Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 16
Understanding American Agriculture: Challenges for the Agricultural Resource Management Survey 2 Contemporary Issues in American Agriculture The Agricultural Resource Management Survey (ARMS) has been called the mirror in which American farming views itself (Economic Research Service, 2007). In its detail, the survey provides myriad observations about farms and farm households. When assembled into aggregate measures, it aims to form a comprehensive picture. The goal of providing in-depth detail that adds up to a multidimensional view of agriculture is a unique and important aspect of the survey. The value of ARMS is in its totality and the interactive nature of the data, not just the value of each data item. It is a comprehensive documentation of many aspects of U.S. agriculture and farm households. In part because it is so comprehensive—and unique in that respect—it has important uses in illuminating contemporary issues in American agriculture. These uses form the basis for assigning priorities of effort, justifying the resources that are devoted to the survey, and largely defining its content and products. This chapter begins with an illustrative catalogue of mandated, programmatic, and research uses of ARMS data by the U.S. Department of Agriculture (USDA) and other federal agencies, the private sector, academic researchers, and the general user community. It continues with a general discussion of contemporary issues in American agriculture and how they generate a demand for information that will help policy makers and private-sector decision makers at all levels understand and deal with those issues. It goes on to discuss in more detail three specific areas of interest in agriculture policy: environmental and resource management, commodities, and the economic situation of farms and farming families. The discussion
OCR for page 17
Understanding American Agriculture: Challenges for the Agricultural Resource Management Survey includes an assessment of the adequacy of the ARMS data to illuminate these issues. PRIORITY USES OF ARMS DATA The ARMS program represents a significant investment of time, talent, respondent burden, and resources. To justify this investment, the survey must be responsive to a set of core requirements that address legislative, programmatic, and analytical needs. These core requirements build on those of the predecessor surveys, which conveyed into ARMS when it was established in 1996, and have been supplemented by more contemporary and changing requirements.1 The task of meeting these core requirements translates into a series of priorities for the ARMS program. The data items needed to meet the core requirements have largely been maintained and protected by making sure these items are included before any other items are added. For the National Agricultural Statistics Service (NASS) and the Economic Research Service (ERS), these priorities affect the content of the questionnaires, which in turn are instrumental to the survey’s ability to meet the core requirements. Mandated Uses USDA is required by Congress, through both authorizing and appropriation legislation, to produce a sizeable portion of the data items that are included in ARMS. Cost-of-production data are required by several pieces of legislation, and one piece of legislation is very specific. The U.S. Code states that the “Secretary of Agriculture, in cooperation with the land grant colleges, commodity organizations, general farm organizations, and individual farmers, shall conduct a cost of production study of the wheat, feed grain, cotton, and dairy commodities under the various production practices and establish a current national weighted average cost of production. This study shall be updated annually and shall include all typical variable costs, including interest costs, a return on fixed costs, and a return for management” (U.S. Code, Title 7). Environmental and food safety legislation call for data on chemical use on field crops. The Food, Agriculture, Conservation, and Trade Act of 1990 and the Food Quality Protection Act of 1996 require NASS to collect data on field crop chemical use and publish those data annually (in the Agricul- 1 The predecessor surveys of ARMS were the Farm Costs and Returns Survey (FCRS) and the Cropping Practices Survey. The predecessors to the FCRS were the Farm Production Expenditures Survey and the Cost of Production surveys. The household component was added to the FCRS in 1988 with the Farm Operator Resource version.
OCR for page 18
Understanding American Agriculture: Challenges for the Agricultural Resource Management Survey tural Chemical Usage Field Crops Summary and the Agricultural Chemical Usage Restricted Use Pesticide Summary). Some data series are used in the preparation of mandated reports. For example, in preparing the Annual Report on Family Farms required by the Food and Agriculture Act of 1977, ERS draws on ARMS data for information on a host of relationships, including (1) farm participation in agricultural programs and the distribution of farm program payments; (2) the structure and organization of farms, including family and nonfamily ownership; (3) the use of new production technologies and other management practices; (4) farm use of credit; (5) farmers’ participation in off-farm employment; and (6) identifying the characteristics of producers purchasing crop insurance. ARMS data are sometimes input to other data series, thus fulfilling their mandate as a derived requirement from another product. Mandated uses of ARMS data include annual estimates of average income for U.S. farm operator households, and annual cost-of-production estimates for over 15 agricultural commodities. These mandated data items generate further requirements, in that a number of data items are necessary to compute these derived estimates. Similarly, ARMS production input data provide annual weights for the NASS computation of the Prices Paid by Farmers Index. In turn, this index is used to calculate parity prices required by the 1933 Agricultural Adjustment Act. Parity prices help administer federal marketing orders for some 45 fruits, vegetables, and nuts. The indices are also required by the 1978 Public Range Improvement Act to calculate annual federal grazing fees on the nation’s western public lands by the U.S. Bureau of Land Management and the U.S. Forest Service. National Accounts USDA is required to support the U.S. national accounts by producing estimates of farm-sector value-added and net income. The agency responsible for the national accounts, the Bureau of Economic Analysis (BEA), reports that the primary source for the input-output (I-O) estimates of farm output is “cash receipts from farm marketings,” which is compiled from ARMS and various sources by ERS. This series is considered to be of higher quality and of more relevance for the I-O estimates than the data collected in the Census of Agriculture. The I-O estimates of farm expenses are based on the ARMS farm costs and returns survey (Horowitz and Planting, 2006). BEA uses USDA’s annual estimates of net farm income to develop its annual estimates of gross domestic product and personal income. ARMS data also contribute to the development of BEA’s regional (state and county) accounts through inclusion of the state-level estimates of net farm income. The Farm
