In this chapter, we identify and clarify definitions and concepts used in the measurement of the farm economy and propose alternatives that may be more useful for measuring the activities of complex operations. The material here bridges the discussions of complexity in the previous chapter and proposals for the statistical framework in the next chapter.
Observing the complexity of modern farms raises the basic question: What makes a farm a farm? It is essential to address this question before turning to others, such as what a complex farm structure is and how such operations should be dealt with in statistics, research, and policy making.
In the United States, even children who have never seen a farming operation firsthand develop an intuitive notion of what a farm is. Their toys are likely to include plastic cows, sheep, horses, and tractors. They are taught songs like “Old MacDonald.” And with the advent of electronic gaming devices and smartphones, it is now possible for anyone to “run” a virtual farm. These experiences tend to reinforce the idea that the typical farm is an entity that perfectly overlaps the activities of a farmer who centrally manages a diverse set of production activities.
In the contemporary economy, however, this heuristic of the farm has become outdated as more complex organizational forms have emerged and become commonplace. For example, some traditional farming activities, like harvesting crops, are now often carried out by hiring specialized companies or using workers who are employed not by the farm but
rather by an employment agency. Some farms are parts of vertically integrated businesses, such as feedlots attached to slaughterhouses or vineyards attached to wineries. And some farms derive income from related activities that are not necessarily farming, such as agritourism, food processing, or energy production.
All of these examples create problems for statisticians the moment they want to classify a company (or operation, business, firm, or holding—for now, we use these terms interchangeably) as either a farm or another type of business. Before turning to that classification question, we therefore address an even more basic question: What is farming?
The Merriam-Webster Dictionary defines farming as “the practice of agriculture or aquaculture,” providing a synonym to be discussed later, in section 4.5. The Oxford English Dictionary (second edition, 1989) defines farming as “the activity or business of growing crops and raising livestock.” This latter day-to-day definition seems fairly close to what social scientists use in some of the disciplines relevant to this study. For instance, the International Accounting Standards Board (IASB)1 describes agriculture (an imperfect synonym of farming) as follows: “Agricultural activity is the management by an entity of the biological transformation and harvest of biological assets for sale or for conversion into agricultural produce or into additional biological assets” (International Financial Reporting Standards Foundation, 2017). The essence of this definition is of a business activity that manages a biological process that leads to either products (such as milk, potatoes, or oranges) or biological means of production (such as animals or seeds).
Definitions Current at Major Statistical Agencies
International statisticians rely on a definition similar to that used by the IASB. Both the Statistical Classification of Economic Activities in the European Community (NACE)2 and the international integrated system of economic classifications (ISIC) managed by the UN Statistical Commission (UNSTAT) group agriculture (again, used as a synonym of farming) together with fisheries and forestry in one category for national accounting purposes. This broad category is defined as “the exploitation of vegetal
1 The IASB is an independent, private-sector body that develops and approves International Financial Reporting Standards (IFRS). The IASB operates under the oversight of the IFRS Foundation. The IFRS Foundation is a not-for-profit public interest organization established to develop a single set of high-quality, understandable, enforceable and globally accepted accounting standards—IFRS Standards—and to promote and facilitate adoption of the standards.
2 Commonly referred to as NACE, based in the French term, “nomenclature statistique des activités économiques dans la Communauté européenne.”
and animal natural resources, comprising the activities of growing of crops, raising and breeding of animals, harvesting of timber and other plants, animals or animal products from a farm or their natural habitats.”3 Note that the addition of natural habitats includes fisheries, hunting, and forestry in the category.
Likewise, the North American Industrial Classification System (NAICS) groups farming together with fisheries and forestry activities:
The Agriculture, Forestry, Fishing and Hunting sector comprises establishments primarily engaged in growing crops, raising animals, harvesting timber, and harvesting fish and other animals from a farm, ranch, or their natural habitats.
The establishments in this sector are often described as farms, ranches, dairies, greenhouses, nurseries, orchards, or hatcheries. A farm may consist of a single tract of land or a number of separate tracts which may be held under different tenures. For example, one tract may be owned by the farm operator and another rented. It may be operated by the operator alone or with the assistance of members of the household or hired employees, or it may be operated by a partnership, corporation, or other type of organization. When a landowner has one or more tenants, renters, croppers, or managers, the land operated by each is considered a farm.4
The sector distinguishes two basic activities: agricultural production and agricultural support activities. Agricultural production includes establishments performing the complete farm or ranch operation, such as farm owner-operators and tenant farm operators. Agricultural support activities include establishments that perform one or more activities associated with farm operation, such as soil preparation, planting, harvesting, and management, on a contract or fee basis.
Excluded from the Agriculture, Forestry, Fishing and Hunting sector are establishments primarily engaged in agricultural research and establishments primarily engaged in administering programs for regulating and conserving land, mineral, wildlife, and forest use. These establishments are classified in Industry 54171, Research and Development in the Physical,
4 Conceptually, this panel agrees with the NAICS position that households that own land and rent to operators should not be considered farm businesses. If a household (or any other sector of the economy, such as government or a business) is engaged only in renting the land, then it should be classified as part of the rental and leasing industry. The farming activity undertaken on such rented land would be measured by the establishment/enterprise using the land, that is, the lessor, not the lessee. Practically speaking, if the lessor and lessee of the land were both considered to be farms, then there would be a risk of double counting.
Engineering, and Life Sciences; and Industry 92412, Administration of Conservation Programs, respectively (North American Industrial Classification System, 2017).
Agricultural activities as classified by NAICS are listed in Annex 4.1, together with forestry, fishing, and hunting activities. Within this activities listing, Codes 111 (Crop production) and 112 (Animal production and aquaculture) are farming activities. Codes 1151 and 1152 include support activities for agriculture, covering the important set of nonfarm businesses that serve as potential substitutes for direct management and operation of production activities by farmers. Indeed, these businesses introduce two possible sources of statistical complexity: some farms purchase agricultural production services from other entities and some farms sell their production services to other farms.
Neither the National Agricultural Statistics Service (NASS) nor the Economic Research Service (ERS) offers explicit definitions of farming.5 However, the Census of Agriculture questionnaire follows an enumeration that is similar to NAICS’s list of “agricultural activities.” The census includes questions on field crops (Section 6); hay and forage crops (Section 7); cut Christmas trees and maple syrup (Section 8); nursery, greenhouse, and floriculture (Section 9); vegetables and melons (Section 10); fruits and nuts (Section 11); berries (Section 12); cattle and calves (Section 13); hogs and pigs (Section 14); equine (Section 15); sheep and goats (Section 16); aquaculture (Section 17); poultry (Section 18); apiculture (Section 19); and other livestock (Section 20) (National Agricultural Statistics Service, 2014).
This emphasis on the management of biological processes in crops and livestock makes it clear that farming is not defined on the basis of the purpose to which products are put: it is not necessary to produce food or animal feed. Biological processes are also managed to produce fiber (flax, cotton, and wool), flowers, fur, pharmaceutical products, and fuel, to give just a few examples. However, there are some activities based on biological processes that are classified not as farming but as industrial—such as the industrial production of yeast or the use of microbes in sewage sludge.
Rather than describing activities on the basis of the purpose of the products or even on the basis of the products themselves, the NAICS objective is to describe a certain production function associated with the way a process creates outputs (products or services) from inputs. It is not a product classification and so differs from the National Income and Product Accounts, which is. When biomass is produced and transformed into ethanol or electricity, this includes a farming activity, very often as a primary
activity, as well as a manufacturing activity, in this case energy production that turns biomass into ethanol.
