Supporting Statement A (1220-0026)

SUPPORTING STATEMENT - IPP IMPORTS (CONT'D)

SUPPORTING STATEMENT - IPP IMPORTS

A. JUSTIFICATION

1. The U.S. Import Price Index together with the U.S. Export Price Index, the Consumer Price Index, and the Producer Price Index, constitute the major outputs of the price programs of the Bureau of Labor Statistics. Although the International Price Program (IPP), which produces the U. S. Export and Import Price Indexes, is the Bureau of Labor Statistics' newest price program, it can trace its origins to the late 19th Century. In 1886 the Aldrich Committee of the U.S. Senate recommended the establishment of a Bureau of Labor to provide statistics on the condition of U.S. workers and the prices of imported goods in the U.S. and other countries. The committee sent staff members to other countries, principally in Western Europe, to collect prices and in 1889 published a report comparing prices in the U.S. with those of Western Europe. This report, which focused on prices for products imported into the U.S., was the precursor of the Wholesale Price Index. Following World War II, the BLS again began a program to develop import and export price indexes. The program advanced to the point where hundreds of prices had been collected from importers and exporters and test indexes had been calculated. Because of a Bureau‑wide 50 percent budget reduction, however, the program was terminated in 1948.

In 1961, a report on Federal Price Statistics prepared by the National Bureau of Economic Research (NBER) for Congress' Joint Economic Committee suggested that responsibility for compilation of import and export price indexes be assigned to a federal statistical agency "to obtain the attention and resources for these indexes that we believe are essential." A further study undertaken for the NBER by Professors Irving Kravis and Robert Lipsey gave greater impetus to the project. In their study, eventually published as Price Competitiveness in World Trade, Kravis and Lipsey outlined both the need for such measures and the feasibility of producing them. In the meantime the BLS, largely because of its expertise in the development of other price measures, had also begun research on the feasibility of producing import and export price indexes. In 1970, Congress provided funds for the construction of import and export price indexes. The legal authority for the collection of import and export data is contained in Title 29, Section 2 of the United States Code. (See Attachment 1).

The first annual import price indexes were produced in 1973. Largely as a response to changing international economic conditions and the need on the part of both the government and the private sector to obtain these data on a more timely basis, collection and publication of the international price indexes were begun on a quarterly basis in 1974. A general index for all-import products was published for the first time in the fourth quarter of 1982.

The expansion of international trade and improvements in the design of the IPP survey led the Office of Management and Budget (OMB) in 1982 to place the IPP indexes on its list of Principal Federal Economic Indicators alongside the Consumer Price Index (CPI) and the Producer Price Index (PPI). Economic indicators placed on this list must be released on schedule and are recommended for use in public and private sector economic analysis.

The increasing importance and value of the IPP's indexes led to requests in 1988 from OMB and several other policy-making government agencies for monthly indexes. To fill this need, the IPP initiated an effort in late 1988 to provide these agencies with monthly indexes for all-imports, all-exports, and certain highly-aggregated import and export product groupings. Using a subset of data from the regular quarterly sample, the IPP began publishing these indexes in February 1989. Because of continuing interest from OMB and other government agencies and because of the need to deflate monthly GDP figures using IPP indexes, the IPP now collects all of its data for products on a monthly basis.

The international services sector has grown tremendously over the last 20 years, creating the need for more comprehensive, reliable, and timely information on price trends of international prices. Services compose approximately 14 percent of cross-border import trade. Highlighting the growing importance of services and the need for expansion of services data, recently a letter from the President’s Export Council Subcommittee requested that the President “give greater importance to the collection and utilization of statistical data on Services” and stating that “as the global engagement of U.S. services companies continues to expand, more resources are required to improve these agencies’ collection and analysis of services data (May 25,2005).” Currently, the IPP is publishing import indexes for air passenger fares, air freight rates, crude oil tanker freight, and ocean liner freight.