OCR for page 19
Understanding American Agriculture: Challenges for the Agricultural Resource Management Survey Production Expenditures by type of organization for county income distribution is produced through a standing request each year from BEA for publication in County Business Patterns. USDA Programmatic Needs ARMS data provide a basis for updating weights in the Prices Paid Index Program,2 as well as for data to implement USDA program requirements and to formulate and evaluate USDA polices and programs. For example, publication of income for farm households was required by a past secretary of agriculture. The departmental definition of a limited resource farm relies on ARMS data. Fuel expenditures data are also used by the USDA Office of Energy Policy and New Uses to assess agricultural fuel usage. General Information ARMS provides data to inform decision makers on a wide variety of farm, farm household, environmental, and other rural issues. For example, ARMS data shed light on important national issues, such as energy use, by supporting analysis of the impact of energy prices on farm production costs and, consequently, on the price of commodities. ARMS data are also used to (1) gather information about the relationships among agricultural production, resources, and the environment; (2) determine the costs to produce various crop and livestock commodities and the relative importance of various production expense items; (3) help determine the net farm income of farmers and ranchers and provide data on the financial situation of farm and ranch businesses, including debt levels; and (4) help determine the characteristics and financial situations of farm and ranch operators and their households, including information on management strategies and off-farm income. Research and Analysis ARMS data support the ERS program of research and analysis, which depends on these data, and contribute to other research and analysis work by providing basic cost-of-production and input use information. The data also support an active and growing program of research and analysis at academic institutions and other organizations that aim to illuminate diverse issues in contemporary American agriculture. Many of these studies are conducted as collaborative efforts between ERS and the research community. 2 Prices paid indices are used to compute parity prices under the Agricultural Adjustment Act of 1938 as amended, Title III, Subtitle A, Sections 301a. The Agricultural Marketing Service uses prices paid indices to determine support prices.
OCR for page 20
Understanding American Agriculture: Challenges for the Agricultural Resource Management Survey ERS is both a producer and a consumer of ARMS data. In addition to its extensive inventory of data sets and recurring publications, the ERS generates demands for ARMS data in its selection of topics for analytical projects. These demands often form the basis for additions and modifications in the data collected in ARMS. To illustrate, ERS work plans for 2006 included several initiatives on topics as varied as farm structure and governance, to food safety, to program participation, to farm household behavior, each using the ARMS as a major data source: Changing structure of livestock and poultry industries. Successive questionnaire versions targeted hogs, dairy, and broiler (chicken raised for meat) producers in 2005-2007 about their operations in the prior year. ERS intends to use the data to summarize structural change in each industry; analyze the sources of production cost differences, such as location or scale economies, as drivers of structural change; derive baseline information on manure management practices; and assess the effects of structural change and manure management practices on excess nutrient loadings. Food safety strategies in the livestock and poultry supply chain. ERS has included questions on production practices related to food safety and their links to processor/contractor requirements in the survey as noted above. ERS intends to use the data to summarize the use of various practices and to identify their impacts on production costs. Producer responses to changes in federal peanut and tobacco programs. The 2004-2006 Cost and Returns Report versions and the 2004 peanut version of ARMS included questions designed to elicit information on how producers responded to the fundamental changes in federal support for those industries, to be used in broader ERS analyses of the market response to the changes. Farm operators’ input procurement strategies. Farmers may procure inputs through purchase, lease/rent, service contracting, on-farm provision, or joint sharing of inputs with other producers. Questions in the Costs and Returns Report allow ERS to identify those several strategies, to generate baseline estimates of their use, and to assess the effects of each strategy on farm efficiency and profits. Farm governance and organization. Large farms may have complex organizational structures; those structures may be an effective means of organizing farm production, but they may also provide a means to avoid USDA program rules. They also complicate the assessment of the role of family farms in American agriculture. Starting with the 2005 Costs and Returns Report, ERS has added questions designed to provide a more precise picture of organizational structures in American agriculture. Farm and farm household linkages to rural communities. The 2004 Costs and Returns Report contained questions pertaining to the linkages
OCR for page 21
Understanding American Agriculture: Challenges for the Agricultural Resource Management Survey between farms and farm households and nearby communities to better understand the importance of farm policy to rural communities and the importance of general economic policy to farm households. Health and health care of rural residents and farmers. The 2005 Costs and Returns Report contained additional questions on the health insurance coverage of each member in the farm operator household, the sources of the insurance (employer, Medicare, private purchase by the household, etc.), and household health insurance costs and out-of-pocket health expenditures. ERS will complement the ARMS information for farm households with information from the Current Population Survey, the National Health Interview Survey, and other sources to characterize outcomes for other rural households. Consumption behavior of farm households. The 2004 Costs and Returns Report reinstated a revised version of questions on household living expenses, and the 2005 report contained a further revised version to calculate consumption flows for the household. The goal was to achieve consistency with the measures of “consumption expenditures” and “current (annual) consumption service flows for the household” calculated with the Consumer Expenditure Survey of the Bureau of Labor Statistics, collected for a nationally representative sample of U.S. households. Design options for green payments programs. Proposed “green payments” would alter federal farm policy to jointly achieve income and conservation goals with the same payment