B-1617 Expanded Marketing Opportunities for Dry Onion Production in Texa An Interregional Analysis of the Spring and Summer Seasons Texas Agricultural Experiment Station Charles J. Arntzen, Director The Texas A&M University System \ a" ‘ 4-,- , , \ . i3" ’ .5 ‘b. [manna wvfiuo In @384 YIWIH] Expanded Marketing Opportunities for Dry Onion Production in Texas: An Interregional Analysis of the Spring and Summer Seasons l ‘Stephen Fuller, H. L. Goodwin, Carl Shafer, and John Schmitzl \ lProfessor, associate professor, professor, and research assistant in the Texas A&M University Department of Agricultural Economics. \. 88 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SaOuaJaJaH 9g . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . suolsnlauog pug Kleululns 88 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Rwanda 1S0D_1Se9,-I uo P9838 uononpold J0 uoneaorl 8g . - . . . . . . . . . - . . . . . . . . . . . - . - . - . . . . . . . . . . . . . - - . . . - . . . . - . - . . . Iz . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . aleqs laxrew S1]: puedxg m sexal Jo OZ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ~ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . sllnsau LI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . [apow J0 uOnBPHBA 91 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . S1809 uonmlodsuell ZI . . . . - - . - - . . . - . - . . - . . - . . - . . - . . . . . . . . . . . . . . . . . . . . . . - . . . . . . . . . . . - . - . . . . OI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SUOIBGH pug pueulaq OI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Iapow J0 Sluauodulog 818G 6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Iealqfireuv L . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . [Banaloaqm L . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SESKIEUV do‘; xloMaureld L . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . asodlnd 8 . . . . . . . . .‘: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Suollgaa gupadulog pug SGJELIS laxlnw uoluo ‘(JG Z . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SBXGL I . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . S91E18 I . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . punolgyloeg I . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . uouanpollul m SLLNHL-LNOO Introduction For over a decade, dry onions have been the most valuable vegetable crop produced in Texas, typi- cally comprising 16 to 20 percent of the state’s vegetable revenues (Texas Department of Agricul- ture 1976-1986). Texas is an important national supplier of fresh onions. During the 1985-87 period, P the share of onions supplied by Texas to the national market averaged nearly 14 percent. This figure compares with a market share which averaged about 21 percent during the 1977-79 period. The decline in market share is the result of a modest decline in Texas shipments and a substantial increase in shipments by competing production regions. The purpose of this study is to analyze the declining share of the dry onion market held by Texas producers and evaluate opportunities to expand the marketing of Texas-produced onions. Background United States The United States Department of Agriculture segregates dry onions by maturity period or harvest period. These include (1) the Spring onion, (2) the Summer non-storage onion, and (3) Summer storage onion crops. The Spring onion crop comprises about 15-18 percent of national onion production and is produced primarily in south Texas, Arizona, and California. Spring onions are the first-harvested of , the