V,* 1 r'o. Division of Agricultural Sciences r\ \\ ;, \ V UNIVERSITY OF CALIFORNIA SOME CHARACTERISTICS OF THE FARM REAL ESTATE MARKET IN CALIFORNIA WITH EMPHASIS ON TRANSACTIONS IN IMPERIAL AND TULARE COUNTIES W. E. JOHNSTON CALIFORNIA AGRICULTURAL DIIIICTIM ftCA EXPER.MENT STAT. ON B ™J!U r-H ^ O O C 00 b- •— 1 i-H i-H O fc «g -3 -c M ■S 9 rt< (N N O - t^ 00 co ^C a 2P 0) 03 - ,o 00 C o 3 !«. aOS IB Oi s 0) to "8 T^ t>- CO O — 8 § co os co ° CN CM CO ^H ,-H •rt 3 5^ a> 5» to tf S4, 0) a* a w in © ": »0 O0 O i-H O 03 g l-H i-H •■z u ss M N M N C^ 00 OS 10 O CO (N CO i-H (M £a > ►» t- -Q -^ ■+-» -Q "^ -^> _o II 3 3 ^ 3 3 'C -O O 73 O Ph 0) +^ TJ -1— T3 -^ 1. +-> *o h— "O ++ C3 0) f_ CU fi x fl -C a) fi43 fl j3 a a) += ^ 3) -P O -P O Ph O Ph c OPhOPhO c _o 03 a 3 F-i ^ o o> 0) o "3 rmer. mfarm rmer. >nfarm 02 c8 c3 pc y Ph 2; o5 o «S5-3 'CSS - M c « 5*0 03 ^ i 03 3 ft 03 T( Wig S- x -^73*o JJ-2 «, o ^ 9 o k .2 a a S 5 03 « o g-0-0 03 O © o> aj -o^-H^ per cent of sales reported the reason for selling as "unsatisfactory investment" com- pared to 26 per cent in this study (table 4). Further, somewhat higher proportions of sales were indicated to have been sold for profit realization (35 versus 24 per cent) and for retirement reasons (30 ver- sus 19 per cent). Other reasons identified were "buy another farm," 9 per cent; "take another job," 2 per cent; and "other," 7 per cent. Types of sellers are also identified in annual survey data for the Pacific farm production region. Over the period 1963-69, active farmers con- stituted about 50 per cent of all sellers, re- tired farmers, 12 per cent; local nonfarm- ers, 13 per cent; absentee owners, 16 per cent; and estate sales, 8 per cent. Age and Education of Buyers Resident buyers were typically younger than nonresidents according to results given in table 7 (see also figure 2). All pur- chases by buyers less than 30 years old were by resident buyers. Buyers 60 years old or older accounted for only 13 per cent of all resident purchases, whereas 39 per cent of the nonresidents purchasing land were 60 years old or older. In general, resi- dent and nonresident buyers appear to be older in the Imperial Valley subarea than in the Tulare County subarea. The age distribution of buyers in both study areas (table 7) differs from that pre- viously reported by the USDA for the Cali- fornia specialty farming area (based on an October, 1964, survey), only in the fact that it is somewhat more skewed in the upper-age classes. The age distribution is quite similar for the three youngest age groups — 3 per cent, less than 30 years of age; 16 per cent, 30 to 39 years; and 35 per cent, 40 to 49 years — but the reported 33 per cent for ages 50 to 59 and 13 per cent Table 7 AGE DISTRIBUTION OF BUYERS FOR 131 TRACTS, TULARE COUNTY AND IMPERIAL VALLEY SUBAREAS Type of buyer Both study areas Resident Nonresident . Imperial Valley Resident Nonresident . Tulare County. Resident Nonresident . Both study areas Resident Nonresident. Imperial Valley Resident Nonresident . Tulare County. Resident Nonresident . Total purchases 131 95 36 58 38 20 73 57 16 100 100 100 100 100 100 100 100 100 20-29 Age of buyer 30-39 40-49 50-59 60-69 number 70-79 80-89 4 24 47 29 18 7 4 19 39 20 9 4 5 8 9 9 3 1 11 16 15 8 5 1 9 13 9 3 3 2 3 6 5 2 3 13 31 14 10 2 3 10 26 11 6 1 3 5 3 4 1 per < zent* 3 18 36 22 14 5 4 20 41 21 9 4 14 22 25 25 8 2 19 28 26 14 9 3 24 34 24 8 8 10 15 30 25 10 4 18 42 19 14 3 5 18 46 20 10 2 19 31 19 25 6 2 6 3 10 *Rounded and therefore may not add up to 100 per cent. nil ,-n 50 r Both study areas >> u a 0) 3 cr u fa 30 20 10 50 40 30 20 10 JZL j Resident buyers :K jgft ^Nonresident buyers I I I I ■ 20-29 30-39 40-49 50-59 60-69 70-79 80-89 Age Imperial Valley 1 Resident buyers II ^ ^Nonresident buyers Ml ■ 20-29 30-39 40-49 50-59 60-69 70-79 80-89 Tulare County ar Age Resident buyers Nonresident buyers 20-29 30-39 40-49 50-59 60-69 70-79 Age Fig. 2. Age distribution of buyers, Tulare County and Imperial Valley subareas. [12] for ages 60 or older as compared to 29 and 27 per cent, respectively, in this study, ap- pears to be significant. (Although this pres- ent study did not ask for estimates of sellers' ages, a survey taken in October, 1965, revealed that 14 per cent of the sellers in the California specialty farming area were less than 45 years of age, 43 per cent were less than 55 years of age, and 73 per cent were less than 65 years old. Buyers were also asked their level of ed- ucation; results are summarized in table 8. The typical buyer in all classifications was a high school graduate, and one-fifth of all buyers had completed at least 4 years of college. Nonresidents were twice as likely to be college graduates than were residents. Buyers' Source of Information Buyers were asked how they heard about the tract of land that they ultimately pur- chased, and their answers are summarized in table 9. Buyers of land in the Imperial Valley subarea had relied somewhat more heavily on information directly from the seller (51 per cent). In contrast, buyers in the Tulare subarea first heard about the tract being for sale more often from a broker or agent (41 per cent) rather than directly from the seller (31 per cent). In both subareas these two sources of infor- mation were the most important for both residents and nonresidents, although non- residents had, understandably, relied more upon brokers or agents rather than direct contact with sellers in learning that the tract of land was for sale. Nonetheless, nonresidents also relied on direct informa- tion from the seller to learn about 35 per cent of all tracts purchased. The unex- pected significance of this source of infor- mation is probably due to the fact that Table 8 EDUCATION OF BUYERS FOR 124 TRACTS, TULARE COUNTY AND IMPERIAL VALLEY SUBAREAS Type of buyer Total purchases Education (in years) 4-7 8-11 12-15 16-19 Both study areas Resident Nonresident Imperial Valley Resident Nonresident Tulare County 124 93 31 54 36 18 70 57 13 7 3 4 4 4 3 3 number 25 21 4 7 5 2 18 16 2 67 54 13 34 26 8 33 28 5 25 15 10 9 5 4 16 Resident Nonresident 10 6 Both study areas Resident Nonresident 100 100 100 100 100 100 100 100 100 6 3 13 7 22 4 5 per cent* 20 23 13 13 14 11 26 28 15 54 58 41 63 72 45 47 49 38 20 16 32 Imperial Valley. . 18 Resident Nonresident Tulare County Resident Nonresident 14 22 23 18 46 *Rounded and therefore may not add up to 100 per cent. [13 Table 9 SOURCE OF INFORMATION ABOUT TRACT OF LAND PURCHASED FOR 169 TRACTS, TULARE COUNTY AND IMPERIAL VALLEY SUBAREAS Total purchases Source of information Type of buyer Directly from seller Broker or agent Mutual friend Community knowledge Advertisement Both study areas Resident Nonresident Imperial Valley Resident Nonresident Tulare County Resident Nonresident 169 123 46 99 68 31 70 55 15 72 56 16 50 37 13 22 19 3 number 60 19 41 11 19 8 31 9 20 5 11 4 29 10 21 6 8 4 9 9 4 4 5 5 9 6 3 5 2 3 4 4 Both study areas Resident Nonresident Imperial Valley Resident Nonresident Tulare County Resident Nonresident 100 100 100 100 100 100 100 100 100 43 46 35 51 54 42 31 35 20 per 36 33 41 31 29 36 41 38 53 cent* 11 9 17 9 7 13 14 11 27 5 7 4 6 7 9 5 5 7 5 3 10 6 7 *Rounded and therefore may not add up to 100 per cent. several nonresident buyers purchased more than single tracts of land over the period of this study. This is particularly evident in the Imperial Valley subarea, where nonresident purchasers evidently first heard of the land for sale directly from the seller in 42 per cent of the trans- fers but nearly one-fifth (6 of 31 sales) of the nonresident purchases in the Imperial Valley subarea were accounted for by one buyer. The source of information from mutual friends of both buyer and seller appeared to have been relatively more im- portant to nonresident buyers than for residents. Community knowledge was use- ful only to resident buyers in finding land for sale. Buyers' Search for Land Buyers were asked if they had actively searched for land prior to their purchase. Of the 124 buyers who responded, 57 or 46 per cent indicated that they had been involved in an active search for land (table 10). Purchasers of land in the Imperial Valley subarea had been somewhat more active in their search for land (49 per cent). Buyers in the Tulare subarea had been less active as a whole (43 per cent), but one-half of all purchases by nonresi- dent buyers in that area had been pre- ceded by active search. A total of 46 respondents revealed the geographical extent of their search. Al- though there was considerable variation between the subareas and among the type of buyers (partly due to the small number of responses), 78 per cent had confined their search to the county where they bought the land (Imperial or Tulare), while 11 per cent had also searched in adjacent counties, and the remaining 11 [14] w PQ o tf Ph 2 O S W ~ W Q ^ s w o T— C5^(NCO(NOCOO s^ 1 rH tH Csl rH i-H <_. W ej Pi. in u C H iC OOiMOO.