1.7): Add option for “not aware of what GDPR is”; add “N/A”	Added options for “Aware of GDPR, but do not know the anticipated impact” and “not aware of GDPR”; firms only see this question if they have chosen “EU” as a market, so “N/A” does not need to be an option
Q48 (Q4.5.1): Emphasize the different measures in section 4	Have bolded the different measures in each part of section 4
Q49 (Q4.5.2): Question references question 15 (Q3.2), firms may forget how they answered	Clarified to indicate that questions refer to all foreign markets
Q55 (Q6.1): Will there be a word limit or ability to attach files?	There is no word limit; we are not able to allow attachments as part of the questionnaire. Firms can submit both public and nonpublic information to the Commission.

Staff also conducted cognitive interviews with eight organizations representing the cloud computing, internet search, financial services, and software development industries. Cognitive testing ensures that a survey question successfully captures the scientific intent of the question and is readily understandable by to respondents, thereby increasing the overall quality of the survey data (table 2).

TABLE 2 – Cognitive testing comments and Commission response²

Comment/Suggestion	Response
Q9 (Q2.1): Did not understand what “within their organization” meant. Does it mean any products used, even if they were other services provided to them?	Updated to say “to maintain internal operations (not including services or products from external vendors)”
Q12 (Q3.1): “Encryption” is missing; “cybersecurity” is missing	Added “encryption” to “Commercial cryptography/encryption regulations” Did not add “cybersecurity” because cybersecurity is not technically a measure but a justification for certain measures.
Q13 (Q5.2, Q5.3), 14 (Q5.4), 17 (Q5.6, Q5.7): Cost questions will be difficult and time consuming to answer. Suggest removing. If not able to remove, suggest moving the questions to the end or only asking Total cost instead of total and variable/only asking revenue.	The cost questions are required for the model. Moved the cost questions to the end, section 5; added language to make clearer that firms may roughly estimate costs.
Q15 (Q3.2): While responding to the question, firms were missing part b and only responding with markets they are currently in	Put “prevented” in bold and added a separate column for prevented entry
Q16 (Q3.3): Revenue would be hard to track by country; it’s easier to track for some markets	Add “Estimated” to question and column header.
Q18 (Q4.1.1): Too many potential selections; “somewhat” and “moderate” are not sufficiently different; “overall” is confusing – should move to top	No change; other scales that were considered were problematic as well.
Q20 (Q4.1.3): Suggest adding in “being able to expand customer base” (or scalability), and “Access to finance/funding.” These are primary concerns for small firms	Added both as suggested
Q20 (Q4.1.3): Prefer the restrictiveness scales rather than impact scales; High and Severe are insufficiently different; Not applicable and No impact are the same	Cannot change scales as they are asking for different things; removed “Severe” column; need “Not applicable” in order for options to be exhaustive
Q20 (Q4.1.3): Firms will answer based on a specific market rather than globally	Added language at the beginning of the section instructing firms to answer based on global operations
Q22 (Q4.1.6): Want fewer choices; space for narrative answers	Kept 5 choices to match similar questions and 5 is a standard survey practice; added an additional narrative question, there is sufficient space for narrative responses throughout the questionnaire

9. Payments or gifts

Not applicable. Questionnaire recipients will not be provided with any payments or gifts for their responses.

9. Assurances of confidentiality

In the questionnaire, the Commission provides recipients with an assurance of confidentiality, indicating the following two paragraphs: “The Commission will designate the information you provide in response to this questionnaire as “confidential business information,” to the extent that the information would reveal the operations of your organization and is not otherwise available to the public. Per USTR’s request, the Commission will not include any confidential business information in its reports. The confidential business information provided may be disclosed and used as provided for in section 6 of the questionnaire: (i) by the Commission, its employees and Offices, and contract personnel (a) for developing or maintaining the records of this or a related proceeding, or (b) in internal investigations, audits, reviews, and evaluations relating to the programs, personnel, and operations of the Commission including under 5 U.S.C. Appendix 3; or (ii) by U.S. government employees and contract personnel (a) for cybersecurity purposes or (b) in monitoring user activity on U.S. government classified networks. We will aggregate the information we receive in your response to this questionnaire with information from other responses to the questionnaire and will not publish it in a way that would reveal the operations of your organization. Aggregate responses will be considered National Security Information, as requested by USTR.”

“Furthermore, the reports based on this survey have been classified as National Security Information (NSI) by the USTR. Unless declassified by the USTR, the reports will not be publicly available for a period of up to 10 years. Consistent with Executive Order 13526, these reports, and analyses based on this survey, are being classified on the basis that it concerns economic matters relating to national security that impact USTR negotiation and enforcement priorities. USTR has stated that it intends to treat these reports as interagency memoranda containing pre-decisional advice subject to the deliberative process privilege.”