instrument. ERS will compare various specific alternatives, using a “screening model” that links ARMS Phase III data to environmental and conservation program data. The screening model will compare alternative green payments to estimates of farmers’ willingness to accept payments for various conservation practices. ARMS is the only source of nationwide data to support research on farmers’ decisions to adopt new technologies and to relate those decisions to (1) the economic performance and structural attributes of farms and farm families, and (2) the subsequent environmental impacts. Key technology adoption decisions being tracked include the choice of bioengineered seed, the selection of waste management practices by livestock producers, the use of chemical and biological pest management alternatives, and the use of information management technologies ranging from precision farming in crop production to marketing commodities and buying inputs via the Internet. Federal Government Agencies Representatives from several federal agencies briefed the panel on the value and uses of ARMS. Common uses in these agencies include analysis of
OCR for page 22
Understanding American Agriculture: Challenges for the Agricultural Resource Management Survey conservation practices, payments, ethanol trade, distribution of commodity subsidies and alternatives, pesticide uses and regulations, and estimates of income and cost of production. Many noted the tie to farm households as a unique benefit of the survey. For the most part, these agencies do not make direct use of the microdata. They have asked for special reports from ERS, used the public-use aggregates available in the online ARMS briefing room, and used information derived from published reports and briefing papers that used the microdata. The Congressional Budget Office (CBO) provides nonpartisan budget information for the agriculture committees in Congress through a small agricultural team (three employees in 2006) in the budget analysis division. This team produces mandated cost estimates for every bill coming out of the full committees, and it does a number of estimates before the bill gets to that point. CBO provides baseline estimates twice a year, including the current year plus 10 years, and tracks changes over time. The team needs data mainly on incentives to adopt and levels of adoption of conservation practices and general details on payments. It obtains its information indirectly through outside reports or conversations with people who know the data. CBO staff, who do not have access to the ARMS microdata, find briefing papers useful due to tight time constraints and stress the importance of ARMS for information on the means testing and diversity of farm operations.3 The President’s Council of Economic Advisers (CEA) has used the ARMS aggregate data directly from the website as well as other information from USDA to assess both short-term and long-term changes in agriculture. Although it is difficult to anticipate its needs, some examples of prior topics include trade of ethanol and distribution of commodity subsidies and alternatives. ARMS cost-of-production data can assist with the latter of these two. In a presentation to the panel, an agriculture specialist formerly on the CEA staff expressed an interest in ARMS allowing stratification of data by government payment, commodity payment (receipt/nonreceipt), and subsidy payment class. He suggested that more value might be gained from the standpoint of commodity programs by decreasing the sampling rate of smaller farms. About half of the farms in the Census of Agriculture have less than $10,000 of gross revenue, and about 9 percent of census farms account for over 90 percent of production. ARMS collects financial and income information in the same questionnaire, allowing analysts to cross-tabulate assets, debts, and net income. This cannot be done with other data sources, including the Census of Agriculture.4 The Environmental Protection Agency (EPA) is concerned with pesticide practices regarding fruit and vegetable crops, which are commodities 3 Presentation by James Langley, Congressional Budget Office, September 28, 2006. 4 Presentation by Joe Cooper, ERS and past CEA staff member, September 28, 2006
OCR for page 23
Understanding American Agriculture: Challenges for the Agricultural Resource Management Survey that ARMS does not cover. Typically the agency needs to determine the impact on agriculture and the total economy if a pesticide is registered or cancelled or an allowable use is changed in any way. In order to accomplish this, information is needed on the use of the pesticide under consideration, the use of pesticides that can be considered alternatives to that pesticide, and the yield and quality impacts of changes in pesticide use. EPA gets most of these data from other NASS sources, as well as Doane Marketing Research, a proprietary data source with 17,000 respondents, primarily from larger farms, that provides estimates comparable in many respects to NASS estimates, although not as complete as NASS on the timing of pesticide application. Data on how much pesticide a farm uses, on what crops it is used, and the timing of pesticide applications and the target pests of those applications are important to EPA. If these basic information items were available from ARMS, they would provide a key framework for many analyses. EPA currently makes extensive use of crop budgets, but they are limited in that they are not able to be cross-tabulated with information about actual conditions—the quantity and condition of crops or the timing of pesticide applications.5 The staff in the USDA Office of the Chief Economist primarily works on short-run, day-to-day issues, drawing on resources throughout the department. They frequently make requests to ERS for ARMS data and analysis to produce farm income and cost-of-production estimates, for example: Calculating farm energy cost-to-output ratios;6 Highlighting the distribution of farm income, household income, and potential problems servicing debt; Explaining the distribution of farm program payments; and Identifying the characteristics of producers purchasing crop insurance. An important benefit of ARMS data is their tie to the households of farm operators. The financial data of ARMS assists the Office of the Chief Economist in evaluating the economic impact of policies, and there is a lot 5 Presentation by Arthur Grube, Environmental Protection Agency, September 28, 2006. 6 USDA chief economist Keith Collins testified to the Senate Agriculture Committee on the implications for U.S. agriculture of higher energy prices using ARMS data to show how energy price increases would affect production costs for different commodities and regions. The testimony provided an estimate to the committee, also relying on ARMS data, of the likely impacts of energy price increases and hurricane damage on farm incomes. Finally, ARMS data were used to suggest likely substitution responses in farm inputs. Statement of Keith Collins, Chief Economist, U.S. Department of Agriculture before the U.S. Senate Committee on Agriculture, Nutrition, and Forestry, November 9, 2005, <http://www.usda.gov/oce/newsroom/congressional_testimony/collins_11092005.doc>.