— i. 1> WNOOOiiONiCO "c8 c o o $4, TJ c 0) n C3 VI ec COOrHCOtOlOO>C cc a3 i-O g CC N^tDNOiO»C»0 O -t CO rH C oq cS Oi > «H O S-. C Qi < t^ oicooioioooooo >r: CO rH » 3 Z2 r "o •- a. >^ o> (. ■+3 O) -^ P +3 ft K Pi ^ « -e c >> a> o3 TT «j St^ So « S CA 'S Ch U "5q fn S "m fl ^ (uOOJoO^a)© C tf 55 »rt £ -S Ph ^ PC H per cent had also searched elsewhere (be- yond adjacent counties) in California. Buyers were not asked how long they searched prior to their purchase, but 25 buyers did indicate that they had an effec- tive time limitation on the period in which they could purchase because of the desire not to be held liable for capital gains taxa- tion on parcels of land sold previously. Most of these purchases were by active or retired farmers. Of the 25 buyers, 19 pur- chased land in the Imperial Valley and 6 in the Tulare County subarea. Buyers were also asked if they had ex- perienced difficulty in finding the type of land or farm they wanted (table 11). There was generally a poor response to this ques- tion, largely because few buyers had much in the way of previous experience in the land market on which to make valid com- parisons. Thus, any response is subjective and conditioned by the degree of past in- volvement in the farm real estate market — buyers who had difficulty would probably be more inclined to answer than those who found no problem in their search. None- theless, the majority of the responses gen- erally indicated no difficulty, although the typical nonresident buyer as a whole, and in Imperial Valley in particular, indicated some difficulty. It is also of interest that about one fifth of resident buyers in Tu- lare County indicated that the search was very difficult, but that only 3 per cent of all buyers in the Imperial Valley re- sponded in such a manner. The Price Determination Process Replies were received from 131 resident buyers and 49 nonresident buyers as to Table 11 OPINION OF BUYERS ABOUT THE DEGREE OF DIFFICULTY IN FINDING THE TYPE OF LAND THEY WANTED, FOR 48 TRACTS ONLY, TULARE COUNTY AND IMPERIAL VALLEY SUBAREAS Total purchases Degree of difficulty in finding land Type of buyer No difficulty Some difficulty Extremely difficult Both study areas Resident 48 32 16 29 18 11 19 14 5 UU7 27 20 7 16 12 4 11 8 3 nber 17 9 8 12 6 6 5 3 2 4 3 Nonresident 1 Imperial Valley 1 Resident Nonresident 1 Tulare County Resident Nonresident 3 3 Both study areas Resident Nonresident Imperial Vallev 100 100 100 100 100 100 100 100 100 per 56 63 44 55 67 36 58 57 60 cent* 35 28 50 41 33 55 26 21 40 8 9 6 3 Resident Nonresident Tulare County Resident Nonresident 9 16 21 ^Rounded and therefore may not add up to 100 per cent. [16] how they and the seller reached agreement on the final sale price. Table 12 summar- izes this information. However, the price bargaining process showing the three de- sired components of seller's asking price, buyer's offer price, and final transfer price is complete for only 131 of the 180 transfers. The final transfer price was equal to the seller's asking price for 82 of the 180 trans- fers. In 76 of the 82 transfers the buyer's initial offer price was also equal to the seller's asking price; in the remaining six cases the buyer initially offered less than the seller's asking price although the buyer subsequently accepted the seller's offer as the final transfer price. The land market in the Imperial Valley appears to have been more a seller's market than in the Tulare County subarea, since the final transfer price was equal to the seller's ask- ing price in 48 per cent of all transfers (as compared to only 42 per cent in the latter area). The dominance of a seller's market in the Imperial Valley is even more appar- ent if only transfers with indicated seller's prices are used for the analysis. Of the 70 responses in the Imperial Valley, 51 trans- fers (73 per cent) had final transfer prices equal to the seller's offer price, whereas that was true for only 51 per cent of the transfers in the Tulare County subarea. The final price was less than the seller's asking price for 49 tracts or 27 per cent of the transfers in both areas. These sales accounted for 37 per cent of all transfers with known asking prices of sellers. The ability of buyers to reduce final transfer prices below the seller's asking price was, as indicated previously, less in Imperial Valley than the Tulare County subarea. Only 18 per cent of all sales in the former subarea had final transfer prices less than seller asking prices, while 41 per cent of transfers were at price levels less than ask- ing prices in the latter subarea. In 63 per cent of the instances where price conces- sions were achieved, final transfer prices were equal to the counter offer prices of the buyers. The remaining transfers (37 per cent) were at price levels which were compromises between initial seller asking and buyer offer prices. It also appeared that resident buyers were more successful in achieving price concessions from sellers. Respondents did not indicate seller ask- ing prices for 26 of the 180 transfers. This could result from the inability of the buyer to recall the initial seller asking price, but was more generally descriptive of those instances in which the seller did not know the approximate market value of his land or did not choose to reveal an acceptable price level to the buyer. Rather, the prospective buyer was asked to first make a firm offer for the tract under con- sideration. The buyer's offer price was sub- sequently accepted in 89 per cent of the transactions emanating from this price de- termination process. The remaining 1 1 per cent of the transactions resulted in a final transfer price at a level above the initial offer price of the buyer. Nonprice Bargaining and Buyer Awareness of Competition Buyers of only 15 per cent of the tracts transferred reported bargaining with the seller on considerations other than price (table 13). Negotiations about terms of sale (other than sales price) constituted the ma- jority of the instances where nonprice bar- gaining was reported, and was relatively more characteristic of resident than non- resident buyers among those buyers report- ing some nonprice bargaining activity. However, as a per cent of total purchases, a higher percentage of nonresident than resident purchasers engaged in bargaining over terms of sale. Buyers were also asked if they had knowledge of other parties with active interest in the same parcel of land which they subsequently purchased. About one-fifth of all buyers were aware of competition for the tract purchased (table 13). On the basis of buyer evalua- tion, it appears that there may have been somewhat more competition in the Tulare County subarea. Methods of Financing The information summarized in table 14 shows that there was considerable vari- ation in methods of financing between the