National Academies Press: OpenBook

Behind the Numbers: U.S. Trade in the World Economy (1992)

Chapter: Appendix B: Canvass of Data Users

« Previous: Appendix A: Sales and Purchases of Goods and Services Between Americans and Foreigners
Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×

B

Canvass of Data Users

Although many of its members are experienced users of foreign trade data, the panel considered it essential to consult with a broad spectrum of data users. It wanted to learn from them what kinds of foreign trade data they use; how they use the data; their evaluation of the quality, cost, and accessibility of the data; their unmet needs; their anticipated future needs; and their suggestions for improving foreign trade statistics. Thus, the panel undertook a canvass of the users of foreign trade statistics.

CANVASS METHOD

The best way to find out how well a program is serving its clients —if the client population is well defined, accessible, and large—is to conduct a formal survey of clients, using probability sampling. There have been numerous attempts, some more successful than others, to survey users of the statistics produced by federal agencies (see, for example, a recent Bureau of Labor Statistics survey of users of Consumer Price Index data [Kamalich and Kwiecinski, 1989]).

The panel examined carefully and then rejected the idea of conducting a formal survey of users of foreign trade data for three main reasons. First, there are so many different sources of foreign trade data that it would be virtually impossible to develop a comprehensive, unduplicated list of users. Second, the development

Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×

of a structured questionnaire that would adequately cover all kinds of users and uses of foreign trade data proved to be extremely difficult. Panel members concluded that a highly structured questionnaire with predetermined response categories would not elicit the detailed and explicit information about user views needed. Third, the panel did not have the resources to conduct a formal user survey whose design and execution would meet generally accepted standards for high-quality survey research.

Fortunately, there were other options available. The panel used three methods of collecting information from users of foreign trade statistics: presentations by users at its meetings, on-site interviews of users by panel members, and solicitation of written statements. (See the invitation letter and the protocol for the interviews at the end of the appendix.)

For descriptive purposes, the users and types of responses have been divided into four groups. Group 1 is recognized key users of foreign trade data, most of them from U.S. federal agencies and international organizations; they were invited to make oral presentations at panel meetings or to submit written statements. Many of the presenters also submitted written statements; for those who did not, the minutes of the meetings served as documentation of their views. (This group does not include any of the agency representatives who made presentations solely as primary producers of foreign trade statistics, who were not considered to be part of the user population.)

Group 2 is data users from several types of organizations who were interviewed on-site, singly or in groups, by members on the panel. Some of the interviews followed a protocol designed to ensure coverage of all relevant aspects of the subject (see end of the appendix). The documentation for these interviews consisted of notes prepared by the interviewers and, in a few instances, written postinterview statements submitted by the person interviewed.

Group 3 is subscribers to the Census Bureau's foreign trade data products. The invitation for written comments was mailed to 589 organizations and individuals on subscription lists provided by the Census Bureau, with a follow-up mailing two months later. The response rate was slightly under 10 percent: after combining multiple responses from the same organization, there were usable responses from 48 organizations.

Group 4 is people who had participated in a November 1989 Conference, “International Financial Transactions: Issues in Measurement and Empirical Research,” sponsored by the National Bureau

Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×

of Economic Research (NBER). The invitation for written comments was mailed to 186 conference participants. A total of 17 responses were received, again slightly less than 10 percent of those invited to comment.

In sum, we received 111 responses. To provide an overall view of the kinds of users whose views we obtained, Table B-1 shows the distribution of the four groups by type of organization, using a slightly modified version of the U.S. Standard Industrial Classification.

About two-thirds of the responses came from the private sector and one-third from users at various government levels. The invited presentations and the responses from Census Bureau subscribers covered a broad spectrum of users from both sectors. We do not know for certain why manufacturers of chemical products were so heavily represented in the latter group; possibly a particularly interested and active industry association encouraged responses by its members. On-site interviews were limited to private-sector users because at that stage of the study the panel had already

TABLE B-1 User Responses By Type of Group

 

Groupa

Type of organization

1

2

3

4

Total

Chemical manufacturers

-

-

16

-

16

Other manufacturers

1

4

11

-

16

Universities

2

1

-

7

10

Membership organizations

1

2

4

-

7

Consulting and research firms

2

3

6

1

12

Other private-sector organizations

3

-

4

-

7

U.S. federal agencies

10

-

-

8

18

Congress, state and local government

12

-

1

-

13

Other countries

1

-

6

-

7

International organizations

4

-

-

1

5

Total

36

10

48

17

111

aGroups: 1, invited presentations; 2, on-site interviews; 3, Census Bureau data subscribers; 4,participants, conference of the National Bureau of Economic Research. See text for more detailed description.

Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×

received presentations from all of the key government organizations. The participants in the NBER conference were mostly university-based researchers who use foreign trade data and representatives of U.S. executive branch agencies.

The user database contains a large number of responses, covering all of the major categories of users of foreign trade data. Many of the responses were detailed, thoughtful, and informative. Virtually all of those who were invited to make presentations to the panel or to participate in on-site interviews agreed to do so, although a small proportion of the Census Bureau subscribers and NBER conference participants responded to our invitation for written comments. Like the subgroups of the general population who write letters to newspapers or to their representatives in Congress, foreign trade data users who are articulate and have strong views or feelings undoubtedly predominated among the recipients of our written invitations for comments who responded.

The panel carried out a content analysis of the user responses and a few simple tabulations based on the responses of Census Bureau subscribers, covering five general topics: acquisition, processing, and uses of foreign trade data; the specific kinds of data used; evaluation of costs and convenience of data access; evaluation of the quality of the data; and current and anticipated future needs for additional data. Few of the users' responses covered every one of these topics. Each of the responses was reviewed and its contents, if any, relative to each of the above subject categories briefly summarized. Then the summaries of individual user comments for each topic were reviewed and synthesized as part of the findings presented in the body of the report. For topics commented on by many of the Census Bureau subscribers, a few simple tallies were prepared. The rest of this appendix presents some of the individual responses.

MERCHANDISE TRADE DATA
SOURCES AND KINDS OF DATA

Among the organizations that responded to the survey, the Census Bureau was the most frequently cited source of foreign trade data. However, this could be misleading, since nearly one-half of the responses came from organizations on the list of subscribers to Census Bureau data products. With a few exceptions, commercial firms that rely entirely on secondary sources for their foreign trade data were not covered directly in our canvass of users. We know

Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×

from our contacts with some of the data retailers that there are a large number of such data users in the private sector.

There is considerable variation in the number of different sources used. Large companies, companies that are primarily in the information business, and government agencies are more likely to use multiple sources. Of 27 Census Bureau subscribers in the manufacturing sector who responded, 10 relied solely on those foreign trade data. Data from the Piers Import/Export Reporting Service (PIERS) were used by 12 of the remaining 17, and for several of these it was the only source used in addition to Census Bureau data.

A few companies, especially those with manufacturing facilities in other countries, are interested in the total international trade picture for specified commodities, not just bilateral trade between the United States and other countries. To obtain information on non-U.S. trade, they must rely on data produced by other countries or by data retailers in the public or private sector.

The kinds of data used are even more variable than their sources, so it is hard to generalize; nevertheless, some patterns do emerge from our user statements.

Except for academic research users, most private sector users of foreign trade data appear to want monthly data. A distinction is drawn between market analysis, which requires monthly data, and strategic planning, which can usually be based on annual data. Some market analysts are aware of the volatility of monthly trade data for specific commodities. Two responses from manufacturers noted that they use rolling 12-month totals for their analyses.