9. Sensitive information

The Commission is not seeking information on issues of a sensitive nature involving persons or firms.

9. Respondents’ projected cost burden

The Commission has reduced the reporting burden on respondents by using an online form with skip-logic and piping and by limiting the length and complexity of the questionnaire. Furthermore, the questionnaire only contains questions that the Commission believes to be readily available from firms’ existing records.

The reporting burden is estimated to be:

Number of respondents: (No.) 13,000

Frequency of response: (No.) 1

Average burden per respondent: (hours) 17

Total burden: (hours) 221,000

These estimates are based on reported response statistics from recently completed, similar surveys that the Commission has executed. It is estimated that completing the questionnaire, including time to gather necessary information, would take approximately 17 hours depending on the size and complexity of the firm. The burden on individual respondents may vary; smaller firms will likely have fewer sections to complete than larger, multinational firms. Moreover, the total burden is likely to be considerably lower because the total response rate will likely be less than 100 percent, and firms without operations abroad will complete only a small portion of the questionnaire.

The Commission has included a notice of the above response burden averages in the questionnaire, along with a request that respondents send comments to the Commission and to OMB.

The combined annualized cost to all respondents for the estimated hour burdens identified above is as follows:

Cost = 221,000 hours x $57.69³ per hour = $12,749,490

9. Annual public response burden

This is a one-time collection of information and therefore the total recurring annual cost burden is zero.

a. Total capital and start-up cost component: The Commission does not expect any capital and start-up costs because all information likely already exists in firms’ records storage facilities.

b. Total operation and maintenance and purchase of service component: The Commission does not expect respondents will need to purchase any services in completing the questionnaire.

9. Federal change in burden

The estimated total cost to the Federal Government is $384,986 as detailed below. No new equipment will be purchased because existing equipment will be used to process the questionnaires.

The estimated number of work hours includes designing the questionnaires, soliciting field test comments, editing results (i.e., contacting respondents after completion of the questionnaires to clarify responses), and compiling and tabulating questionnaire responses.

Total personnel cost⁴ = $359,986

Operational costs⁵ = $ 25,000

Total cost = $384,986

9. Program change justification

The Commission currently imposes no reporting burden on firms with respect to digital trade. The burden on firms increased because of a request from the USTR to provide two reports on this topic. Such data are not publicly available. This is a one-time collection for such data. The USTR’s request letter is attached as a supplemental document.

9. Project plan and schedule
   
   

After receiving completed questionnaires, Commission staff will review and edit each response for accuracy, resolve any questions with the respondent, and tabulate the responses. Data will be analyzed, compiled in a form that will not reveal the individual operations of any respondent, and prepared for publication. A letter providing the link to the online questionnaire is scheduled to be mailed on or before February 6, 2018. The respondents are requested to respond by March 8, 2018. Incorporating questionnaire information, the Commission will deliver the results of its investigation into the business-to-business market to the USTR by October 29, 2018 and its investigation of the business-to-consumer market to the USTR by March 29, 2019.

9. Non-display of expiration date

Not applicable, the expiration date appears at the beginning of the Commission questionnaire.

9. Exceptions to certification statement to form OMB 83-I

Not applicable.

1 Question numbers referenced are based on the version of the questionnaire that was issued for public comment on November 9, 2017. Unless the question has been removed, updated question numbers corresponding to the attached questionnaire are in parentheses.

2 Question numbers referenced are based on the version of the questionnaire that was issued for public comment on November 9, 2017. Unless the question has been removed, updated question numbers corresponding to the attached questionnaire are in parentheses.

3 The hourly figure was approximated by dividing the Commission’s average salary level ($120,000) by the number of work hours per year (2,080), which is equivalent to an average wage of $57.69 per hour. The Commission projects that this is an accurate hourly cost estimate for personnel who will likely complete the questionnaire.

4 The Commission estimates that 6,240 personnel hours (156 personnel weeks) will be spent on the questionnaire (at average wage), which is approximately 14 percent of the total personnel hours the Commission has budgeted for both studies combined. Personnel costs include three weeks of two full time staff members to follow up with respondents. This time also includes calls to questionnaire recipients from Commission staff to ensure that organizations reply to the questionnaire and that responses received are accurate.

5 Operational costs include printing and mailing. Costs for printing and mailing postcards, letters, and reminders provided by a mailing house are estimated at $25,000.

11