two subareas and among the two classes of buyers. Of the 176 purchases, 28 or 16 per cent were paid fully in cash. More than one- third (36 per cent) of all nonresident pur- [17] 2 < C5 P « W GO J 1—1 (< PL, < — . ca a) c > « - a W^OJfflOcON^M MMOO^MOlOiNN >> a> •eg ft ° MH N N r It-It- 1 O i— I I 1 II 1 1 1 1 1 03 > c o 01 s-s > 'm 01 m .2 O U hi -^'C ft ^a MiOOOOONffiiOMIN 1 1 1 1 1 1 1 1 1 M *3j! N T— 1 I— 1 1— 1 1 1 1 1 1 1 1 1 1 C .agvg 3 CO fe O O fc 3 (OtOOOCCNtOMM ■*NOOlNCON"5N o N H H 55 H i-H t-H C^ i-i rH (M ^H H M JH T3 ec J ■*■* 0> G OOW»fl©^(N(NOM 1 1 1 1 1 1 1 1 1 >> 3 c "5 C ft bfi 3 # co 03 3.1 <13 g | 1 ft g ^HQOeocOi-H s o « _ C ^_ la a OOHNfO^OINN 1 1 1 1 1 1 1 1 1 e8 3 hi NiOIN^Mr^INN 1 1 1 1 1 1 1 1 1 3 0) 0> ft "3 c "3 C la ca OhOSCDiOh^cOOO O O O O 1 1 O 1 1 O o OOW^ONCONiOh o o o o 1 1 o 1 1 1—1 T— 1 1— I T-H T-H r-H i-H T-H 03 re 6 03 ■+J 03 ^ t>> +3 a 03 ^3 0J -p >> +3 s 03 d s d "S d 03 d a £ "S d XI 03 o3 03 d 03 03 o3 03 d 03 >> +=> T3 t> -^ T3 d -^ T3 73 d '5? "^ d -=5 O d •£ "S 1 '1 % 1 1 6 1 1 *o 01 ft d « 8 ^ ,2 SO" S >> m "53 g "S '£ d g "E3 d m -53 d C "5S g 2^ '« g ^cyOOJaiO^cpO r rH03O03cpO(T3a3O ■g tf £ £tf £ -g # £ -grt^; g-tf^^Pn^ PQ h- 1 H « (— 1 H Table 13 SUMMARY OF NONPRICE BARGAINING AND BUYER AWARENESS OF COMPETITION FOR 180 TRACTS, TULARE COUNTY AND IMPERIAL VALLEY SUBAREAS Type of buyer Total purchases Nature of nonprice bargaining between buyer and seller Buyer was aware of competition for tract purchased Total Terms of sale Mineral rights Other* Both study areas. . . Resident Nonresident Imperial Valley. . . . Resident Nonresident Tulare County Resident Nonresident number 180 131 49 106 75 31 74 56 18 number 27 14 13 17 8 9 10 6 4 per cent 15 11 27 16 11 29 14 11 22 number 14 9 5 9 6 3 5 3 2 number 9 3 6 8 2 6 1 1 number 4 2 2 4 2 2 number 38 29 9 18 13 5 20 16 4 per cent 19 22 18 17 17 16 27 29 22 "Included bargaining over size of cotton allotment, proportion of down payment, and determination of current crops to be included in sale agreement. chases were cash sales compared to only 9 per cent of all resident purchases. Cash purchases of land were concentrated in the Imperial Valley subarea (only three were in the Tulare subarea) and about two-thirds of the cash transactions there involved nonresident buyers. In fact, cash payments were associated with one-half of all nonresident purchases in the Imperial Valley. Mortgaging the purchased tracts was the most common single method of buyer fi- nancing in both subareas and was the basis for 35 per cent of all purchases. It was rela- tively more importance to resident buyers. The second most common single method of financing was purchasing on contract; this accounted for 15 per cent of all pur- chase financing arrangements. It was used as a method of financing relatively more in the Tulare subarea, and in fact was the basis for about one-third of all nonresi- dent purchases in that area. In 9 per cent of the purchases an as- sumed mortgage was the chief single fi- nancial instrument used by buyers. It was used relatively more by resident buyers of Imperial Valley land than by any other group of purchasers. The last single method of financing was use of mortgages on other real estate owned by the buyer. This source of investment funds made up the basis of 5 per cent of all sales. Use of multiple forms of financing ap- peared to be relatively more common among resident than nonresident buyers (24 versus 13 per cent). Both assumed mortgages, and mortgages on the pur- chased tracts were heavily relied upon by buyers using more than a single method of financing. Nearly one tract out of every ten re- ported that the purchase was accomplished with no cash down (16 of the 176 tracts). These were associated with purchases with multiple forms of finance discussed above and were typically associated with resident buyers. Buyers were also asked to indicate the source of finance for noncash sales (table 15). Federal Land Banks, insurance com- panies, savings and loan associations, and banks, were the single most important source and were the only basis of financial assistance for 53 per cent of all sales. Sellers were the sole source of financing for 32 per cent of all sales. Some variability is apparent between subareas and among types of buyers. Credit is important in the transfer of farm real estate from seller to buyer. Pre- viously published USDA data for the Pa- [19] < H pc3 2 Ph i— i Q H 55 fc> O o a PS § 3 H OQ H O «! « ® «-5 g>Jta M^ « C^(MOOiOiOCOCOO II go® 1 1 1 1 1 1 1 1 1 o el? M S °S 13 8, C ® 2 1 si en O THrHCOOOrH-tfC^lM IN(N rH i-l rH 1 1 1 1 1 1 1 1 1 _c <1S M (3 o c O ±5 53 w CJ -^ 03 -£ is N (O h iO ■* r- i 0-*COrH(NtD^(M 1 1 1 1 1 1 1 1 1 "E~ 5 "* t>» »0 Ol h^MOSININ^N^ "o O CO CO .t^ 3 "£ Si rH rH rH rH ^ ^-i <16 c 21 -S.H N0500IN05M>OOiO lOtDNINCCOiHOOiC (M rH ,— 1 ^ t-^ ,-H ,_( ,—1 ,_, ,_, (N H M C +2 o-c Zs M 8P| OOiHCONtOMMiO lOXCOINQOOOOOOiO M-Q CO-* h COIN lOrH H 3 i — 1 rH r-( a to QQ £ c3 +3 CP ■^ >> -r5 03 ■+J cp ■^ >t -1-3 3 03 pi s el -^ C 03 pj a s "S Pi ,fi cs

)+> T! ^ -^ T3 d-t-^T3 T3 p! ■=? ^ fl ■=< O pi *S "o 1 U2 -f-i _, "P .i-H —, CP "H -^ .jHcpOcJcuOjrtaiO -grt^; ^Ph^-5^^ -gtf£ £#£-#£ pq 1— 1 H pq HH H cific farm production region for the last decade indicate that three-fourths and more of all sales have been financed by credit, and that the percentage has in- creased from 74 per cent of all sales in 1960 to 87 per cent in 1969. The same source also estimates that the ratio of debt to purchase price has increased from 70 per cent in 1960 to 75 per cent in 1969. Although this study was not able to estab- lish the relative proportions of the total volume of credit extended by financial in- stitutions and sellers, previous evaluations have established the importance of seller- financing in the farm real estate market. During the year ending October 1, 1967, it was estimated that sellers provided 78 per cent of the total dollar volume of credit in the Pacific farm production region. Credit extended by sellers in the West is at interest rates less than, or comparable to, those of many financial institutions and is generally shorter termed than that of financial institutions; the average term of seller credits ranged from 12 to 15 years, while average terms for insurance com- panies and Federal Land Bank financing are 22 and 33 years, respectively. Table 15 SOURCES OF BUYER FINANCING FOR 136 TRACTS, TULARE COUNTY AND IMPERIAL VALLEY SUBAREAS Type of buyer Both study areas Resident Nonresident. . . Imperial Valley. . Resident Nonresident. . . Tulare County. . Resident Nonresident. . . Both study areas Resident Nonresident. . Imperial Valley. Resident Nonresident. . Tulare County. . Resident Nonresident. . Total purchases 100 100 100 100 100 100 100 100 100 Source of financing Financial institution only Seller only number per cent* 53 53 54 49 45 70 56 59 43 32 29 42 35 36 30 29 25 50 Both financial institution and seller 136 72 43 21 112 59 33 20 24 13 10 1 57 28 20 9 47 21 17 9 10 7 3 79 44 23 12 65 38 16 11 14 6 7 1 "Rounded and therefore may not add up to 100 per cent. [21] ADDITIONAL DESCRIPTION AND ANALYSIS SPECIFIC TO THE IMPERIAL VALLEY SUBAREA A total of 324 transfers of land were adjudged to be possible bona fide transac- tions in the Imperial Valley subarea on the basis of information available from transfer documents. These transfers had deed dates from January 1, 1960 through June 30, 1966. The total number of pos- sible bona fide transactions was reduced to 274 transfers on the basis of further infor- mation obtained from buyer response to mail surveys and personal interviews. Of the 274 transfers, 135 were definitely veri- fied as bona fide transactions; the remain- ing 139 transactions were possible bona fide transactions, but additional survey in- formation which would permit definite verification was not available for this group. Of the 135 varified bona fide transac- tions, 106 (79 per cent) were purchased by resident buyers, as defined previously and shown in table 1; the remaining 29 trans- fers were to nonresident buyers. Detailed information from personal interviews with resident buyers and responses to the same questionnaire mailed to nonresidents was obtained for 87 of the 135 verified trans- fers, of which 65 involved resident buyers (75 per cent). 