In the public sector, frequency requirements are more varied. Users of monthly data include policy analysts, such as those at the Federal Reserve Board and units of the International Trade Administration and Agricultural Marketing Service of the Department of Agriculture that provide current market analysis for their constituencies. Other researchers in the Federal Reserve System or the Department of Labor 's Office of International Labor Affairs rely mostly on annual or quarterly data. State agencies that promote production for export rely on annual data, since no monthly or quarterly data on exports by state of origin are available.

Most users, other than those interested only in broad economic indicators, need data for selected commodities or commodity groups. Users looking only at international trade are generally content with commodities as classified under the Harmonized System (HS), but those who want to look at both domestic and foreign compo-

Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×

nents of production and consumption usually find it necessary to convert the HS data to a system compatible with the U.S. Standard Industrial Classification (SIC), or vice versa.

Finally, there are users in the transportation industry with entirely different needs. Mode of shipment—vessel, air, or other—is their primary interest, along with data on port of entry or exit, point of origin within the exporting country, and final destination in the importing country.

USES

In asking foreign trade data users to tell us how they use the data, we did not try to constrain their uses to any specified set of categories. However, earlier attempts at classification were of some value in organizing the information they gave us (see Committee on National Statistics, 1976). We discuss uses in three broad categories: private sector, excluding research; public sector, excluding research; and research in both the private and public sectors.

In the private sector, nonresearch uses of merchandise trade data that were reported fell into three broad categories: market analysis, strategic planning, and activities intended to influence government policy and administrative decisions. Of these three, market analysis, often called market-share analysis, was by far the most frequently mentioned. Such analysis focuses on past, present, and future (forecasted) supply and demand balances for specific products of interest to the user. Producers want to track their own market shares and those of their competitors, both at the country and individual company level. A business researcher for a chemical manufacturer said that PIERS was a powerful tool and that, in combination with Census Bureau data, it provided a detailed picture of product movement. With these sources and a knowledge of the industry, a user can usually determine “how much of a particular chemical is imported or exported and by which company.”

Companies usually do market analysis only for their own use, but research and consulting firms and in some instances even manufacturers share information with their customers or clients. Thus, for example, an aircraft manufacturer reported that it provides information to airlines on international commodity movements by air to help the airlines in planning their freight operations.

Strategic planning is a longer range activity and does not necessarily require up-to-the minute monthly data. It does require the

Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×

analysis of trend data for specific commodities and country combinations. Such analyses are used to support decisions on investments in new facilities and their location and on the development and marketing of new products.

Companies in the private sector try to influence government decisions in a variety of ways and often use statistics on various aspects of foreign trade to support their cases. Manufacturers who believe that foreign competitors are using unfair trade practices, such as dumping, will marshal the evidence and present it to Congress and the appropriate executive branch agencies. One large corporation stated that the foreign trade data it acquires and processes are useful to company officials who serve on various government committees concerned with trade policies.

Another interesting example of this type of use was described by the manufacturer of a specialized product that requires a particular type of wood that is not widely available. Some countries placed import restrictions on this product in the belief that it could be just as well manufactured domestically, using any available type of wood. The company provided customers in these countries with U.S. export data showing that it was selling its product to countries all over the world, allowing some of the customers to appeal successfully to their governments to remove the import restrictions.

We also heard from some law firms that assist clients in their dealings with government agencies on trade regulatory matters. Understandably, these users did not provide specific information on their clients or the issues they dealt with, but they had some interesting comments on other aspects of the data (covered below).

In the public sector, we identified three main categories for nonresearch uses of foreign trade data: policy guidance, administrative decisions, and trade promotion. Aggregate trends in foreign trade, as presented in the Census Bureau's monthly trade balance series and the Bureau of Economic Analysis's quarterly balance-of-payments trade balance series, are important elements of the economic data that guide broad economic policies at the national and international levels. Detailed information on trade for specific countries and commodities is used to evaluate U.S. international trade policy and to predict the likely effects of new legislation and regulations. Primary concerns are the international competitiveness of U.S. exports and the effects of imports on domestic employment and earnings. Agencies such as the International Trade Administration, the International Trade Commission, and the Labor Department's Office of International La-

Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×

bor Affairs are frequently asked to prepare analyses of these issues for policy makers in the executive branch and Congress.

Administrative decisions that depend on foreign trade data include decisions on which foreign countries are eligible for various trade preference programs and which U.S. companies are eligible for assistance to counteract the effects of import competition. At the federal level, the International Trade Commission is the primary user of data for these purposes. Some of the program eligibility requirements have precise cutoffs for eligibility, with the result that the accuracy of the data can become critical in borderline cases. The ability to develop comparable data for imports and domestic production is also an important consideration.

To promote exports, and in some instances the substitution of domestic products for imports, federal and state agencies monitor foreign trade and conduct market analyses for the benefit of current and potential exporters. At the federal level, the Commerce Department 's International Trade Administration and the Agriculture Department 's Agricultural Marketing Service are important users of foreign trade data for these purposes. At the state level, departments of commerce or industry and other state agencies assist exporters by providing technical assistance, market information, export financing, and state representatives to foreign countries. To support and monitor the effectiveness of these activities, states have strongly urged the federal government to provide detailed state-by-state data on exports and imports.

Quantitative research on international trade was reported to us by foreign trade data users in executive branch agencies, the Federal Reserve System, international agencies, universities (some supported by federal funding) and commercial and nonprofit research organizations. Most of the research is policy oriented; much is aimed at better understanding the effects of foreign trade on the U.S. economy. Some examples of research topics and studies are the development of models to predict the outcomes of changes in foreign trade policies; analysis of the response of specific industries to global exchange rate disturbances; identification of the determinants of trade patterns, including cost, production, and demand factors, for specific commodities; analysis of the determinants of the commodity composition of trade for various foreign countries; cross-sectional and longitudinal analysis of the economic determinants of variation in the volume of aggregated and disaggregated bilateral trade flows; and the testing of hypotheses that trade enhances cooperation and deters conflict between countries.

Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×

The majority of the researchers who work on these topics combine data on foreign trade with data on domestic production and consumption; they therefore tend to use SIC-based classifications for foreign trade data by commodity.

DATA ACQUISITION AND PROCESSING

Most of the Census Bureau subscribers who commented on costs were satisfied with the fees charged for trade data, and some thought they were quite low. Some subscribers who used data from both the Census Bureau and other sources noted that data from private sources (such as the Journal of Commerce, PIERS, or Tradstat) or other countries were more expensive. One respondent observed that the charges for Japanese data on imports and exports were about 20 times the charges for comparable U.S. data. Another, who uses data from the Census Bureau, PIERS, and Tradstat, noted that PIERS data were more expensive than Census data but believed that PIERS was a cost-effective service. The same user noted that Tradstat was twice as expensive as PIERS but still regarded Tradstat as quite affordable. Two Census Bureau subscribers said they would be willing to pay more for data of higher quality.

Most of the comments on costs of acquisition obtained from sources other than the Census Bureau's subscriber list (groups 1, 2, and 4; see Table B-1, above), also indicated satisfaction with Census Bureau charges for foreign trade data. One group of academic users took a strongly contrary view, however, pointing out that the annual costs of obtaining the Census Bureau's import and export tapes, along with the concordance tapes needed to convert the commodity data to other classifications, would total $1,200. Their statements also commented on the high costs of purchasing international bilateral trade data from the United Nations or the Organization for Economic Cooperation and Development (OECD) and expressed their unhappiness that purchasers are not permitted to share these data with other institutions, so that acquisition through data retailers is precluded.

Users had numerous suggestions for making the data more convenient to work with, which in most cases was equivalent to saying that they would like data suppliers to add value to their product in ways that would reduce users' processing costs. These suggestions covered four topics: mode of dissemination, time-series data, seasonal adjustment, and classification.