OCR for page 24
Understanding American Agriculture: Challenges for the Agricultural Resource Management Survey of interest in the off-farm income of household members. In a presentation to the panel, the office representative stated that it would use the data more extensively if they were available longitudinally; he also noted the value of distributional information from ARMS and data on the new technology adoption and how it affects the cost of production.7 The National Resources Conservation Service (NRCS), previously the Soil Conservation Service, is a USDA agency that provides leadership in a partnership effort to help conserve, maintain, and improve the country’s natural resources and environment. It analyzes the interplay between rural land and land under development. NRCS has used ARMS data on production costs, input use, and technology adoption in selected regions to assess the performance of conservation programs, and it has used information on manure and fertilizer use and management practices to calculate nutrient mass balances for selected crops, regions, and management systems. The agency has consulted ARMS for hog production characteristics, practices, and costs to assess the economic feasibility of proposed nutrient management policy alternatives. It used ARMS data about incentives for farmers’ use of resource management practices to assess the effect of their programs. Use of ARMS data by other USDA agencies is both periodic and topical. USDA reports the following examples of uses by other USDA agencies: The Agricultural Marketing Service uses ARMS data in deriving its monthly cost-of-production estimates for milk production for the United States and five regions. The Risk Management Agency has used special tabulations from ARMS to understand levels of farm income and risk management tools used by farmers. The Agricultural Research Service has used special tabulations from ARMS to better understand the structural and production characteristics of farms and the demographic characteristics of farm operators for each of its research planning regions. The Farm Service Agency uses the annual burley and flue-cured tobacco cost-of-production estimates derived from ARMS data to help set tobacco price support levels. The Cooperative State Research Education and Extension Service has used ARMS information about adoption of alternative pest management strategies by farmers in its program planning. The Rural Business-Cooperative Service has used ARMS data to obtain information about the use of cooperatives by farmers. 7 Presentation by Joseph Glauber, U.S. Department of Agriculture, September 28, 2006.
OCR for page 25
Understanding American Agriculture: Challenges for the Agricultural Resource Management Survey Interest Groups There is a wide variety of interest group users with a wide variety of data needs. Two of the groups, the American Farm Bureau Federation and the National Cotton Council, participated in a panel discussion of data needs and the adequacy of the ARMS program for meeting those needs. The American Farm Bureau Federation (AFBF), an independent, nongovernmental organization governed by and representing farm and ranch families, considers ARMS one of the most important USDA products. It sends news releases to about 75,000 members and strongly encourages its members to respond to NASS surveys. With the reauthorization of Farm Bill legislation scheduled for 2007, ARMS may become more vital in terms of measuring costs of production and net farm revenues. The AFBF expressed an interest in more information on contracting and information on individual commodities on a more frequent basis.8 The National Cotton Council is typical of many commodity-oriented interest groups. It uses the ARMS on a regular basis, as it is the best source for many of its data needs. The cost-of-production information is used most often and gets a lot of attention in the industry. With a 1997-2003 gap in the ARMS rotating collection of cotton cost-of-production data, there are doubts as to whether the adoption of new practices (e.g., biotechnology) is appropriately captured as estimates are updated in the intervening years. Besides wanting more frequent collection of cotton cost-of-production data in the ARMS rotation, the council would like to see the ARMS sample size increased to enable the provision of data on as local a geographic level as possible, in order to have distributional information beyond averages. The National Cotton Council does not conduct any elaborate in-house surveys, although they do conduct a survey on acreage intentions at the beginning of each year. Like the AFBF, they encourage members to respond to all NASS surveys and also signed on to a NASS letter supporting participation by their membership.9 DATA RELEVANCE Contemporary Issues That Generate Farm-level Data Demands The core requirements for ARMS data have been generated with a view of the kinds of data that decision makers and other users require in order to understand contemporary American agriculture. Understanding the U.S. agricultural sector is a daunting challenge: it has characteristics 8 Presentation by Robert Young, American Farm Bureau Federation, September 28, 2006. 9 Presentation by Gary Adams, National Cotton Council, September 28, 2006.