6 Roughly, the same propor- tion between resident and nonresident buyers was also true of the nonverified set of transactions. Resident buyers accounted for 103 (74 per cent) of the 139 nonverified sales. Verified and Indicated Sales Prices Prices verified by buyers were compared with prices indicated by Internal Revenue Service stamps on transfer documents to ascertain whether the latter could be used as an accurate indicator of sales prices in the Imperial Valley subarea. 7 This was an important consideration because indicated sales prices were available for most of 139 nonverified transfers. If they were found to give accurate estimates of actual market prices, then the analysis of farm real estate prices in the last section of this report could be based on a possible 274 observa- tions, rather than only on the set of 135 verified sales. However, it was found that indicated sales price was not a statistically satisfac- tory estimator of verified sales price. The statistical model based on the 135 verified bona fide transactions suggests that the in- dicated sales price based on Internal Rev- enue Stamps overestimates actual verified sales price on the average for observations of transactions in the Imperial Valley sub- area. 8 That is, parties involved in the transactions appeared to have purchased more revenue stamps than justified by actual verified sales prices. (Actual verified sales prices averaged about 75.3 per cent of indicated prices). Therefore, indicated sales price does not appear to be a valid proxy for actual sales price, and the possi- 6 Of the 65 transfers involving resident buyers, 62 of them were made by farmers. The oper- ating units of 31 farmers involved in 59 of these transactions are the basis for information analyzed in Johnston (1971). indicated sales price was estimated on the basis that each $1.10 of I.R.S. represented $1,000 of market sales value. 8 This conclusion follows from the following linear regression with verified sales price (VSP) as the dependent variable and indicated sales price (ISP) as the independent variable: VSP = .753 ISP R 2 =.909 134 d.f. The null hypothesis that the regression coefficient (/3) is equal to unity is rejected on the basis of the student's ^-statistic, t = 14.02. The same model was also estimated for observations by size of tract transferred given in table 17 to determine whether deviations associated with large dollar transactions might have affected the regression line for the entire set of observations. The null hypothesis /3 = 1 was rejected for all six classes except for transactions involving tracts of land ranging from 161 to 320 acres in size where it was not rejected at the 90 per cent confidence level (t= 1.76). As a result, it is concluded that the relationship between VSP and ISP is not sensitive to size of tract (and, hence, indirectly to the size of total purchase price). [22] bility of expanding the total set of obser- vations by the 139 nonverified transactions is precluded for the subsequent analysis of farm real estate prices in the Imperial Val- ley subarea. Sales Frequencies During the period covered by the survey of the land market in the Imperial Valley subarea, 15 per cent of the transfers were for tracts of land involved in more than a single real estate transaction. Of the 274 transfers, 42 were associated with 18 tracts of land involved in 2 transfers and with 2 tracts of land observed to have changed ownership three times during the study period. The remaining 85 per cent of the transactions involved 232 tracts which had single changes of ownership. The frequency with which buyers par- ticipated in the land market was also of interest. Based on the name of the buyer (grantee) given on transfer documents, the observed frequency was as follows: 163 buyers purchased only 1 tract of land, 38 buyers purchased 2 tracts of land, 4 buyers purchased 3 tracts of land, 5 buyers purchased 4 tracts of land, 1 buyer purchased 6 tracts of land, 1 buyer purchased 7 tracts of land, and 1 buyer purchased 10 tracts of land. Therefore, the 274 transfers involved 213 individual buyers. However, the above dis- tribution is based solely on the identifica- tion of the new ownership of each tract, and does take into account the fact that individual buyers may have been involved in more purchases through various distinct ownership identities, e.g., single owner- ship, partnerships, corporations, and the like. 9 The ownership of land does not neces- sarily correspond to the way in which land is actually combined into farming units. For example, the 31 farming units de- scribed in more detail in Johnston (1971) accounted for 59 transfers as summarized below: 17 farm units were associated with single purchases only, 6 farm units were associated with two purchases, 4 farm units were associated with three purchases, 2 farm units were associated with four purchases, and 2 farm units were associated with five purchases. The Imperial Valley subarea survey then demonstrates by comparison that land ownership and operation are not neces- sarily the same because of the apparent fact that the several parties involved in the farm operation may own land under more than a single distinct ownership identity. Distribution of Transfers Over Time Table 16 summarizes the distribution of farm land transfers in the Imperial Valley by month and year according to deed dates given on transfer documents. Of the 274 transfers observed over the period January 1, 1960, to June 30, 1966, 211 transfers occurred during the first four years 1960-1963. However, total number of transfers ranged only from 50 to 54 annually in each of the first four years, lesser numbers of 31 transfers were identi- fied for 1964, 27 for 1965, and only 5 during the first 6 months of 1966. Repre- sentatives of real estate and lending in- stitutions suggested that the reduced sales frequency might be due to the uncertainty about whether farms in the Imperial Ir- rigation District were subject to the 160- acre limitation. Annual patterns of sales activity are similar for both the verified and nonverified sets. The only apparent difference between the two sets are evi- denced by unlike relative frequencies identified for 1965 (figure 3). Seasonal activity indicated by the fre- quency of transactions during the calen- dar year appear to be highest from April through July. The months of March, August, and September were the least im- portant, as evidenced by lower relative fre- quencies (fig. 4). 