Most users were aware that the Census Bureau is phasing out

Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×

its microfiche releases and shifting to the use of CD-ROM, so comments on the difficulty of working with microfiche are now of limited relevance. For the record, however, an academic respondent said that “microfiche is probably the most ‘user-hostile' manner in which this information could be stored,” citing the fact its resolution was not sufficient to use it for optical scanning or even to make readable hard-copy prints.

A majority of the users who commented on the Census Bureau's switch to CD-ROM appeared to be looking forward to it, even though some recognized that the costs of acquisition and processing would be higher. However, these views were by no means unanimous. Some users objected to what they anticipated would be substantial increases in costs to acquire and process data in CD-ROM format that were previously available on microfiche. Some mode-related comments came from Census Bureau subscribers who use the microfiche releases who may not have been aware of the impending switch to CD-ROM. Some had been using spreadsheet utilities for their analyses and would have liked to receive the data in a form more directly compatible with that approach.

The development of usable time-series data is complicated by two factors: revisions made to correct reporting and processing errors and changes in commodity classifications and definitions used for key variables. The shift in January of 1989 from the classifications of the Tariff Schedule of the United States Annotated (TSUSA) for imports and the Statistical Classification of Domestic and Foreign Commodities Exported from the United States (Schedule B) for exports to the Harmonized System brought significant advantages in terms of increased international comparability of foreign trade data, but it created difficulties for users attempting to maintain time series for specific commodities.

Some users' comments on revisions to correct errors came from Census Bureau subscribers. Four of them want the Census Bureau to provide more complete information about revisions or expressed unhappiness about the lack of such information. A U.S. agency that uses Census Bureau foreign trade data and also compiles trade data from its own sources compares these data and submits information about suspected errors to the Census Bureau. That statement said that revisions based on their feedback were being made at a slower pace than in the past. The respondent attributed this change to shortage of staff in the Census Bureau's Foreign Trade Division and problems associated with the changeover to the Harmonized System. An international organization expressed concern about the size of revisions in Census Bureau data

Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×

and questioned whether the same revisions were being applied to all data series.

We received several comments on the shift to the Harmonized System. Most of those who commented approved of the change or at least recognized the need for it, but some raised questions about how best to bridge the break in commodity time-series data. A representative of an international organization, recognizing the cost of providing a bridge between the old and new systems, nevertheless urged that Census Bureau recompile prior years' data in the HS commodity classifications. Users also mentioned other factors that have complicated the development and use of time-series data, such as the substitution of Canadian import data for U.S. export data that took effect at the start of 1990 and the frequent changes in commodity definitions within existing classification systems.

Comments on the seasonal adjustment of foreign trade data came mostly from trade policy and research-oriented users in federal agencies and universities. One agency user believed that the seasonal adiustment factors were not being revised often enough and complained that the older seasonally adjusted data were not being revised to reflect recalculated factors. Other respondents, including a data retailer and a federal agency, wanted seasonally adjusted data at a finer level of detail.

Problems associated with the use of foreign trade and domestic production or employment data in the same analyses were mentioned by several users, including both researchers and persons doing market share analyses for a company, commodity, or industry. A federal agency user called for an improved and automated, current and consistent master concordance that could be used to link trade classifications to domestic output and employment classifications.

It is difficult to identify any consensus from the users' comments on acquisition and processing of data. Federal agency users are in a privileged position and tend to be fairly well satisfied. One of them said: “The availability of the COMPRO data base to federal users greatly facilitates access to trade information; hopefully, the new National Trade Data Bank will perform the same service to the general public.” Researchers in universities had varying opinions. For those who are working under government contracts or grants, the acquisition of data is often facilitated by access to COMPRO. One individual researcher who did not use this source said that the U.S. foreign trade data he needed for his project “was easy to find and use.” At the other end of the spec-

Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×

trum, a university research group that creates and maintains extensive economic models said that most of the other economic data sets they were using were much less expensive and easier to process than the U.S. merchandise trade data.

It is even more difficult to summarize the experiences and comments of commercial users on data acquisition and processing, and we are limited by the fact that most of our responses in this group came from subscribers to Census Bureau data, rather than those who rely entirely on data retailers or other sources. One thing that did emerge clearly in the comments from Census Bureau subscribers was their concern about the timeliness of the data (see next section).

TIMELINESS

Broadly speaking, timeliness is a component of data quality. Timeliness is measured by the length of time that elapses between the end of the reference period for a particular data set and the time when the data are available to users. Data are considered to be more or less timely depending on whether this elapsed time is short or long. For many kinds of uses, the utility of data for a particular time period declines as time passes.

We asked users of foreign trade data to comment on timeliness, and nearly half of the respondents did say something about it. It seems reasonable to assume that most of those who did not comment are not greatly concerned with the timeliness. However, of those who did comment, few were fully satisfied. Thus, timeliness is clearly a matter of concern for many users.

Concerns about timeliness ranged from mild to severe. It was not always possible to relate the users' comments on timeliness to specific data sources and publications or other data products. However, a substantial number referred specifically to the microfiche versions of the Census Bureau's IM145, IM146, and EM545 monthly reports. Most of these users were tracking specific commodities, and some had arranged to receive data on selected commodities from computer printouts, which are available earlier than the microfiche reports. Some users of the Census data commented that delays in availability of the IM145 and EM545 microfiche reports had been increasing. One user who had tried to circumvent these delays by telephoning the Commerce Department for specific information was finding this increasingly difficult to do. Clearly, some users were more aware than others of the variety of means of access to the Census Bureau data. (As noted, the panel

Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×

obtained users' comments prior to the change from microfiche to CD-ROM; user comments on this change are discussed above.)

Only a few of the users who complained about lack of timeliness explained, even in general terms, why they thought they needed the data sooner. Those who did mentioned factors such as the seasonality of their business, the need to follow the dynamics of the market closely, and the need for timely data to prevent unfair trade practices such as dumping.

Several users compared the timeliness of the Census Bureau foreign trade data with data from other sources or commented on changes in the timeliness of Census Bureau products. With respect to other countries, their views were somewhat mixed. One user singled out Japan as having a better record, and another said that most developed countries did better than the United States. However, another said, that delays in availability of Canadian data were equal to those experienced for U.S. data. Some users pointed out that data from private sources, such as those from the PIERS system (which, however, covers only vessel trade at selected major ports), were available sooner than the Census Bureau data.

Two consulting and research organizations commented on trade-offs between timeliness and accuracy. The result was a standoff. One was willing to wait a little longer for more accurate data, but the other was willing to sacrifice some accuracy and detail for more timely data. Another user said: “We would pay some premium for prompt information in the areas of our greatest interest.”

A review of the comments on timeliness suggests that there are certain categories of users more likely than others to demand quick access to data. Most “impatient” users of foreign trade data are in the private sector, and the majority appear to be doing market analysis, including monitoring of competitors, for themselves, clients, or members of industry associations. Users making short-term decisions involving international financial markets may not have an absolute time requirement, but according to one data retailer, his clients in this category want the data “5 minutes” before anyone else.

Government and university researchers who use foreign trade data for basic economic research tend to be much more patient. Most work with annual or less frequent data and are much less concerned with timeliness than with the availability of extended time-series data and issues of comparability between foreign trade and other economic data series.

Government users who supply inputs to the country's short-

Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×

term trade, monetary, and fiscal policies belong in the impatient group. A representative of the Federal Reserve System observed: “The issue of timeliness is nearly as important, from the Fed's perspective, as that of accuracy . . . efforts to enhance the timely reporting of monthly trade data should also be given high priority. ”

A statement prepared in 1986 by the U.S. International Trade Commission, which furnishes Congress and the executive branch information and analysis based on federal economic statistics for use in making trade policy decisions, emphasized problems created by “untimely reporting, ” including the carryover of data from one reporting period to the next. The scope of the carryover problem was subsequently reduced by shifting the initial release date from 30 to 45 days after the end of each reporting period. The same statement was critical of delays in the availability of reports on domestic output in manufacturing by the Standard Industrial Classification.