In some cases, the production activity is very different from farming, even if plants or animals are used. Examples include petting farms, dude ranches, or farms offering yoga with goats. Although these operations may have been conventional farms in the past, focused strictly on crop or livestock production, they do not necessarily fit farm definitions any longer, even if their animals are registered for animal health reasons.
A side remark is that statisticians employ a variety of approaches for dealing with “gray” or illegal activities, for example, when the goal is to assess the total size of an economy. The United Kingdom has added income from illegal activities, such as sex work and illegal drug sales, to its gross domestic product (GDP) calculations to conform with reporting rules from the European Union. In Canada, now that recreational marijuana use will soon become legal, Statistics Canada is preparing to add estimates of the plant’s production and sale to assess its economic impact.6 The U.S. Bureau of Economic Analysis, which produces GDP estimates, only includes legal activities, so in the United States marijuana growing would only be counted as a farming activity in states, such as Washington and Colorado, where marijuana production, sale, and use are currently legal.
These examples also show that to operationalize the concept of farming as the management of a biological process, an agreed-upon list of activities is needed that can determine the exact borders where farming stops and nonfarm production and services begin (or, as we argue in section 4.5, where it may make sense to introduce a new category of agriculture that lies between farming and industry). Such a list of activities could make clear that a tree nursery and an energy plantation (where miscanthus or willows are grown, for example) are still farming, while a forest is not, even if it undergoes some pruning. Fish farming (aquaculture) is considered an agricultural activity, as is the growing of seaweed (algae),7 but the capture of wild fish from the ocean and harvesting of kelp from the open ocean are not.
Defining Agricultural Support Activities
Essential to this definition of farming is that biological products must be managed. Simply harvesting (as a contractor) is not farming if it is done
7 NAICS defines algae harvesting as an agricultural activity (112519).
as a service but is rather an agricultural support activity.8 The less technical Organisation for Economic Co-operation and Development (OECD) definition is “the activity or business of growing crops and raising livestock,” one could add “and their preparation for the primary market” to make clear that harvesting, storing, and packaging are included.
Another example of agricultural-related activities that has posed problems for classification purposes is the making of wine and cheese. Are the biological processes in cheese making and wine making characteristic enough to call them farming activities instead of industrial activities? NACE has tried to solve this dilemma by introducing the idea of a primary market of a farm:
Agricultural activities exclude any subsequent processing of the agricultural products classified under divisions Manufacture of food products and beverages and Manufacture of tobacco products, beyond that needed to prepare them for the primary markets. The preparation of products for the primary markets is included here (that is in farming). The division excludes field construction (e.g. agricultural land terracing, drainage, preparing rice paddies etc.) classified in the section Construction, and buyers and cooperative associations engaged in the marketing of farm products, classified in the section Wholesale and retail. Also excluded is landscape care and maintenance, which is classified in the class Landscape service activities [text slightly altered for readability by deleting codes].9
In agriculture, one frequent situation where the breakdown of the value added presents difficulties is when the unit produces grapes and manufactures wine from the own-produced grapes, or when it produces olives and manufactures oil from the own-produced olives. In these cases [. . .] these vertically integrated activities would generally lead to classification of the units under agriculture (Federal Statistical Office, 2008, p. 18).
In other words, by NACE standards, wine and cheese making are not part of farming but of manufacturing. However, because these activities are sometimes not fully separable from the farming activities, for a given entity they may in practice be counted as farming. As discussed in Chapter 3, this is one characteristic that makes farms more complex, because they have the
8 However, custom work performed by a farmer is a category of income on the Schedule F federal tax form that farmers complete. There are many farms that engage in such work, and often they do not have a separate business for it; it is just lumped under their farming income, as the equipment that they utilize is first and foremost for their farming purposes.
management skill and technical capabilities to run a significant value-added component as part of their farming business.
Under code 115, NACE also includes in its definition of agriculture activities incidental to agricultural production and activities similar to agriculture but not undertaken for production purposes (in the sense of harvesting agricultural products) that are done on a fee or contract basis. Also included are post-harvest crop activities aimed at preparing agricultural products for the primary market, such as contract work for sorting or grading products or packaging them. Such contract work may be viewed as an agricultural support activity, and this is one of the differences between farming (this section) and agriculture (see section 4.5). Companies specializing in such activities are agricultural support firms that perform farming activities on farms. They are engaged in a farming activity—that is, farming (managing the whole biological process from inputs to outputs)—which is different from being a farm.
To summarize, farming is the characteristic activity that takes place on a farm, and typically it involves the management of a biological process, such as growing crops or raising livestock, for the purpose of harvesting products or reproducing a biological means of production. A list of activities and products such as those included in NAICS industry codes 111 and 112 is useful for precisely delineating between farming, agriculture, and manufacturing activities. Conceptualizing farming activities in this way does not imply a change in data collection by NASS and ERS, but it may help both agencies clarify and separate discussions about complex farms as businesses (farms), locations (farms, fields), and activities.
Having defined farming, or at least described it, the next step is to define what a farm is. In principle, this is the entity that carries out farming, whether it is a firm, business, holding, or operation. Applying this principle to the modern economy, however, turns out to be less straightforward than one might perhaps expect.
The Merriam-Webster Dictionary defines a farm as “a tract of land devoted to agricultural purposes;” the Oxford English Dictionary widens this to “an area of land and its buildings used for growing crops and rearing animals.” However, as argued above, farming does not necessarily demand either land or buildings: pigs and poultry farms may consist only of buildings with no crop production present, but they are still farms; as are greenhouses and buildings used to grow chicory or flowers from roots and bulbs, sometimes cultivating their crops in a substrate instead of soil.
And the vertical farms10 which are the latest development in urban farming are certainly farms. The more scientific definitions, such as those used by statistical agencies, use terms, such as “entities,” “businesses,” “holdings,” and “units,” that engage in farming activities.
Examining the Principal Activities of a Farm Business
More importantly, the organizational forms that farming activities can take create definitional problems. Statistical agencies that count and describe farms must classify organizations instead of activities; businesses receive a census form and report multiple activities whose operating costs often cannot be disentangled. This is not a problem when businesses are fully specialized in farming, narrowly speaking, and have no other activities, whether agricultural, industrial, retail, or service.
Even in these simpler cases, however, the results of such a classification may raise questions if farms have moved production activities that were once commonly carried out on the farm to other producers downstream in the production pipeline. Historically, for example, cheese and butter making were farm activities but, beginning during the industrial revolution of the 19th century, such activities were increasingly transferred to the food industry. On the input side, support tasks involving contract work or machines, advice from risk-management firms, and work now done by other specialized companies either on the farm or elsewhere are all examples of activities that have displaced comparable ones that were once commonly handled “on the farm.” This shift toward specialization means that functions previously classified as farming are now more typically classified in industrial sectors. Therefore, as captured by economic statistics, the farm sector has contracted, relatively, while industry and service sectors have grown. This evolution suggests that, for some purposes, it may make sense to view agriculture as a sector that is broader than farming (see section 4.5).
