Supporting Statement A FINAL

U.S. Department of Commerce

U.S. Census Bureau

OMB Information Collection Request

Monthly Retail Surveys (MRS)

OMB Control Number 0607-0717

PART A. JUSTIFICATION

The Census Bureau plans to request a revision of the current Office of Management and Budget clearance for the surveys known as the Monthly Retail Trade Survey (MRTS) and the Advance Monthly Retail Trade Survey (MARTS). The MRTS and MARTS are related collections sharing the same initial sampling frame and jointly collect data that are published in conjunction with each other. These two surveys, currently cleared separately under control numbers 0607-0717 and 0607-0104, respectively, will therefore be combined under one control number and will be collectively called the Monthly Retail Surveys (MRS).

1. Necessity of Information Collection

The MRS are administered monthly to a sample of employer firms (i.e., businesses with paid employees) with establishments located in the United States and classified in retail trade and/or food services sectors as defined by the North American Industry Classification System (NAICS).

The MRTS provides estimates of monthly retail sales, end-of-month merchandise inventories, and quarterly e-commerce sales of retailers in the United States. In addition, the survey also provides an estimate of monthly sales at food service establishments and drinking places.

Sales, inventories, and e-commerce data provide a current statistical picture of the retail portion of consumer activity. The sales and inventories estimates in the MRTS measure current trends of economic activity that occur in the United States. The survey estimates provide valuable information for economic policy decisions and actions by the government and are widely used by private businesses, trade organizations, professional associations, and others for market research and analysis. The Bureau of Economic Analysis (BEA) uses these data in determining the consumption portion of Gross Domestic Product (GDP).

The MARTS, a subsample of MRTS, began in 1953 as a monthly survey for activity taking place during the previous month. MARTS was developed in response to requests by government, business, and other users to provide an early indication of current retail trade activity in the United States. Retail sales are one of the primary measures of consumer demand for both durable and non-durable goods. MARTS also provides an estimate of monthly sales at food service establishments and drinking places.

Retail and Food Services Sales during 2016 were estimated at $5.5 trillion. The estimates produced in the MRS are critical to the accurate measurement of total economic activity. The estimates of retail sales represent all operating receipts, including receipts from wholesale sales made at retail locations and services rendered as part of the sale of the goods, by businesses that primarily sell at retail. The sales estimates include sales made on credit as well as on a cash basis, but exclude receipts from sales taxes and interest charges from credit sales. Also excluded is non-operating income from such services as investments and real estate.

The estimates of merchandise inventories owned by retailers represent all merchandise located in retail stores, warehouses, offices, or in transit for distribution to retail establishments. The estimates of merchandise inventories exclude fixtures and supplies not held for sale, as well as merchandise held on consignment owned by others. The Bureau of Economic Analysis (BEA) use inventories data to determine the investment portion of the Gross Domestic Product (GDP). We publish retail sales and inventories estimates based on the NAICS.

Retail e-commerce sales are estimated from the same sample used to estimate preliminary and final U.S. retail sales. For coverage of the universe of e-commerce retailers, research was conducted to ensure that retail firms selected in the MRTS sample engaged in e-commerce. Total e-commerce sales for 2016 were estimated at $395 billion.

Sales data for select industries are released in the press release “Advance Monthly Sales for Retail Trade and Food Services,” approximately 10 business days after the close of the reference month, which also includes more detailed estimates for the prior month. Advance inventory estimates for 3 aggregate levels are released in the “Advance Economic Indicator Report” approximately 19 business days after the close of the reference month and the preliminary estimates for inventories data are released in the “Manufacturing and Trade Inventories and Sales” approximately 43 days after the reference month. E-commerce sales estimates are released quarterly as part of the “Quarterly Retail Ecommerce Sales” report, approximately 45 days following the reference period. Copies of the December 2016 releases mentioned above are included in Attachment 1.

The U.S. Census Bureau conducts these voluntary surveys under the authority of Title 13,

Sections 131 and 182 of the United States Code.

The MRTS and MARTS results are published on the Census Bureau’s website, http://www.census.gov/retail

Below are the retail form numbers along with a description of each form.

MRTS Forms

Series Description

SM-44(17)S Non Department Store/Sales Only/WO E-Commerce

SM-44(17)SE Non Department Store/Sales Only W E-Commerce

SM-44(17)SS Non Department Store/Sales Only/Screener

SM-44(17)B Non Department Store/Sales and Inventory/WO E-Comm.

SM-44(17)BE Non Department Store/Sales and Inventory/ W E-Comm.