Users of foreign trade data at the state level face a somewhat different situation. Most data of interest to them—for example, exports by state of origin—are only published on an annual basis. At present, these users appear more concerned about provision of additional commodity detail and improvements in accuracy than about timeliness. Some commented on the general lack of timeliness of publication of results from the Annual Survey of Manufactures, which are issued with a lag of 2-3 years.

In summary, a substantial minority of users who responded to our invitations for comments, mostly those using monthly data, wanted more timely foreign trade data. Prominent among this group were users who represented manufacturers or economic consulting and research firms and were using the data for market analysis. Only a few said why they wanted more timely data, and the consequences of delayed availability were only hinted at in one or two instances. Government users who analyze current economic data to aid decisions on short-term international and domestic economic policies also wanted more timely data. Researchers, on the other hand, whether in government or in universities, seemed to be satisfied with this aspect of the foreign trade data.

It will be of considerable interest to see whether two recent changes in the procedures for dissemination of foreign trade data—the Census Bureau's phase-out of microfiche and switch to CD-ROM for many of its reports and the development of the National Trade Data Bank by the Commerce Department's Office of Business Analysis—make it possible for private-sector users to get the data they want in a more timely manner.

Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×
ACCURACY

It is more difficult for users to evaluate the accuracy of foreign trade data than it is for them to evaluate timeliness. Nevertheless, nearly half of the users who made presentations, wrote, or were interviewed by panel members had something to say about accuracy. Only a few of those were fully satisfied. Some of the concerns about accuracy were expressed in moderate terms, but other users were more critical. Some users said they had no basis for judging the accuracy of foreign trade data, but others mentioned various ways in which they attempt to evaluate the data. Some, especially those who have few competitors for particular products, compare their own data on export shipments with those of the Census Bureau. A manufacturer of a specialized wood product had compared the EM-545 data with his own records of shipments for all of 1989 and submitted a lengthy listing of the errors found. He felt certain that if he had been able to detect so many errors in his firm's shipments, there must have been enough additional errors to make the statistics virtually useless. The nation 's other manufacturer of this product had similar comments.

Another method of checking is to calculate unit values and review them for reasonableness. A chemical manufacturer who routinely does this found that exports of a specific commodity were being reported in pounds instead of kilograms. Another chemical manufacturer found the Census Bureau's data on shipments of materials under a particular tariff code showing extraordinarily high values per kilogram for several months. A commodity specialist at the Census Bureau was able to check the records and discovered that the commodity had been misclassified.

Some users have compared Census Bureau data on vessel imports and exports with data from the PIERS system and identified what they believe to be errors in one or the other system. A unit of the U.S. Department of Agriculture (USDA) regularly monitors Census data for specific commodities, countries, ports, and customs districts, using data from other USDA units on grain inspections and export sales of agricultural commodities. Its feedback to the Census Bureau frequently leads to revisions of the merchandise trade data. The Energy Information Administration, which regularly includes Census data on coal imports and exports in its publications, has found it necessary to submit the data received from the Census Bureau to an edit procedure in order to eliminate what it considers to be incorrect or misleading information (Energy Information Administration, 1989).

Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×

Comparisons of imports or exports by country with aggregate figures often reveal discrepancies. A representative of the Bureau of Statistics for the International Monetary Fund wrote that his staff had been unable to account for large discrepancies between aggregate trade data and the total of exports to and imports from partner countries that were observed for some years.

In addition, a Minnesota official compared the agricultural exports for the state, as reported in the Census Bureau's country-of-destination series, with estimates published by the Department of Agriculture. The official found that for 1988 the agricultural exports reported for Minnesota in the country-of-destination series were $259 million, compared with $1.8 billion estimated by the Department of Agriculture. For the United States as a whole, the relative difference in agricultural export statistics was smaller but still substantial, with the Census Bureau reporting U.S. agricultural exports of $23.4 billion in 1988, compared with Agriculture's $35.3 billion. Clearly, there was a problem with the allocation of exports to states.

Some users gave their views on the reasons for specific errors or, more broadly, on what they believed to be the main sources of error. We consider first the errors that exist in the official trade documents (or computer records) when they are received by the Census Bureau. The importance of reporting errors, in contrast to processing ones, was highlighted by a chemical manufacturer. He conjectured that most accuracy problems of U.S. trade data occur in data collection rather than in analysis or manipulation of the data by government analysts.

The problem of reporting amounts in pounds instead of kilograms has already been mentioned. Misclassification of commodities is clearly a problem, especially at the more detailed levels of classification. A chemical manufacturer said: “In general I have found the reports to be very helpful for the analysis of major commodity chemicals . . . As the product becomes more specialized . . . the numbers begin to become more and more inaccurate.” A statement from the U.S. International Trade Commission (ITC) (written before the changeover to the Harmonized System) pointed out that the Customs Service has little incentive to look beyond the 5-digit level of the TSUSA, the level at which duties are assigned. Consequently, ITC believed that the 7-digit statistical categories were not receiving the attention needed to ensure accurate reporting.

Some users mentioned problems with reports of value. In some instances, it was hard to separate conceptual and operational re-

Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×

porting problems. A university economist, for example, believed that the rising share of intrafirm trade was making present methods for calculating trade values and prices less and less reliable because there is a large arbitrary element in the prices and values reported for such transactions when multinational firms carry out transfer pricing practices.

A more clear-cut reporting problem, undervaluation of trade with Canada due to missing inland freight, was described in a presentation by a representative of the Bureau of Economic Analysis. He explained that the value of U.S. exports is supposed to include all inland transportation costs. Unfortunately, this is not the case, and for some time BEA has included an estimate of this undervaluation in its balance-of-payments series. BEA's estimate is based on a sample survey conducted by the Census Bureau in 1969 for documents filed in July 1968, which remains the only possible basis for the estimate. In 1988 BEA added $1.8 billion to exports to Canada. A statement by an official of the Commerce Department's International Trade Administration expressed doubt as to whether the U.S.-Canada data exchange, which began at the start of 1990, would fully resolve the problem of undocumented exports. He expected that Canadian customs would be likely to take less interest in duty-free imports from the United States.

Information on state of origin for exports is of special interest to state and local officials interested in the promotion of exports and the use of port facilities. A state official reported that state-of-origin information is missing from approximately 25 percent of all shippers ' export declarations (SEDs). When the information is reported, it may not be correct. The same official asserted that the Census Bureau 's Origin of Movement series is of limited use because the series inaccurately assigns exports to a state or region that may not have produced the exported item. The series showed that Louisiana, for example, exported 44 percent of total U.S. crop exports, although it produced only 1.6 percent of the total U.S. crop output. Some users believed that significant numbers of shipments, especially exports, were being missed altogether. A former Commerce Department official noted that Census' recent port audits had found underreporting of air freight to be a significant problem. He believed that seaborne exports, especially those moving out in containers, might also be subject to substantial underreporting. Another user believed that information about small items combined with other items in a single container could easily be lost.

Shifting to the question of how the accuracy of foreign trade

Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×

data is affected by the Census Bureau's processing operations, we return to the case of the two specialized wood product manufacturers, mentioned above. One of them had sent us a listing of suspected errors in Census's 1989 data for exports of their product, with a copy to Census. Their earlier attempts to notify Census of these problems had apparently gone astray, but this listing reached the appropriate commodity specialist in the Foreign Trade Division, who determined that the main cause of the errors was incorrect or outdated values of the parameters used in the computer edit checks of value-to-quantity ratios. These parameter values can be changed monthly, as needed, but, to a large extent, feedback from users, such as occurred in this instance, seems to be the primary source of information used to keep them current.