calendar year, and typically they move directly to the fresh market. Texas, New Mexico, and Wash- ington are the primary producers of Summer non- storage onions which, on the average, account for about 10 percent of national production. The Summer storage onion constitutes about 70-75 per- cent of total onion production in the United States. This crop is harvested during the August through October period with subsequent shipment con- tinuing through the fall and early spring (April). The storage onion receives no competition from new onion production in the United States until Spring onion harvest commences in March. Im- ported new onions from Mexico during the winter and spring, however, offer some competition for the storage onions. There are about twelve states involved in the production of storage onions. Lead- ing states include Oregon, New York, Colorado, Idaho, Michigan, and Washington (USDA 1976- 1987). During the 1975-77 period, per capita consump- tion of onions averaged about 14 pounds (USDA, Vegetable Situation and Outlook Yearbook, 1978). The increase in per capita consumption of fresh onions and other salad-vegetables (broccoli, cauli- flower, tomatoes, and lettuce) is often attributed to health-conscious consumers who are increasing their consumption of fresh vegetables and away- from-home consumption that often includes both fresh and processed onion products. Per capita consumption of onions has also benefited from the increased consumption of ethnic foods, in particular Mexican foods, which often include liberal quanti- ties of onions. An estimated trend line, based on 1970-85 per capita onion consumption, and expected population growth were used to project mid-1990 onion consumption. The trend line showed per capita consumption to increase by .29 pounds per year and population to increase 19 million over the 1985-95 period. With this scenario, onion consump- tion could increase about 10,500 cwts over this period, or approximately 25 percent. Texas About three-fourths of the harvested dry onion acreage in Texas is located in south Texas, in particular, in the Rio Grande Valley (62%), San Antonio-Winter Garden (10%), and Laredo (3%) regions. The remaining onion acreage in Texas is located in the High Plains (16%) and Trans-Pecos (9%) regions (Texas Department of Agriculture 1976-1986). The south Texas area is the primary source of the state’s Spring onion production, while the Plains and Trans-Pecos areas are producers of Summer onions (nonstorage). During the past decade, about 94 percent of the south Texas onions have been shippped to the fresh market with the remainder merchandized as ringers or chopped, frozen, and sold to food processors (South Texas Onion Committee Report 1975-1985). The shipment of Spring onions from south Texas commences in March and peaks in April and May. On the average, about 63 percent of the dry onion shipments from Texas are in three months: March (4%), April (29%), and May (30%). The remaining shipments are concentrated in June (10%), July (16%), and August (10%). The March-May shipments are Spring onions that originate in south Texas while the J une-August shipments tend to be Sum- mer onions that originate in the remainder of the '8100np01d 82x01 Kq p0dd111s su01u0 111p 10 0u1100p 0111 u1 0101 1112011 111318 2 p01121d 011211 001x011 111011 8110d1111 0111 12111 aauaPlAa QMESHIQUOQ o" S! a-Iaql '(L86I-LL6I VCISII) 8u01120013u188010 82x01 2111 101110 180u1 pu2 M0pu1M 0unf-1101211 0111 u1 p01110001 012 001x011 111011 8110d1u1 0111101u0010d 51-09 1n0q2 ‘03210112 0111 11g 801213 p011u11 0111 u1 s1u0u1d111s