More problematic from a classification perspective is the low level or even lack of specialization in mixed enterprises, such as mixed farms and integrated companies. For entities engaged in several activities, some of them farming and some of them not, there is a need for a rule to classify them as belonging to a certain sector or industry (a class of activities). The
10 Vertical farming is the practice of growing food or medicine in vertically stacked layers, vertically inclined surfaces, or integrated in other structures. The modern idea of vertical farming uses Controlled Environment Agriculture technology, where all environmental factors can be controlled. These facilities utilize artificial control of light, environmental control (humidity, temperature, gases) and fertigation. Some vertical farms make use of techniques similar to greenhouses, where natural sunlight can be augmented with artificial lighting. The Association of Vertical Farming has its own typology; see https://vertical-farming.net/vertical-farming/integration-typology.
horticulturalist who occasionally sells Christmas trees at the roadside is still a farm and not a retailer, but conceptually we enter a fuzzy area where, at the opposite end of the spectrum, there might be a retail garden center that also supplies a few percent of its sales in December from self-grown Christmas trees (a farming activity). Such examples indicate the need for a criterion and a threshold whereby entities may be classified by sector.
In macroeconomic statistics, guided by such frameworks as the NAICS or ISIC/NACE, it is recognized that a unit may perform one or more economic activities described in one or more categories. According to the OECD Manual on Business Demography Statistics:
In such cases, the principal activity of a statistical unit is the activity, which contributes most to the total value added of that unit. The principle activity is identified according a top-down method and does not necessarily account for 50% or more of the unit’s total value added. A secondary activity is any other activity of the unit, whose outputs are goods or services, which are suitable for delivery to third parties. The value added of a secondary activity must be less than that of the principal activity. A distinction should be made between principal and secondary activities, on the one hand, and ancillary activities, on the other. Principal and secondary activities are generally carried out with the support of a number of ancillary activities, such as accounting, transportation, storage, purchasing, sales promotion, repair and maintenance, etc. Thus, ancillary activities are those that exist solely to support the principal or secondary economic activities of a unit, by providing goods or services for the use of that unit only.11
By applying such a rule and threshold, mixed enterprises such as the Christmas tree sellers described above can be classified as farms (if the tree nursery is the principal activity) or as retail (if the tree nursery is a secondary activity). Adoption of such a method would bring agricultural statistics in line with statistics covering other sectors of the economy, which the panel finds to be an important consideration guiding decisions on the agricultural statistical framework. The value of statistical agencies taking this approach would be generated because the modern agricultural sector is an integral element of the economy, with many medium-sized businesses and much linkage with the rest of the economy. A status aparte in statistical methods is therefore to be prevented as much as possible.
RECOMMENDATION 4.1: In line with statistics for other parts of the economy for classifying a business as a farm or as an entity operating in a nonfarming sector with secondary activities in farming, the
National Agricultural Statistics Service and the Economic Research Service should apply clear rules based on the nature of the business’s principal productive activities.
This recommendation does not imply that only entities classified as farms with farming as a primary activity are of interest to NASS and ERS. On the contrary, the agencies should be interested in all businesses engaging in farm activities, even those for which it is a minority activity. Counting farming activities in businesses that are classified in sectors other than farming is necessary to accurately and consistently estimate totals for the sector when there is a change in the role of large companies in farming. There are good reasons to try to survey agricultural production that takes place within these companies, especially if a large share of the production takes place in them. This may be appropriate because, among other reasons, such mixed production businesses are affected by agricultural policies (see Chapter 2). For reporting on entities engaged in farming as a secondary or tertiary activity, specific classification categories may be needed, such as part-time farms and multi-functional farms. At the sector or macro level, the size of agricultural business complexes can be calculated with input-output tables (see section 4.5).
Applying this method of classification requires careful specification of how entities or units are defined. For a company like Smithfield Foods Inc.,12 which owns both slaughterhouses and farms, should the farms be classed as different units, or should the full multinational company be classified under its principal activity as a slaughterhouse and therefore part of the food industry? Are Walmart stores individual units or is all corporate activity one business? The answers to such questions—presented in detail in Chapter 5—determine the number of farms counted in an agricultural census.
Distinguishing Between Firms and Establishments
In economic statistics, the answer to the above-posed question is dictated by the difference between firms and establishments. A firm is “an organization conducting a business . . . . A firm may operate one place of business or more.” An establishment is “a single physical location where a firm’s business is conducted” (National Research Council, 2007).13 For our
12 Smithfield Foods, Inc., is a meat processing company and wholly owned subsidiary of the WH Group.
13 In line with Bureau of Economic Analysis’s November 2017 update to its handbook Concepts and Methods of the U.S. National Income and Product Accounts: “Companies consist of one or more establishments owned by the same legal entity or group of affiliated entities. Establishments are economic units, generally at a single physical location, where business is conducted or where services or industrial operations are performed (e.g., a factory, mill, store, hotel, movie theater, mine, farm, airline terminal, sales office, warehouse, or central administrative office). Establishments are classified into an industry on the basis of their principal production method, and companies are classified into an industry on the basis of the principal industry of all their establishments.”
example, this means that Walmart is a firm with many establishments; and that the farms of Smithfield Foods can be counted as farms, if we define a farm as an establishment and Smithfield Foods is organized in such a way that data from the farms can be separated from other activities.
Eurostat has followed this accounting approach, stating “A farm is a single unit, both technically and economically, which has single management and which produces agricultural products . . . either as its primary or secondary activity.”14
As covered in Chapter 2, the official definition of a farm used by NASS is as “any place15 from which $1,000 or more of agricultural products were produced and sold, or normally would have been sold, during the year.”16 In line with the day-to-day use of the word “farm” and the current practice in NASS and Eurostat, a farm should be defined in a way that focuses on the productive entity as a business engaged in clearly specified types of activities.
RECOMMENDATION 4.2: For conceptual purposes, the National Agricultural Statistics Service and the Economic Research Service should define a farm as an establishment (single unit with a legal or informal management structure) that (1) has its principal or secondary activity in farming with the production of agricultural products and biological assets such as seeds and animals; and (2) for which full economic data on key business variables, such as costs and revenues, can be collected and made available.
This recommendation is intended to help NASS and ERS unravel the struc-
14 From Eurostat’s online Glossary, under “Agricultural Holding,” see http://ec.europa.eu/eurostat/statistics-explained/index.php/Glossary:Agricultural_holding. The online text further states: “An agricultural holding, or holding or farm is a single unit, both technically and economically, operating under a single management and which undertakes agricultural activities within the economic territory of the European Union, either as its primary or secondary activity. Other supplementary (non-agricultural) products and services may also be provided by the holding.”
15 USDA uses “place” in a nonstandard way. A “place” in its usage does not imply contiguous land parcels where ownership and management overlap.
16 For the USDA/ERS Glossary, see https://www.ers.usda.gov/topics/farm-economy/farmhousehold-well-being/glossary, which covers the following terms: farm; farm operator and principal farm operator; family farm; farm operator household; farm operator household income; farm operator household wealth; farm typology; commodity specialization; disposable personal income of farm and nonfarm residents.
ture of some common types of complex holdings. When a person or businesses has two clearly distinguishable farms with separate accounts, such as one farm in Wyoming and one in California, or one on either side of the same county, the farms should be treated as two farms that happen to have the same owner. This definition is quite close to the USDA’s approach, which defines a farm as a management unit.
Regarding the practicalities of implementing Recommendation 4.2, the panel recognizes that there may be scant political will to change the definition of a farm, regardless of how antiquated the dollar threshold may be, due to the implications for federal funding and other reasons. Indeed, the transfer in 1997 of the data collection for and publication of the Census of Agriculture from the Census Bureau to NASS resulted, in part, from congressional concerns about a proposal to modify the definition of a farm.