The deliberate exclusion of low-value import and export transactions from processing by the Census Bureau is not, in itself, a source of error. However, Census publishes monthly estimates of the total value of these excluded shipments by country, and some users questioned the accuracy of these estimates. A specialist in market development for an aircraft manufacturer felt that the Census Bureau's estimates of low-value shipments were very unrealistic. A member of a group of foreign trade data users commented that the frequent increases in the cutoffs for low-value shipments were making it extremely difficult to monitor changes in the rapidly growing international air freight industry.

Data on exports by state of origin are produced by the Census Bureau in its Origin of Movement series, based directly on the SEDs, and its Exports from Manufacturing Establishments, based primarily on the Annual Survey of Manufactures (ASM). State users were critical of the quality of data in both series. For the first of these two series, the Massachusetts Institute for Social and Economic Research (MISER), in cooperation with Census, has developed procedures to allocate the state-of-origin information missing from about 25 percent of all SEDs. A representative of the Michigan Department of Commerce criticized the ASM-based series as too dependent on untested assumptions and adjustments.

To some extent, the problems with these series do not arise from reporting or processing errors; they occur simply because the exporter is not asked to report (and would probably not be able to report accurately in many instances) the state where the exported commodity was produced or grown, rather the state from which it started its export journey. In one of our on-site interviews, a representative of a consulting and research firm said he did not think that the state data for imports and exports would be very

Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×

useful because of such definitional problems. Only a few users addressed the question of how their uses were affected by errors in foreign trade data, and some of those who did merely indicated that the data had to be used with caution. However, more serious consequences were reported by a few.

In the private sector, a chemical manufacturer said that accurate and timely import statistics were necessary to prevent dumping by foreign firms. A law firm concerned with trade regulatory matters provided a very specific example in which a questionable data adjustment by a Census Bureau employee caused one of its clients to lose continuing eligibility for duty-free treatment under the Generalized System of Preferences, which depended on the country's share of total imports in a specified category remaining below 50 percent.

In the public sector, the 1986 statement by a representative of the U.S. International Trade Commission was quite explicit about the effects of errors in foreign trade data. He explained that in the agency's import-injury investigations, inaccuracies in the statistics could mean the difference between an affirmative and a negative determination, thereby affecting the granting or withholding of import relief. In addition, errors in reporting the country of origin of imports have erroneously caused products of some countries to become ineligible for preferential trade programs and, in other instances, have resulted in the misapplication of quota agreements.

As indicated above, some user statements addressed tradeoffs among timeliness, accuracy, and cost. Three statements from consulting and research firms identified quality as their primary concern. One was willing to surrender detail for accuracy and would even be willing to work with quarterly data if they were more accurate. A representative of the International Trade Administration expressed concern that recent efforts to facilitate movement of shipments through ports despite staffing cuts would lead to more errors in commodity classification: “It is clear that Customs' automated programs are designed to promote processing but will do little to overcome problems of misreporting.”

Most of the users' recommendations for improving the accuracy of foreign trade data were fairly general. Some chemical manufacturers who had expressed concerns about misclassification of commodities called for more checking of information supplied by exporters, active Census Bureau efforts to solicit users' reactions about the quality of the data, and, in one instance, increasing and enforcing penalties for misclassification or nonreporting.

Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×

A user in the Department of Agriculture recommended improved communication “among the various government agencies using and/or contributing to each other's data.”

Other users recommended that the Census Bureau make increased efforts to evaluate its merchandise trade data and provide users with more information about its quality. One of them recommended expansion of the recently initiated port audit program. Another called for more information “about how the data are gathered and how the numbers are affected by reporting problems and bottlenecks in the system. ” One respondent interviewed by panel members believes that current impressions about the quality of the data come primarily from “disaster stories” about substantial anomalies uncovered by users and that lack of systematic information about quality “contributes unnecessarily to the impression that all the data have very substantial errors embedded in them.”

The Census Bureau, which tries to detect and correct at least some errors by the application of a series of edit checks, also comes in for its share of criticism. Some users who follow the data for specific commodities told us that they had notified the Census Bureau of apparent errors that they had identified. In some instances the problems had been dealt with adequately; but in others, they believed that Census had been unresponsive or had taken too long to eliminate the problem. A specific problem identified was the lack of any mechanism to systematically update values of parameters, such as acceptable ranges for unit values, for the thousands of commodities that the Census Bureau deals with. There were a few mentions of “disasters, ” that is, large errors that Census failed to detect or correct prior to publication. What is not clear was what happens when these disasters come to the Census Bureau's.attention: To what extent are users notified of the revisions made? Is information about such errors used to develop improvements in the processing system, rather than just to make corrections?

Concerns about the accuracy of the data appear to be strongest in two quarters: agencies such as the U.S. International Trade Commission, which make administrative decisions based on the data, and companies that use the data for up-to-the-minute market share analysis. Both groups work with data for detailed commodity categories and often with monthly data. At this level of detail, well over half of the nonempty data cells are based on only one or two shipments, so that even a single error can have a large impact. Given the Census Bureau 's relative lack of control over the quality of the incoming data and its limited resources for

Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×

detecting and correcting errors, one has to ask whether it is reasonable to expect Census to meet demands for merchandise trade information at the current level of detail with data of acceptable quality.

UNMET USERS' NEEDS

Many users said they would like to have additional foreign trade data, and the kinds of data they would like varied widely. We have not tried to estimate the cost of providing all of the new data that users would like to have, but it would clearly be very high. The more costly data needs identified included: expanded coverage of low-value transactions, especially those shipped by air; substantial expansion of data on state of origin for exports and final state destination for imports; expansion of the samples used in the Bureau of Labor Statistics foreign-trade-price-index program to support more detail by commodity group and trading partner; and a survey to compile information on export orders.

Several users in chemical and other manufacturing industries called for finer commodity detail that would make it possible for them to monitor trade in the specific commodities that they produce and export. For example, one such request was for reporting of horseshoe nails separately from other categories of nails. Many of these users also called for elimination or reduction of the use of “basket categories ” that group miscellaneous commodities or transactions that cannot be coded precisely within a broad group. As might be expected, similar comments were received from membership organizations and from consulting and research firms that serve manufacturing industries.

Leaving aside major changes, such as the recent shift to the Harmonized System, commodity classifications and definitions are frequently changed in order to adapt to the development of new products and to meet special needs expressed by users. Some users would like additional assistance in keeping up with these changes and making the necessary adjustments to the time series that they are maintaining. A consulting and research firm asked that the Census Bureau provide not only a list of such changes, but “some explanation of why the category changed and specifically what products are affected.”

Two users asked that commodities be classified by the level of technology required to produce them. This might be interpreted as a request for the producer to add value to the basic data, since one way to satisfy this need would be for users to place detailed

Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×

commodities into “level of technology” categories according to their own classification schemes. In addition, there were several comments about the need to provide better compatibility between merchandise trade and domestic production commodity data. This area is another in which users' needs might be met either by the Census Bureau's producing new data or by users' adding value to existing data. A membership organization suggested that the United States give serious consideration to making the commodity codes used for merchandise trade more directly compatible with those used for domestic production.

Representatives of air carriers and aircraft manufacturers expressed considerable concern about the exclusion of low-value shipments from the regular monthly processing of official trade documents and about the lack of detail provided and questionable accuracy of the estimates provided for this category of shipment. They pointed out that the excluded shipments encompass the whole air express and most of the air cargo industry. They clearly would like to see the Census Bureau resume processing all or a sample of the low-value shipments.