SM-44(17)BS Non Department Store/Sales and Inventory/Screener

SM-45(12)S Department Store/Sales Only/WO E-Commerce

SM-45(12)SE Department Store/Sales Only/W E-Commerce

SM-45(12)SS Department Store/Sales Only/Screener

SM-45(12)B Department Store/Sales and Inventory/WO E-Commerce

SM-45(12)BE Department Store/Sales and Inventory/W E-Commerce

SM-45(12)BS Department Store/Sales and Inventory/Screener

SM-72(17)S Food Services/Sales Only/WO E-Commerce

SM-20(17)I Non Department and Department Store/Inventory Only

MARTS Forms

Series Description

SM-44(17)A Non Department Store/Sales Only/WO E-Commerce

SM-44(17)AE Non Department Store/Sales Only W E-Commerce

SM-44(17)AS Non Department Store/Sales Only/Screener

SM-72(17)A Food Services/Sales Only/ WO E-Commerce

Each MRS form has two versions; one with an “E” suffix and one with an “A” Suffix. The forms are identical, except that those with the “E” suffix are sent to smaller firms (which we refer to internally as “EINs”), while those with the “A” suffix are sent to larger firms, which we refer to internally as “alphas”. Thus, there are a total of 36 variants of forms along with their fax counterparts. Forms can be found at https://www.census.gov/retail/get_forms.html

Please note that with the start of mailing for our next business sample revision for the December 2017 statistical period, we will be removing the leased department forms: SM-45(17)S, SM-45(17)SE, SM-45(17)SS, SM-45(17)B, SM-45(17)BE, and SM-45(17)BS. Please see Part A, Section 5 for more information on the removal of leased departments.

2. Needs and Uses

The U.S. Census Bureau tabulates the collected data to provide, with measured reliability, statistics on United States retail sales. These estimates are especially valued by data users because of their timeliness.

The sales estimates are used by the Bureau of Economic Analysis (BEA), Council of Economic Advisers (CEA), Federal Reserve Board (FRB), Bureau of Labor Statistics (BLS), and other government agencies, as well as business users in formulating economic decisions.

BEA is the primary Federal user of data collected in the Monthly Retail Surveys. BEA uses the information in its preparation of the National Income and Products Accounts (NIPA), and its benchmark and annual input-output tables. Data on retail sales are used to prepare monthly estimates of the personal consumption expenditures (PCE) component of gross domestic product for all PCE goods categories, except tobacco, prescription drugs, motor vehicles, and gasoline and other motor fuel. These estimates are also published each month in the Personal Income and Outlays press release. If the survey were not conducted, BEA would lack comprehensive data from the retail sector. This would adversely affect the reliability of the NIPA and GDP. Production of the NIPA figures also require inventory figures in order to publish the monthly inventory to sales ratios. Additionally, they use MRS inventory figures to measure changes in inventories for estimates of gross output in the annual Input-Output Accounts tables, as well as for computing annual and quarterly GDP-by-industry statistics.

The Bureau of Labor Statistics (BLS) uses the data as input to their Producer Price Indexes and in developing productivity measurements. The data are also used for gauging current economic trends of the economy. BLS uses the estimates to develop consumer price indexes used in inflation and cost of living calculations.

CEA, other government agencies, and businesses use the survey results to formulate and make decisions. CEA reports the retail data, one of the principal federal economic indicators, to the President each month for awareness on the current picture on the “state of the economy”. In addition, CEA’s Macroeconomic Forecaster uses the retail sales data, one of the key monthly data releases each month, to keep track of real economic growth in the current quarter.

Policymakers such as the FRB need to have the timeliest estimates in order to anticipate economic trends and act accordingly.

Private businesses use the retail sales and inventories data to compute business activity indexes. The private sector also uses retail sales as a reliable indicator of consumer activity. In addition, businesses use the estimates to measure how they are performing and predict future demand for their products.

Information quality is an integral part of the pre-dissemination review of the information disseminated by the Census Bureau (fully described in the Census Bureau’s Information Quality Guidelines). Information quality is also integral to the information collections conducted by the Census Bureau and is incorporated into the clearance process required by the Paperwork Reduction Act.

3. Use of Information Technology

The Census Bureau offers and encourages electronic submission of responses via the Internet by using its Centurion instrument. Responding via the Internet provides companies with a convenient reporting method and reduces respondent burden. This method of data collection also reduces the amount of data that has to be keyed; therefore, reducing data capture costs. No additional software is needed by the respondent. In addition, all data submitted through Centurion are encrypted. A cover letter is included in the mailing package with instructions for reporting online (see Attachments 2 and 3 respectively).