2. The most critical uses of the IPP indexes are found in the public sector. Major public-sector uses of the IPP indexes include Deflating Monthly Import and Export Trade Statistics, deflating the foreign trade component of Gross Domestic Product, determining monetary and fiscal policy, determining trade and commercial policy, negotiating trade agreements, and escalating government contracts. The prices provided by respondents form the foundation of information necessary to assure that the IPP indexes accurately reflect conditions in the international marketplace. U.S. policy makers must have reliable, accurate statistics to insure that appropriate actions are taken, especially during periods of economic difficulty. Policy decisions based on inadequate information can adversely affect the U.S. economy and consequently individual companies, including those asked to provide IPP price data. When public policy makers have reliable statistics on international trade, they are in a better position to make sound decisions on the regulation and promotion of international trade. These decisions can benefit all internationally active companies, as well as the U.S. economy as a whole.

IPP produces monthly indexes in order to provide information with which to deflate the monthly merchandise trade data issued by the Department of Commerce. (Attachments 2 and 3 are examples of trade balances issued monthly in "United States Department of Commerce News: U.S. Merchandise Trade".) The resulting real trade flows, obtained by using monthly international price indexes as deflators, enable measurement of real output and provide a more comprehensive understanding of the underlying dynamics of international trade.

The Commerce Department also uses international price indexes to adjust for inflation in the foreign trade sector of its quarterly National Income and Product Account (NIPA). (Attachment 4 is the constant dollar tabulation of imports and exports from the U.S. Department of Commerce, Survey of Current Business).

In addition to serving as a tool for the public sector, the Import Price Index has a variety of other private sector uses by the media, bankers, financial analysts, academic researchers, and corporate managers. These uses include market analysis, forecasting future price trends, estimating for contract escalation and replacement cost accounting, measuring import price and income elasticity, and estimating exchange rate pass-through values and the effect of currency fluctuations on prices by specific countries or regions.

The Import Price Indexes (MPI) in combination with Export Price Indexes (XPI) can also be used in various ways to measure a country’s international competitiveness. One method for indicating international competitiveness is through the use of terms of trade indexes. A terms of trade index I s defined as an XPI divided by the respective MPI. Because demand for imports and exports are tied to import and export prices, a change in the terms of trade will lead to a change in the trade balance. Another way to look at international competitiveness is by expressing Import and Export Price Indexes in foreign currency terms. Foreign currency import price indexes measure fluctuations in the revenue by foreign sellers in the U.S., and foreign currency export price indexes illustrate how U.S. export prices vary from the perspective of buyers of U.S. products.

2. Historically, the primary interaction between the Bureau and the respondents is through the repricing form. Prior to the anthrax problems, which curtailed the mail service in October 2001, nearly all IPP repricing data were collected via the mail. Forms were mailed out to respondents and they were returned in BLS-supplied envelopes. As a result of the curtailment of incoming mail, the program discontinued this method of data collection and switched entirely to mailout/fax-back of nearly all forms.

In 2003, IPP introduced a web application for monthly data collection. This tool permits respondents to directly update their data online via the internet. Web collection has expanded rapidly since the IPP started soliciting respondents in 2003. As of August 2006, the Program had solicited 79 percent of all respondents, exceeding its goal of 75 percent solicitation by September 2006. The Program plans to have solicited 95 percent of all respondents for web repricing by September 2007. Through August 2006, 45 percent of IPP respondents are actually providing prices via the web application or have agreed to start using this method. Currently, Field Economists offer this option to all new respondents, and at initiation, it is the preferred method of collection offered to companies. Over the next few years, the IPP will solicit all its respondents for interest in web repricing.

Of the remaining respondents, the majority reprice using the mailout/fax back process while a small percentage provide data via non-automated phone, email, or other special arrangements between the analysts and the respondents. In addition, e-mail repricing has the possibility of expanding, depending upon how data security issues are handled in the Bureau.

Over time, these various electronic data collection methods for repricing have permitted, and will continue to permit, the Program to more rapidly collect and publish monthly information. Each month the IPP is the first of the 3 BLS price series to be published.