9 For example, single transactions in which two individuals were each identified as grantees and a sale to a partnership of the same two individuals were evaluated as being transfers to three separate buyers describing distinct ownership identities. [23] Relative f requ (per 25 ten ce nt) n n 20 ^ 1 15 I 10 - l 5 1 1 L960 L961 ^Verified bona fide transfers *•" Nonverified transfers 1962 1963 1964 1965 1966 Year Fig. 3 Relative frequency of transfers, by year, for 135 verified bona fide transfers and for 139 nonverified transfers, Imperial Valley subarea, 1960-1965, and first 6 months of 1966. Distribution of Transfers by Size Table 17 shows the distribution of tracts by size categories on the basis of deeded acreage. Over 70 per cent of all sales were 160 acres or less in size. Again, there ap- pears to be little difference between the two sets of observations. The set of 135 verified bona fide transactions engrossed 23,184 acres and had an average tract size of 173 acres. Distribution of Transfers by Value The typical land transfer among verified bona fide transactions in the Imperial Valley subarea had a gross sales price of $60,000 to $79,999 (see table 18). Trans- actions with gross sales prices of less than $100,000 accounted for 68 per cent of the total; 27 per cent of the transactions had gross sales prices of more than $100,000 but less than $300,000. A total of seven land transfers were made which had gross sales prices in excess of $300,000. Five of these sales were for more than one-half million dollars and one was in excess of one million dollars. The average price for the seven land transfers was $642,000. Distribution of Transfers by Price Per Acre The typical verified price paid per deeded f Nonverified transfers Jan. Feb. Mar. Apr. May June July Month Aug. Sept. Oct. Nov. Dec. Fig. 4. Relative frequency of transfers by month for 135 verified bona fide transfers and for 139 nonverified transfers, Imperial Valley subarea. [24] o i>- CD *o co CO rH M - CO o > ^ Si -r 3 -G o oi>. r-oo oo ff ffo oh .h cn nm m »3- vr m mvc v (dollars) 135 verified bona fide transfers, Imperial Valley subarea. (See also table 19 for data on distribution of inflated prices.) [29] Fig. 6. Isoprice map based on per acre land prices inflated to January 1, 1967 values, Imperial Valley subarea (dollars). [30] due largely to purchases by nonresidents. The difference between prices paid by resident and nonresident buyers is in- cluded in a statistical analysis later in this report. An Isoprice Map of the Imperial Valley Subarea Figure 6 is based on inflated per acre prices (January 1, 1967, values) and repre- sents an attempt to roughly depict their general spatial distribution within the Im- perial Valley subarea. Isoprice (equal) contours, drawn about verified sales data only, delineate areas wherein per acre land prices were generally less than $400, $400 to $800, $800 to $1,200, and $1,200 and more per acre. The area adjacent to the Salton Sea and in the northernmost por- tion of the subarea is characterized by heavier textured (Type IV) soils and ap- pears to be still in early development stages (the basis of the relatively low per acre prices revealed by this survey). Heav- ier soils are also associated with the two smaller areas with per acre prices of $400, and less, in the east central portion of the study area. Established vegetable produc- tion areas coincide with enclosures by $1,200 contours — near Brawley, El Centro, and Holtville; however, the high price ridge between El Centro and Imperial may also be due, in part, to long-term non- agricultural development potential, al- though there was conscious effort to elim- inate sales with apparent near-term non- agricultural use intention from the survey. Cotton Allotments Transferred Buyers of 72 of the 135 tracts indicated that they had acquired cotton allotments ranging from 1.5 to 280 acres with the pur- chased tract. A total of 1,724.6 acres of cotton allotments was transferred. The average over the tracts which had cotton allotment transfers was 12.2 per cent of deeded acreage. The remaining 63 tracts were trans- ferred without cotton allotments. The average cotton allotment for the total of 23,184 acres of land transferred amounted to 7.4 per cent of all deeded acreage. 11 Estimated Cash Value of Nonland Improvements Records in the office of the county assessor were available for all but 16 tracts out of the total of 135 included in the verified set of transactions. Assessment records for the 16 tracts did not exist for the tax year prior to the transaction because the tract transferred was not an identifiable parcel in the preceding year. This was usually due to the tract being a portion of a larger parcel of land previously identified on the tax rolls. Nonland improvements were not large in value (dollar) magnitudes. Of the 119 tracts, 91 (76 per cent) were without assess- able improvements. The assessed value of improvements ranged from $70 to $3,170 on the 28 tracts with improvements. The total assessed value of improvements for all tracts was only $31,130. Assuming that an assessment ratio of 20 per cent is ap- propriate, the estimated cash value of im- provements is $155,650 and averages only about $5,560 per tract. The relatively minor importance of im- provements is also reflected in the fact that the average gross price per acre for the 135 transfers was $657.04, while the average price (net of improvements) was $648.17 per acre. Thus, over the entire 135 transactions the estimated cash value of improvements averaged only $8.87 per acre. Post-Purchase Improvements In Land The quality of land in the Imperial Valley is closely dependent on the quality of water in the Colorado River, which is the source of irrigation water for the Imperial Irrigation District. Soils in the Imperial Valley are characterized by high salt ac- cumulations; soil quality has been ad- versely affected by the degradation of water quality in the lower reaches of the Colorado River. Under these conditions, 11 This is in sharp contrast to the estimates resulting from the Tulare County subarea where it was found that transferred cotton allotments accounted for 32.1 per cent of all deeded acreage. Corresponding estimates for Tulare and Lower Tule River Irrigation Districts were 33.4 and 30.6 per cent, respectively. [31] extensive investments in drainage systems are commonplace and leaching is widely practiced. In 1948, only 71,000 acres of land had tile drains. Recent estimates indicate that there are now more than 13,000 miles of tile drains on over 300,000 acres — which is about three-fourths of all the cultivated land in the Valley (Scott and Johnston, 1968). A deterioration in the quality of land could be reflected in lower land values in the future. Therefore, owners invest in land improvements to offset the possible effects of poor water and soil qualities on productivity, and to insure against eco- nomic losses on their investments. Several farmers expressed concern about the non- awareness of these problems by absentee nonresident buyers and the resultant ad- verse effect on landlord-tenant relations. It was felt that nonresidents were rela- tively less aware of the effect of degraded water quality upon soil salinity (and pro- ductivity) and, therefore, were reluctant to invest in drainage systems to offset the degradation of water quality during their sometimes short periods of ownership. The magnitude and incidence of invest- ments in land improvements were inves- tigated in the survey. Buyers were first asked to subjectively evaluate the drain- age of purchased tracts at the time of acquisition. Less than average drainage was associated with 55 tracts transferred, average drainage with 65 tracts, and only 15 tracts were subjectively evaluated as having above average drainage at the time of transfer (table 20). It is possible that resident farmer buyers might provide a more accurate measure of the drainage on purchased tracts. A total of 95 tracts were purchased by residents who indicated that they intended to farm the land them- selves. Almost one-half evaluated pur- chased tracts to have less than average drainage, 40 per cent average, and 12 per cent above average drainage. Buyer's were then asked to indicate whether they made improvements in land after purchase. There was no attempt to identify the types of practices or invest- ments in the questionnaire; rather, buyers were asked what sorts of investments had been made which they thought led to im- provements in land. Buyers' responses ranged from leveling, leaching, and slip- plowing to rather costly investments in additional or (in some cases) complete underground tiling systems. Of the 135 tracts, buyers indicated that improvements had been made on 66 tracts — roughly one-half of all tracts transferred and 10,185 or 44 per cent of the total of 