We also continue to use other options such as facsimile technology and telephone to retrieve data from our respondents as well as other Census Bureau technologies, such as automated check-in to collect and process data. A 24-hour facsimile machine connected to a toll free telephone line permits respondents to fax data to our collection facility in Jeffersonville, Indiana. This process expedites the receipt of questionnaires, thus decreasing the number of telephone follow-up inquiries.

The National Processing Center (NPC) in Jeffersonville, Indiana performs telephone follow-up for all firms that have not responded by the due date, as well as those firms that have reported incomplete or questionable data. NPC will call from the third working day to the seventh working day of each month for MARTS and will call during the latter 2 weeks of the month for MRTS.

An automated system for check-in of returned questionnaires allows for timely identification of responses and removal from the delinquent workload. The automated telephone follow-up call scheduling system ensures that firms are contacted when the data are likely to be available, based on past experience with each firm. This process decreases the number of phone calls necessary to obtain data from respondents. Businesses may also respond by mail.

Below you will find a table indicating the percentage of response by each method:

Month	Survey	MRTS Check-In (%) by Type of Receipt (details may not equal total due to rounding)
		Mail	Internet	Fax/Clerk	Analyst	Total %
Jun-16	MRTS	12%	68%	19%	1%	100%
Jul-16	MRTS	12%	68%	19%	1%	100%
Aug-16	MRTS	11%	68%	20%	1%	100%
Sep-16	MRTS	12%	68%	19%	1%	100%
Oct-16	MRTS	11%	68%	20%	1%	100%
Nov-16	MRTS	11%	68%	20%	1%	100%
Dec-16	MRTS	11%	68%	20%	1%	100%
6 Mos Avg	MRTS	11%	68%	20%	1%	100%

Month	Survey	MARTS Check-In (%) by Type of Receipt (details may not equal total due to rounding)
		Mail	Internet	Fax/Clerk	Analyst	Total %
Jun-16	MARTS	9%	68%	22%	1%	100%
Jul-16	MARTS	6%	68%	25%	1%	100%
Aug-16	MARTS	8%	68%	22%	2%	100%
Sep-16	MARTS	8%	70%	21%	1%	100%
Oct-16	MARTS	6%	69%	24%	1%	100%
Nov-16	MARTS	7%	68%	24%	1%	100%
Dec-16	MARTS	9%	67%	23%	1%	100%
6 Mos Avg	MARTS	8%	68%	23%	1%	100%

4. Efforts to Identify Duplication

Consultations with other governmental agencies, trade associations (including the International Council of Shopping Centers (ICSC) and the National Retail Federation (NRF)) and government and private sector data users via telephone conversations, meetings, trade journal articles, and written correspondence indicate that these data are not available from other sources on an ongoing timely basis.

Annual sales and inventories by retailers are collected in the Annual Retail Trade Survey (ARTS) (OMB Control Number 0607-0013), and retail sales are collected in the quinquennial Economic Census. While the annual and quinquennial census data provide levels of sales for retail trade, they do not provide the data, on a timely basis, needed to monitor the current state of the economy. The latest data available from ARTS is for statistical year 2015. The monthly estimates are the only available data source for use as a trend series for evaluating current business conditions.

The MARTS sales estimates are based on early reporting of sales by a subsample of firms in MRTS. Rather than burdening the entire MRTS sample to report early enough to produce an advance estimate, we select a subset to produce the MARTS. Firms that report to the MARTS survey are not required to provide sales and ecommerce sales information for MRTS. Rather, sales and ecommerce data collected for MARTS are transferred into the database used for MRTS. This ensures that respondents do not have to provide duplicate responses. In addition, if companies have their end-of-month inventory estimates available during the MARTS collection, they can provide those figures at the same time.

5. Minimizing Burden

MARTS is a subsample of approximately 4,700 firms in the MRTS which in total contains approximately 12,000 firms.  Rather than burdening the entire MRTS sample to report early enough to produce an advance estimate, we select a subset to produce the MARTS. The design used to select the samples for both MARTS and MRTS use the least number of sampling units required to produce national level estimates with the desired level of reliability, thus minimizing respondent burden.  Sales data are collected from the largest firms and from a sample of small and medium-sized firms.  The selection of a new MRTS sample every five years replaces about 98 percent of the small and medium-sized firms that participate in the survey, thus minimizing respondent burden by redistributing reporting burden.  The selection of a new MARTS sample every two and one-half to three years replaces a portion of the small and medium-sized firms that participate in the earlier survey, thus redistributing the earlier response date burden. The current sample of retailers was introduced in April 2013 for MRTS and in November 2015 for MARTS.  We will begin collection on the next new sample for the December 2017 statistical period with publication on the new sample basis in Spring 2018.