OCR for page 26
Understanding American Agriculture: Challenges for the Agricultural Resource Management Survey unlike other sectors of the U.S. economy, and, in part because of that uniqueness, the sector has been thought to pose special challenges in measurement and analysis. Agriculture has long experienced a trend of increasing productivity. It has been an engine of growth and a source of net exports in international trade. Technological and structural changes have led to the largest producers becoming less distinct from other small-business sectors of the U.S. economy than in earlier years. Agriculture is becoming dramatically more concentrated. Just 15 percent of all census-defined farms produce 85 percent of agricultural product value. Despite a shift toward greater concentration in agriculture, a large segment of the smallest producers have significant agricultural production (Economic Research Service, 2006a). Smaller family farms (smaller farms are defined as units with annual gross revenue less than $250,000) represent 91 percent of total U.S. census-defined farms, according to 2003 data, holding 71 percent of all farm assets and 70 percent of the land owned by farms. These smaller family farms make substantial contributions to the production of hay, tobacco, vegetables, cash grains and soybeans, dairy products, beef cattle, Christmas trees, and greenhouse production. Abounding in rich variety, they are classified by ERS into a typology as limited-resource, retirement, residential/lifestyle, and farming-occupation/low sales, and farming-occupation/high sales farms. Smaller farms continue to play an important role in natural resource and environmental policy, accounting, for example, for approximately 82 percent of land enrolled in the Conservation Reserve and Wetlands Reserve Programs. Environmental and conservation issues are also high on the public agenda and generate new and complex demands for data. Large-scale producers have in many cases adopted minimum till or no-till production systems to minimize environmental impacts from runoff of agricultural fertilizers and chemicals. In general, environmental quality interacts with agricultural production, but precisely how and when are not well understood. Nonetheless, regulatory and other policy decisions have been made and are likely to increase in scope and effect. Incentives for legislatively favored management of agricultural lands have been increasingly incorporated into the annual Farm Bill. EPA regulations have impacted the livestock sector in particular in recent years. Water quality issues are a major concern throughout the United States, and agriculture plays a significant role in water pollution, whether from nutrients, pathogens, pesticides, salts, or sediment (Economic Research Service, 2006b). Further complexity arises from the confluence of environmental and marketing issues in the production of farm products, as is evident in the increased interest on the part of producers and consumers in organic products. Related issues involve food production practices, for example, free range production
OCR for page 38
Understanding American Agriculture: Challenges for the Agricultural Resource Management Survey purchasing power of the net income per person on farms and the income per person not on farms that prevailed during the five-year period August 1909-July 1914” (U.S. Department of Agriculture, 1944). The lack of income parity between the farm and nonfarm population was a central component of the “farm problem” as defined at that time. There were several early sources of statistics on the income of farm households. The first statistical series compared the disposable personal income per capita of farm residents with that of nonfarm residents for the years 1910-1943. The goal of this series was to build on USDA’s aggregated sector-level estimate of net farm income to get farm income per farm, by dividing by the number of farms in the Census of Agriculture and supplementing the resulting estimate of farm income per farm with information on off-farm income available occasionally from the Census of Agriculture and other sources. USDA published this derived series until 1983. A later historical series, on the incomes of principal farm operator households, began in 1960. This series was the sum of the annual estimate of the net farm income of the sector and off-farm income of the farm operator household based on Census of Agriculture data. A major deficiency of this series was inconsistency in the treatment of cash and noncash income items, and it was replaced by one that treated income items in a way consistent with the treatment of the income statistics for U.S. households based on the Current Population Survey. For 1984 and later years, estimates were based on the Farm Costs and Returns Survey (FCRS) data. The series was later refined with the 1988 FCRS data in a variety of ways, including recognizing that not all farms are family farms and that not all farm business income went to the farm operator household. This series is the basis of the current ERS statistical series on farm operator household income and is compared with the incomes of the average U.S. household and published annually. This is not a straightforward matter, since comparing farm household income from ARMS with the income of the average U.S. household from the Current Population Survey (CPS) requires calculating an estimate of farm household income from ARMS that is consistent with CPS methodology. The CPS defines farm self-employment income as net money income from the operation of a farm by a person on his or her own account, as an owner or renter, and includes income received as cash but excludes in-kind or nonmoney receipts. Farm self-employment income from ARMS is the sum of the operator household’s share of farm business income (net cash farm income less depreciation), wages paid to the operator, net rental income from renting farmland, and other farm-related income (net income from a farm business other than the one being surveyed, wages paid by the farm business to household members other than the operator, and commodities paid to household members for farm work) (Economic Research Service 2005). Although USDA’s estimates of income from farming are based on survey data, they have an important feature that is not followed in estimating
OCR for page 39
Understanding American Agriculture: Challenges for the Agricultural Resource Management Survey nonfarm self-employment income of households: farm operators are not asked directly about their net income from farming. Instead, ARMS collects data about the various components of farm income—revenues from selling products and services and costs of production. ERS then calculates each farm’s net income in several ways, using cost concepts that fit with different income concepts. For example, cash income does not account for depreciation of capital, while net farm income does. It is important to provide different measures of farm income because the use of data derived directly from survey responses varies among them. To obtain an estimate of a farm household’s off-farm income, ARMS does ask directly about net income received, even for off-farm business income that may have returns and costs complications similar to those of farming.13 The data show a striking trend of farm operator household income overtaking the incomes of nonfarm households over the past two decades. The collection of farm income data on ARMS is an important input to the USDA sector-wide set of income accounts. The cash revenue received in the sector is estimated, generally, by multiplying an average price times an estimate of the quantity marketed. Data on the direct government payments to the sector come from the government agency making the payments. In general, the data on cash expenses come from the ARMS program. There are exceptions—for example, the estimate of interest expenses comes from agencies that make loans to the agricultural sector. The depreciation is calculated based on the value of buildings and the value of machinery and equipment reported by the ARMS program. One contribution of the ARMS microdatabase is the ability to disaggregate the sector-wide estimates of farm income in order to show the flows to and from each of the major stakeholders in the agricultural sector. To do this requires detailed enumeration by the ARMS program. Specifically, the ARMS questionnaires determines the revenue flows and expenditure flows for the following groups: Family farms—generally, these are proprietorship farms, partnership farms, and incorporated farms operated by family members. Nonfamily farms—generally, these are farms operated by widely held corporations. Landlords—generally, these are individuals or companies that rent land to farmers—but sometimes they lease cows or machinery to farmers—and often they contribute part of the chemical and fertilizer expenditure for the operation of the land that they rent to their tenant. 13 The 2005 Core Phase III survey asks for “the principal operator’s pay from operating any other business” and for “the cash income the principal operator, spouse, and all other household members received in 2005 from” a list of categories including “other off-farm sources of income,” which is where business income other than the principal operator’s pay from such business would be reported.