23,184 acres covered in the survey. Buyers of 56 tracts (9,138 acres) also provided esti- mates of their investments in land im- provements which ranged upwards to $400 per acre. The total investment was $885,- 800, an average of $85.52 per acre. Although 66 tracts (49 per cent) had had post-purchase improvements in land, there was obvious variability between the incidence on tracts purchased by residents and by nonresidents. Resident buyers car- ried out improvements on 54 of 106 tracts, whereas some improvement was noted for only 9 of the 29 tracts purchased by non- residents, two of the nine tracts belonged Table 20 BUYERS' SUBJECTIVE EVALUATION OF DRAINAGE QUALITY AT TIME OF PURCHASE FOR 135 TRANSFERS, IMPERIAL VALLEY SUBAREA Evaluation of drainage Less than average Average Above average. . . Total Evaluation by All buyers number 55 65 15 135 per cent 41 53 12 106 Resident farmer buyers number 46 38 11 95 per cent 48 40 12 100 [32] to buyers who intended to farm the land themselves. The relatively low rate of im- provement by nonresidents is more pro- nounced because the typical nonresident purchase was made earlier in the time period (as shown previously). The nine nonresident buyers reported that they had invested $94,500 on tracts constituting a total acreage of 1,280 acres. Their average investment in land improvements was about $74 per acre, or somewhat less than the over-all average, but it is possible that tenants may have performed some addi- tional practices (e.g., slip-plowing) which could have been classified as leading to land improvement. INFERENCES FROM REGRESSION MODELS The final phase of this report employs multiple regression models in an attempt to use quantitative data from the survey to explain observed variability of agricul- tural land prices. Preliminary analysis re- vealed that the most reliable results were obtained from analyses using the 87 tracts in the Imperial Valley subarea and the 80 tracts in the Tulare County subarea for which detailed information was available from personal and mail surveys. Table 21 summarizes results of the mul- tiple regression models adjudged most sat- isfactory in their power to explain ob- served variability in farm real estate prices in each of the two subareas. The depen- dent variable — the phenomena whose variation the model seeks to explain — is verified price per acre, net of estimated nonland improvements. Only one-third to one-half of the observed variation in net prices per acre is explained by the selected sets of independent variables included in models 1-3. The Durbin-Watson statistic indicated that there was no serial correla- tion in the regressions reported in table 21. The low explanatory power of the models, evidenced by relatively low values for coefficients of multiple determination (R 2 's), results from the inability to quan- tify all considerations which cause land prices within even rather homogeneous areas to vary among observed transactions. This endeavor — i.e., the attempt to model the factors influencing land prices — makes apparent the difficulty of explaining land price variability from only quantitative data. Other nonquantitative or extremely difficult to measure factors such as buyer (or seller) motivation, the influence of al- ternative forms of financing, completeness of knowledge about productivity and ex- pected incomes from ownership, etc., in- fluence the determination of prices and create price variability among individual tracts transferred in the farm real estate market. 15 However, despite the apparent inability of the regression models to ex- plain relatively large proportions of total variability, statistically significant rela- tionships between the dependent and in- dependent variables available to this anal- ysis are of interest. Model 1 is a regression for the Im- perial Valley subarea. Statistically signifi- cant variables include percentage of crop land, weighted average Storie Index for the tract, a monthly time trend, and a dummy (0-1) variable which identifies the buyer's residence. Two variables associ- ated with drainage quality are also in- cluded because of their utility in model 2, but they are not statistically significant in this "all buyers' " model. Despite the rela- tively low explanatory power of model 1 (R- = .331), this and subsequent models demonstrate the importance of the cited independent variables in their relation to variations in observed net per acre prices. Using only statistically significant vari- ables, model 1 can be interpreted as fol- lows: 13 The net per acre price of the typical acre of land with average drainage in the Imperial Valley sold to nonresident 12 For an interesting and more detailed presentation of agricultural land prices and individual price variability of tracts about "market value" see Harris and Allee (1963, pp. 7-12). 13 Means of independent variables for model 1 are: percentage crop land, 0.842; weighted average Storie Index, 52.46; and month, 33.54. [33] H pa < w o « H H * « g ° o 2w wg Q 5 ^ Q gq «< o PL, * G °°° OS CO <1> !S to i-H 2& no 3 CO CO 1 1 o -~ ' 1 * * * 8*1 S G G CO to O l> CO ">* !OH ^H Os r^ <" ^ v^ »0 "— ' 3*-S 1 + * gas g eg o3 O (h rt .Q 0S.S CO O O ^h 00 to O0 Tfl 1 ^ -i O0 M 1 03 2^&^ + + * * "2 ft... 00 co r- O r>. co -hh co CO Tfl 1— 1 tO CO c~ 8 a CO CO rf tO OS to 2 o OS ^ CO s — ' CO ^ S t- CO to OS Ph ° + + + CO CO o G -2 Tt< OS ^ TT< »0 "* fi O 00 OS O - rH CM G ^ OS^ b- CO CO J O M 01 M -2 « -" g-g 2 CD « ^ *-2 2* as 03 g >> 13 O s ft >, +3 § o O 03 03 "an >> 1^ hH « I < J 3 G co 03 >, » 1 ca q co S I ^ t. . 01 as a > u G — as H a) S 2 fe >- q, G o 01 ^3^ a>__as 3 c"o3 5i,^J2 J 3 S> 03 ^ ^ O-g'C G co 0J.3 o3 0> 03 M o3 >"d ^^2 M o g § II I 3 as c3 2 c° Si 3 » C ft ,>o > 1IS1S =2-5 G « 3 ?ft«5a as buyers is equal to $204.46 (the constant), plus $331.57 for the average percentage of crop land, $145.84 for the mean weighted average Storie Index, and $89.22 for sale in the average month. The resultant total of $771.09 per acre holds for purchases by nonresident buyers; resident buyers ob- tained the comparable typical acre for $567.50, which is $203.59 less than what nonresident buyers paid at the same point in time for tracts with similar percentages of crop land and weighted Storie tract rat- ings, and with average drainage. The statistically significant coefficient for the resident-nonresident dummy vari- able is of interest, as is the magnitude of that estimate. The distribution of trans- fers discussed previously (table 19) sug- gested that nonresident buyers had paid relatively higher prices than resident buy- ers. However, that observation in itself could not be interpreted as though non- residents paid more for comparable land since croppable acreage, soil quality, time of sale, etc., varied among tracts pur- chased. But the regression analysis does suggest that even after the above factors have been taken into account, resident buyers appear to have purchased com- parable land for an estimated $203.59 per acre less than did nonresidents. Nonresi- dents may have been less informed about the productivity of the land they pur- chased, less successful in bargaining about price, or influenced by other factors which led them to pay more per acre than their resident counterparts. Model 2 results from an analysis of the 65 transfers purchased by residents only. If the above noted disparity between prices paid by resident and nonresident buyers is valid, model 2 is an attempt to explain price variability among a more homogeneous and, perhaps, more knowl- edgeable group of buyers. The fact that the percentage of price variability ex- plained by the model increases from 33.1 for model 1 to 48.4 per cent for model 2 using the same independent variables and that the standard error of the regression is reduced by $55 per acre appears to sup- port the hypothesis that the set of resident buyers were more consistent in their range of purchase prices than the larger set of all buyers described by model 1. Results for model 2 also differ in the respect that the subjective evaluation of drainage on purchased tracts is a statis- tically significant variable in the model of resident buyers' purchase prices. This is apparently due to the extent that tracts with above average drainage appear to have had a $278.80 per acre advantage over tracts purchased at the same point in time with comparable percentages of crop land and weighted Storie tract ratings, but subjectively evaluated as having only aver- age drainage. Model 2 would then esti- mate a price of $514.55 per acre for land with average drainage and $793.35 per acre for land with above average drainage as the expected price paid by resident buy- ers when all other variables in the regres- sion model are evaluated at their means. 