In 1999 the IPP implemented the use of direct data entry for initiating new reporters into the survey. In early 2003, the IPP implemented a further enhancement of this direct data entry process that allows staff in the National Office to review the initiation data in the same system as the data entry system. This enhancement has further reduced the lag time between the collection of price data and their use in IPP indexes, ensuring more accurate and timely IPP indexes.

The IPP has implemented a few changes in recent years to reduce burden on IPP reporters, especially those companies which are major traders and account for a significant portion of international trade. We now provide our field economists with more accurate information about the potential overlap between establishments that are in both the IPP and the Producer Price Index. This information allows the Field Economists to better co-ordinate visits to establishments to obtain new items for repricing, ensuring that we are adhering to requests from establishments about the timing of our visits. We also implemented an enhanced refinement process that provides Industry Analysts the ability to reduce the burden for a respondent when it is needed. Finally, we modified our second stage selection algorithm to lower the percentage of infrequently traded ELIs that are sampled, since they are more likely to be out-of-scope for the IPP. We believe that these improvements should reduce the overall burden on respondents and will improve IPP’s overall response rate at initiation.

4. The U.S. Customs Service collects data on the value of all U.S. imports and exports. Until 1989, the Department of Commerce used these data to construct unit value indexes. These indexes have been shown to be inadequate and were discontinued in October 1989. Since then, the IPP Indexes are the sole comprehensive price indexes for imports and exports.

Generally, similar data which exist in the field of international prices cannot be used in lieu of the data collected by the IPP survey because the only "similar" data (trade journal prices and the former Department of Commerce unit value indexes) are the same data whose deficiencies prompted the creation of the IPP survey. However, where feasible, and in order to reduce costs and duplication, the Program does use secondary source data. For example, the IPP survey does incorporate U.S. Department of Agriculture, U.S. Department of Energy, and certain other published market data in selected product and service areas.

5. The sampling procedures used by the IPP tend to select firms that are high‑volume, regular traders in a product or service area. This technique minimizes the chances of small organizations being selected to report data for more than one or two items.

6. The International Price Program indexes are closely followed statistics which are viewed as a sensitive indicator of the economic environment. Federal policy‑makers in the Department of Treasury, the Council of Economic Advisors, the Bureau of the Census, the Bureau of Economic Analysis (BEA), and the Federal Reserve Board utilize these statistics to form and evaluate monetary and fiscal policy and the general business environment. These agencies use the monthly data to deflate trade statistics to produce real, as opposed to the current nominal, trade flows. These real figures help to improve the agencies' formulation and evaluation of monetary and fiscal policy and the general business environment. Failure to provide timely data would tend to extend recognition and adaptation time to economic events.

7. Quarterly published services price indexes were recently transitioned to monthly production and publication. At this time, all IPP data for products and services are collected and published on a monthly basis. This monthly collection and publication of price data enables the Department of Commerce to produce monthly merchandise trade flow figures adjusted for inflation.

So that we can meet our publication deadlines, the IPP requests that its respondents return the monthly price information forms within a week after receipt. Currently, the IPP Press Release is typically during the second week of the month following the reference period.

The International Price Program does not request duplicates of any document.

The IPP does not require respondents to retain records of any kind, for a period of any duration.

The IPP is designed to produce valid and reliable results that can be generalized to the universe of study.

The IPP indexes are based on established classification systems.

The IPP collects confidential price data. These data are for internal BLS use only, to construct price indexes.

8. The IPP survey reflects inputs that have been provided by a wide range of organizations and individuals over the years. The original recommendations for the IPP survey grew out of the 1961 report sponsored by the Joint Economic Committee of the Congress. This information has been updated and maintained via regular contact with the BLS’ Business and Labor Research Advisory Councils (See Attachments 5 and 6), Federal Statistical Users Conferences, numerous international conferences, and ongoing meetings with the various Federal Agencies which use the IPP data for analysis. Users include offices of the Departments of Labor, Commerce, Treasury, and Energy, as well as the Congressional Budget Office and the Federal Reserve Board.