OCR for page 40
Understanding American Agriculture: Challenges for the Agricultural Resource Management Survey Contractors—generally, these are businesses that establish contracts with farming operations in order to participate (or “control”) some or all of the decision making in crop or livestock production. For example, in egg production, the contractor may own the birds and pay the farm operator a management fee to compensate for the feed, the barn, and the labor. Alternatively, the operator may own the birds but the contract specifies the number, quality, and quantity of eggs to be delivered. There is a wide range in the types of contracts and the types of incentive schemes in these contractual relationships. Importantly for the purpose of analysis, these businesses (“contractors”) receive part of the return to labor and capital (i.e., the net farm income) in the sector. The task of determining the size of the various flows associated with each enumerated farm adds “enumeration complexity” to an already complex survey environment. Nonetheless, as a result of collecting these flows, USDA can show the role of each stakeholder group. Specifically, for 2005, estimates provided by ERS indicated: $74 billion aggregate sector-wide net farm income; of which $49 billion was received by family farms $9 billion was received by nonfamily farms $2 billion was received by landlords $16 billion was received by contractors $2 billion was discrepancy The panel requested documentation of the discrepancy, which requires detailed analysis to reconcile (see Box 2-1). ARMS collects cash revenue for each observation, and the sum across all observations does not replicate, exactly, the results of the methodology for estimates of cash revenue for the sector-wide accounts (described above). In addition, there were numerous other sources of discrepancies—this is not surprising if one is generating statistics from different sources. The panel found the detailed reconciliation to be instructive as an indicator of the size of the differences in the estimates from the ARMS program and the published sector-wide estimates (Box 2-1). Making this reconciliation available to users on an annual basis would enable them to understand the different sources of data in the different accounts and to appreciate the magnitude (or lack thereof) of the discrepancies. In the final analysis, significant discrepancies exist between the expenditure estimates and the income estimates in the national accounting framework for estimates of gross domestic product. In addition to providing a disaggregation by type of stakeholder in U.S. agriculture, ARMS calculates a farm-level estimate of net value-added,
OCR for page 41
Understanding American Agriculture: Challenges for the Agricultural Resource Management Survey BOX 2-1 Reconciling the Apparent Gap in Net Farm Income and Farm Household Earnings Aggregate net farm income for the agricultural sector was $82.5 billion in 2004. “Earnings of the operator household from farming activities” was $14,201 per household in 2004. Assuming 2,060,822 “family farms,” the aggregate calculates to be about $29.3 billion. ISSUE: How to get from the $82.5 billion for the agricultural sector to the $29.3 billion for “family farms.” POSSIBLE EXPLANATIONS: The net farm income line of Table 2-2 for 2004 provides the main elements of the gap as estimated by ERS. They are net farm income received by landlords who are not farmers, nonfamily farms (notably corporations in farming), and contractors who pay farmers to produce commodities but are not themselves farmers. RECONCILIATION: In order to compare aggregate sector-wide estimates of net farm income with an estimate of farm operator household income from farm and nonfarm sources, several adjustments must be made to the estimate of aggregate sector net income. These include (1) distributing aggregate sector-wide income to stakeholders, as depicted in Table 2-2; (2) the remaining net farm income of operators, which is a measure of profit earned, must then be converted to an estimate of net cash income; (3) the estimate of net cash income of operators must be distributed to households that participate in the business operation (many farms, particularly, but not exclusively, larger businesses, have multiple operators); and (4) the business measure of net cash income that accrues to the primary operator must be converted to a census-based measure of money income. All these steps result in an estimate of “income from farming activities that accrues to the principle operator households” not aligning with aggregate sector-wide estimates of net income, even when the per farm estimate is expanded to account for one principal operator household for each farm. SOURCE: T. Covey et al., 2005, p. 1. which allows the tabulation of value-added by such variables as type of farm, size of farm, and geographic area (Table 2-2). The approach to farm income measurement is predicated on the selection of an appropriate statistical indicator for several possible measurements of interest. In a presentation to the panel, ERS provided the information shown in Box 2-2, which ties the appropriate statistical indicator for farm income to the concept to be measured.14 14 Presentation by James Johnson, Economic Research Service, September 28, 2006.