Users also mentioned several other types of data needs that could be met by adding items to the SEDs and import entry summaries or by processing and disseminating existing information on these forms in new ways. Some of these needs might be met by the users themselves through manipulation of data already available from Census, but they are noted here for the sake of completeness. Some chemical manufacturers expressed a desire for additional data on weights or volumes or the use of different units. A user from a company associated with container shipping asked for data that would require the addition of several new items, such as container size and mode of domestic transit, to the import and export documents. A university researcher proposed the use of a new commodity classifier based on the price of individual items. A representative of BEA proposed the addition of an item to the Customs Bureau import entry summary form to indicate whether or not the additional transaction was billed in a foreign currency. Two users called for additional improved data on affiliated trade, that is, intrafirm trade within multinational enterprises. At present, Census does not provide any data on affiliated trade: BEA derives a limited amount of this information from its enterprise-based surveys of U.S. direct investment abroad and foreign direct investment in the United States. The recent inclusion of the exporter's employer identification number on the SED opens up the possibility of distinguishing affiliated and nonaffiliated exports

Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×

through a linkage process involving the Foreign Trade Division's export file, the Census Bureau's Standard Statistical Establishment List (SSEL), and the BEA surveys. A feasibility study was undertaken in 1986, and recent legislation giving BEA limited access to the SSEL is expected to facilitate the development of this kind of data.

In addition, in a “grab bag” category, several users expressed an interest in obtaining data in a form that identifies individual exporters and importers (or consignees), such as are presently available for vessel shipments from the PIERS system. Some of them recognized that it might not be possible for the Census Bureau to provide data at this level of detail. Included in this group of users was a unit of the Canadian government mainly concerned with the promotion of domestic production to substitute for imports to Canada. The reports published by this unit include lists of importers who are potential buyers of domestically made products. They would like to be able to use U.S. trade data to develop lists of U.S. importers and manufacturers who would be potential customers for Canadian exports.

So far, we have covered user-identified needs that could be met primarily through enhancements of the Census Bureau's foreign trade statistics program on the basis of processing official trade documents. Now we turn to user needs whose satisfaction would require the use of other sources of foreign trade data or, in at least one instance, the creation of a new data source. Two users called for expansion of the Bureau of Labor Statistics' monthly trade price index program to develop bilateral trade price indexes. One of them also proposed a survey of export orders. Such a survey would presumably be based on a sample of manufacturers and wholesalers. In his view, the data from such a survey would be valuable for the analysis and projection of export developments. Another proposal, from a user in the Canadian government, has implications for the censuses and surveys conducted by the Census Bureau's Industry Division. This user recommended that data on the production of U.S. firms be collected or compiled using the Harmonized System, so that it would be possible to estimate apparent domestic consumption for export commodities. This user also noted that Statistics Canada would be reporting production on this basis, beginning in 1990. The chief economist of a large U.S. corporation wants BLS to provide data on productivity for more countries as an aid in understanding international competitiveness and comparative advantage. He specifically mentioned Malaysia and Thailand as countries for which he would like to be

Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×

able to monitor productivity levels achieved, rather than just growth rates, especially in manufacturing.

Several users want access to more complete and centralized information on tariffs and nontariff trade restrictions. Some wanted the information for operational rather than statistical uses, but others want it in connection with their research. An official of a foreign economic research unit in the Department of Labor called for the development and maintenance of a tariff-rate file, with inclusion of information about product and country eligibility for special tariff rates. With regard to nontariff trade restrictions, a university researcher and an economist in the Federal Reserve System were clearly frustrated by their inability to obtain detailed information. The lack of such information on a historical basis made it difficult, if not impossible, for them to analyze the structural characteristics of protection and its effect on country-specific trade patterns. The user from the Federal Reserve System pointed out that the United Nations and the World Bank maintain extensive sets of data on nontariff barriers but that, because of the political sensitivity of these data, these agencies are reluctant to allow researchers access.

Most of the users who gave us their views focused their attention on current needs for data and did not respond to our invitation to comment on anticipated future needs for additional or different data as the international trade environment changes. There were some notable exceptions, however. The corporate economist and chief statistician of a major data wholesaling corporation recommended that the panel step back from the existing system and take “a more comprehensive look at what is really needed for longer term policy making.” He mentioned several ongoing and anticipated changes in the environment for international trade. Among them were the greatly expanded role and flexibility of multinational corporations and the attendant increase in the volume of intracompany transfers, trends toward multilateral reductions in trade barriers and the establishment of regional trading blocs, shifts to market-based economic activities in the Eastern European countries, and the discontinuities in long-standing trade time series created by the recent adoption of the Harmonized System. The corporate economist stated: “A fundamental redesign of information gathering about international transactions is called for . . . . ” With respect to merchandise trade, he said: “A redesigned trade data system could be developed by using well-known principles of sampling and devoting the same resources that are currently expended (for 100 percent processing of foreign

Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×

trade documents) for a much more sophisticated and intensive analysis of the characteristics behind the data.”

Others described what are seen as the main challenges facing the present U.S. system of foreign trade statistics: the possibility that starting in 1992 it may no longer be possible to identify country of origin or destination within the European Economic Community; the full implementation of the Canada-United States Free Trade Agreement, which will lead to the end of filing of import documentation on both sides of the border; the increasing commingling of goods and services, making the collection of detailed merchandise trade statistics less meaningful and relatively more costly; and the arbitrary nature of the valuation of commodities in affiliated transactions. In view of these emerging trends, some doubt that the U.S. system of merchandise trade statistics could be maintained indefinitely in its present form and therefore that large investments to make it more efficient would be misdirected. With respect to the scheduled 1992 changes in the European Economic Community, a business analyst for a chemical manufacturer had a specific recommendation, namely, that the U.S. should press the EC to continue reporting exports and imports by individual countries within the EC.

Several users commented on the difficulty of using existing foreign trade statistics to understand the increasingly complex flows involved in production by multinational companies. They noted that parts of final products are often made in several countries and shipped back and forth before final assembly. Lack of information about these processes, they believed, might distort perceptions of U.S. bilateral trade problems and economic and political responses to them. On the marketing side, they wanted better access to data on bilateral trade transactions not involving the United States.

INTERNATIONAL SERVICES TRANSACTIONS AND CAPITAL FLOWS

Only about one-third of the more than 100 respondents to our canvass commented on statistics for international services transactions or financial flows. This was not unexpected because nearly one-half of these organizations were reached through the list of subscribers to the Census Bureau's publications of merchandise trade statistics. Furthermore, data on merchandise trade have historically received the lion's share of attention and interest from the general public, and only a relatively small number of special-

Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×

ists follow trends in other kinds of international economic transactions.

The past decade, however, has seen a rapid growth of the service sector, both domestically and internationally, with a consequent increase in the demand for data needed by the government to establish and evaluate national strategies and priorities and by the private sector to monitor trends and analyze market shares. To ensure that these areas were properly covered in its study, the panel invited several experts on statistics on international services transactions and financial flows, representing both producers and users of data, to make presentations. Primary attention was given to international services transactions.

INTERNATIONAL SERVICES TRANSACTIONS

Some users commented on unresolved problems associated with the concepts and definitions used in the collection and dissemination of data on trade in services. Two users said that the dividing line between goods and services is often not clear. One pointed out that multinational companies have considerable flexibility in valuing services associated with goods and may be expected to use this freedom to change the domestic or foreign content of their products to their own advantage. He believed that there should be a standardized way to allocate such costs.

One official from the Office of the U.S. Trade Representative pointed out that the terms “exports of services” and “invisibles,” which are used in analyzing services transactions, are frequently misunderstood. He recommended that total trade in services be labeled as “U.S. sales of services to foreigners,” rather than as “exports”: the former applies to both cross-border sales and sales through affiliates; the latter bears the connotation of cross-border trade only. Similar recommendations were made by both a staff member of the Senate Subcommittee on Government Information and Regulation of the Committee on Government Affairs and the Coalition of Services Industries.