Discussions of the politicized aspect of the farm definition often take place in the context of proposals to increase the farm-size threshold (e.g., O’Donoghue et al., 2011), which would invariably reduce the number of farms. The recommendation above, which is expanded on in Chapter 5, is quite different: it recommends that NASS and ERS use a farm enterprise concept and a farm establishment concept when collecting data in order to (i) make the concepts as consistent as possible with those used by other statistical agencies, (ii) provide clarity to survey respondents, and (iii) identify business units that are likely to correspond better with the way respondents organize their own data.
The establishment concept is not fully equivalent to the current farm concept used by USDA to estimate the number of farms, but it probably comes close. The majority of farms, especially the smaller ones, have a simple structure and will report their current operations as establishments, resulting in the same number of farms. For complex farms, it is currently unclear whether they would report one of their establishments or report a complex farm corresponding to a statistical enterprise. If they were to report the latter, the number of farms (farm establishments) would increase; otherwise it would stay unchanged. Recommendation 4.2 should therefore be interpreted as a recommendation to help obtain a clearer picture of the farm structure and reduce the administrative burden for complex farm structures, and not as a recommendation that intends to reduce the number of farms.
However, if administrative or political reasons required that there be a check with the traditional farm-count number, ERS/NASS could develop a mapping from enterprises and establishments to farms. It could then provide counts at the county, state, and national levels that would be consistent with those that have historically been produced. Such a mapping could be done to preserve the distribution of farms across space or in total in a given base year.
Counting and Measuring Farms and Fields for a Complex Farm Business
The definition of a farm as an establishment means that there are businesses that may be involved in two or more farms. (The logic of this is elaborated on in Chapter 5, where it is also clarified why in statistical terms businesses are equated with firms.) Even defined in this comparatively granular way, a single farm may engage in several different activities, such as growing different crops or raising different types of animals or doing both. A cropping farm will have different fields that are not necessarily adjacent to each other, and so, as noted above, the “place” that defines a farm need not be contiguous. Likewise, an operation that raises livestock or fish does not necessarily have all of its barns, ponds, and so forth on the same site. A farm may engage in relatively minor activities in other industries, such as providing support services to agriculture or being involved in retail. Farmers sometimes use the terminology “different enterprises,” if they produce more than one type of crop or livestock within the farming operation; for example, corn, soybeans, and a herd of beef cows or calves could each be classified as a separate enterprise.17
Another issue is that a farm, as defined above, is an economic concept. That is, a farm might be organized as a complex legal structure for reasons of tax, inheritance, or otherwise. For instance, a father and son might run a farm wherein the former owns most of the land while both own the machinery together. Or some of the land might be part of a family trust. In any case, the use of these different legal entities does not necessarily change the status of a farm from a single farm to more than one farm.
Defining the Farm as an Establishment
Defining a farm as an establishment means that, as measured by a statistical agency, the number of farms as well as the size of the average farm both depend on how farmers organize their activities. The farmer who buys out a neighbor, adds that land to his own property, and then rents out the purchased farm building or uses it for his contract work operations or for an agritourism business is enlarging his establishment. He (or she) still has one farm, though he also may now have other income sources or may own some other establishments, such as the agritourism business. But, if a similar farmer also buys out a neighbor but instead uses the farm buildings for another farm activity—such as when pig breeding continues on the original farm and hog finishing occurs on the recently acquired farm—proper categorization then starts to depend on the way the business is organized.
17 As in general statistics, the term “enterprise” is used for the level of units (firms) that include several establishments; this report does not use the term.
In the second scenario, if the two farms are or can be separated in a technical and economic sense, such as by having different management accounts each with its own profit-and-loss account and balance sheet, then there are two establishments. This conceptualization is in line with the day-to-day language: the farmer owns two farms. This can of course easily become a source of complexity in data collection if statistical agencies view such a situation as two establishments even though the farmer manages them as one—or the other way around. The solution is to take the farmer’s reality as a point of departure, even if this reality is mainly shaped by the history of the farm or by the legal, fiscal, or risk-management considerations of the farm.
Giving farmers clear guidance concerning the fact that a farm is an establishment and that the data collection aligns with how they see and organize their own farms is important for reducing confusion among respondents about how to report. This implies that NASS needs to be clear to farmers about what “a place” means in its definition of a farm. Where the farm business’s establishment and legal entity structures align, opportunities for exploiting linkages with administrative (e.g., tax) data will exist.
The Census Bureau faces a similar challenge, although on a much larger scale, when it measures the number of “businesses” in the U.S. economy overall. Although it never formally defines what a business is in its technical documentation, the bureau nevertheless reports statistics on businesses in the U.S. economy. This is possible only because it uses the word “business” (synonymous with “company” and “firm”) to simplify its presentation of the more nuanced concept of an “institutional unit,” such as an establishment or firm, which it does describe in its technical documentation. An institutional unit exercises decision-making control and has rights to residual profit; it can be a single-location corporation or LLC, or a multilocation (perhaps multinational) conglomerate with many subsidiary units, each operating under a distinctive legal form of ownership.
The Census Bureau attempts to measure some of this organizational complexity using a mix of administrative and survey data, but of course it cannot capture it all. Two elements of the Census Bureau’s approach that may offer helpful guidance for NASS are (i) the use of a “company organization” instrument to identify and link related reporting units prior to conducting its census enumeration and mailing; and (ii) the sampling of smaller units to free resources for more intensive focus on larger units. We elaborate on this in Chapter 5 in the context of data collection, where it becomes crucial to distinguish between entities from which data are being collected and entities in the statistical structure that we want to measure. The latter should be well defined and set in stone; in contrast, the collection entity may have to conform more to the availability of data as maintained by the businesses.
Refining terms in this way can have an influence on the number of farms that are reported and therefore also on the average size of farms, although it is not clear in which direction. As argued above, most likely the number of farms would increase and the average size would decrease, since this approach invites complex holdings to report separately for each of their component establishments. In judging this advice, the agencies should consider the fact that statistics will also have to report on the number of farming businesses that own more than one farm (see below, and Chapter 5).
Defining the farm as an establishment would be an important step toward improving data collection on complex holdings. It would imply that some farmers own, operate, or are the residual claimants of more than one farm, instances that indicate a type of complex farm holding, that is, management structures that extend over a combination of establishments. Grouping multiple establishments under a single management unit for the purposes of data collection and reporting is a common feature of measuring business activity outside the agricultural sector: as mentioned earlier, Walmart is a single firm (business), but the different stores it owns are separate establishments.
If the farm is defined as an establishment, research into the organization of farming, agriculture, and the food chain would benefit from the use of a consistent definition of the firm in the agricultural context. This could be one business that owns several farms (establishments); a business that owns one farm as well other nonfarm establishments; or a business for which the farmer is involved in several farms that have complex ownership and managerial structures. Such situations can increase the complexity of a farming business, at least relative to the classical situation in which the establishment level and the business level are equivalent.
In Chapter 5, we discuss the value of implementing in the measurement infrastructure a definition of business entities based on shared management structures. This includes both cases in which one business owns and operates one establishment (a simple farm business) and cases in which one business owns and operates a group of establishments (a complex farm business). Using this type of definition would facilitate collecting and reporting statistics at both the “establishment” level and the “farm business” level. Chapter 5 discusses how this can be implemented.
Sharecropping and Noncommercial Farming
Having defined the terms farm and farm business, we end this section by discussing two phenomena that also complicate measurement and reporting. One is the existence of sharecropping, and the other is farming for purposes other than selling agricultural products.