14 The regression model for the Tulare County subarea (model 3) is also summa- rized in table 21. Only three quantifiable variables — percentage of crop land, weighted average Storie Index, and time — are included in the models based on purchases by all buyers. Unlike results for the Imperial Valley subarea, quality of drainage and buyer residence were not found to be statistically significant factors associated with variation in land prices in the Tulare County subarea. The typical per acre net price is estimated to be $782.70 per acre when the independent variables are evaluated at their means. 15 A hypothesis in the early stages of this study was that the percentage of cotton allotment on individual tracts would re- veal a direct and measurable influence on per acre sales price in multiple regression models. However, in the many alternative model formulations investigated in this study, variables reflecting the presence of cotton allotments on tracts transferred were statistically insignificant; that is, there was nothing to be inferred statis- 14 Means of independent variables for model 2 are: percentage crop land, 0.838; weighted average Storie Index, 54.78; and month, 34.51. 15 Means of independent variables for model 3 are: percentage crop land, 0.946; weighted average Storie Index, 68.10; and month, 37.25. [35] tically about the value of cotton allot- ments. In the case of the Tulare County subarea, this result may have been be- cause most tracts had rather uniform allot- ments in the magnitude of \/ A to i/ 3 of crop land acreage. It is the dominant crop in that subarea and historically has been in- cluded in the operation of most of the farms there. The consequence is then that there was not much variability in the in- dependent variable to associate with ob- served differences in per acre prices. In contrast, there was wide variability in the proportion of cotton allotment found on tracts in the Imperial Valley, but in that area there are other high value crop alter- natives available to compete with cotton on farm production units. Thus, for dif- ferent reasons in the two subareas, one is not able to infer statistically from the regression analysis that cotton allotments differentially influence per acre sale prices. A further indicator perhaps underlying the statistical insignificance of cotton allot- ments is that buyers only gave the motiva- tion of purchase for larger cotton allot- ment for 2 of 141 tracts purchased in both study areas (see table 2). The three models in table 21 can also be used to give some insight into the aver- age rate of price appreciation in the two study areas. The rate of price appreciation for resident buyers in the Imperial Valley is estimated at an average of .47 per cent per month from model 1 and .66 per cent per month from model 2. The rate of ap- preciation averaged .5 per cent per month in the Tulare County subarea. It would then appear that the average rate of land value appreciation in the two study areas was very close to the USDA estimate of 5.9 per cent annually over the period, 1963-69, for the Pacific farm production region. SUMMARY This report is based on a survey of farm real estate transfers in a Tulare County subarea consisting of the land lying within the Tulare and Lower Tule irrigation dis- tricts and approximately the northern 80 per cent of the Imperial Unit of the Im- perial Irrigation District (Imperial Valley subarea). Personal interviews with resi- dent buyers supplemented by information from mail-out questionnaires returned by nonresident buyers provide the data un- derlying the reported characteristics of the farm real estate market in the two study areas. Major findings include the follow- ing facts. 1. One-half of all purchases were for the primary purpose of farm enlargement and nearly one-third were viewed chiefly as in- vestments by buyers. Among resident buy- ers, two-thirds of all purchases were for farm enlargement and less than one-fifth were chiefly for investment purposes, whereas nearly three-fifths of the total pur- chases by nonresidents were primarily for investment purposes. Nonresidents also indicated that about one-fourth of the tracts purchased by them were to replace land sold previously outside the two study areas. Farm enlargement as a primary mo- tivation for land purchase was relatively more important in the Tulare County subarea than in the Imperial Valley sub- area. 2. Buyers expect that sons or relatives will continue farming operations in the future on about two-fifths of the tracts purchased. 3. Financial pressure and realization of profit each accounted for about one-fourth of the sellers' motivation to sell as subjec- tively assessed by buyers. Retirement was the primary basis for the transfer of an additional one-fifth of the total sales. Sellers in the Tulare County subarea were relatively more motivated by financial pressure and less by realization of profit than those in the Imperial Valley subarea. 4. Farmers accounted for almost two- thirds of all sales. About one-third of all u , sales by farmers were thought to be due to financial pressure; a comparable percent- age was due to combined retirement and health-age motivations. Over two-fifths of the tracts sold by nonfarmers were ad- [36] judged to have resulted from the oppor- tunity to realize profit; financial pressure and unsatisfactory renting experience were next most important reasons for sales by nonfarmers. 5. Resident buyers were typically younger than nonresidents. The typical buyer was a high school graduate and one- fifth of all buyers had completed at least four years of college. 6. Over two-fifths of all buyers learned that the tracts they purchased were for sale directly from the seller and about one- third were so informed by a real estate broker or agent. The former source was relatively more important than the latter for both resident buyers and for Imperial Valley subarea transfers. 7. Nearly one-half of all buyers had ac- tively searched for land, mostly in the same county, prior to their purchase. A time limit imposed by possible liability for capital gains taxation on land previ- ously sold was responsible for some of the purchases. On the basis of a somewhat limited response, the typical buyer felt that he had experienced no difficulty in finding the type of land he wanted, al- though this was not true for nonresident buyers in the Imperial Valley. 8. A sellers' market was more evident in the Imperial Valley subarea than in Tu- lare County. In those cases where the seller's initial asking price was reported, about three-fourths of the sales in the former area had final transfer prices equal to the seller's asking price as compared to one-half in the latter subarea. Where price concessions were achieved from sellers, resident buyers were relatively more suc- cessful than nonresidents. About two- thirds of the transfers characterized by price concessions had final transfer prices equal to counter-offer prices of the buyer. For those cases not reporting seller asking prices, the buyer's initial offer price was accepted for nearly nine-tenths of the transactions. Elements of nonprice bar- gaining (terms of sale, mineral rights, etc.) was reported for only one-eighth of the transactions. About one-fifth of all buyers were aware of competition for the tract purchased. 