Michigan State University Extension
Ag Experiment Station Special Reports - SP000001
07/28/98
Michigan State University Agricultural Experiment Station INTRODUCTION
In his landmark study, Types of Farming in Michigan(1) (1939), Professor E.B. Hill observed that before World War II, Michigan agriculture was extremely diverse, due to wide variations in soil types, typography, climate and economic conditions found within the state.
While physical factors still have a major affect on the structure of Michigan agriculture, non-farm factors, technical change, and economic conditions have played an increasingly important role in shaping Michigan agriculture since 1939.
The expansion of urban-suburban communities into rural areas has had a major impact on land use. Advances in crop and livestock breeding, research on plant and animal nutrition, the development of labor-saving equipment, the availability of inexpensive sources of nitrogen and new agricultural chemicals, and improvements in transportation have transformed crop and livestock production and made agriculture more dependant on the non-farm economy. In recent years, macroeconomic factors and international trade developments have contributed to the impetus of change.
The net effect of these changes has been a significant downsizing of the farm population, changing land use patterns, expanding farm size, and the evolution of a bimodal agricultural structure--with an increasingly smaller number of highly specialized and technologically sophisticated farms accounting for an increasing share of agricultural marketings.
Yet agriculture remains a major contributor to Michigan's economy. In 1990, the agriculture, food, and forestry sectors contributed an estimated $53-54 billion(2) to the state economy.(3) Of this total, approximately 17% was generated from farm income, 51% from food processing, 3% from input sales, and 29% from forest product sales. In terms of value added, the agriculture and food sector contributed an estimated $10.5 billion--equal to 10.5% of the value added by all primary industries (manufacturing, agriculture and forestry) and equal to 47% of the contribution of motor vehicles and equipment (Ferris, 1991).
Furthermore, the agricultural sector directly accounts(4) for approximately one-eighth of total employment in primary industries (manufacturing, mining, agriculture, forestry and fisheries)--roughly equal to one-half of the employment associated with the manufacture of automobiles and parts.(5) In recent years, the value of farm exports ranged from nearly $900 million (1980-81) to about $500 million (1984-85), equivalent to about 20-30% of farm cash receipts from farm marketings. In addition, Michigan food processors exported approximately $305 million (1983), equal to 3.7% of the total value of shipments from Michigan food processors (Ferris, 1987).
This report reviews the long-term trends in the structure of Michigan agriculture in order to provide the historical perspective needed to better understand the present and anticipate the future.
U.S. Census of Agriculture data are analyzed to characterize changes in Michigan agriculture over the period 1950 to 1987. Base year data from the 1950(6) census are used to document Michigan agriculture before the rapid changes that followed WW II. The year 1987 was selected as the terminal year since this is the latest year for which census data are available. While using two years as a basis for describing changes over a 37-year period fails to take into account change within the period considered, it does provide an overview of the magnitude of the long-term changes that have characterized the period.
STRUCTURAL CHANCES IN MICHIGAN AGRICULTURE, 1950-1987
Overview Trends in the share of land in agriculture, the farm population (number of farms), and average farm size highlight the magnitude of the changes in Michigan agriculture since 1950.
Land Use Since 1950 total farmland area has declined by 40%, from 17,269,922 to 10,316,871 acres in 1987. While farmland accounted for 47% of Michigan's land in 1950, it represented only 28% in 1987.
Farm Population Data on the number of farms are the best indicator of changes in the farm population.(7) While Michigan's total population increased from 6.372 to 9.145 million (+70%) from 1950 to 1987, the number of farms declined 67%--from 155,589 in 1950 to 51,172 in 1987.
Average Farm Size A major consequence of these trends has been an increase in average farm size (+82%), from 111 acres per farm in 1950 to 202 acres per farm in 1987.
These and related changes in the structure of Michigan agriculture are reviewed in detail in the sections that follow. The first section analyzes general changes in the farm sector, including land use, farm size, operator characteristics, and types of farms. The second section reviews changes in crop production. The third section analyzes changes in livestock and poultry production. The final section reviews changes in agricultural marketing.
General Changes in the Farm Sector Key indicators of change in the farm sector include shifts in land use, a decline in the number of farms, an increase in farm size, changing operator characteristics, and shifts in the distribution of farm types.
Agricultural Land Use With the decline in farmland to 60% of the 1950 level, agricultural land has shifted from less intensive to more intensive uses (Table 1, Figures 1 & 2).
Greatest declines in most extensive uses. Although acres in all agricultural uses declined over the period, the most significant changes occurred in the more extensive uses. For example, pasture and woodlands acres fell 81% and 66%, respectively, compared to -26% for cropland acres (Table 1). By 1987 cropland was increasingly being used for crop production (75% vs. 71%), woodland was increasingly allocated to non-pasture uses (82% vs. 42%), and pasture was increasingly drawn from cropland (53% vs. 38%).
Increasing importance of harvested crops on cropland. Compared to 1950, in 1987 cropland was increasingly used as harvested cropland (from 71% to 75%) and for other uses (from 11% to 18%), but less for pasture (from 18% to 6%) as shown in Figure 3. In 1987, the most important other cropland uses were idle land (55%) and cover crops (30%).
Significant increase in irrigated land. Data on irrigated cropland provide yet another indicator of a trend towards intensified land use. In 1950, only 995 farms irrigated 13,901 acres--only 0.1% of Michigan's cropland. By 1987 3,755 (+277%) farms irrigated 1.2 million acres--equal to 14.7% of Michigan cropland.
Number of Farms and Farm Size Over the period, the number of farms declined from 155,587 to 51,172 (-67%), while average farm size increased from 111 acres to 220 acres (+82%). As a result, both the number of farms and the percentage of Michigan's farmland in each farm size class shifted dramatically. These overall changes are summarized in Lorenz curves for 1950 and 1987 (Figure 4), showing the percent of farms accounting for the percent of farmland, respectively.
Increasing number of larger farms. Small farms (1-49 acres) accounted for about the same share of Michigan's farms in 1950 (29%) as in 1987 (29%), as shown in Table 2 and Figure 5. In contrast, the percentage of 50-179 acre farms declined sharply (-29%) over the period, from 55% in 1950 to 39% in 1987, shifting the farm size distribution towards larger units. For example, from 1950 to 1987 the share of 180-499 acre farms increased 42%, the share of 500-999 acre farms rose 865%, and the share of 1,000 and more acre farms increased 1,801%.
Although the share of farms in the largest size class grew rapidly over the period, the 50-179 acre farm remained the most common (modal) farm size (55% in 1950 vs. 39% in 1987). In contrast, in 1987 large commercial farms (1,000 acre and more) accounted for only 3% of Michigan farms.
Increasing share of farmland in larger farms. Analysis of changes in the percentage of farmland in each size category provides an alternative perspective on farm size dynamics (Table 2 and Figure 6). While about 30% of Michigan farms (1987) were 1-49 acre units, they accounted for only a small and relatively constant share (6.5% in 1950; 3.4% in 1987) of farmland. In contrast, over the period the share of farmland in the 50-179 acre size class fell from 51% to 20% (-61%); and farmland in the 180-499 acre size class declined slightly from 36% to 32% (-11%).
In contrast, land in larger farms expanded rapidly over the period. By 1987 farms of 500 acres or more accounted for over 45% of the farmland area, compared to only 6% in 1950; and farms of 1,000 acres and more operated 21% of the farmland in 1987, compared to only 2% in 1950.
Farm Operator Characteristics The socioeconomic characteristics of farm operators were quite different in 1987, compared to 1950 (Table 3).(8) Over the period, census data indicates an increase in part ownership, a rise in nonresidential operators, a growing importance of nonfarm employment, and an increase in operators' years in residence. On the other hand, there was little change in the age distribution or racial composition of Michigan farmers.
Increasing importance of part ownership. In 1950 about 73% of Michigan's 155,587 farms were operated by full owners, with the remaining farms operated by part owners (18%), and tenants (9%). Although full owners were still the most important tenurial group (61%) in 1987, an increasing share of Michigan's 51,172 farms were operated by part owners (+89%) and a declining share by full owners (-17%) and tenants (-33%).
Increasing importance of non-residential operators. While most farm operators resided on their farm in both 1950 (93%) and 1987 (82%), by 1987 an increasing share of Michigan farms were operated by nonresidential operators-- up from 4% in 1950 to 13% in 1987.
No change in racial composition of farm operators. As was the case in 1950, in 1987 farm operators were almost totally white (99.5% vs. 99.4%).
Increasing importance of nonfarm income. Over the period, off-farm employment became an increasingly important source of income for farm households. In 1950, only 33% of the farm households earned more income from nonfarm sources than from farm sales. By 1987 almost one-half (49%) of farm households reported that farming was not their main occupation. Similarly, while only 31% of the households reported working 100 or more days off-farm in 1950, by 1987 this share rose to 49%--with 43% reporting working 200 or more days off-farm.
Increase in years operating the same farm. Census data indicates that over the period, the average years operators have lived on their present farm increased by 33%, from 15 years to 20 years. This was largely do to a decline in the percentage of operators in all categories except 10 or more years, which accounted for only 57% of the operators in 1950, but 72% in 1987 (+27%).
Little change in age of farm operators. Over the period, the average age of farm operators increased only slightly, from 49.9 years in 1950 to 50.9 years in 1987. In both years, the largest percentage of operators were in the 45- 54 years category (22-23%). On the other hand, while the percentage of operators in the younger age categories (44 years and below) declined slightly, the percentage of operators 65 years and older increased from 16% to 19% (+15%).
Types of Farms The Census of Agriculture groups farms into types of farms (standard industrial classification, SIC), based on the major source of production income (Table 4).(9) Caution is necessary in comparing these data for 1950 and 1987 since the definitions/categories used differed somewhat between the years. For example, in 1950 a large share of farms were classified as miscellaneous, which included farms with over 50% of their sales from forest products, horses, fur animals, bees/honey, and horticultural products.(10) In 1987, miscellaneous was dropped as a category and these farms were included in other categories or not reported in the data. Yet the data provide a general indication of the changes in the relative importance of types of farms over the period. Overall, these data show an increase in the importance of specialized and more land intensive production activities and a decline in the relative importance of dairy and general farms.
Increasing number of field crop and livestock farms. From 1950 to 1987 there was an increase in only the number of farms classified as other field crops (+183%) livestock, excluding dairy and poultry (12.5%), and cash grain (+5%) farms. The largest declines (absolute number) occurred in the SIC farm type categories: poultry (-91%), dairy (-89%), and general (-86%) farms.
Increasing share of field crop and livestock farms. In terms of relative changes (percent of farms in each SIC), the largest share increases were for other field crops (+762%, from 1% in 1950 to 11% in 1987), livestock, excluding poultry and dairy (+242%, from 7% to 24%), and cash grain (+218%, from 10% to 31%).
The largest share declines occurred in the SIC categories: poultry (-74%, from 3.4% to 0.9%), dairy (-65%, from 29% to 10%) and general farms (-57%, from 12% to 5%).
Although comparable data were not reported for 1950, in 1987 the SIC categories specialty livestock and horticultural specialties accounted for 6% and 3% of Michigan farms, respectively.
Crop Production Indicators of changes in crop production include changes in harvested acres, the size of crop enterprises, input use, yields, and total production.
Harvested Acres Between 1950 and 1987, acres in harvested crops declined 20%, from 7.950 million to 6.359 million acres (Table 5). Yet acres in several crops increased significantly, while others declined in area. As a result, important changes occurred in the relative importance (percent of harvested area) of major crops. In addition, these changes served to increase the crop diversity of Michigan agriculture.
Major decline in oats, hay and wheat acres; significant increase in soybeans, and corn acres. In terms of acreage reductions, oats (-1,082,164 acres), hay (-981,141 acres), wheat (-894,139 acres) and dry beans (-113,113 acres) suffered the greatest declines (Figure 7). In sharp contrast, the period saw major expansions in soybeans (+955,830 acres), corn (+677,575 acres), and sugar beet (+67,133 acres) acres.
In terms of percentage declines in acres, flax (-100%), buckwheat (-95%), oats (-81%), rye (-75%), wheat (-72%), barley (-55%), Irish potatoes (-48%), hay (-41%), trees/fruits (-29%), and dry beans (-23%) fell most dramatically. Among major crops,(11) soybeans (+1,389%), sugar beets (+94%), corn (+42%), and vegetables (+27%) showed the largest percentage increases in harvested acres.
Increasing crop diversity. These trends in harvested area had a major impact on the relative importance (percentage share of harvested area) of major crops. In 1950 the five major crops--hay (30%), corn (20%), oats (17%), wheat (16%), and dry beans (6%)--accounted for almost 90% of the states' harvested acres (Figure 8). By 1987 the top five crops' share had fallen to 82% of the harvested area (Figure 9). Corn--up from 20% of harvested area in 1950 to 36% (+78%) in 1987--replaced hay as the state's dominant crop; as hay declined from 30% to 23% (-26%) of harvested area. Yet soybeans showed the most dramatic share increase (+1,761), rising from the 11th (1% of harvested area) to the third (16%) most widely planted crop in 1987. While dry bean's share remained at about 6% over the period, the decline in both the oats and wheat area reduced these crops' share from 17% to 4% (-76%) and from 16% to 6% (- 64%) of harvested area, respectively.
Increasing importance of sugar beets, vegetables, and nursery crops. In 1950 twelve minor crops/commodity groups accounted for about 10% of the harvested area (Figure 8). By 1987, these minor crop's share of harvested area rose to about 18% (Figure 9). Among this group, harvested area shares declined the most for rye (-68%, from 0.7% to 0.2%), barley (-43%, from 1.3% to 0.7%), and Irish potatoes (-35%, from 1.1% to 0.7%). In contrast, harvested area shares increased for sugar beets (+142%, from 0.9% to 2.2%), other (+97%, from 0.2% to 0.5%), vegetables (+59%, from 1.4 % to 2.2%), and nursery crops (no data for 1950, 0.4% in 1987).
Size of Crop Enterprises Since 1950 an increasingly smaller share of Michigan farms operate an increasingly larger share of harvested acres. This trend is evident both in terms of the share of farms in each crop enterprise size class and the share of acres accounted for by farms in each acreage size class (Table 6).
Increasing prevalence of 100 or more acre crop enterprises. In 1950, over 70% of Michigan's farms harvested only 1-24 acre of any single crop. Although small acreage were particularly common among farms producing Irish potatoes (95%), fruit, nuts, berries and grapes (over 92%), and sugar beets (90%), farms with small acreage also dominated oat (85%), corn (81%), wheat (80%), dry bean (76%), and hay (71%) production (US Census of Agriculture, 1950). In contrast, in 1950 less than 1% of the farms harvested 100 or more acres of a single crop.
By 1987 the share of farms harvesting only 1-24 acres of a crop had declined substantially, but farms with these small acreage still accounted for the majority of farms producing oats (71%), vegetables (67%), fruit/orchards (63%), wheat (57%), and Irish potatoes (49%) as shown in Figure 10. On the other hand, by 1987 the share of farms harvesting 100 or more acres accounted for a major share of the sugar beet (33%), dry bean (31%), Irish potato (29%), soybean (24%), and corn (22%) farms.
Increasing share of harvested acres in 100 or more acre crop enterprises. While farms with 1-24-acre crop enterprises remained quite common in 1987, they accounted for a decreasing share of the acres in major commodities compared to 1950 (Figure 11).
By 1987 farms harvesting 100 or more acres cultivated over two-thirds of the area in Irish potatoes (83%), dry beans (71%), sugar beets (71%), corn (70%), and soybeans (69%); and over one-half of the area in vegetables (63%), orchards (60%), and hay (57%). In contrast, farms harvesting less than 100 acres continued to account for a major share of acres in oats (86%) and wheat (70%).
Input Use, Yields, and Crop Production. While increased input use and improved varieties led to higher yields for all major crops, total production only increased for crops which also experienced an increase in harvested area.
Increasing input use. Even in the early 1950s, most farms (71%) reported using commercial fertilizer (Table 7).(12) While data on the amount applied are not available, 4,529,286 acres received fertilizer--equal to 28% of total farmland or 59% of harvested cropland. Surprisingly in 1987 only 73% of the farms reported using commercial fertilizer, possibly due to the increase in specialized livestock farms without crop production enterprises. On the other hand, in 1987 fertilizer was applied to 5,300,902 acres--equivalent to a much larger proportion of farmland (51%) or harvested cropland (86%).
While most farms reported owning four-wheeled tractors (72%) in 1950, ownership increased (+30%) to 93% by 1987. In contrast to trends in fertilizer and tractor use, farms hiring labor declined (-39%) from 49% in 1950 to only 35% in 1987.
Increasing crop yields. For all major field crops, yields increased over the period (Table 8 and Figure 12). Sugar beet yields increased the most (102%), followed by corn (89%), wheat (82%), soybeans (68%), oats (68%), potatoes (36%), and dry beans (28%).
Increasing soybean, corn, and sugar beet production. Over the period, total production increase for soybeans (+2,735%), sugar beets (+292%), and corn (+195%)--the only field crops for which harvested area expanded and three of the four crops with the highest yield increases (Table 8 and Figure 13). Despite previously noted yield increases for other major field crops, total production declined for oats (-67%), wheat (-48%), potatoes (-29%), and dry beans (-1%) due to a decline in their harvested area.
Animal Production Indicators of changes in animal production include changes in animal inventories, the size of herds and flocks, animal product yields, and total production.
Animal Inventories From 1950 to 1987, the livestock/poultry sector experienced a major transformation (Table 9). In 1950 almost all farms held animal inventories. By 1987 production had become highly specialized and concentrated on far fewer farms-- with inventories falling sharply for all except hogs/pigs and chickens. On the other hand, herd/flock sizes increased substantially over the period.
Declining number of livestock and poultry farms. From 1950 to 1987, the number of farms raising all types of livestock declined dramatically--especially farms with inventories of chickens (-101,154 farms),(13) milk cows (-99,491), cattle/calves (-94,718), and hogs/pigs (-50,466), as shown in Figure 14.
In terms of percentage changes, the number of farms raising chickens declined the most (-97%, from 67% to 7%), followed by farms with milk cows (-94%, from 68% to 13%), turkeys (-92%, from 1.4% to 0.3%), hogs/pigs (-90%, from 36% to 11%), cattle/calves (-83%, from 73% to 38%), horses/ponies (-81%, from 31% to 18%), and sheep/lambs (- 78%, from 6% to 4%). While comparable data is not available for 1950, in 1987 16% and 2% of the Michigan farms held beef cow and broiler inventories, respectively.
Declining number of animals, except hogs/pigs and chickens/broilers. Similarly, from 1950 to 1987 census data show that livestock populations declined for all species, except for hogs/pigs and chickens/broilers (Figure 15 and Table 9).
In terms of percentage change, the greatest declines were in the number of sheep/lambs (-74%), followed by milk cows (-57%), horses/ponies (-46%), and cattle/calves (-31%). In sharp contrast, hog/pig numbers expanded rapidly (+78%) over the period, while the number of chicken (-0.3%) and turkey breeding hens (-9%) remained relatively constant. While comparable data are not available for 1950, broiler and beef cow inventories stood at 158,461 and 110,156 animals in 1987.
Size of Animal Herds and Flocks The trend towards livestock specialization is vividly illustrated by changes in herd/flock sizes (average inventory levels per farm), as shown in Table 9.
Increasing size of herds/flocks. Chicken inventories per farm increased most sharply (+2,840%, from 81 to 2,374 birds per farm), followed by hogs/pigs (+1,687%, from 12 to 220 animals), turkey breeding hens (+1,109%, from 29 to 356 breeding hens), milk cows (+607%, from 7 to 53 animals), cattle/calves (+303%, from 15 to 60 animals), horses/ponies (+189%, from 2 to 6 animals), and sheep/lambs (+17%, from 42 to 49 animals). While not reported as a separate category in 1950, in 1987 broiler and beef cow inventories averaged 182 birds and 13 animals per farm.
Continuing prevalence of small livestock/poultry farms. While comparable data are not available for 1950, very small and small livestock operations remained quite common in 1987 (Table 10). These small livestock enterprises accounted for almost all farms with beef cows (1-49 animals, 96%) and sheep/lamb (1-99 animals, 89%); and over one-half of the state's hog/pig (1-99 animals, 65%) and dairy (1-49 cows, 59%) farms (Figure 16). Similarly, small chicken (1-99 birds) and small broiler (farms selling 1- 1,999 birds) operations accounted for about 88% and 97% of farms with chicken inventories and broiler sales.
In contrast, very large (200+ animals) dairy and beef cow farms accounted for only 2.3% and 0.2% of these farms; very large (1,000+ animals) hog/pig and sheep/lamb enterprises accounted for only 11.4% and 0.4% of these farms; and very large chicken (20,000+ birds) and very large broiler (selling 30,000+ birds) farms accounted for only 2% and 1% of these farms.
Increasing importance of large dairy and very large hog/pig and poultry operations. In 1987 very small and small operations continued to account for the largest share of beef cow (73%) and sheep/lamb (46%) inventories (Table 10 and Figure 17). In sharp contrast, very small and small operations accounted for only a minimal share of inventories of hogs/pigs (1-99 animals, 7%), milk cows (1- 49 animals, 25%); chickens (1-99 birds, 1%) and broilers (farms selling 1-1,999 birds, 10%). For dairy, farms with intermediate and large (50-199 cow) herds held the largest share of inventories (63%). On the other hand, very large farms accounted for the largest share of hog/pig (1,000+ animals, 69%) and chicken (20,000+ birds, 91%) inventories; and broiler sales (30,000+ birds, 77%).
Animal Product Yields and Production Despite significant declines in animal inventories (excluding hogs/pigs), animal/animal product production increased for several output categories over the period, largely as a result of increased productivity.
Increasing milk cows and layer productivity. While data are only available to assess changes in milk cow and layer productivity, similar trends characterize livestock and poultry production in general (Table 11). For example, milk yield increased by 253% over the period, from 4,321 to 15,213 lbs. per cow. Similarly, layer output increased by 175%, from 7 to 19 dozen eggs per hen.
Increasing hog/pig, dairy, egg, and turkey production. Sales data, a proxy for output (Table 11), show a major increase in hog/pig production (+130%)--as would be expected given the strong increase (+78%) in inventories (Table 9). Despite the major decline in dairy cow inventories (-57%), milk production increased 53% due to the significant increase in output per cow. Egg production increased 133% against a relatively constant inventory level. Finally, turkey production increased (+331%), despite a slight decline (-9%) in breeding hen inventories.
In contrast, over the period sheep/lamb output fell (-69%), as did cattle/calves (-10%) and chicken (-23%) sales.
AGRICULTURAL PRODUCT SALES The previously described changes in harvested acres and livestock/poultry inventories led to shifts in the share of agricultural marketings associated with crop verses livestock and the share of market value associated with enterprises within these categories (Table 12).
Crop Marketings Changes in crop marketings are reflected in both the share of farm sales from crop production and the relative contribution of specific crops.
Increasing share of agricultural marketings from crops. Over the period, the crop share of agricultural marketings increased by 27%, from 39% of total agricultural marketings in 1950 to 50% in 1987 (Table 12).
Increasing share of crop sales from other field crops, horticultural crops, and vegetables. While grains (corn, wheat, and oats) accounted for the largest share (33%) of crop sales in 1950, by 1987 grain sales fell to 27% and other field crops (primarily soybeans, dry beans, potatoes, and sugar beets) surpassed grains as the major contributor (28%) to crop market value (Table 12, figures 18 and 19).
In percentage terms, horticultural crops experienced the greatest increase in share of sales (+96%, from 9% to 17%), followed by vegetables (+14%, from 9.5% to 10.8%), as shown in Table 12. In contrast, fruits/nuts showed the greatest declines in shares of sales (-22%, from 17% to 13%).
Increasing importance of corn and soybeans. Over the period the relative importance of major field crops shifted dramatically. In 1950 the two most important crops--wheat and dry beans--accounted for 37% of crop sales (Table 13, Figures 18 and 19). By 1987 two different field crops-- corn and soybeans--accounted for 36% of the market value of crops sold. In percentage terms, this represented a 1,068% increase in soybean's (from 1.2% to 13.6%) and 242% increase in corn's (from 6.5% to 22.4%) share of market value. In sharp contrast, wheat's market share fell sharply (-85%, from 23% to 3.3%), followed by dry beans (- 66%, from 15 to 5%), oats (-49%, from 2% to 1%), and Irish potatoes (-36%, from 7% to 4%).
Livestock Marketings Changes in livestock marketings are reflected by changes in both the share of farm sales from animal production and the relative contribution of specific types of animal products.
Declining share of agricultural marketings from livestock and poultry. From 1950 to 1987, the animal share of agricultural marketings declined by 18% (Table 12). Whereas in 1950, livestock products accounted for 61% of agricultural sales, by 1987 their share fell to 50% of total agricultural market value.
Increasing share of animal product sales from hogs/pigs and cattle/calves. Although dairy product sales accounted for the largest share of animal product sales in both 1950 and 1987, their share of the total sales declined (-11%, from 50% to 45%) over the period (Figures 20 and 21). Similarly, share contributions declined for sheep/lamb products (-78%, from 2.0% to 0.4%) and poultry/poultry products (-37%, from 14% to 9%). In contrast, animal product share contributions increased the most for the all inclusive category, "other products" (+155%, from 1.1% to 2.8%), followed by hogs/pigs (+65%, from 11% to 18%), and cattle/calves (+16%, from 22% to 25%).
III. CONCLUSION
Over the past 37 years, Michigan agriculture has undergone a major transformation. While it is inconceivable that the next 37 years will bring such dramatic changes in the structure of Michigan agriculture, many of the trends identified are likely to continue--if at a somewhat slower pace. To the extent to which we can anticipate the future, such as the impact of the GATT negotiations and the North American Free Trade Agreement on Michigan agriculture, the better prepared we will be to absorb the impact of such developments.
Of major concern is the impact of the changes highlighted in this report on the viability of rural Michigan communities. In many formerly agricultural communities, agriculture is no longer the dominant source of income and employment. Consequently, in planning for the future, policy makers and the research community must not only envision strategies to strengthen the state's production base, but also develop creative strategies for revitalizing our declining rural communities.
1 E.B. Hill. 1939. Types of Farming In Michigan. Special Bulletin 206, Michigan State College, Agricultural Experiment Station. June.
2 These estimates include direct and indirect impacts and take into account the wholesaling, retailing, service, and transportation industries.
3 Estimated by Professor John N. Ferris, Department of Agricultural Economics, Michigan State University, 1991.
4 This estimate excludes the indirect employment impact of agriculture on the transportation, wholesaling, retailing, and farm services.
5 Ferris, J.N. 1987. "A Description of Michigan Agriculture and the Food Industry". Agricultural Economics Staff Paper No. 87-36. Department of Agricultural Economics, MSU, East Lansing, MI. May.
6 In the Census of Agriculture, inventory data generally refers to the year of the census and sales data refers to the previous year.
7 The Census Bureau has revised its definition of a farm several times. In the 1950 Census of Agriculture, places were counted as a farm "if the annual value of agricultural products, exclusive of home-garden products, amounted to $150 or more, and...all land under the control of one person/partnership was included as one farm." Since 1974 the census definition of a farm has been any "place from which $1,000 or more of agricultural products were produced and sold or normally would have been sold during the census year." The farm definition presents two problems. First, data between 1950 and 1987 are not directly comparable due to differing definitions of a farm. Second, the census definition includes many small units that do not depend on agriculture as a major source of family income.
8 In 1950, 492 farms were operated by managers. The 1987 Census does not include managers as a tenure category.
9 Only commercial farms are included in the type of farm typology. In 1950 all farms with sales of $250 or more were classified as commercial. In 1987, this minimum was $2,500. Farms are placed in the respective categories if the respective category represented 50% or more of all farm products sold.
10 Since the 1974 census, farms with greater than 50% of sales accounted for by forest products are not included in this typology.
11 Excluding sorghum which expanded from 24 (1950) to 8,598 (1987) acres.
12 The 1950 Census of Agriculture does not report fertilizer use, so these data are drawn from the 1954 census.
13 In 1950, chicken inventories refer to birds 3 months old or older. In 1987, the census reported birds 4 months old or older.