Sharecropping is an institutional form of renting out land that has
engaged economists since Adam Smith. If a landlord leases land to a farm owner, that activity is an investment activity, which does not make that landlord (or a company like an LLC acting as a landlord) either a farmer or a farm business (simple or complex). However, if a share-contract is in place, the rent is not charged as a fixed-dollar value but rather is charged in the form of the products produced, that is, as a share of the crop. In this case, the landowner becomes a risk-bearing managing partner, and one could argue that the operation is a (separate) farm and part of a complex farming business.
However, in line with the Farm Service Agency’s payment eligibility criteria and the Internal Revenue Service (IRS) definition,18 if someone is receiving rental income or farm income, sharecropping arrangements should not been seen as separate farms unless the landowner is “actively engaged in farming.” Actively engaged here means that all participants, whether individuals or legal entities (such as partnerships or corporations), must provide significant contributions to the farming operation. These contributions may consist of capital, land, equipment, active personal labor, active personal management, or some combination of these. If it is a management contribution, it must be critical to the profitability of the farming operation, and the contributions must be at risk.
Another phenomenon that requires attention is the question whether a business has to be actively engaged in the sale of agricultural products to be considered a farm. For subsistence farms in developing countries, the answer is “no,” but in developed countries it is unusual to include hobby-activities for leisure, like gardening or unpaid housekeeping work, in official economic statistics. From an agricultural policy perspective, activities that are outside markets and not sensitive to agricultural policies are also not very interesting, even if some types of registration can be useful, such as registering all horses in a country for animal health reasons.
Therefore, for most purposes, it makes sense to define a farm as an establishment that at least has the objective to market its produce. In Europe, Eurostat uses a lower threshold of at least 1 hectare (2.5 acres), and includes farms with less than 1 hectare provided they market a certain proportion of their output or produce more than a specified amount (regionally specified in Euros); this latter threshold would normally exclude household gardens providing incidental sales. The Eurostat method guarantees that farms that do not sell products in a certain calendar year, for example due to extreme weather events or harvest loss or if they store their harvest to sell later, are still considered farms.
18 See IRS Publication 225, Farmer’s Tax Guide.
Now that the terms farming and farm have been defined and elements of their complexity revealed, we turn to the question, who is a farmer? This is a relevant question because researchers and policy makers are not only interested in the farm itself but also in the farmers behind that business. Understanding their decision making, for example regarding trade-offs between food production and environmental or other public aspects, is crucial, and monitoring their well-being and any income problems associated with farming is equally crucial.
Not all persons that work on a farm are farmers. Many farm workers are employees or are hired through an agricultural work agency or contractor, as in the horticulture sector. On some farms, family members provide a helping hand in busy periods or on weekends. In most cases, while all of these persons are active in farming, they are not typically treated as “the farmer” either in commonplace discussion or from a statistical agency perspective. According to the Merriam-Webster Dictionary, a farmer is “a person who cultivates land or crops or raises animals (such as livestock or fish).” The Oxford English Dictionary defines farmers as persons “who own or manage farms.”
Owners or Managers?
The Oxford English Dictionary definition exposes a quandary for statisticians: Are farmers owners or managers? Both categories are relevant to various research and policy questions. While they overlap and in many cases are the same person, on some farms they are different persons, creating additional complexity in data collection and interpretation.
Defining the farmer as the owner of the business entity and the person who makes management decisions (or appoints managers to do so) signals the importance of those who are responsible for the operation and who bear all the financial risks. They are the entrepreneurs, the residual claimants, in a view that sees the farm as a bundle of contracts (Allan and Lueck, 2002). Risk management is an important aspect of American farm policy, implying that this is an important dimension when identifying farmers.
Defining the farmer as the owner of a farm means that there will be some farmers (as owners of a farm) who do not work there and are absentee owners. It also means that there may be family members working occasionally on a farm who will be designated as owners for reasons of income tax, marriage, or with an eye to intergenerational transfer and inheritance taxes. This may be broad but at least it is a jurisdictionally precise criterion, and the distribution of profits would make clear who is and who is not an owner.
In principle, both categories—owners and decision makers—are of interest. They could be folded into one group under the term producer, a term that NASS introduced for 201719 to indicate any person involved in farm decision making (governance structure), from day-to-day decision making to the work of absentee owners who may only make investment decisions once a year. The term also includes hired managers.
Compared with ownership, there is less clarity with the term decision maker: Are these the persons who are involved in day-to-day operational decision making concerning, for example, spraying and environmental management?20 The term farm operator also hints at a level of active working and decision making. Or are these the persons who make the important strategic decisions on the choice of marketing channels or about investments and finances? Or are they the owners, while it is a farm manager who “runs the show”? From a user point of view, the appropriate designation of the farmer will depend very much on the type of (policy) research to be informed.
If a farm is a corporation, it could be argued that, from a legal perspective, the corporation is the farmer, and the officers of the corporation (the CEO and others) are the persons who express decisions on behalf of the business. In several ways, however, this seems not to be a very informative approach. A farmer or producer is always a person, even in the case of a corporation that does farming, where most likely at least the CEO is the farmer/producer.
What if a Farm Has Multiple Farmers?
It follows from the analysis above that a farm may be associated with more than one farmer, and similarly with more than one manager and more than one owner. It makes sense that NASS and ERS are now asking respondents to the Census of Agricultural to identify multiple producers at a farm and farming businesses, where the farm has more than one producer. In the 2017 Census of Agriculture and Agricultural Resource Management Survey (ARMS), respondents may list multiple men and women who have been engaged in making decisions for the operation; they are then asked to
19 See Publication of Agriculture Census Data on Farm Operator Demographics, a report by the National Institute of Statistical Sciences Technical Expert Panel, October 12, 2017. There it was recommended to replace the label “Operator” with “Producer” in all publications. The 2017 Census of Agriculture and future censuses use these terms: “All Producers,” “Principal Producers,” “non-Principal Producers.” These terms span the breadth of agriculture and are seen as consistent with current terminology used by producers and by professional agriculture organizations.
20 The 2017 Census of Agriculture and ARMS were broadened to capture farm decision making beyond the day-to-day.
provide more detailed information on up to four people per farm. Of those four people, the respondent may choose to identify one or more of them as a principal operator or senior partner on the census, which is different from previous census surveys, which only allowed one principal operator to be identified. However, the term principal operator is being used in the 2017 census for bridging purposes only and will cease to exist in the future. The ARMS survey continues to ask for a single principal operator to be identified for the farm household.
This raises the question whether or not a single person can be designated as most important in operating the farm or in share of ownership. In larger, more complex farm businesses it can be difficult to rank the “most important manager” as more or less important than the most important owner. But even within one of these subgroups of producers, persons can have more or less equal decision-making or ownership rights. In more traditional family farms, there have often been cultural gender issues and generational issues that have led to misrepresenting the reality of who is contributing to operating the farm, and in this regard data could very much be influenced by who responds to a survey in any given year.21 The maximum that can be done is to ask, in ARMS or special surveys, for data on each farmer/producer that identifies how many hours each worked and who was involved in which decisions, but such questions are potentially time-consuming and thus costly.
In the case of multiple operators, during the data collection process currently it is the respondent for the farm who identifies the principal farm operator. This identifier also serves as the link to a household for the collection of data on income and other household variables. Such an approach, however, does not lead to an accurately defined household population. (We return to the issue of the relevant household population in the next section, and then again in Chapter 5.)