u01u0 111p 12101 10 1u0010d 9-3 1n0q2 1u0s01d01 8110111111 080111 801213 p011u1] 0111 01u1 p0110d1111 su01u0 111p 0111 10 1u0010d ()3 1n0q2 p0121113110 8211 001x011 ‘11112011018111 'M0pu1M 11211 0111 10 1u0010d 3 01 1 p01u1210 1111201d111 8211 3351 00u1s 0101 p0s2010u1 812131001) ‘10110M011 fpu011 011u110p ou p0Mo118 8u01301 Bu110d11100 101110 '1 01n31d u1 uM0118 s1 po110d 13-1151 0111 10110 10>112111 u01u0 118011 0111 u1 82x01 pu2 21111011129 10 010.1 Bu18u2110 0111 "1u0010d 11> 01 59 u1011 p0u1100p 11211 u1 82x01 Aq p1011 012118 10>112u1 0111 ‘1110810/11109 ‘1u0010d If SBM GIBQS GLIZ} ‘ZHIOOJGd SBM‘ 012118 0.8210112 51-1151 8‘21u101112[) ‘K211 .81111n(1 '8100np01d 21u101112Q Kq p0u11210 u00q p211 1snBnV pu2 ‘111nf ‘aunf ‘K211 u1 012118 10>112111 Bu1u1100p 182x011, 10 110nu1 12111 p0M0118 8181112u2 10111111 '81111211su00 1201301012111110 0021 8u01301 3u110d11100 u011M po110d 2 ‘111dv u1 p0dd1118 0q 01 su01u0 82x01 10 012118 B11182010u1 112 8M0118 110111M ‘1 01q21 u1 u0112111101111 0111 11q p01210q01100 ‘112d u1 ‘81 B111pu11 81111 '10)112u1 12u0112u 0111 101u0010d 53-59 p0u11210 1111u01818u00 810dd1118 82x01 ‘111dV Bu11nq "111dV1d00x0 8111u0u1 112 u1 p12MuM0p 3u1pu011 0q 01 82x01 u1 01211810112111 1€1111uou1 p0M0118 081010x0 81111 po110d 13-1151 0111 Bu11np 8u01B01 3u110du100 pu2 82x01 101 p0121n0120 82M 1sn18nv pu2 ‘111nf ‘aunf ‘K211 ‘111dv 101 012118 10>112u1 111111u011 ‘(1861-1161 VGSfl) 19112111 181131111 @111 w; 9101111100 ‘s0121s Bu10np01d u01u0 0321018 1111M u0110un_fu00 u1 ‘801218 080111 01111M ‘M0p111M10>112u1 111m f 0111101 0111 001x011 M011 pu2 ‘21u101112Q ‘u013u11182M 82x01 111011 8u01u0 0321018-u0u 10u1u1n3 '1u0u1111001d 012 001x011 M0 N pu2 801218 080111 ‘11211 u1 8012118 10>112u1 1u2110du11 u11210 2u0z11V pu2 ‘21u101112Q 82x01 111011 su01u0 0321o18-u0u 81111d3 ‘10110M011 ‘111dV 11g '012u1u10p (11011 M011 pu2 ‘u2.81110111 ‘0p21010Q ‘180M11110N 0111023) 8u01301 3u10np01d u01110 0321018 10u1111n3 0111 ‘10q11101d03 pu2 1101211 B1111nq '111u0u1 1111 012118 10>112111 Bu1u1u12x0 Aq pado -10110p 81 ‘Mop111M 10>112111 10qu101d03-1101211 110111 3u11np ‘s100np01d 82x01 11q p01111210 012118 10>112u1 B111u1100p 0111 110 u0112111101u1 0101d11100 01011 112111 p112 2131003 ‘81u211u0 M0u sn1d ‘u013u11182M pu2 ‘u0301Q ‘0p21010Q ‘21u101112Q Aq 0p2u1 010M 012118 10>112111 u1 8u12B 12111121811118 180w @111 ‘POI-lad 18-9861 01 81-1161 0111 Bu11np 1u0010d 19 01 31 1n0q2 111011 p0s2010u1 u01B1111182M pu2 ‘112111 ‘1103013 ‘2131003 ‘0112p1 ‘0p21010Q ‘21111011123 1o 012118 101112111 0111121nu1n0 0111 ‘18211u00 u1 '01218 11112 10 0u1100p 180.8121 0111 ‘L8-986I u! W991“ 9'71 01 8L-LL6I u! TWQQ-Iad L'8Z u1011 p0u1100p 10>112111 12u0112u 0111 10 012118 182x01 ‘11ns01 2 8V "1u0010d ()1 1n0q2 p0u1100p s1u0u1d111s 82x01 ‘121n01112d 111 £p0u1100p s1u0u1d111s 12n102 82x01 pu2 u231110111 111 p0u1100p 012118 10>112111 11101 M011 0111 ‘11ns01 2 8V p0s2010u1 81110111d111s 12101 01111M po110d 81111 10110 1u218u00 111011112101 u12u101 01 p0pu01 >110 A M0 N u1011 s1u0u1d111s u01u0 111g "1u0010d 53 10 01211s 13-9351 110111 1111M 818211 -u00 81111 i1u0010d 11 10 012118 10>112111 0111121nu1n0 2 p211 82x01 p112 >11011 MGN ‘u231110111 ‘su0s20s 31 -1151 0111 u1 'p0182111100 u0111 pu2 81108208 13-9351 pu2 31-1151 0111 101 p0121n0120 010M 8012118 10>112111 12n11u2 ‘801218 Bu10np01d sn011211 0111 10 8012118 10>112111 .8111Bu2110 0111 110 u0112111101u1 u12.8 01 "0s2010u1 1u0010d 917 2 ‘su0s20s 13-9351 0111 111 spunod u01111q 333 01 su0s20s 31- 1 151 0111 u1 spunod u01111q 15'1 10 03210112 u2 u1011 s1u0u1d111s 118011 111 0s2010u1 01121112.1p 0111 8012118n111 u0112111101u1 81111 '3 01q21 u1 