OCR for page 42
Understanding American Agriculture: Challenges for the Agricultural Resource Management Survey TABLE 2-2 Net Income Derived for ARMS Using Sector-wide Account Procedures, 2004 and 2005 (dollars in billions) Operators 2004 Family Farms Nonfamily Farms Landlords Contractors ARMS Total Estimates Sector-wide Income Accounts Value of crop production 91.5 14.6 5.9 1.3 113.0 125 Value of livestock production 69.6 13.7 0.0 39.7 123..0 124 Revenues from services and forestry 21.1 1.1 11.4 0.0 33.7 34 Value of agricultural sector production 182.2 29.5 17.4 41.0 270.0 283 Purchased inputs 96.6 14.8 1.2 26.2 138.8 137 Farm origin 34.6 8.2 0.3 19.6 62.7 58 Manufactured inputs 28.6 2.5 0.7 31.9 32 Other intermediate expenses 33.5 4.0 0.3 6.6 44.3 48 Net government transaction 4.6 −0.1 −0.7 3.9 5 Farm gross value-added 90.2 14.6 15.5 14.8 135.1 152 Capital consumption 16.3 1.3 2.8 20.4 23 Farm net value-added 73.9 13.3 12.7 14.8 114.7 129 Payments to stakeholders 32.7 5.7 38.5 43 Net farm income 51.5 8.2 1.8 14.8 76.2 85
OCR for page 43
Understanding American Agriculture: Challenges for the Agricultural Resource Management Survey Operators 2005 Family Farms Nonfamily Farms Landlords Contractors ARMS Total Estimates Sector-wide Income Accounts Value of crop production 83.8 17.2 5.0 1.8 108.0 113 Value of livestock production 71.5 11.1 0.3 38.4 121.2 127 Revenues from services and forestry 22.0 1.8 12.3 0.0 36.1 36 Value of agricultural sector production 177.2 30.1 17.7 40.2 265.2 275 Purchased inputs 102.0 14.5 1.4 24.6 142.6 147 Farm origin 32.4 5.6 0.4 15.7 54.1 58 Manufactured inputs 31.7 3.2 0.7 0.2 35.8 36 Other intermediate expenses 37.9 5.7 0.3 8.7 52.7 53 Net government transaction 10.0 0.4 −1.8 8.6 16 Farm gross value-added 85.2 16.0 14.4 15.6 131.2 145 Capital consumption 15.6 1.3 2.7 19.5 24 Farm net value-added 69.7 14.7 11.7 15.6 111.7 120 Payments to stakeholders 32.3 7.3 39.6 46 Net farm income 46.7 8.4 1.8 15.6 72.0 74 SOURCE: James Johnson, Reconciling Sector-wide Net Farm Income and Farm Household Income Data Series, ERS, September 28, 2006.
OCR for page 44
Understanding American Agriculture: Challenges for the Agricultural Resource Management Survey BOX 2-2 Income Measures and Statistical Indicators Measure Statistical Indicator Farming’s contribution to state and national economies Farm sector’s net value-added Earnings of farming’s risk takers Sector-wide net income Net income of farm business establishments Farm business income Income earned by farm households Money income of farm households from nonfarm and farm sources Income available for household use Farm household’s disposable (after tax) income ARMS data on farm income are an important contributor to the national income and product accounts. In fact, the need to populate the national accounts has driven decisions on the content of the ARMS program over the years. The content of the questionnaires has been written to generate data consistent with the accounting requirements of the national ac-counts.15 The data collected for this purpose include production expenses, imputed rents, sources of farm-related earnings other than the value of production and receipts, capital items such as farm office equipment, and data by the organizational structure of farms that can be used to distribute wages and other items by legal form of business organization. Beyond the content of questionnaires asking explicit questions for use in the farm income accounts, ARMS produces a farm-level estimate of value-added that is as consistent as possible with sector-wide measures of net value-added and farm income. Weighted estimates of firm-level value-added can be compared with sector-wide estimates produced from multiple sources of data as a consistency check. The firm-level data can be disaggregated to show where value-added and net income were generated by type and size of operation and geographic area. Other means of comparison are available. Data edit, analysis, and calibration programs have been written that enhance the ability to identify and 15 The questions that are used for the national accounts include Sections A, B, and C of Phase III (data on the whole picture of the farm), Sections D and E (income flows), Section F (expenses), and Section G (management). Sections I and J are used to fill in areas of farm and operator characteristics.
OCR for page 45
Understanding American Agriculture: Challenges for the Agricultural Resource Management Survey check potentially problematic responses while the survey is ongoing. The calibration routines result in ARMS data matching official USDA estimates for farm numbers by size of operation and for crop acreages and poultry and livestock on farms. ARMS has been expanded to generate directly observed estimates of farm expenditures for the 15 largest states as measured by volume of farm sales—called the 15-state oversample. For those states (Arkansas, California, Florida, Georgia, Illinois, Indiana, Iowa, Kansas, Minnesota, Missouri, Nebraska, North Carolina, Texas, Washington, and Wisconsin), directly observed annual data can be used in establishing estimates that feed into the national accounts. Closely associated with issues of farm income are issues of the well-being of farm families. Farm household well-being, however, is a more comprehensive measure than income, involving off-farm income, which depends on family composition, wealth, labor force participation, and other factors. The concepts and measures used to determine well-being were designed for comparability with Census of Population and other major survey definitions. The primary measure of household income is money after-tax income. In comparing the farm household measures from ARMS with those calculated from Current Population Survey, the two areas of comparison are population and the actual income measures. Since 50 percent of areas with farms are not in the CPS (primary sampling unit) sample, there are issues of sampling in these comparisons. In addition to comparisons with CPS data, ARMS measures have been compared with the Survey of Consumer Finances and Internal Revenue Service (IRS) data. The comparison with IRS is complicated by the difference in populations covered, the conceptually different income measures, and the fact that IRS has documented substantial underreporting of farm income on tax records. There are several peculiarities in the ARMS data related to measuring facets of well-being. Some of the survey questions require considerable recall; for example, the questions on household living expenses ask respondents to recall, in the winter of the reporting year, expenses incurred over the prior year. The point-in-time reporting of expenses may result in misleading measures of facets of well-being. Expenses are episodic, subject to seasonal fluctuations, so annual estimates cannot be calculated by multiplying the reference week by 52. These issues are grist for consideration in a formal methodological research and development program, which we recommend in Chapter 3. MAINTAINING RELEVANCE BY DEVELOPING NEW DATA SOURCES The above discussion indicates the value of collecting conceptually correct data for individual farms and households in order to understand the structure of returns in U.S. agriculture. The issue of returns is illuminated