It is also clear that, based on the definitions above, a farmer may be involved in more than one farm and even in more than one farm business. In the case of a farmer who owns or co-owns a complex farm business that includes more than one farm, the farm business is a holding company. However, this firm is often a conceptual construct: the presence of a company that oversees the whole structure might be lacking, in practice, even for cases where a farmer owns or co-owns several farms or is involved in LLCs with family members or others. It is these constructions that are especially
21 In the past, as documented by NASS, see https://www.usda.gov/media/blog/2016/01/19/counting-all-farmers-capturing-many-faces-agriculture-2017-census, there has been bias as to who has been reported as “operator,” and even with the new questions and approach on the Census it is unlikely one will see a total change, due to the “culture” of agriculture. Although a woman may be the decision-maker on a farm, she may not be reported as such by some families because women are not viewed as “the farmer.”
complex for data collection. For agricultural statistics, in cases like these the thinking should be clear, namely that the farm establishment is often the point of entry from which higher-level structures, like farm businesses, as well as households can be profiled.
Finally, it is worth remarking that the producer/farmer (whether owner or manager) identified in the Census of Agriculture or ARMS is not necessarily the respondent to the survey. Especially in complex farm businesses, the respondent could very well be an administrator, an outside accountant, or another staff member involved in farm management. Conceptually, the issue of who the respondent is should not influence the view of who the farmer is, but we will return to this issue in Chapter 5, where survey designs are discussed, because this issue also has consequences for what can be asked and to whom.
Policy makers and researchers are not only interested in farms and farmers, but also in farm households, because the household situation can influence the behavior of the farmer and the activity of the farm. A well-known example of this influence is the way investment decisions by farmers over long time horizons, such as 30 years, are influenced by whether or not they have a successor, which affects supply responses to policies (Gasson and Errington, 1993; Calus, 2009). Investment decisions can also be influenced by income from sources other than the farm, because such incomes can reduce the cash flow needed for consumption or even be allocated to farm investments. Agricultural policies are also sometimes justified as needed to sustain the family farm in times of low production or income, making the interest in family households a legitimate question.
This interest in the total income and well-being of the farmer and the farm household—which factors into ERS mandates—is especially relevant for the family farm. On very large farms and in complex farm businesses, the family dimensions are relatively less important to the functioning and stability of the operation. Such organizations are more like big family firms in other sectors: it is hard to imagine that family income from other sources plays as big a role in the investment decisions made by Walmart as it does in decisions made by small retail businesses.
The family farm is a dominant concept in the public mind and in political debates on agriculture. Historically, family farms are what brought most of the United States into cultivation. In 1930, one of the most iconic American paintings (now in the Art Institute of Chicago), was popularly called “Iowa Farmer and Wife,” although the painter Grant Wood originally named it “American Gothic,” after the unusual window in the building pictured. In reality, the couple in the painting was not a couple, nor was
either of them a farmer (the man was Wood’s dentist).22 Perceptions of the family farm in agricultural statistics create a similar distortion: already in 1977, James Bonnen, in his Assessment of the Current Agricultural Data Base: An Information System Approach, questioned the concept:
The idea of the “family farm,” with all its value and organizational assumptions, constitutes the central concept around which most of our food and fiber statistics are designed and collected. Yet, it has become an increasingly obsolete representation of the reality of the food and fiber sector. . . . The world has changed and the concept has not. (p. 387)
Given that this observation is 40 years old, it seems even harder to adapt an ingrained image, such as “American Gothic,” to current reality. Adapting the concept to reality would nevertheless follow a tradition. Reinhardt and Bartlett (1989) point out that the concept of the family farm was originally used for homesteading farms, which had no outside labor or capital nor used contractors, and over time it was broadened to keep up with changes in the organization of farming.
The interest in family farms and farm households raises the question of how to define them. The fact that a farm can have several farmers (producers) means that it can be associated with several different households, although this is not necessarily so: a man and his spouse can both be producers, as can one or more of the children, while they are living in the same household. But just as typically, children or brothers may set up their own households, which results in several households being associated with a single farm. Such households may also contain persons who are not necessarily direct relatives of the farmer: children from an earlier relationship of their spouse, interns, and so on. Meanwhile, some family members, such as children away at college, are only members of the household for part of the year.
All these factors make it useful to have a clear definition of both family and household. The Census Bureau defines a family as “a group of two people or more (one of whom is the householder) related by birth, marriage, or adoption and residing together; all such people (including related subfamily members) are considered as members of one family.” Beginning with the 1980 Current Population Survey, unrelated subfamilies (referred to in the past as secondary families) are no longer included in the count of families, nor are the members of unrelated subfamilies included in the count of family members. The number of families is equal to the number of family households, but the count of family members differs from the count
of family household members because the latter count also includes any nonrelatives living in the household.
According to the Census Bureau, a household consists of all the people who occupy a housing unit. A house, an apartment, a group of rooms, or even a single room is regarded as a housing unit when it is occupied or intended for occupancy as separate living quarters; that is, when the occupants do not live with any other persons in the structure and there is direct access from the outside or through a common hall. A household includes both the related family members and all the unrelated people, if there are any, who share the housing unit, including lodgers, foster children, wards, or employees. A person living alone in a housing unit is counted as a household, and so is a group of unrelated people sharing a housing unit, such as partners or roomers. The count of households excludes group quarters. There are two major categories of households: “family” and “nonfamily.”
NASS and ERS use these same definitions for household and family, which are used broadly across the statistical system and should continue to be so used. For NASS and ERS, a family is a group of two people or more related by birth, marriage, or adoption and residing together. A household consists of all of the people who occupy a housing unit, that is, a house, an apartment or other group of rooms, or a single room, occupied or intended for occupancy as separate living quarters.
Even with precise definitions for farm and family, however, linking the two in the term family farm is problematic. First, there are many cases in which a farmer lives alone, unmarried or as a widow(er), and such farms are nevertheless called family farms. More problematic is that an overwhelming number of farms, including those that are part of complex farm businesses, are owned and operated by one or two related families. Therefore, the terms simple farm and family farm are not synonyms. Indeed, in its publications, ERS highlights the fact that family farms can be very complex, with multiple households sharing in the farm’s income.
In practice, data collection from household units can be restricted to the households of farmers who are owners, assuming that the nonfarm activities of salaried managers do not affect farm management. Data collection on those owner-households, for instance regarding nonfarm income, can also be restricted to those persons who are farmers/owners and their spouses, assuming that the nonfarm income of children who might live at home is used for their own personal expenses and savings and not for financing the farm or for reducing the amount of farm income needed for household expenditure. However, this last assumption may be questionable if the child is the potential successor on the family farm.
Off-farm income, under this approach, consists of all income from sources other than farming. This may include other agricultural (support) activities, self-employed income, wage income, and capital income, among
other things. Neither is it particularly relevant if this income is earned at a location on the farm or elsewhere; whether it is income from a tractor repair shop or bed and breakfast on the farm premises or income earned at a job in the town does not matter.
From the above line of reasoning, it follows that the definition of a farmer is not influenced by his or her income streams from other, nonfarm sources, nor by his or her age. A 75-year-old farmer who receives 75 percent of his income from a retirement pension is still a farmer, and so is the 40-year-old farmer who earns 80 percent of her income from working outside of agriculture or from owning another enterprise. Even if a farmer earns all of his or her income outside farming and makes a loss running a farm—perhaps to enjoy the living environment, or to benefit from certain social security payments or tax facilities—he or she is still a farmer (producer).