Since the Program involves a continuing rotation of industries and sampling units, contacts are conducted in person with trade groups and a number of individual businessmen. The fact that the IPP survey is voluntary requires that ideas on survey design, operations and data presentation offered by these sources be studied carefully and instituted when possible.

No comments were received as a result of the Federal Register notice published in Volume 71, No. 114 on June 14, 2006.

9. The IPP does not provide any payment or gift to its reporters.

10. The Bureau of Labor Statistics Commissioner's Order 3-04 "Confidential Nature of Bureau Records," explains the Bureau's policy on confidentiality:

In conformance with the Confidential Information Protection and Statistical Efficiency Act of 2002 (Title 5 of Public Law 107-347) and other existing laws and Departmental regulations, it is the policy of the BLS that:

“Data collected or maintained by, or under auspices of, BLS under pledge of confidentiality shall be treated in a manner that will assure that individually identifiable data will be used only for statistical purposes and will be accessible only to authorized persons.”

The pledge on the collection instrument states that “The Bureau of Labor Statistics, its employees, agents, and partner statistical agencies, will use the information you provide for statistical purposes only and will hold the information in confidence to the full extent permitted by law. In accordance with the Confidential Information Protection and Statistical Efficiency Act of 2002 (Title 5 of Public Law 107-347) and other applicable Federal laws, your responses will not be disclosed in identifiable form without your informed consent.”

The Confidential Information Protection and Statistical Efficiency Act of 2002 (CIPSEA) safeguards the confidentiality of individually identifiable information acquired under a pledge of confidentiality by controlling access to, and uses made of, such information. CIPSEA includes fines and penalties for any knowing and willful disclosure of individually identifiable information by an officer, employee, or agent of the BLS.

11. Prices and the company trade data required to assign measures of size for product or service disaggregation may be regarded as sensitive by some respondents. When it is explained that the disaggregation process is intended to identify a single (or very few) specific products or services for pricing, this appears to present no serious problem, especially in view of the BLS record for confidentiality. This conclusion is borne out by experience with the International Price Program.

The question relating to affiliate relationships (asked during the initiation process and on form 3007D) may also be treated as sensitive by some reporters, but collection of this information has presented no difficulty in the past. The issue of how to handle transfer prices (the price of a good shipped between related establishments) has long been a focus of the IPP because transfer pricing represents a substantial portion of U.S. trade. Prior to 1994, the IPP only collected prices between related parties if they were considered to be arm’s length (that is, the price trended with the market). Beginning in 1994, the IPP began collecting price data for all intrafirm transactions, but excluded non-market based prices from index estimation until enough data could be collected to determine if their inclusion would introduce bias. After collecting enough data for testing it was determined that there was no significant difference in the trends for non-market based transfer prices and those at arm’s length, so the IPP began using all transfer prices in index calculation beginning with the February 1998 indexes. Their inclusion was consistent with the trade statistics that the import and export price indexes were designed to deflate.

In August 2004, the IPP began conducting a pilot field test for transfer prices. The goal of the pilot is to determine whether it is possible to improve the market reflectivity of transfer prices collected by the IPP to better match program objectives. Specifically, the pilot aims to gather information on transfer price practices and methods currently used in the international trade community, study the feasibility of using market reflective proxies as replacements for non-market reflective transfer prices, and develop collection policies and procedures to foster these goals. The program hopes that studying real world situations will provide a guide to the best policies and procedures for collection and repricing of transfer price data.

12. Average person‑hours per response is estimated separately for initiation and for repricing. For initiation, which requires

an interview with a BLS data collector, the information is entered directly into a laptop computer. (See screen shots in Attachment 7). Form 3008 is a worksheet used by BLS data collectors (See Attachment 8). The response burden estimate is based on field collection experience. Response burden varies depending on the size of the company, the number and variety of products or services produced in the establishment, and the types of records kept. Thus far in the survey, which has been carried out at small, medium, and large size establishments, the respondent burden for initiation averaged approximately one hour.