9. Methods of financing varied between the two subareas and among the two classes of buyers. Mortgages on purchased tracts were the single most common method of financing in both subareas, and were associated with more than one-third of all transfers. It was relatively more im- portant in the Tulare County subarea and among resident buyers. Cash pur- chases, accounting for one-eighth of all transfers, were concentrated in the Im- perial Valley subarea and among nonresi- dent buyers — nearly one-third of all non- resident purchases and more than one-half of Imperial Valley nonresident purchases were cash sales. One-fifth of all sales util- ized two or more types of financing — typi- cally, an assumed mortgage plus a new mortgage on the purchased tract — one- tenth of all transfers appeared to require no cash down. Slightly more than one-half of the transfers relied solely on financial institutions as the source of financing of non-cash sales; sellers were the sole source of financing for one-third of the trans- actions. Some additional characteristics of the farm real estate market in the Imperial Valley subarea were examined in more depth. The following are summarized from that more specific analysis. 1. Resident buyers accounted for four- fifths of all transfers. Purchasers intended to farm the acquired agricultural lands in three out of four cases. Resident buyers gave farming as the intended use over renting out by a ratio of 9 to 1, whereas the ratio was 1 to 4 for tracts purchased by nonresidents. 2. About one-sixth of all tracts trans- ferred were involved in more than a single real estate transfer during the study pe- riod. The remainder had only single changes of transfer. 3. Annual sales activity (in numbers of real estate transfers) was nearly constant during the years 1960 through 1963, but dropped sharply thereafter. Nonresident buyers were involved in 1 out of every 4 transfers in the former period, but in only one-tenth of the activity in the latter period. Seasonal activity was highest from April through July and lowest in March, August, and September. 4. Nearly three-fourths of all transfers involved tracts of land of 160 acres or less. 5. The typical land transfer had a gross [37] sales price of $60,000 to $79,999 and two- thirds of the transactions had gross sales prices of less than $100,000. 6. The typical verified price per deeded acre (net of nonland improvements) was between $550 and $649 per acre. The typical price paid by residents was also between $550 and $649, but the median price for nonresident purchases was be- tween $850 and $949 per acre. An adjust- ment for the appreciation of land values over the lengthy 61/9-year study resulted in an even more marked divergence between typical resident and nonresident purchase prices because of relatively heavier non- resident activity in the early portion of the study period. 7. Three-fourths of tracts for which assessment records were available were without assessable improvements. The average estimated cash value of improve- ment on the remaining tracts was only $5,560 per tract. Over all transactions the estimated cash value of improvements averaged less than $9 per acre. 8. Resident buyers evaluated the drain- age at time of purchase as less than aver- age on one-half of the tracts. Post-purchase improvements in land averaging about $85 per acre had been carried out on one- half of the tracts. Improvements were made on more than one-half of the tracts purchased by residents, while only one- third of nonresident-purchased tracts had post-purchase improvements. The use of regression models to explain the observed variability in net prices per acre led to the following additional in- ferences: 1. Using available quantitative data, only one-half, or less, of the observed price variability is explainable. 2. Nonresidents appeared to have paid about $200 per acre more for comparable lands than residents in the Imperial Val- ley subarea. 3. Resident buyers appeared to have paid about $280 per acre more for tracts subjectively evaluated by them to have above average drainage than for compa- rable lands with average drainage. 4. Residence of buyers and drainage characteristics of purchased tracts did not appear to have influenced price in the Tulare County subarea. 5. The rate of land value appreciation in the two subareas appeared to have averaged 0.47 to 0.66 per cent per month over the time span of this study. ACKNOWLEDGMENTS The author gratefully acknowledges the cooperation of those whose response to per- sonal interviews and mail surveys provided most of the primary data used in this study. Assistance and cooperation were also freely given by the Imperial and Tulare County Assessor's offices and the staff of the Tulare, Lower Tule, and Imperial Irrigation Districts. Mr. Doyle Kauk and Mr. Phil Pierre, former graduate students in the Depart- ment of Agricultural Economics at Davis, assisted in the design of the questionnaire, supervised the surveys, and performed much of the initial summarization of the data. The final analysis was carried out with the able assistance of Mr. Geoffrey Allen and Mrs. Sue March. Lastly, the author would be remiss if he did not acknowledge the typing of early drafts by Mrs. Micki Eagle and the final manuscript by Mrs. Barbara Huffine. Financial assistance for this study was provided in part by the U. S. Department of Agriculture, Economic Research Service, Farm Production Economics Division, under Cooperative Agreement No. 12-17-03-2-114. [38] LITERATURE CITED California State Board of Equalization. 1967. Annual reports, annual issues 1960-67. Harris, Curtis C, Jr. and David J. Allee. 1963. Urbanization and its effects on agriculture in Sacramento County, California. 2. Prices and taxes of agricultural land. University of California, Agricultural Experiment Station, Giannini Foundation Research Report No. 270. Johnston, Warren E. 1971. Economies of size and imputed values of farmland in the Imperial Valley of California. University of California, Agricultural Experiment Station, Giannini Foundation Research Report No. 314. Scott, V. H. and W. E. Johnston. 1968. "Renewable resources in the development of a desert area of California," paper in Desert Agriculture in California. A task force study by the University of California, Davis. Pp. 7-28. U. S. Department of Agriculture, Economic Research Service. 1969. Farm real estate market developments, periodic issues, 1964-69. 6m-l,'72(P7908L)VL [39] ME! A FARM product; Well, not exactly — you can't grow auto- mobiles on farms, but farm products are essential in manufacturing them. Consider the annual agricultural needs of just one major automobile company. or, in terms of approximate acreage: 900,000 bushels of corn 736,000 bushels of flax- seed 74,000 bales of cotton 1 5,000 acres of com 80,000 acres of flax 78,000 acres of cotton During the same period this company used products derived from 364,000 sheep and 36,000 cattle — plus many other items such as hog bristles and beeswax. In all, produce equivalent to the output of 1,000 good-sized farms is needed yearly. No wonder a top executive in the automotive industry has said: "Our plants, here and throughout the world, would have to close their doors in a few days if their flow of agricultural materials were to stop." Supplying America's countless industries — and feeding the nation bountifully — makes agriculture America's biggest and perhaps most important business. That is one reason why anything which affects agriculture affects everybody.