Farming and agriculture are often taken to mean the same thing. In section 4.1, we provided several examples of agencies that use agriculture as a term to describe farming, but we also argued, based on the NAICS classification, that it is useful to restrict the term farming to the management of biological processes that excludes support activities to farming (NAICS 115000)23—such as harvesting, cotton ginning, and farm management—if they are provided as services by other firms. In other words, if farmers harvest their own grain, it is an activity that is part of farming, but if these activities are carried out by another firm or farm as a service it is a support activity to farming (see section 4.1).
The essence of the problem, and an important source of complexity in agricultural production, arises from the way farming activities are located within the larger food and agriculture supply chain. Compare the following stylized economies, where gray shading denotes activities undertaken by farmers:
In the most extreme case (diagrammed above), farming is a matter of subsistence for household members, and the activities of such farmers
23 NAICS lists them as support activities to Agriculture and Forestry, but if this difference is made between agriculture and farming, they should logically be called support activities to farming.
span the entire supply chain. Households in such cases are responsible for all steps in the production of farm output, and there is no marketintermediated transfer of intermediate goods between input suppliers, farmers, and processors, prior to final consumption by consumers. If such a household were farming wheat, the members of the household would be responsible for collecting seed and manure, preparing the fields, planting, harvesting, threshing, winnowing, grinding, baking bread, and ultimately consuming their own product. This is a model of agricultural production devoid of any trade or specialization and is perhaps approximated by very poor societies in developing parts of the world.
Less extreme is the case where farmers specialize in one part of the production process. This supply chain (diagrammed below) represents the case in which farm businesses only undertake the agricultural production activities and transfers of intermediate goods that arise from inter-firm transactions. Returning to our wheat example, the farm purchases seed and fertilizer from elsewhere, and the harvested wheat is sold to a flour mill.24 The rest—all of the planting, cultivation, and harvesting—is undertaken under the direct management of the farm.
Of course, there are definitional challenges in precisely demarcating the distinction between production and processing. Are threshing and winnowing grain the former or the latter? What about the drying of tobacco leaves? Or the packing of citrus fruit? Nevertheless, for the purposes of discussing complexity in agricultural production, these distinctions may be taken at face value, and under the current NAICS system all the aforementioned examples are classified as agricultural production activities (see section 4.1).
Agricultural production is constantly evolving, and it has generally increased in complexity over the course of human civilization. In particular, three phenomena stand out. First, trade across economies has existed for millennia, but the breadth and depth of such trade are greater today than ever before and occur at every stage of the supply chain. Second, the number of links within the supply chain has grown tremendously. Third, the boundary of the farm increasingly fails to align with the definition of agricultural production activities.
24 There may be additional steps; for example, the farmer may sell to a cooperative or dealer who does not process the wheat but sells it to flour mills.
In the supply chain diagrammed below, the farm business is no longer responsible for all the agricultural production activities:
Instead, in this example the farm hires a specialized service provider who assumes responsibility for soil preparation and planting (NAICS 115112). When the crop is ready, a potentially different service provider is contracted to harvest (NAICS 115114). And while some management tasks are undertaken by the farm business itself, others are delegated to a specialist provider (NAICS 115116).
The increasing decentralization of production activities through the use of specialist service providers is of particular interest because, while these businesses are not farmers themselves, they engage in on-farm production, covering the whole range of activities from soil preparation, planting, and cultivating, to harvesting, packing, and management. More than 50 years ago, the USDA and the Census Bureau recognized the increasing importance of these businesses to the overall farm economy and conducted the first Census of Agricultural Services as a follow-on survey to the 1969 Census of Agriculture. The following paragraph from the 1974 Census of Agriculture succinctly summarizes the motivation for expanding its scope to include such service providers:
Until the 1940s, agriculture in America was largely self-reliant in regard to many production and harvesting practices now available from off-farm sources in the form of agricultural services. During the last three decades agricultural services have become an increasingly specialized industry. The technological and scientific changes in American agriculture have been directly related to the development of the agricultural service industry. A census of this industry is essential to provide facts necessary for
- a broader view of today’s farm production,
- a better understanding and interpretation of long-term agricultural changes and trends, and
- a more meaningful analysis of the interrelationships of agriculture and agricultural services.25
251974 Census of Agriculture, Volume III: Agricultural Services, Appendix A, pg. A-1.
Despite these well-intentioned exhortations, the Census of Agricultural Services was discontinued in 1978 and no current Census Bureau or NASS survey program specifically targets these businesses. While a dedicated data collection effort has long since disappeared, the importance of specialist service providers to farm production and their contribution to farm complexity has only continued to increase. Indeed, these businesses fall into a coverage void between NASS, which surveys farms, and the Census Bureau, which surveys nonfarm, nonagricultural businesses. A critical component of the modern agricultural supply chain has simply fallen between the statistical cracks.
This situation is problematic for two reasons. First, key measures of the overall agricultural production sector are increasingly being mismeasured. For example, we currently cannot construct a reliable estimate for the number of individuals employed on farms. From the Census of Agriculture, we know how many workers farms hire and how much they spend on contract labor, but the actual number of farm workers, how much they are paid, and their take-home wages at the national, state, or county level are all unavailable in any current economic survey. The increasing concentration of land and output is a commonly recognized source of intra-firm complexity, but the vertical disintegration of farm production activities is another source of complexity that also requires attention.
Second, service providers may themselves be farm operators. The Census of Agriculture currently asks farmers to report related income from the provision of specialist services, but only if these are not stand-alone businesses. Indeed, in the past, explicit dollar thresholds were employed to define farm-related versus stand-alone businesses. Documenting the relationships within farm businesses between the constituent establishments—farm establishments and nonfarm establishments—is key to overcoming the challenge of reporting on complex farms.
To the extent that USDA should be reporting on agricultural production activities in the United States, regardless of the business entity carrying out that activity, these agricultural production activities should, in future, be surveyed.
RECOMMENDATION 4.3: A program akin to the defunct Census of Agricultural Services, perhaps undertaken as a follow-on survey to the Census of Agriculture, should be developed to collect and report economic activity undertaken by establishments and firms engaged in agricultural production activities through the provision of support services to farms. To accomplish this task, such providers must be identified and included in a Farm Register (described in detail in Chapter 5).
Defining and Enumerating Secondary Activities
The other boundary issue, discussed earlier in this chapter, is that farms may engage in secondary activities that are not farming but rather food processing or retail. Cheese making and produce selling by the roadside or at a farmers market stall are examples of activities closely linked to farming. Notice that, in the last supply chain diagrammed above, the farm business engages in economic activity beyond agricultural production. As an example, if the farm business is a berry farm, it might process some fruit on-site to produce jams or preserves (manufacturing). Some of the fruit may also be used in baking pies and muffins (manufacturing). These items may be sold at an on-site store (retailing) to patrons who visit the farm to pick their own berries and enjoy recreational activities (services). It follows from the recommendations in section 4.2 (and as illustrated with the wine and cheese examples) that these are very often secondary activities of the farm, because they are not organized in a separate establishment; nor do they account for the majority of the work (or value added) in the farm. However, it also follows that, if activities of this kind are organized in a separate establishment or where farming is not a primary activity (such as a garden center that grows some Christmas trees itself), the establishment is not a farm and these are not agricultural support activities. Often, an establishment of this kind will be classified as a food processor or retailer.
These sectoral spillovers mean that a census of farms (farm establishments) is not a census of all farming or agricultural activities, because some of the included farms also engage in a subset of activities outside farming; meanwhile, some farming and agricultural activities are carried out by establishments that are not classified in a statistical framework as farms but instead as agricultural support firms, food processing companies, or retailers. This is the direct consequence of the fact that the Census of Agriculture and ARMS seek to survey organizations and their managers and are interested in the decision making in those organizations.