For repricing, which is an update to price data previously provided by the respondent to either a mailed shuttle form (see Attachment 9), or an online data collection application (see Attachment 10), the burden estimate is based on BLS experience in earlier samples. The burden varies from one minute for routine updates of prices for unaltered products or services, to thirty minutes for reporting changes in product or service specifications or substitution of models within a product or service line. Since these types of changes are infrequent, a generous estimate of nine minutes was selected.

Companies and establishments of all employment sizes, including those with fewer than 100 employees, are covered in the samples. This comprehensive coverage is necessary to avoid bias and to assure that the sample is representative of the universe of importers. Small companies, collectively, have substantial weight in the price‑forming universe, and the evidence suggests that the pricing behavior of small companies is different from that of large companies. Therefore, the smaller units must be directly surveyed.

The sample sizes and estimated annual respondent burden for FY 2007, FY 2008, and FY 2009 are shown on the following page.

	Number of Reporters (end of FY)1		Frequency of Response Per Year		Total Annual Responses		Estimated Avg # of Hrs Per Response		Estimated Total Hrs of Annual Burden
Fiscal Year 2007
Initiation2	2000	x	1	=	2000	x	1	=	2000
Repricing3	3700	x	6.44	=	236805	x	.65076	=	15409
Total Burden	5700				25680				174097
Fiscal Year 2008
Initiation	2000	x	1	=	2000	x	1	=	2000
Repricing	3700	x	6.4	=	23680	x	.6507	=	15409
Total Burden	5700				25680				17409
Fiscal Year 2009
Initiation	2000	x	1	=	2000	x	1	=	2000
Repricing	3700	x	6.4	=	23680	x	.6507	=	15409
Total Burden	5700				25680				17409

Respondent burden cost for monthly import data collection for the periods covered by this clearance package are as follows:

Fiscal Year	Total Hours Burden	Average Hourly Pay		Annualized Cost of Burden
2007	17,409		$30.96		$538,983
2008	17,409		$31.81		$553,780
2009	17,409		$32.68		$568,926

The figure for 2007 was estimated by multiplying the estimated total hours of annual respondent burden for import data collection (17,409 hours) by an estimated average hourly total compensation for white collar workers of $30.96. This rate is derived from the published December 2003 value for total compensation for white-collar employees in private industry, the last year this number was published. Data was accessed from the National Compensation Survey - Compensation Cost Trends website of the Bureau of Labor Statistics at http://www.bls.gov/ncs/ect/home.htm. The December 2003 value of $27.92 was updated to $30.13 for March 2006 using the 7.9 percent increase in the Employment Cost Index of the BLS (ECI). Estimates for 2007 and subsequent years are derived by calculating the percentage change in ECI for white-collar total compensation between March 2005 and March 2006 and applying that percentage change (approximately 2.75 per year) to subsequent years.

13. Reporters need no special equipment or technology for collection of this information; the company’s methods for maintaining its records are incidental to the IPP survey. Reporters’ total annual capital costs (both the total capital and start-up cost component and the total operation and maintenance and purchase of services component) relative to the IPP survey are $0.

14. For FY 2007, the collection and publication of data for the IPP Survey (both imports and exports) will cost approximately $15.795 million.

15. Total respondent burden has increased, primarily due to a shift to collect more data for imported items than for exports. This shift reflects the change in the relative value of U.S. imports compared to U.S. exports. However, the average burden per respondent has remained about the same.

16. As of May 2006, the merchandise price indexes are published using four different classification systems: The Harmonized System (HS), the BEA End Use System (End Use), the Standard

International Trade Classification, Rev.3(SITC), and the Foreign Trade North American Industry Classification System (NAICS). Due to budget cutbacks, indexes published using the SITC classification will be discontinued in July 2006, while indexes published by HS category will be dropped in 2007. Price indexes for internationally traded services are published using two other definitions, since services are not covered in the published classification systems used for merchandise trade: the Balance of Payments (BOP), which represents transactions between U.S. and foreign residents; and International services indexes, which represent transactions “inbound to” and “outbound from” the U.S.