The Census Bureau accepts data prepared on a company’s own form. This relieves the respondent of the burden of posting data to a report form.

Furthermore, firms are not required to maintain additional records. The data requested are generally maintained in existing company records. Carefully prepared estimates are acceptable if book figures are not available. Finally, interviewers use computers with online edits of response data to conduct follow-up and delinquent interviews, thus reducing the number of callbacks to respondents.

Effective with the next MRTS sample, which begins collection in December of 2017, we will be removing the Leased Department question and all impacted forms, and will no longer provide monthly estimates for this data series moving forward.  This follows the lead of the Economic Census and Annual Retail Trade Survey.  All historical series including these estimates will still be available to data users.  Leased departments are broadly defined as operations of one company conducted within the establishment of another company.  Currently, leased department sales are collected in only two industries, North American Industry Classification System (NAICS) 452111 (Department Stores) and 452112 (Discount Department Stores).  Removing this question will reduce the burden for the companies that are currently requested to provide this information.  The Census Bureau evaluated the collection and dissemination of this data series and determined that data from this inquiry may no longer be desired by external users and that the published estimates resulting from this question did not outweigh the added burden of requesting this data from the respondents.  We have therefore decided to remove both the leased department question from the survey as well as the resulting published estimates for the leased department series from the publication on the new MRTS sample.

6. Consequences of Less Frequent Collection

Estimates from MRTS and MARTS are used extensively by government and private economists and others to evaluate current economic conditions. The monthly retail sales and inventories estimates are important economic indicators and provide timely input for BEA’s computation of the National Income and Product Accounts. Less frequent data collection would create a serious gap in the economic information available to evaluate current economic conditions and to formulate economic policy. Moreover, these statistics provide current-to-previous month and year-to-year trend data that are essential for evaluating current conditions and for formulating economic policy.

7. Special Circumstances

There are no special circumstances.

8. Consultations Outside the Agency

A pre-submission notice was placed in the Federal Register on January 12, 2017 (82 FR

pg 3717-3719), inviting the general public and other Federal agencies to comment on the information collection. We received two letters, both from the Bureau of Economic Analysis, supporting the MARTS and the MRTS, copies of which are included as Attachments 4 and 5 respectively.

9. Paying Respondents

The Census Bureau does not provide any payment or gifts to respondents for participating in either the MARTS or the MRTS.

10. Assurance of Confidentiality

Data collected in these surveys are confidential under Title 13, Section 9 of the United States Code. We inform respondents of the voluntary and confidential nature of this survey in the initial mailing letter that we send to our contacts. This letter is available to respondents on our website at http://www.census.gov/retail/marts/get_forms.html. The initial mailing letter is included in Attachment 6. On a monthly basis, we inform respondents of the voluntary and confidential nature of this survey on the questionnaires/forms we provide as well as on our website. Screenshots of our electronic reporting instrument can be found in Attachment 7. The survey questionnaires/forms are included as Attachments 8 and 9. As mentioned above in Section 1:

Necessity of Information Collection, please note that since each MRS form has two versions; one with an “E” suffix and one with an “A” suffix, which are identical, we have included only the “A” suffix versions of the form in the attachments.

11. Justification for Sensitive Questions

The monthly retail surveys request only routine business information that is generally available from existing company records and is not of a sensitive nature.

12. Estimate of Hour Burden

Approximately 12,000 retail businesses are requested to report in the MRTS each month, which includes the MARTS subset of approximately 4,700.

According to responses given by a sample of respondents, we computed an average of five minutes per response for MARTS and seven minutes per response for MRTS. However it should be noted that to eliminate duplication, firms that report to the MARTS survey are not required to provide sales and ecommerce sales information for MRTS, as previously referenced in Section 5: Minimizing Burden.

This results in an annual burden of 16,799 hours (see chart).

Sample Size	Annual Responses	Total Responses	Hours	Total Burden Hours
12,000	12	144,000	0.11666 (7 min)	16,799

We used the MRTS sample size and burden hours to estimate the overall burden for the MRS since the MARTS sample is a subsample of MRTS. The MRTS survey has the maximum number of questions for any given survey respondent as well as the longest estimated burden (at 7 min).