uM0118 012 801218 3u10np01d 10_f2u1 0111 101 s1u0u1d111s 1101110 111p 12nuu2 13- 1 151 0111 811013011 B11110du10Q pu2 8012113 10111211 u01u0 111g 's111u0111 3u1dd1118 101110 101 p011108q0 010M 8pu011 01q2u10081p 0N ‘1u0010d 5 01 1u0010d 11 u1011 p12MuM0p p0Bp0 8211 0un1= u1 8100np01d 82x01 Aq 0p2u1 81u0111d1118 12nuu2 10 012118 0111 ‘111081011u0Q ‘812011 1u0001 u1 p11111 -0u0 111120u 01 (81) 151 0121) 1u0010d 93-53 1n0q2 u1011 p0s2010u1 8211 111dV u1 s1u0u1d111s 12nuu2 10 012118 0111 '8p12Mdn Bu1pu011 81 111dV Bu11np 82x01 111011 s1u0u1d111s 12nuu2 10 u01110d 0111 ‘1 01q21 u1 u0112u1 -101u1 0111 1§q uM0118 8V '8u01B01 u0110np01d 8101218 Table l. Dry onions: percent of Texas’ annual shipments per each March, April, May, June, July, and August shipping period, 1977-1987.‘ K Year March April May June July August 1977 0.0 21.7 33.6 14.4 19.2 10.4 1978 4.8 24.7 31.0 12.5 18.0 8.6 1979 0.4 23.3 39.9 7.6 11.1 15.1 1980 5.4 29.3 25.9 11.3 19.3 8.1 1981 2.1 29.2 28.5 10.9 18.9 10.3 1982 12.7 34.1 20.6 10.1 14.2 8.2 1983 5.6 29.5 27.4 9.9 12.4 12.7 1984 0.4 31.5 31.7 7.9 17.3 10.5 1985 3.0 31.2 36.1 7.0 12.5 9.9 1986 5.4 35.4 23.6 9.6 19.0‘ 6.8 1987 2.2 35.6 35.1 9.0 12.3 5.8 ‘Yearly shipments may not total 100 percent because of September and October shipments. In 1979 and 1983 these months included about 2.5 percent 0f annual shipments. Source: U. S. Department of Agriculture, Agricultural Marketing Service, Fruit and Vegetable Division, Fresh Fruit and Vegetable Shipments by Commodity, States and Months, FVUS-7, 1977-87. Table 2. Dry onions: estimated U.S. domestic shipments, exports, imports, 1977-87 , by calendar year (1000 cwt). Origin 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 Arizona 437 377 496 482 665 638 802 554 538 530 California 3,044 2,616 3,534 3,189 3,682 4,879 4,065 4,226 4,714 4,892 5,343 Colorado 1,532 1,994 1,694 1,616 2,244 2,126 2,573 2,817 3,404 3,178 3,312 Georgia --- -—- 150 407 170 165 280 369 Idaho 1,559 1,806 1,563 2,041 2,178 1,944 2,188 2,431 2,597 2,776 2,910 Michigan 1,513 1,404 1,749 1,410 1,500 1,533 1,455 1,723 1,910 1,287 1,160 New Mexico 1,038 1,119 948 1,165 1,210 1,631 1,225 1,331 1,318 1,609 1,656 New York 2,462 2,732 3,545 3,616 3,499 3,808 2,999 2,908 3,156 3,144 2,617 Oregon 2,418 2,774 2,587 3,638 3,029 2,672 3,054 3,066 3,773 4.106 4,710 Texas 4,252 5,104 3,522 5,071 4,168 4,382 5,254 4,116 3,321 4,803 3,565 Utah 13 4 12 24 15 13 4 323 578 473 655 Washington 86 1,019 985 1,350 1,006 857 1,477 1,300 1,307 1,619 2,013 Other 30 20 17 6 7 5 4 1 4 0 9 U.S. Domestic Shipments 18,384 20,969 20,489 23,622 23,020 24,665 25,343 25,214 26,801 28,705 28,849 U.S. Exports 1 2 191 421 1,977 343 724 1,870 634 1,104 898 Texas Exports 1 2 8 28 0 0 0 141 77 105 0 Total U.S Shipments 18,385 20,971 20,680 24,043 24,997 25,008 26,067 27,084 27,435 29,809 29,747 Total U .S. Imports 1,438 1,365 1,548 862 917 1,246 1,431 1,839 1,768 1,577 2,434 Source: USDA, AMS, Fruit and Vegetable Division. Fresh Fruit and Vegetable Shipments By Commodities, States and Months. FVUS-7. 1977-1987. ‘SJEQS IGXJBUI SBXGL PUB ‘BIUJOJHBQ U! SPUGJL GJHBIJ sexei + egwomeg ___ tfitfififilvfi%§lfiflfllfilildllflo I§D$§I>YQ§§> lfi I 61.61 Old» L1G o I T I I l£fi%filvfi%l¥ll£%fitfiliiflltflo £$9§fi7§%%ll§¥l6l6l9£6lu6l I 1 I I I fi I T I I T I I W I I I I Purpose Shifts in the relative importance of Texas as a national supplier 0f dry onions has generated a need to learn more about its competitive position. To study this issue requires a complex of economic information on competing regions and associated production costs, regional demands, and transfer costs. The