The HS system is also used for sampling, weighting, and the collection of data. Each published product group is composed of classification groups, constructed from homogeneous or related product categories in the Harmonized TSUSA or Schedule B classifications. Index aggregation weights are now revised on a yearly basis which more accurately reflects the constantly changing patterns of international trade. For the IPP’s goods indexes, the aggregation weights at the stratum and detailed classification group levels consist of the universe trade dollar value totals that are published by the Bureau of the Census. Changes affecting the weights of products in the basket of goods bought and sold in foreign markets are now made every January, beginning in 2004, and reflect shifts in trade patterns from two years earlier. The IPP also plans on evaluating the need of reweighting Services indexes on a more frequent basis. Services indexes currently have their aggregation weights recalculated once every five years.

The IPP also publishes Locality of Origin (LOO) indexes, which are import price indexes broken out by country or region and aggregated using the NAICS. Beginning in 1991, indexes were calculated and published for Japan, Canada, the European Union, and Asian Newly Industrialized Countries (these indexes were also grouped according to status as “developed” or “developing”). The Latin American region was added in 1998, and with the January 2005 press release, the list of localities for the Import Price Indexes by Locality of Origin was expanded to include the following regions: France, Germany, the United Kingdom, Mexico, the Pacific Rim, China, the Association of Southeast Asian Nations, and Asia Near East Countries.

Repricing schedules are mailed to respondents who are using the mailout/fax-back collection two business days prior to the first day of the pricing month. Respondents providing data via the web receive a notification to reprice on the second business day of the reference month (see Attachment_11). Data collection continues for five weeks; the indexes are released approximately one week later. Schedules which are sent to the Regional Offices for initiation may remain in the Regional Offices for up to a year.

The IPP data are published in a monthly news release that includes a description of some of the highlights of import and export price movements over the past month. The format and content of the published data appear in the attached tables from the IPP Press Release. (See Attachment 12). The release also includes tables that detail aggregate price indexes for each of the published classification systems. In addition to the news release, the IPP publishes more detailed tables that contain indexes and percent changes over the past four months for each of the program’s published indexes. The IPP also offers full historical tables that show the index values for each published stratum dating back to when the series was first published. IPP outputs are available to the public by e-mail, on diskettes, by fax, or on the Internet. Detailed analyses using international prices are also published periodically in the Monthly Labor Review (See Attachments 13 and 14).

The overall data collection timetable is as follows:

Sample initiation began: 1970
Sample initiation ends : Continuing
Time series repricing began: 1970
Time series repricing ends: Continuing

The historical publication timetable is as follows:

IPP Press Release (quarterly and monthly updates of time
series):
Quarterly since 1974
Monthly since 1989
Publication MLR summary articles:
Periodically since 1983

17. The International Price Program requests an extension of authorization to not display the expiration date for OMB approval on the Product Information repricing form and instructions (BLS 3007D). OMB has granted the IPP this exemption since 1989, and the exemption has resulted in a substantial savings in printing costs and personnel time.

18. Because it is a voluntary survey and because it imposes no recordkeeping requirement for reporters, the IPP does not indicate a retention period for recordkeeping requirements.

1 These numbers are estimates subject to change due to differing relative values of U.S. imports and exports and to variations

in response rates.

2 Initiation refers to the initial collection of data to be used in repricing. Totals include the Field Economist’s visit to the

company as well as the time spent to select items for repricing using the disaggregation sheet (form 3008, attachment 8) .

3 Repricing refers to the update of price information previously provided by the respondent. The repricing form (3007D,

attachment 9) and the web application (attachment 10) are the primary means of repricing but all collection types

(telephone, e-mail, etc.) are included in these totals.

4 During initiation, the respondent determines how many months he/she will need to supply data in a given year based upon

how often the company changes its pricing information. The average import company is requested to supply information

6.4 months out of every year.

5 Total number of annual responses = 3700 (respondents) x 6.4 (months).

6 Time to reprice is based upon 9 minutes of response time per item x 4.338 items = 39.042 minutes/60 = .6507 hours.

7 Rounded to the nearest hour.

13