OCR for page 46
Understanding American Agriculture: Challenges for the Agricultural Resource Management Survey by the currently integrated ARMS program. The success of its pioneering integration of data collection about farms and farmers prompts exploration of further opportunities for integrated data collection. Conservation Effects and Assessment Project A recent attempt to produce an integrated database in order to quantify the environmental effects and benefits of conservation programs is the Conservation Effects Assessment Project (CEAP), sponsored by the National Resources Conservation Service. CEAP has two main components: (1) watershed case studies; and (2) national assessments of cropland, grazing lands, wetlands, and wildlife. The goal of the program is to provide farming communities, the public, legislators, and other stakeholders with an interest in environmental policy issues with an accounting of the environmental benefits obtained from these conservation programs. By design, the CEAP cropland survey borrows heavily from the ARMS program. To conduct the survey, a subsample of NRI points, as described earlier in this chapter, is selected from points that are classified as cultivated cropland in the most recent annual NRI survey. These points numbered 10,000 in 2003, 2004, and 2005 and were cut to 6,000 in 2006. The NASS interviewers fan out to interview operators of the farm or field associated with NRI sample point. The CEAP database is enriched with the addition of USDA Field Office records, which are used to identify program participation and practice application.16 This is a unique aspect of CEAP, in that these records are not currently collected in the ARMS program. The CEAP farmer survey overlaps considerably with Phase II of ARMS and could be considered in competition with that survey operation. Like the ARMS Phase II survey, it is administered late in the year, after harvest, by NASS. It asks questions going back three years on cropping patterns/ rotations, field characteristics, double cropping, cover crops, and soil test results; applications of fertilizer, manure, pesticides; timing and equipment used for all field operations; tillage, cultivation, chemical application, and harvest; conservation practices associated with the field; from the respondents and the Field Offices, conservation plan and program participation; general characteristics of farming operation and the operator; irrigation practices; and some subjective information on wildlife. While NRCS plans to continue the analytical aspect of the CEAP with the data generated through the program, no further CEAP survey has been planned at the time of this writing. A decision has been made to forgo the farmer surveys in 2007 and to look at the potential of using NRI, CEAP, weather, and other data in an integrated fashion to paint a picture of the 16 Presentation by Jeff Goebel, September 28, 2006.
OCR for page 47
Understanding American Agriculture: Challenges for the Agricultural Resource Management Survey environment and its impact on soil quality and water quality and quantity, as well as on-site and off-site environmental impacts; the idea is that this approach would become a standard part of the NRI survey process after 2007 for grazing lands, wetlands, and cropland. The “time-out” affords an opportunity for considering whether it makes sense for NRCS, NASS, and ERS to work on an integrated approach for surveying agricultural lands. In the panel’s view, there are serious unanswered questions about survey methodology and cost-effectiveness that need to be answered before a decision can be made that some integrated effort should be launched. Recommendation 2.1: NRCS, NASS, and ERS should engage in a focused research and testing program and use experience with integrating the Conservation Effects Assessment Project and ARMS to assess the feasibility of integrating ARMS with other surveys and data sources. Longitudinal Data from ARMS In 2006, ERS and NASS submitted a proposal for funding to begin to develop the Agricultural and Rural Development Information System (ARDIS). ARDIS has several objectives, one of which is to establish and maintain data collection on the demographic characteristics, employment, and income sources of rural households over time. ARDIS would collect information on participation in farm, rural development, and other USDA and federal programs, enriching it with household well-being information, in a new longitudinal survey of nonfarm rural households and rural-based farm households. The agencies envisioned that ARDIS would allow researchers to isolate the effects of rural development, farm conservation, and marketing programs from one another and from the myriad other forces affecting the economic well-being of farm and rural households. The close interaction among farm and nonfarm rural households is critical to understanding how rural America adjusts to changing economic circumstances or policy over time. For example, farm operators who report receipt of government payments in one year could be followed to determine the effect on management practices and income in subsequent time periods, as a means of evaluating the effectiveness of these programs at the individual farm level. The unique, previously nonexistent, longitudinal ARDIS data would help discern these patterns of adjustment. As proposed, ARDIS would be a simultaneous collection of ARMS data and information from separate panels of farm and nonfarm households over time in the same rural area. The goal is to gauge how farm and nonfarm rural households respond differently to economic change and to enhance the ability to assess the impact of farm policy on the rural economy. The
OCR for page 48
Understanding American Agriculture: Challenges for the Agricultural Resource Management Survey new program would track critical indicators over time with special-focus modules to address specific, emerging policy issues; monitor farm and rural household adjustments to changes in policy, prices, and technologies by building on ARMS; and enhance research capacity to analyze, interpret, and apply new agricultural and rural development information. At the time this report was prepared, the funding for development and implementation for ARDIS was not available to the agencies. This initiative bears watching as it would provide a unique data source for enriching analysis of the interplay between programs and performance results. Recommendation 2.2: In preparation for funds becoming available for a longitudinal design of ARMS, ERS and NASS should systematically conduct research and explore the need for and feasibility of obtaining panel data from ARMS. Furthermore, as a test of the power of such information, more use should be made of the existing longitudinal microdata from the Census of Agriculture. One possible approach would be to create a pseudopanel of such data. Another would be to make a retrospective link between the Census of Agriculture and a year of ARMS.
Representative terms from entire chapter: