I E> RA RY OF THE U N IVERSITY OF ILLI NOIS 630.7 U6b 7)0.676-700 AGRICULTURE NOTICE: Return or renew all Library Materials! The Minimum Fee for each Lost Book is $50.00. The person charging this material is responsible for its return to the library from which it was withdrawn on or before the Latest Date stamped below. Theft, mutilation, and underlining of books are reasons for discipli- nary action and may result in dismissal from the University. To renew call Telephone Center, 333-8400 UNIVERSITY OF ILLINOIS LIBRARY AT URBANA-CHAMPAIGN L161 O-I096 /tb FARM REAL ESTATE SALES in Illinois Of DEC 41963 FolkeDovring HVasiTY OF AIWM.S "d William H. Scofield Bulletin 697 UNIVERSITY OF ILLINOIS AGRICULTURAL EXPERIMENT STATION in cooperation with FARM ECONOMICS DIVISION, ECONOMIC RESEARCH SERVICE, U. S. DEPARTMENT OF AGRICULTURE Urbana, Illinois November, 1963 Publications in the Bulletin series report the results of investigations made or sponsored by the Experiment Station CONTENTS Background Data for the State 6 Long-term trend in market value of farm real estate and relationship to farm income 6 Rate of transfers by type 8 Sellers and buyers of farm real estate 10 Financing land transfers 11 Dollar Levels and Trends in Sales Prices 12 Sources of data and limitations 12 General characteristics as to size and level of prices per acre, all sales 13 The time series and the census data 15 Price differentials between improved and unimproved properties, by size class 19 Size of tracts sold in relation to size of farms the parcel market versus the whole-farm market 21 Price variation by size of tract and range of price 29 Farmland value areas in Illinois 34 Variations in land earnings 35 Summary 38 Appendix A: The Data and How They Were Treated 39 Appendix B: Data by State Type-of-Farming Areas 46 This study was initiated by the Farm Economics Division, Economic Research Service, U.S. Department of Agriculture, as a part of its continuing research program in farmland values and valuation. Later the Department of Agricultural Economics, University of Illinois, assumed major responsibility for completion of the study. The basic data for the study were provided by the Property Tax Division, Department of Revenue, State of Illinois. This office obtains transcriptions of all bona fide transfers of real property from county deed records as a part of its periodic review of assessment levels and procedures. The authors wish to express special thanks to Richard J. Lyons, Director of the Department of Revenue, and E. L. Maynard, Supervisor of the Property Tax Division, for making the data available. Their guidance and counsel in technical matters regarding the use and interpretation of these data is gratefully acknowledged. Farm Real Estate Sales in Illinois FOLKE DOVRING and WILLIAM H. ScoriELD 1 ILLINOIS HAS LONG BEEN NOTED for its productive and high-valued farmlands. With a current (1963) market value of $9.9 billion, the level and trend in prices of farm real estate are of concern to many thousands of landowners, lenders, and tax officials. Sales of land pro- vide the most concrete and objective evidence of the prevailing prices being paid and of the process by which participants in the market ex- press their judgments with respect to the many factors that determine the price of land. Land transactions also provide opportunities to adjust the size of operating units to the family life cycle and to technological develop- ments. The ready market for land in the last two decades has made it possible for owners to realize substantial capital gains on farmland that had been acquired at lower prices. The availability of sales records for practically all bona fide farm- land sales in the state for a five-year period provided a unique opportunity to document and analyze the level, trend, and size charac- teristics of land sales in the various areas of the state. These records contained close to 16,000 sales of farm property for the period April 1, 1952 through March 31, 1957. The sales of farm property covered 1,261,000 acres of farmland, with an aggregate sales price of about $265 million. Several comparisons also were possible with other independent estimates of market values. Comparisons of the size characteristics of lands sold with farm operating units help document the extent of the parcel market and the key role it plays in the steady trend toward fewer but larger farms in the state. Some new information is also present on the relations between gross and net returns from farmland and sales prices. 1 Folke Dovring, Professor of Agricultural Economics, University of Illinois, and William H. Scofield, Leader of Farm Real Estate Values Investigations, Farm Economics Division, Economic Research Service, U.S. Department of Agriculture. Anthony M. Grano, now on the staff of the U.S. Department of Agriculture at Oregon State University, assisted Professor Folke Dovring with the study. BULLETIN NO. 697 [November, PERCENT OF 1947-49 AVERAGE 175 150 ILLINOIS 125 \ / 100 ' 75 50 25 1912 1920 1930 1940 1950 I960 Index numbers of average value of farm real estate (farmland and buildings) per acre, Illinois and United States, March 1, 1912-1962, 1947-49 = 100. (Fig. 1) Background Data for the State Long-term trend in market value of farm real estate and relationship to farm income The long-term trend in farm real estate values in Illinois has been remarkably similar to that in the United States as a whole (Fig. 1). In 1920, the national index did not rise quite as high as the Illinois index, and in the early 1930s, it did not drop quite as low. The Illinois curve thus tends to go a little farther than the U.S. curve in the more extreme changes, for in a more homogeneous area like Illinois, there are fewer contrasting trends canceling out each other. Some difference may be noted in the trend in farm real estate values during the last decade. In the 1940s, the general rise in values was quite similar, but between 1950 and 1960, the Illinois index advanced more than the U.S. index. In 1961, the difference again disappeared. Although the difference between 1950 and 1960 was moderate, it is of some interest. The 1940s were a period of inflation, and as a consequence, the rise in land values during that period was more apparent than real. In real (deflated) terms, land values advanced only moderately. The close agreement of the Illinois index with the national data in the 'forties is noteworthy. In the 1950s, by contrast, the value of the dollar changed slowly, and the rise in land values was real for the most part. The somewhat faster rise in Illinois than in the United States as a 1963] FARM REAL ESTATE SALES IN ILLINOIS - 400 - 3OO - 200 - IOO 1930 1935 I94O 1945 1950 1955 I960 Net farm income and farm real estate values, Illinois, 1929-1961. Net farm income includes rents paid to nonfarm landlords and interest on farm mortgage debt. Real estate includes farmland and buildings, March 1 of following year. (Fig- 2) whole during the 'fifties becomes more significant against this back- ground. The land price development in Illinois thus offers a good opportunity to analyze land price changes in the United States. The changes in farm real estate values are more meaningful when compared with changes in farm income (Fig. 2). During the 1940s, farm incomes rose more than real estate values, and in some of these years, returns from farm real estate in the United States were higher than those from common stocks. 1 By contrast, the rising trend of land values during the 1950s, in real terms, was accompanied by a decrease in the proportion of net farm income to real estate values. During the last few years, farm real estate has not even returned an interest equal to the current interest rate on mortgages. In both changes in farm real estate values and in farm income, the development in Illinois has generally been the same as in the United States as a whole. The average level of value per acre of farmland and buildings, according to the 1959 Census of Agriculture, is shown for each county in Illinois in Figure 3 (left). The percentage change between the 1959 and 1954 census data is shown in Figure 3 (right). 1 W. H. Scofield, "Returns to Productive Capital in Agriculture," The Farm Real Estate Market, February 1960 (July-November 1959). BULLETIN NO. 697 [November, DOLLARS PER ACRE STATE AVERAGE' 39 Average value per acre, 1959 Census of Agriculture (left), and increase in average value per acre in the 1959 census as percent over the value in the 1954 census (right). (Fig. 3) Rate of transfers by type An important background factor in changes in farm real estate values is the rate of transfers. Some main features of the rate of transfers are given in Figure 4, distinguishing between voluntary sales, forced sales (including both foreclosures and tax sales), and other sales. The latter group includes administrator and executor sales as well as sales resulting from inheritances and gifts and other mis- cellaneous and unclassified sales. There was a high total rate of transfer in the 1930s, especially in the beginning of the decade when the price of farmland was still going down. This high rate of transfer reflected the depressed condition of agriculture at that time. Forced sales were the largest category, while voluntary sales were less than one-third in some years less than one- fifth of all transfers. Apart from the direct effects of the depression upon incomes and hence upon land values, this high rate of forced sales is likely to have contributed to depressing land prices by creating a large supply of farmland on the market. 1963] FARM REAL ESTATE SALES IN ILLINOIS 100 80 2 Q 60 40 20 ^ J VOLUNTARY SALES AND TRADES 1925 1930 1935 1940 1945 1950 1955 I960 Estimated number of farm ownership transfers per 1,000 of all farms, by main type of transfer, Illinois, 1926-1958 (year ending March 1). (Fig. 4) This situation was drastically reversed during the 1940s. The number of forced sales decreased as prosperity came to agriculture, and voluntary sales and trades again increased and became the bulk of all transfers. The relatively constant land price level, in real (deflated) terms, as mentioned on page 6, is explicable if the high rate of trans- fers in the 'forties is seen in part as the aftermath of the depression and the high rate of forced sales in the 'thirties. The many forced sales led to an involuntary increase in the market supply of farmland. As long as the effects of this increase were felt, there was little tendency to bid up farmland in real terms. The abnormal conditions during a period of running inflation may also have contributed. Only to keep up with the declining purchasing power of money, land prices had to be raised every year. These increases in price may have been all that land buyers were willing to risk, since no one knew whether a period of deflation would follow the wartime boom. In the 1950s, the rate of transfers fell far below the levels of the 'thirties and 'forties. Forced sales accounted for only a tiny fraction of all transfers. Voluntary sales and trades also fell sharply and ac- counted for only about half of all transfers or about the same magni- tude as the relatively constant "other" categories. Since the 1950s, the 10 BULLETIN NO. 697 [November, farmland market has not been fluid, but has been governed by the rigidly restricted market supply. Land buyers have reacted by bidding up land, even in real terms. Around 1960, there seemed to be a new reversal of trend, probably a stabilization of the level reached. How- ever, it may yet be too early to judge the meaning of this. Rates of transfer are given as per 1,000 of all farms. Thus the absolute figure corresponding to a certain rate has been declining as the number of farms decreases (from 230,000 in 1930 to 221,000 in 1940, 203,000 in 1950, and 176,000 in 1958). * Moreover, the signifi- cance of the rate depends in part on the relative proportions of sales of entire farms and of land parcels for farm enlargement. 2 In the land market during the 'fifties, land purchases for farm enlargement were relatively more numerous than ever before. In many cases, the land parcels for farm enlargement were parts of previously existing farms which were divided at the time of transfer. In some other cases, they were isolated parcels changing hands between nonfarm investors (the "fleeting parcel market"). The rates of transfer in Figure 4, like the data on land sales used in this bulletin, refer to acquisitions of land rather than offerings. The rate of complete farm units being offered for sale probably fell even more sharply than the rate of purchases shown in the chart. Sellers and buyers of farm real estate To analyze the farm real estate market it is important to know who supplies land for sale. Data on sellers are available from market sur- veys since 1949. ~The~number of sales in these surveys is quite small, however, and the data can be used only to show some main features. Sellers were" classified as active farmers, retired farmers, estates, and nonfarmers. During the 1950s there were only slight changes in the proportions between these categories. Active farmers accounted for upwards of one-third of all sellers, but their selling activities appeared to be somewhat less frequent in the later part of the decade. Retired farmers, on the other hand, accounted for 15 to 20 percent of all sellers and were somewhat more active later in the decade. Estates were one- fourth to one-third of the land sellers on the market and non- farmers were from 15 to over 20 percent. 1 The 1959 census enumerated 154,600 farms. The minimum criterion of a farm was, however, somewhat more restrictive than before. J See Dollar Levels and Trends in Sales Prices, page 12. 1963] FARM REAL ESTATE SALES IN ILLINOIS 11 Data on buyers from similar surveys show tenants to have declined in relative importance from one-third to one-fourth, while owner- operators (one-third and over) and nonfarmers have gained. Part owners, in this case, are included with owner-operators. And since part owners are an increasing proportion of all farmers in Illinois, such a shift in buyer categories was to be expected. The slight increase in the role of nonfarmers agrees with the most recent trend, shown by the censuses 1950-1959, toward somewhat more land under lease in Illinois. 1 Financing land transfers A restriction on the farmland market is in the mode of financing and the degree to which buyers are able (and willing) to obtain credit to finance a purchase. Other recent survey data, available since 1944, show that the rising trend in value has been accompanied by an increasing tendency for land purchases to be credit financed. In the mid-forties, less than 40 percent of all sales were credit financed. The percentage rose through the late 1940s and has exceeded 50 percent since 1953. In the late 1950s, nearly two-thirds of all sales were credit financed, while the proportion was slightly lower (around 60 percent) in 1960-1961. The average ratio of debt to consideration on credit-financed sales has risen somewhat in*recent years. It remained below 50 percent throughout the 1940s and early 1950s, but has varied between 55 and 60 percent in the last few years. Behind both of these movements is the emergence of a method of financing which previously was not much in use installment land contracts. Although infrequently used in the early" 1940s, contract sales accounted for between 5 and 10 percent of all farmland sales in Illinois in the later years of the same decade. During the 1950s, their share increased rapidly. In 1960 and 1961, more than one-fourth of all farmland sales in the state were by contract. 2 1 For data on owner-operated and tenanted land in 1959, see Economics for Agriculture, Tenure and Appraisal, TA 12, by F. J. Reiss, Univ. of 111. Col. of Agr., Dept. Agr. Econ., 1962. 2 On installment land contracts as a method of transfer of farmland, see R. Vern Elefson and Philip M. Raup, Financing Farm Transfers with Land Contracts, Minn. Agr. Exp. Sta. Bui. 454, 1961 (also, North Central Reg. Pub. 122). Also, see Fred L. Mann, A Comparative Study of Laws Relating to Low- Equity Transfers of Farm Real Estate in the North Central Region, University of Missouri Res. Bui. 782, 1961 (also, North Central Reg. Pub. 136). 12 BULLETIN NO. 697 [November, Dollar Levels and Trends in Sales Prices Sources of data and limitations The commonly used indices of farmland price are based mainly on two types of data, both of which reflect the opinion of informed persons: census returns 1 and crop reporters' estimates. 2 Both farmers and crop reporters draw on a variety of commonly available information which should allow them to closely determine the realities of the farmland market. The time series derived from these two sources (Farmers and crop reporters) are also in fairly good agreement between themselves as regards the trend of change in farm- land values. The absolute level of price, however, shows some differ- ence between the two sources. This measure of agreement does not show how close the estimates are to the facts. The two sources draw on the same information and may also influence each other. The crop reporters have access to the census results, and the farmers may know the opinions of the crop reporters in their districts. This does not, however, invalidate any of the major findings on long- or medium-term trends in price change. But for short-term trends and detailed analyses, a source of information is needed that is independent of opinion data and that more directly reflects actual sales pricesjDf farms. In this report land sales data from 1952/53 through 1956/57 are used to show the character of the farmland market. Nearly 16,000 sales were analyzed, involving 1*4 million acres of farmland, sold for an aggregate price of about $265 million. (For detail on the data and their treatment, see Appendix A.) 1 The Census of Agriculture, taken every five years, asks farmers (among many other things) about the market value of their farms. Since 1950, the question on farm sales value has been asked only on a sample of approximately 20 percent of all farms. The rate of response to this question is high, but it is far from 100 percent. In 1954, the rate of response in Illinois was 77 percent of the sample farms ; thus, about 15 percent of all farm operators had answered this question. This results in possible sampling errors, even apart from any general tendency among farmers to understate or overstate the value of the farms they operate. 2 The crop reporters in each of the nine crop-reporting districts in Illinois estimate the current average selling value of farmland three times a year. They base their opinions principally on sales of farmland in recent months before each estimate. They also try to assess the influence of individual factors, such as soil productivity, on the selling price and to adjust for these in estimating what land of average quality would sell for. Separate estimates are made for land with and without buildings. 1963] FARM REAL ESTATE SALES IN ILLINOIS 13 Table 1. Sales and Acres Sold by Size of Tract, 1952/53-1956/57 Size class in acres Number of sales Acres sold Size group as percent of total* Number of sales Acres sold Under 3... 174 488 5,382 36,824 134,187 72,558 291,018 207,873 243,622 94,053 66,848 68,125 16,650 12,555 5,385 5,352 1,260,920 1.1 5.8 12.9 21.8 _ 7.9 22.8 11.3 9.8 3.0 1.8 1.4 .3 .1 100" .04 .4 2.9 10.6 5.8 23.1 16.5 19.3 7.5 5.3 5.4 1.3 1.0 .4 .4 100* 3-9 924 10-29 2,035 30-49 3,433 50-69 1,244 70-99 . . 3,604 100-139. 1,790 140-179. 1,545 180-219 477 220-259 281 260-379. . 223 380-499 . 40 500-699 . 22 700-999 . 7 1,000-4,999 4 5,000 and over Total, all sizes. . 15,803 1 Detail may not add up to exactly 100 because of rounding. General characteristics as to size and level of prices per acre, all sales Sales were distributed by size in acres as shown in Table 1. The average size of tract per sale was nearly 80 acres, less than half the average size of farms in Illinois (173 acres) at the time of the 1954 Census of Agriculture. The largest group of sales, both in number and acres transferred, was the size group 70-100 acres, comprising close to 23 percent of both. The second largest group in terms of numbers was 30-50 acres. High percentages of acreage were found also in the groups 100-140 and 140-180 acres. Sales of large farms were relatively few. The consequence of this on the analysis of the parcel market versus the whole- farm market will be discussed later. Average price varied with the size of tract in a manner inconsistent with the per-acre values by size of farm shown in the 1950 and 1959 Censuses of Agriculture. 1 The census valuations do not reflect much relationship between size and per-acre value. Only the smallest sizes had values higher than the general average, evidently because of the greater importance of buildings. The sizes between 140 and 500 acres 1 The 1954 census did not report per-acre values by size of farm. 14 BULLETIN NO. 697 [November, (which have the bulk of the farm acreage) all had values close to the general average, while the sizes above 500 acres had lower values. Those between 50 and 140 acres showed lower values in 1950 but not in 1959. When differences in cropland percentage were taken into account these differences largely disappeared; then all size classes above 50 acres including the largest farms were close to the average. 1 The sales prices show a more complex pattern (see Appendix Table 4). Taken without further analysis, they would indicate (among other things) that the largest tracts sold for average prices below the average value of large farms in 1950, while at the same time the general (unweighted) average rose by 20 percent between 1950 and 1954. In the group over 1,000 acres, this might be fortuitous since there were only four sales. But there were 29 sales in the group 500-1,000, index 84, and 263 in the group 260-500 with index 94. In the groups 180-220 and 220-260 acres, with indices well below average, the number of sales was also substantial. Data by areas within the state show analogous traits. Despite the substantial number of sales, the low average prices in these groups cannot be accepted as characteristic of the value of all farms in the same groups. The discussion above revealed the effect the percentage of cropland had on the values reported in the census. The most logical interpretation of the low values of large farms in the sales data is that they reflected lower cropland percentages in the large farms sold as units than among the universe of farms in the same size group. To some extent, this variation in price by size of tract was influenced by the distribution of sales between areas of the state. Straight average sales prices per acre by counties for all five sales years in the sales data, in comparison with the county average values in the 1954 census, are shown in Appendix A, Figures 13 and 14. The two series of data are in fairly close agreement. Only about 15 counties showed more than 20 percent difference. Among these counties, the sales value was higher than the census value in more cases than it was lower. These upward deviations included a number of counties with strong urban influence, such as Du Page, Lake, St. Clair, and Madison. This could be expected, since urban influences on the land values were more likely to be reflected in market sales than in farmers' responses to a census question. Otherwise, there is a reason- ably close correlation between sales prices and census values. 1 For this adjustment, average value per acre was multiplied by average total acreage per farm in the size group, and this total value was divided by average cropland acreage per farm in the size group. 1963] FARM REAL ESTATE SALES IN ILLINOIS 15 When the farm acreages in each county were used as weights, the state average of sales prices was increased from $211 to $239 for all five sales years combined. For sales year 1955 (April 1954-March 1955), which corresponded most closely to the 1954 census, the average weighted by county farm acreages was $234.5 (against $220 as the unweighted average for the same year). 1 This is encouragingly close to the census data. The average in the 1954 census was $231.47. Data from the USDA, based on censuses, crop reporters' estimates, and other sources, indicate $242 as average for the years 1953 to 1957. Apparently, the census is quite good as a source of knowledge about land values. County averages were weighted also by township farm acreages. (See Farmland value areas in Illinois, page 34, and Appendix A.) In some cases, as in the northeastern counties, it is quite clear that sales for nonagricultural uses have raised sales prices above those that apply to farmland. The same explanation applies to some extent also in southwestern Illinois (East St. Louis, Carbondale). However, several counties in this area had sales prices very much above the census level. These included rural counties like Clinton, Washington, Perry, and Union. It is, therefore, probable that the census values were actually on the low side in these parts of the state, for one reason or another. 2 West-central Illinois, with very nearly the same level in all the data, is probably the most purely agricultural part of the state. The time series and the census data The time series of state data are shown in Table 2, data for each of the areas in Figure 5. Sales data weighted by county farm acreages from the five sales years were compared with census averages from 1950, 1954, and 1959. x The five-year average was rather close to the center year (1955) average. This is of interest as a condition for analyses based on all five years added together. Analyses where township data are used must perforce utilize totals for all five years, since yearly sales in a township are often very few. 1 In certain counties the deviations of sales prices from the 1954 census average were such that substituting the sales prices for census values in a com- parison with 1959 census data would eliminate extremes in the price trend from 1954 to 1959. Thus, for instance, Massac County had sales price averages of 157 (unweighted) or 160 (weighted) percent of the 1954 census average; the apparent intercensal value increase from 1954 to 1959 was 84 percent. That is more than twice the statewide average value increase between the censuses (39 percent). Conversely, Scott County, for instance, had a sales-to-census ratio as low as 76-74, and a value rise for 1954-1959 of only 4 percent. These may be individual indications of weak spots in the 1954 census. There were, however, also cases to the opposite effect, such as in Putnam and Schuyler counties, and a scatter diagram failed to show any statistically significant relation between the sales-to-census ratio and the 1954-1959 value increase. 16 BULLETIN NO. 697 [November, Table 2. Dollar Levels and Index Numbers, 1952/53-1956/57 Value per acre Index numbers" Sales year Sales values USDA esti- mates Sales values USDA esti- mates Un- weighted Weighted 13 Un- weighted Weighted b 1953.. . $181.70 $222.60 227.70 234.50 243.40 265 . 20 238.70 $225 . 70 229.70 233.55 248.50 274.65 242.40 86.1 93.1 104.2 106.3 110.3 100.0 93.3 95.4 98.2 101.9 111.1 100.0 93.1 94.8 96.3 102.5 113.3 100.0 1954 . 196 60 1955... . 220 00 1956... . 224 40 1957 . 233.40 Average . . . 211.20 Using the average of the five sales years as base = 100. b By farm acreage in each county, 1954 census. c Unweighted average of the five yearly figures. The area weighting has considerably increased the agreement between the sales values and the USDA estimates, as indicated in Table 2. This applies to all the dollar figures. The change from 1953 to 1955 was greater in the sales data than in the USDA estimates. It represented, however, less than two-fifths of the difference between the census averages of 1950 and 1954. This confirms that there was a lull in the price movement from 1953 to 1955, even though perhaps somewhat less marked than indicated by the USDA figures. The change in the sales data from 1955 to 1957 was almost exactly two- fifths of the difference between the average values in the censuses of 1954 and 1959. The index numbers also showed greater agreement when the weighted sales data were compared with the USDA estimates. In most areas, the trend of census averages forms an almost straight line. And in most areas, the sales averages also can be inter- preted as representing the same trend and level as indicated by the census data. More sizeable divergencies were at hand in two or three areas. In area 1, which consists of a few counties to the north and west of Chi- cago, the sales data indicate that the census trend between 1950 and 1954 ought to be curvilinear. That is quite possible in view of the influence from a large urban land market. On the whole, these observations confirm that the census data and the sales data were in rather close agreement. Such differences that occurred were in most cases moderate and may well have depended on the size of the sample. 1 1 The number of sales in an area in a given year tends to be between 200 and 300, but this frequency varied sufficiently to permit incidental circumstances to influence some of the averages. 1963] FARM REAL ESTATE SALES IN ILLINOIS I960 '52 '54 '56 '58 '59 1950 '52 '54 '56 '58 '59 17 150 100 50 150 100 50 150 100 50 150 S 100 cc H! 50 150 IOO 50 I 1 I l 1 I I I - SALES DATA O CENSUS DATA --- INTERPOLATION BETWEEN CENSUS DATA .-O AREA I I I l I l I l ii ---- o-- AREA 3 i 1 l i i l i l ii AREA 4b i l l i i i I i i l .-O AREA 6 I l l l I .-O AREA 8 l l i l I l l l l l --- AREA 2 I l I I I I I I l l O AREA 4a i l l I I i i i l I cr" AREA 5 l i l l l l l I l i O ' AREA 7 I l l l i l i i l l >0 O AREA 9 i l i i i i i l l l 1950 '52 '54 '56 '58 '59 1950 '52 '54 '56 '58 '59 Index numbers of the sales data, five-year averages, and census data, 1950, 1954, and 1959, by type-of-farming areas. In each area the sales data average for five years (1952/53-1956/57) is the base = 100. (Fig. 5) 18 BULLETIN NO. 697 [November, Type-of-farming areas in Illinois. (Fig. 6) BY COUNTY BOUNDARIES - BY NATURAL BOUNDARIES This measure of agreement does not extend to the direct crop- reporter estimates as far as the level of price is concerned. The crop-reporter estimates, which are not published as dollar figures but used as a basis for index numbers, showed on the whole a somewhat higher price level than both the census and the sales data. As index numbers, the crop-reporter estimates were closer to the sales data than they were as absolute figures. The period covered by the sales data contained only a part of the decennial change from 1950 to 1960, and precise differences of trend between areas were hard to pin down. Over the longer run, since 1910 for example, census data indicate that the value movements in 1963] FARM REAL ESTATE SALES IN ILLINOIS 19 the type-of-farming areas have been rather parallel and on the whole similar to that of the state (Fig. 1). Marked differences between areas occurred principally in the latter half of the 1950s. For the state as a whole, the price level (in current dollars) rose about 65 percent in the period 1950-1959. Rates of increase not very far from the average were found in most areas of northern and central Illinois. Higher rates (doubling or more) were seen in the Chicago area and most of southern Illinois. Price differentials between improved and unimproved properties, by size class More than two-thirds of the farmland parcels in the sales data were classified as improved with buildings, and these included over four- fifths of the acreage transferred in these bona fide sales. The totals are as follows: Number Acreage Average size of sales sold of tract, acres Improved tracts 11,956 1,036,823 87 Unimproved tracts 3,847 224,097 58 Total 15,803 1,260,920 80 The average price was higher on improved than on unimproved tracts. The unweighted average price was $217 on the former and $184 on the latter. To understand this, some differences in structure between the two groups of sales should be noted. Sales classified by improvement class and size class are shown in Table 3. Unimproved tracts were smaller on the average, showing a more marked preponderance for the size groups 30-49 and 70-99 acres. This is as could be expected, since these classes consisted almost entirely of land parcels intended to be added to existing farms. There were only 48 sales of unimproved tracts above 220 acres in size. Separate average prices by improvement class and by size class are shown in Table 4. Both improvement classes showed the same trend of highest price in the smallest size classes, the price falling thereafter as size increased. This trend was interrupted by markedly higher prices in the size class 70-99 acres than in the preceding class and also, among the improved tracts, in the size class 140-179 acres. One possible explanation for this interruption might be that parcels of these sizes were placed on the market relatively often because of subdivision at the time of sale of farms having a high per-acre value. The price difference was moderate, however, and the average prices in these sizes still did not fully reach the weighted average based on 20 BULLETIN NO. 697 [November, Table 3. Percent Distribution of Improved and Unimproved Tracts, by Size Class, 1952/53-1956/57 Size class in acres Percent distribution of Number of sales Acres sold Improved Unimproved Improved Unimproved Under 3.. 1 3 .5 3.6 19.1 34.7 8.4 22.0 5.6 3.9 1.0 .5 .5 .1 .02 .05 100 .04 .4 2.2 7.9 5.2 21.5 17.6 21.3 8.3 6.0 6.0 1.4 1.2 .4 .5 100 .03 .4 6.1 23.2 8.4 30.3 11.1 10.3 3.4 2.2 2.8 .7 .3 .7 100 3-9. 6 6 10-29 10 9 30-49 17 5 50-69 7.7 70-99 . . 23.0 100-139. . .. 13.2 140-179. 11 7 180-219. 3 6 220-259. 2 2 260-379 1 7 380-499 .3 500-699 .2 700-999 ... .04 1,000 and over.. .03 Total,* all sizes. 100 * Detail may not add up to exactly 100 because of rounding. county farm acreage (page 15). The explanation may well be, there- fore, that these parcel sizes were somewhat more frequent in high- value than in low-value areas. The price of unimproved land was a relatively small percentage of the improved land price in the lowest size groups, and thereafter rose with increasing size. This should thus reflect the decreasing share of buildings in the total farm value (Table 4, column 5). Less easy to anticipate is the fact that this value ratio climbed to 90 percent and over in the size classes between 30 and 100 acres, and then went down again. A clue to this may be in the assessment of land and buildings on improved land. Building assessment usually is 20-30 percent of total value, and the same or even higher ratios are applicable to properties between 30 and 100 acres. The closeness of the prices paid for unimproved and improved land in these size classes rather indicates, then, that the presence of buildings in these size classes was of little concern to buyers, most of whom bought land to enlarge an existing farm. The price ratios on larger farms, on the other hand, showed more difference between improved and unimproved land than would be ex- pected from the ratio of building assessment to total assessment, which is often quite low in these size groups. The decrease in price per acre with increasing size of tract was sharper on unimproved than on im- 1963] FARM REAL ESTATE SALES IN ILLINOIS 21 Table 4. Average* Price per Acre of Improved and Unimproved Land, by Size Class, 1952/53-1956/57 Size class in acres All sales Improved tracts Un- improved tracts Col. 4 as percent of col. 3 Under 3.. $2,824 $3,067 $983 32 3-9 1,364 1,500 629 42 10-29 . . 272 297 228 77 30-49. 197 204 184 90 50-69 189 192 181 94 70-99 223 227 213 94 100-139 195 200 163 82 140-179 220 228 148 65 180-219 190 195 129 66 220-259 . ... 202 206 144 70 260-379. 170 180 69 38 380-499 . 145 155 b b 500-699 154 145 b b 700-999 b b b b 1,000 and over b b Average*. ... 211 217 184 85 a Unweighted averages. b Less than 10 sales. proved land. This might be explained if the tendency to subdivide the better tracts at the time of sale and to sell the less fertile ones un- divided is assumed to be more marked on unimproved land than on land with buildings. The over-all ratio of building assessment to total assessment of improved land in the sales material was close to 22 percent. This is not altogether inconsistent with the ratio between value of improved and unimproved land shown in the table. If unimproved land sells for 85 percent of the price for improved land, this would indicate an average building value of 15 percent of the value of the improved properties. The difference against the building-assessment ratio may be explained, as discussed above, by the fact that buildings become superfluous in many farm enlargement situations. Size of tracts sold in relation to size of farms the parcel market versus the whole-farm market The proportion of farmland sales that were for farm enlargement can to some extent be gauged from the sales data. The size distribution of sales permits certain comparisons with census data to show how the rate of transfers relates to the rate of disappearance of farms in the process of farm enlargement. 22 BULLETIN NO. 697 [November, Table 5. Size Distributions of Sales and Intercensal Changes in Farm Numbers by Size of Farm Size class in acres Number of sales, 1952/53- 1956/57 Change in farm numbers, 1950-1954 Change in farm numbers, 1954-1959 Intercensal change in farm numbers as percent of number at beginning of intercensal period 1950-1954 1954-1959 Under 3.. 174 + 6791 -3,463/ -3,8161 -2,267/ -1,644 -3,416 -4,090 -3,355 - 961 + 124 + 1,851 + 615 + 18 -19,725 -4,980 a -3,215 - 920 -3,029 -4,520 -5,701 -2,568 - 985 +3,185 + 1,686 + 148 -20,899 a + 25.51 -30.5/ -24.3) -18.5/ -19.3 -16.6 -16.4 -10.6 - 5.0 + 0.8 + 6.7 + 15.8 + 4.4 -10.7 -44. 4 a -14.7 -13.4 -17.7 -21.6 -20.1 -13.9 - 6.0 + 10.8 +37.4 +34.7 -11. 9 a 3-9. 924 10-29. 2,035 30-49 . . 3,433 50-69 . . 1,244 70-99 . . 3,604 100-139 . . 1,790 140-179.. . . 1,545 180-219 477 220-259 281 260-499 . . 263 500-999 29 1,000 and over 4 Total, all sizes 15,803 a Partly because of change in definition of a farm. The number of bona fide sales by size of tract in the five-year period under study is shown in Table 5 in comparison with the changes in farm numbers in the census periods 1950-1954 1 and 1954- 1959. The total disappearance of farms was larger than the number of bona fide sales. Changes in farm numbers were, of course, also in- fluenced by the market for farmland rentals. Bona fide sales, more- over, were only a part of the transfers in fact, less than half. The total number of transfers in these five years was around 35,000, or less than twice the disappearance of farms in each of the two five-year intercensal periods. This point needs further clarification. When two small farms are added together, it usually leads to a diminution by two in the number of farms in the lower size classes and to the addition of one unit to a higher size class. Thus there is a net disappearance of one. When entire farms are bought and added to existing farms, the diminution in farm numbers should equal the number of such transactions. If a farm is enlarged by buying two or more other farms to add to the existing one, the total disappearance of farms would still equal the number transferred for such purpose. 1 The 1950 census was taken April 1, the 1954 census in November. The intercensal period was thus closer to five than to four years. 7963] FARM REAL ESTATE SALES IN ILLINOIS 23 On the other hand, when a large farm is sold as several parcels which are used for enlarging existing farms, only one farm would dis- appear in the total, although several others become larger (that is, sev- eral units are subtracted from lower and added to higher size groups). The rate of transfer (that is, of purchase) will be higher than the rate of disappearance of farms. The total number of all transfers in a five-year period, or 35,000, may be compared with the rate of farm disappearance in a five-year period, or about 20,000 (possibly somewhat less because of change in definition). The rate of transfers of entire farms continuing as entire farms would thus be approximately 15,000 or less. Both in 1950-1954 and in 1954-1959, farms above 140 acres in size diminished in number. In other words, the rate of disappearance of medium-sized farms was higher than the rate of increase among the large farms. In 1950-1954, the number of farms between 140 and 220 acres diminished by 4,316, while those over 220 acres increased by 2,608. In 1954-1959, farms between 140 and 260 acres decreased by 9,254 units, while those over 260 acres increased by 5,019. Thus, even medium-sized farms were enlarged. Also, many of them were used for farm enlargement, either by being added undivided to an existing farm or by being subdivided into smaller parcels. When subdivided, they show up in the sales material among the smaller size groups. If it is accepted that out of 35,000 transfers some 20,000 (or rather more) were for farm enlargement, and 15,000 (or less) for con- tinued existence as farms, then it is clear that rather more than half the bona fide sales must have been for enlargement purposes. The pro- portion is likely to be tilted towards enlargement purchases among the bona fide sales, since family transactions would tend to maintain farms as undivided somewhat more often. This conclusion is supported also by the size distribution of sales (see Figure 7 for the state as a whole and Table 6 for type-of-f arming areas). Most of the sales were in the small size classes, those in which farms are getting fewer. Only a few hundred sales were in the size groups where farm numbers are increasing. Over 11,400 (72 percent) were under 100 acres in size. As far as numbers of transfers are concerned, the sales data reflect, above all, the parcel market. As regards area, the proportions are, of course, different. Sales of tracts under 100 acres accounted for 43 percent of the acreage in the sales data, and tracts 100-220 acres in size accounted for another 43 percent, leaving barely 14 percent in the sizes over 220 acres (see Table 1). In the 1954 census, these three main groups occupied 8, 35, o I I I 1000000000 c\J c\J ro ro *- O CM CO O f O ^" CO CO t- Ci ' OO CM CO oo St^.'^'IOOCO COOOCO -t^ OO ** OO CO CM ^H OOOOi-" to ^H CMt>-COt^. W3COCM ^ CM C^OCOtO-*-* tMCS^J-OO*-*- sssgss Sliisl 1 I-l + ggggSgg o ^-1 l-l s Ot>-^CO-^eO CMt^CMCOOO * O Os OS OS t-- t>* *-* OO O CO i-t OO W5 O CO r-t O CM r- CO OS CO OO I-- CO O CD CO -H 2 ^* O O CO CM -^ CO OO O O OS *< OO i 'CO OS CM CO CO i-l OS O CO OS CO OO i-l -C t- OO O I-H CO~CO(M 0*OCM~H <<}< W) CM^tOSW3CM O^POO**' *O 0> SSo?cot222 S^oco * " " S s c OSO- 2 ig area to fecOCMi-HCCCOCO CO CM i-t O i- ^ te co'^ ""CM S COCOCOCMS CO H ^ + "^ CO CM CO OO CO r CO Tf U5 CO -sf OS t- CO Tj- oo xr oo r 1>- ^cooo-^fot^- "^ MH o I ^-i-toococMb* ot-^eoco- O ja OOOCOCMO OO^OCOO*-O "- o O -^ CO CM ^^ 3 CM 4r -O CO OS T CO -^ 5 OTJ*OOOCM oocooccoo os -a-35ss Oi 0) u CO CO 1^- OO iO UdCOCOOOiOlOOO -^ eS siliii sfHis ' I i^ -^ CM -^ - -l f- 4> bD OOSCMCMOS^O O OS OS O OS O N sislsi 1 -