It is therefore often necessary to extend survey coverage to include these secondary or smaller activities. Farms could be asked how big (and what) these other activities are, and other types of firms could be asked if they engage in farm activities and farm support activities. Based on such an estimate of activities, statistics could be generated to identify the size of farming or agriculture in the total economy.
Two key features of agricultural production in the United States are missing in the supply-chain diagrams above depicting stylized economies. First, the diagrams ignore the central role of international trade. While long-distance trade in agricultural commodities in either their raw or manufactured state is nearly as old as human civilization itself, the world economy has never been more integrated that it is today. Second, the diagrams
ignore the relative contribution of each link in the supply chain. The contribution of agricultural production activities to total economic value added in the United States is now lower than at any point in history. Although food is no less important to human well-being, the sources of value added have shifted to other links in the supply chain—toward companies like Monsanto, ADM, and John Deere on the input side and ConAgra, Coca-Cola, and McDonalds on the processing side.
In addition to agriculture, an agribusiness complex has been created in the United States and other large economies (Davis and Goldberg, 1957). That also means that agricultural policy, along with environmental policies that target farming, has effects on other sectors than farming and agriculture. To give policy makers and the public insight into these interdependencies, statistics on the agribusiness complex are needed. This can be done with a methodology based on input-output tables of the national accounts (and its satellite accounts) that link farming to activities in other sectors.
RECOMMENDATION 4.4: The National Agricultural Statistics Service, the Census Bureau, and the Bureau of Economic Analysis should all report on the size of the agribusiness complex and its components in terms of income, employment, and environmental impacts and develop a program that harnesses existing data collection efforts to create a new satellite account for reporting on the food and agriculture industries.
A farm is an establishment and should be economically identifiable. This implies that any farm either has a profit-and-loss account and a balance sheet or, at least, that the farmer should be able to create them for a survey. In the real world, terms like income and net worth are not always very clearly defined, and many farmers restrict their accounting to fiscal accounts that satisfy the tax authorities.
Accounting standards as issued by the International Financial Reporting Standards (IFRS) Foundation give clear guidance on how to create a balance sheet, a profit-and-loss account, and a flow of funds for the farm based on accrual accounting. Their accounting standard IAS41 gives clear definitions on valuations in agriculture that ARMS could use. For farm households, an income statement can be added. Off-farm activities generate substantial income and thus contribute to the well-being of farm households. Income at the household level thus may originate from different sources: farm income and nonfarm income. Nonfarm income can be defined as the net income from all nonfarm businesses, wage and nonwage categories.
An important issue in complex farm businesses is that of transfer pric-
ing between different farms and other establishments that are part of the farm business. IFRS accounting standards provide guidance for transfer pricing, but the complexity of this accounting could be a reason to create consolidated accounts for complex farm businesses.
Providing information to policy makers and the public on the real structure of agriculture and the inequalities in the farm sector makes it necessary to provide data not only at the farm (establishment) level but, even more, on the financial situation of complex farm businesses. This means that it is preferable to select farms for ARMS from a register that includes both simple and complex farm businesses. Alternatively, if census data are only available at the establishment (farm) level, ARMS could use these data as a basis for selection, but whenever a farm is part of a complex farm business it should collect data for the complex business as a whole.
To find a solution to the issue of data collection on complex agricultural holdings, as specified in Chapter 3, this chapter analyzed concepts from general statistics and accounting. The analysis suggests that complex agricultural holdings should not be fitted into a definition of a farm (counting them all as one farm); nor should such holdings be pressed into a definition imposed on them that fails to recognize the juridical or fiscal organization or the informal management arrangements that are present in such complex holdings.
Adopting a generalized statistical framework and integrating agricultural statistics into it could help solve the problems in data collection and interpretation created by the presence of complex farm operations. Central to this framework is the recognition that a farm business may have either a simple organization, in which it is one establishment—the “classical” farm—or a more complex organization, in which it consists of several establishments.
For the simple farm businesses, not much needs to be changed in data collection. Such farms can be questioned through census forms that query all that NASS wants to know about them, their farm activities, their farmers, and their households. The current structure of ARMS measures land uses, farming practices, input use, and financial statements well. The use of administrative means or surveillance to collect data on these farms will enhance the quality of the information collected and reduce the burden for the farmer. For data collection from complex farm businesses, the differences between farms and farm businesses and the decisions about whom to ask about farming activities, farmers, and households all have to be taken into account. Farms, farm businesses, and farm households—and even farm fields—each inhabit their own universes that can be represented in registers
or list frames designed to capture them. This aspect of data collection is discussed in detail in Chapter 5. The presence of these distinct universes also implies that different data items should be associated with different reporting units.
The main objective of the Census of Agriculture is to report on the structure of farming in terms of the number of holdings, their activities, and their size. That distribution is fairly stable over time, which means that a yearly survey may not be needed for most purposes and that the current five-year interval may be sufficient. As argued in section 4.2, complex farm businesses could benefit from more guidance in profiling their activities, revenues, land, and assets for the Census of Agriculture. In addition to the structure of farms, this census could also be adapted to report on farm businesses.
Information on yields and prices has to be reported comparatively more often to improve the functioning of markets. However, on this score, experts from upstream and downstream industries who visit multiple farms are often more knowledgeable than individual farmers; and remote sensing technology is another valuable source of information about fields, cropping activities, and yields. As a result, there is a diminishing need to include such variables in the Census of Agriculture.
By contrast, to report on revenues and income for the purpose of carrying out policy evaluations, farm businesses (both simple and complex) are the most important level of analysis and should be a major focus of reporting for ARMS. In addition, as argued above, attention must be paid to agricultural support activities and agricultural business complexes. This increased attention to complex farm businesses will require additional resources, which may be freed up by reducing the number of questions in the Census of Agriculture (especially concerning monetary aspects), by reduced attention to very small farms, and by the deployment of modern information technology solutions. Chapter 5 provides additional guidance on these strategies.
AGRICULTURAL ACTIVITIES LISTED IN NAICS SECTOR 11: AGRICULTURE, FORESTRY, FISHING AND HUNTING
|NAICS Code||Type of Activity|
|111100||Oilseed and grain farming|
|111200||Vegetable and melon farming|
|111300||Fruit and tree nut farming|
|111400||Greenhouse, nursery, and floriculture production|
|111900||Other crop farming|
|111990||All other crop farming|
|112000||Animal Production and Aquaculture|
|112100||Cattle ranching and farming|
|112110||Beef cattle ranching and farming, including feedlots|
|112120||Dairy cattle and milk production|
|112200||Hog and pig farming|
|112300||Poultry and egg production|
|112400||Sheep and goat production|
|112900||All other animal production|
|112920||Horses and other equine production|
|112990||All other animal production|
|113000||Forestry and Logging|
|114000||Fishing, Hunting and Trapping|
|115000||Support Activities for Agriculture and Forestry|
|115110||Support activities for agriculture|
|115112||Soil preparation, planting, and cultivating|
|115113||Crop harvesting, primarily by machine|
|115114||Post-harvest crop activities (except cotton ginning)|
|115115||Farm labor contractors and crew leaders|
|115116||Farm management services|
|115210||Support activities for animal products|
SOURCE: U.S. Census Bureau, see https://www.census.gov/eos/www/naics/2012NAICS/2012_Definition_File.pdf.
This page intentionally left blank.