The burden hours were previously calculated for MRTS and MARTS separately as the surveys were cleared under separate control numbers. The burden estimate for MRTS (Control Number 0607-0717) was previously 14,427 hours and the burden estimate for MARTS (Control Number 0607-0104) was previously 4,900 hours, totaling 19,327 requested hours for both surveys. As these are related surveys and the collection of data in the MARTS feeds into the MRTS, we have consolidated the calculation of the burden estimate resulting in an adjustment of -2,528 hours requested for the two surveys combined. This adjustment is due solely to the change in which the burden is estimated for the two surveys together and not due to any program changes.

The annual cost to respondents is estimated to be $542,609 based on the median hourly salary of $32.30 for accountants and auditors. (Occupational Employment Statistics- Bureau of Labor Statistics May 2015 National Occupational Employment and Wage Estimates, $32.30 represents the median hourly wage of the full-time wage and salary earnings of accountants and auditors SOC code 13-2011)

http://stats.bls.gov/oes/current/oes132011.htm

13. Estimate of Cost Burden

We do not expect respondents to incur any costs other than that of their time to respond.

The information requested is of the type and scope normally maintained in company records and no special hardware or accounting software or system is necessary to provide answers for this information collection. Therefore, respondents are not expected to incur any capital and start-up costs or system maintenance costs in responding. Furthermore, purchasing of outside accounting or information collection services, if performed by the respondent, is part of usual business practices and not specifically required for this information collection.

14. Cost to Federal Government

The total cost to the Federal Government for the Monthly Retail Surveys (MARTS and MRTS combined) is expected to be relatively fixed over the upcoming three years at approximately $4.3 million per year, all borne by the Census Bureau. This estimate includes the cost for such things as data collection, processing, review of tabulated data, publication, equipment, overhead, printing, support staff, etc.

15. Reason for Change in Burden

This burden now incorporates the burden for both the MRTS and MARTS surveys. As explained above in Item 12, we used the MRTS sample size and burden hours to estimate the overall burden for the MRS since the MARTS sample is a subsample of MRTS. The MRTS survey has the maximum number of questions for any given survey respondent as well as the longest estimated burden (at 7 min).

The burden hours were previously calculated for MRTS and MARTS separately as the surveys were cleared under separate control numbers. The burden estimate for MRTS (Control Number 0607-0717) was previously 14,427 hours and the burden estimate for MARTS (Control Number 0607-0104) was previously 4,900 hours, totaling 19,327 requested hours for both surveys. As these are related surveys and the collection of data in MARTS feeds into MRTS, we have consolidated the calculation of the burden estimate resulting in an adjustment of -2,528 hours requested for the two surveys combined. This adjustment is due solely to the change in which the burden is estimated for the two surveys together and not due to any program changes.

We also anticipate that the annual reporting burden will be reduced as a result of our new sample for 2017. Business sample revisions are further discussed in Supporting Statement B. We anticipate the new MRTS sample to have approximately 11,300 respondents which will result in a new burden estimate of 15,819 hours.

16. Project Schedule

The Census Bureau mails the MARTS forms to the respondents roughly 6 business days prior to the end of the reference month and mails the MRTS forms to the respondents on the last workday of the reference month. The return date requested is the second business day after the close of the reference month for MARTS and the 8^th business days following the close of the reference month for MRTS.

An automated system screens the questionnaires for completeness and consistency when the data are keyed. Data are tabulated and edited, summary estimates are analyzed, and data tables are prepared.

Sales data for select industries are released in the press release “Advance Monthly Sales for Retail Trade and Food Services,” approximately 10 business days after the close of the reference month, which also includes more detailed estimates for the prior month. Advance inventory estimates for 3 aggregate levels are released in the “Advance Economic Indicator Report” approximately 19 business days after the close of the reference month and the preliminary estimates for inventories data are released in the “Manufacturing and Trade Inventories and Sales” approximately 43 days after the reference month. E-commerce sales estimates are released quarterly as part of the “Quarterly Retail Ecommerce Sales” report, approximately 45 days following the reference period.

17. Request to Not Display Expiration Date

We wish to continue to display the expiration date.

18. Exceptions to the Certification

There are no exceptions to the certification statement.

19. NAICS Codes Affected

The following are the 3-digit NAICS codes for the retailers affected by the information collection:

NAICS

Code Description

441 Motor Vehicle and Parts Dealers

442 Furniture and Home Furnishings Stores

443 Electronics and Appliances Stores

444 Building Material and Garden Equipment and Supplies Dealers

445 Food and Beverage Stores

446 Health and Personal Care Stores

447 Gasoline Stations

448 Clothing and Clothing Accessories Stores

451 Sporting Goods, Hobby, Book, and Music Stores

452 General Merchandise Stores

453 Miscellaneous Store Retailers

454 Non-store Retailers

722 Food Services and Drinking Places