A representative from the Office of the U.S. Trade Representative (USTR) also commented on the completeness, timeliness, and comparability of U.S. statistics on international service transactions. With respect to completeness, he stated that “several important sectors, particularly financial services and transportation, are underrepresented” and that certain types of financial services are not covered at all. On the subject of timeliness, he noted that

Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×

the need to obtain the data through establishment surveys precludes the development of monthly data, and he pointed to the fairly substantial lag in obtaining information through such surveys.

On the question of international comparability, the USTR representative said that many countries do not collect balance-of-payments data in the same detail as the United States. To improve comparability for those who do, international agencies are attempting to establish standard classifications and formats for compiling and presenting the data. Two university researchers who are adding services trade to a general equilibrium model covering several industries and countries confirmed the need for such standards; they said they were facing enormous difficulties in putting the data for different countries on a comparable basis.

A former Commerce Department official pointed to several possible sources of error in the statistics on services transactions. He believes that the methods now used in estimating receipts and expenditures related to international travel, including transportation expenditures, should be thoroughly evaluated. U.S. residents are not questioned about how much they spend abroad after they return; rather they are asked when they depart from the United States how much they intend to spend. The sample excludes travelers on chartered planes, and the sample data are processed by a private company subject to relatively limited control.

He also noted that for several types of services, data are now collected that previously had not been available. Yet the minimum amounts of individual transactions for which reports are requested are so high that presumably some types of transactions, for example, legal services, are substantially underreported. For some kinds of transactions, no data are collected or estimated: for instance, receipts and expenditures by news organizations or incomes from the sale of foreign newspapers and magazines that are printed in the country where they are sold but obtain their content from abroad through electronic transmission. On the subject of completeness, a statement from the Coalition of Services Industries noted that BEA surveys have begun to capture information on some service industries previously ignored but that many service industries still are not covered by any survey. The statement also asserted that reporting requirements and the complexity of the surveys have limited company response.

A BEA official alluded to the effects of the U.S.-Canada data exchange agreement on the valuation of U.S. exports to Canada, pointing out that one consequence would be a complete lack of

Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×

data on cross-border receipts and payments by U.S. and Canadian carriers because Canadian import data are valued on a free-on-board (f.o.b.) basis.

Several users commented in broad terms about needs for more and better data on trade in services. An official of the U.S. Trade Representative noted the need for putting together a good information base on service transactions for use in connection with the Uruguay round of multilateral negotiations on services trade and said that much more detailed data may be needed as the negotiations progress. An official of the Department of Statistics of the International Monetary Fund also alluded to interest in an expanded list of service items in the context of these negotiations.

User statements included recommendations covering several aspects of statistics on international service transactions. A statement from the Coalition of Services Industries expressed substantial dissatisfaction with the data currently available and recommended clarification of concepts, more detail on different types of services, and monthly reporting. Other statements by a rather diverse group of users called for a variety on enhancements and improvements of services trade data, including more data on the educational component, more analyses of the data, development of price indices, more detail by country, and data by state.

CAPITAL FLOWS

Comments about statistics on financial flows were included in 10 of the presentations and statements received by the panel. The majority dealt with perceived deficiencies in the quality of existing data. In contrast to their perceptions of recent advances in the quality of data on international service transactions, the users believe that the quality of data on financial flows has been declining (see also Lipsey, 1990).

Specific problems with the existing reporting systems were discussed by researchers of the Federal Reserve Board and the Department of the Treasury. Both emphasized coverage problems and inaccurate reporting. Concerning the coverage problem, the Treasury International Capital (TIC) system concentrates on large filers and tends to miss the small filers. Given the development of the 24-hour international market, it is believed that the system will have to be modified to encompass the smaller players if they become more prevalent. Concerning inaccurate reporting, respondents said filers are devoting less resources to accurate reporting

Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×

and do not exercise adequate quality control. In addition, data collection agencies have inadequate quality control. Users believe that the accuracy of the data on U.S. capital flows and investment income cannot be improved without the devotion of additional resources and efforts to convince filers that accuracy in these reports is important.

Others noted that large quantities of physical cash move in suitcases and boxes outside normal banking channels. Electronic transfers among banking institutions have ballooned, and the movement of financial transactions is increasingly difficult to monitor. One proposed solution is for transnational banking authorities and national central banks to undertake a careful review of the types of financial flows occurring inside and outside of current measurement systems. One respondent believes this to be of paramount importance because the current information is woefully inadequate to address the key policy issues that will be important in the 1990s and beyond.

A researcher at the Federal Reserve Bank of New York also recognized increasing difficulties in measuring financial flows. The most important future problem he could foresee was the globalization of financial markets. Noninterest income has been growing in size relative to total services trade, but, it has become increasingly difficult to measure these transactions, even in nominal terms.

Some users called for additional detail on certain kinds of financial transactions and for greater consistency with other kinds of data. A representative of an industry association wanted foreign investment data to be available at finer levels of disaggregation and on a consistent basis with other data series. A bond market specialist from an investment firm was interested in better information on interest income received and paid, portfolio investment incomes, and factors driving foreign direct investment and foreign takeovers in the United States. A Treasury Department representative said that criticisms of the crude breakdown used in the TIC reporting system were common and that users wanted more information on purchaser characteristics and coverage of instruments, such as futures swaps, that could be used as hedges.

Finally, as was the case for merchandise and services trade, users of state data wanted information on some types of financial transactions at the state level. A representative from the Minnesota Department of Trade and Economic Development told the panel that data for the state from the BEA surveys of U.S. affili-

Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×

ates of foreign firms were of limited value due to disclosure problems. No information is provided on the value of Japanese affiliates in Minnesota, for example, because it could potentially reveal proprietary data of a single firm. Yet the Japan External Trade Organization reports two dozen firms in Minnesota owned in part or entirely by Japanese entities. The Minnesota representative also said that a second source of information on foreign investments at the state level is the transactions data produced by the Commerce Department 's International Trade Administration, which provides an annual listing of individual foreign investment transactions in the United States. Because the information is collected from secondary sources, it is fully disclosed. But the Minnesota official found the information from this source to be of limited usefulness, because the value was reported for fewer than half of the transactions and because the listing provided only a limited amount of summary information by state.

INVITATION FOR WRITTEN COMMENTS

Reproduced below is the text of the letter sent by panel chair Robert E. Baldwin to solicit comments from users of international trade data.

Accurate, timely and relevant data on U.S. foreign trade are necessary for many purposes, especially in the present era of trade deficits, greater competition from abroad, and increasing globalization of production. Under the auspices of the National Academies of Sciences and Engineering—National Research Council, the Committee on National Statistics is undertaking a two-year panel study to evaluate the quality of U.S. foreign trade statistics and to develop recommendations for improving them. The study, which has been funded by the Bureau of the Census and the Bureau of Economic Analysis of the U.S. Department of Commerce and the Customs Service of the U.S. Department of the Treasury, covers international trade in both merchandise and services.

For our evaluation, it is essential that we obtain the views of a wide spectrum of users of foreign trade data. As part of the panel study we will be interviewing users in several categories, but our resources are insufficient to meet or speak with all of you. Therefore, as Chair of the Panel on Foreign Trade Statistics, I would like to invite you to submit written statements describing what foreign trade data you are using, where you obtain the data, how you use the data, what your views are on the adequacy of the data for your purposes and how it might be improved, and what type of trade data will be most useful to you in the coming years.

Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×

Please send your statements to Dr. Anne Kester, the Study Director of the Panel on Foreign Trade Statistics, at the address shown below. Information on any or all of the following subjects will be useful:

  • The kinds of data that you are using. (Specify level of detail such as geographic and commodity types for merchandise trade data and services categories for services trade statistics.)

  • The sources — governmental or non-governmental — from which you obtain the data?

  • How you process and use the data.

  • Your views about various aspects of the data you use, including relevance, level of detail provided, frequency of publication, timeliness, accuracy, costs of obtaining the data and compatibility with other kinds of data, such as domestic production, as well as your views on how they might be improved.

  • Your anticipated future needs for additional or different data as the international trade environment changes.

To be of maximum benefit to our study, please send your statements to us by April 30, 1990. We look forward to hearing from you.

PROTOCOL FOR INFORMAL INTERVIEWS WITH DATA USERS

Reproduced below is the protocol for panel and staff members in interviews with data users.

  1. Introduction

    Review purpose of interview and how the information will be used.

    If agreed to by respondents, start recording.

  2. User/respondent characteristics.

    1. Organization or business.

      Name and address.

      Type of organization.

    2. For each person being interviewed, name, telephone number and position in organization or business.

    3. Are those present answering questions for the entire organization or only part of it? If the latter, may want to identify other important users in the organization.

  3. Acquisition of foreign trade data by the organization.

    1. Source agencies/organizations: Census, BEA, BLS, ITA, USDA, EIA, United Nations, OECD, EUROSTAT, private companies, etc.

    2. For each source, modes of acquisition.

      Hard copy publications.

      Electronic (magnetic tape, CD-ROM, on-line, other).

Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×

Microform/microfiche.

Other.

  1. Specifics. Obtain name, source, frequency and cost of each acquisition. For frequency, is it regular (monthly, quarterly, or annual) or occasional?

  1. How are data converted to usable information?

    1. Which of the following are done?

      Extract data items for citation in article or report.

      Extract data items for use in models.

      Calculate indices.

      Create new tables.

      Time series analyses.

      Other

    2. If yes to any of the items in C,1, how are the data prepared for input to perform the necessary calculations?

    3. Are data adjusted to conform to concepts or classifications different from those used by the source organization? If yes, what adjustments are made?

    4. Are foreign trade data processed or analyzed in conjunction with other kinds of data, e.g., data on domestic production or sales?

  2. Specific data items used.

    1. Are the same items used regularly or do they vary over time. If the latter, what is the nature of the variation?

    2. Broad categories of data used: imports, exports, prices, service trade, others.

    3. Variables, e.g., quantity, value (Customs import value, C.I.F. import value, F.A.S. export value).

    4. Classifiers.

      1. Commodities. Which classification systems are used?

        What is the finest level of detail and what are the groups or items of interest at that level?

      2. Geography.

        Region/country of origin or destination. What is finest level and which regions or countries are of interest?

        Port or district of entry/exit. Which are of interest?

        State of origin of exports. Which ones?

      3. Mode of transportation.

  3. Outputs containing foreign trade data.

    1. Publications of user organization or individual. For each one identified obtain:

      1. Name and type of publication (e.g., newsletter, industry or trade publication, economic report or bulletin, article for scholarly journal, etc.).

      2. How foreign trade data are presented (e.g., tables, charts, cited in text).

Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×
  1. Frequency (periodicity or one-time).

  2. Size and nature of audience, method of distribution.

  1. als. Identify specific medium, e.g., magnetic tapes, diskettes, CD-ROM, on-line access.

  2. general nature and purpose, frequency, recipients and how the foreign trade data are presented.

  1. Intended effects/uses of outputs. Identify and discuss all that apply.

    1. Influence trade policies or negotiations.

    2. Influence legislation or regulation relevant to foreign trade.

    3. Analysis and forecasting of general economic conditions.

    4. International market analysis.

    5. Investment decisions.

    6. Provide general information to secondary users.

    7. Other.

  2. Satisfaction with data currently available.

    [NOTE: If respondents are unsatisfied with any aspects, probe concerning adverse effects of perceived deficiencies on the uses identified in G. If they express satisfaction with respect to items 1 to 4 below, ask how their uses of the data would be affected if the amount of detail or frequency of publication were reduced, the delays following the end of reference periods were increased, or subscription prices were increased?]

    1. Content. Does it meet all needs? If no, what else is needed?

      Are there problems of comparability with other kinds of data?

    2. Costs and convenience of acquisition and user processing of foreign trade data. If not satisfied, what improvements are desired?

    3. Frequency and timeliness. Ask how respondents' use of the data would be affected if data are published less frequently (e.g., for merchandise trade statistics, from a monthly to a quarterly or annual basis.) Ask specifically about reactions to February 1987 shift from 30 to 45 days for publication of Census monthly data on merchandise trade.

    4. Are there problems associated with revisions? Describe.

    5. Accuracy. Have respondent's uses been affected by errors in data? How? How were the errors detected? Do they inform source organizations? Have they received satisfactory responses?

    6. Information on data concepts, definitions and methodology.

      What sources of information do they use and are these adequate for their needs?

    7. What are the other limitations of the data?

Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×
  1. Future data needs.

    Do respondents expect that their needs for foreign trade data will change in the future? If yes, how and why?

  2. Access to users of foreign trade data supplied by this organization.

    If this organization supplies foreign trade data to other organizations and individuals through publications or other media, inquire about their willingness to allow use of their mailing list for distribution of an invitation for users of their data to submit written statements to the Panel on Foreign Trade Statistics.

Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×
Page 187
Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×
Page 188
Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×
Page 189
Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×
Page 190
Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×
Page 191
Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×
Page 192
Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×
Page 193
Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×
Page 194
Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×
Page 195
Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×
Page 196
Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×
Page 197
Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×
Page 198
Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×
Page 199
Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×
Page 200
Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×
Page 201
Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×
Page 202
Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×
Page 203
Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×
Page 204
Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×
Page 205
Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×
Page 206
Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×
Page 207
Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×
Page 208
Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×
Page 209
Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×
Page 210
Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×
Page 211
Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×
Page 212
Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×
Page 213
Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×
Page 214
Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×
Page 215
Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×
Page 216
Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×
Page 217
Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×
Page 218
Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×
Page 219
Suggested Citation:"Appendix B: Canvass of Data Users." National Research Council. 1992. Behind the Numbers: U.S. Trade in the World Economy. Washington, DC: The National Academies Press. doi: 10.17226/1865.
×
Page 220
Next: Appendix C: A Comparison of U.S. Export Statistics With Those of Major Trading Partners »
Behind the Numbers: U.S. Trade in the World Economy Get This Book
×
Buy Hardback | $29.95
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

America's international economic decisions rest to a large degree on the information available to policymakers. Yet the quality of international trade and financial data is in serious doubt. This book reveals how our systems for collecting and analyzing trade data have fallen behind the times—and presents recommendations for new approaches to accuracy and usefulness of these economic data.

The volume traces the burgeoning use of international economic data by public and private analysts at a time when the United States is becoming increasingly integrated into the world economy. It also points out problems of capturing new transactions, comparing data from different sources, limited access to the data, and more. This is the first volume to review all three types of U.S. international data—merchandise trade, international services transactions, and capital flows. Highlights include:

  • Specific steps for U.S. agencies to take.
  • Special analyses on improving the accuracy of merchandise trade data, filling data gaps on the fast-growing international services transactions, and understanding structural changes in world capital markets.
  • Comments, complaints, and suggestions from an original survey of more than 100 key users of trade data.

This practical volume will be invaluable to policymakers, government officials, business executives, economists, statisticians, and researchers.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    Switch between the Original Pages, where you can read the report as it appeared in print, and Text Pages for the web version, where you can highlight and search the text.

    « Back Next »
  6. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  7. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  8. ×

    View our suggested citation for this chapter.

    « Back Next »
  9. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!