objective is to determine the cost-com- petitiveness of Texas onions in the national dry onion market and identify windows where Texas shipments may be increased. Although the analysis is accomplished with a national model, the focus of the research is on Texas and its competitors. Framework for Analysis Theoretical In a competitive environment, the type and quantity of commodities produced in a region reflect an efficient utilization of resources. Efficient re- source use implies a geographical distribution of production that satisfies market requirements at the lowest possible cost of production and transfer. Often the observed changes in location and volume of production may be considered an adjustment toward a long-run equilibrium (Bressier and King 1970). Analytical The study uses a spatial equilibrium model to address the competitiveness issue, in particular, a transportation model. The model includes each surplus region’s monthly producer price, which is used as a proxy for cost, and monthly transportation charges, which link surplus producing regions and consumption regions. Based on this information, the model determines the most efficient regional trade patterns, i.e., the producing regions which can serve demand regions at lowest cost as well as the flow between regions. To the extent these flows approximate historic flows, a tool is available to measure changes in shipping patterns that result from an induced stimuli (e.g., transportation rates or producer costs). The model can be expressed in mathematical notation as follows: Let: i = onion producing region (i = 1,.....n) j = onion consuming region (j = 1,.....m) k = months of year (k = 1, 2, 3,.....,12) Qsik = quantity of onions produced in region i 1n time period k QCJ-k = quantity ofonions consumed in region j 1n time period k Sijk = quantity of onions shipped from region i to region j in time period k TCiJ-k = transfer cost linking region i to region j in time period k Pik = shipper price (cost) in region i in time period k The objective is to determine the SiJ-k’s which minimize, n r? 12 i221 F1 :21 Pik + SijkTCijk Subject to, Sijk - O m Qsik J f1 Sijk n Qcjk i 5 1 Sijk n m E Qsik _ _§ Qcjk 1-1 1-1 Data Components 0f Model Substantial data and background information are required to construct the transportation model. There is a need to ( 1) delineate production and consumption regions; (2) estimate available supplies in each producing region; (3) estimate consumption in each consumption or demand region; (4) estimate regional production costs or prices; and (5) estimate transportation charges that link production and consumption regions. Demand and Supply Regions The developed trade model includes 34 regions: fourteen supply regions and 20 consumption or demand regions. Monthly onion supply for each of the regions was based on an average of 1983-85 shipments. The estimated monthly supplies for each producing region are included in the model and are identified in Table 3. These data originate with the Agricultural Marketing Service (AMS). The AMS report provides information on monthly domestic shipments by truck, rail, and piggyback (trailer on flat car) from each major producing state (USDA 1977-87). The consumption of fresh onions was estimated with data from the Nationwide Food Consumption Survey (N FCS) and the Census of Housing. The Food Consumption Survey contains comprehensive and detailed data on food consumption patterns. The survey data indicates the frequency and quan- tity of fresh onion consumption (lbs/week/house- hold) in four regions of the United States (West, South, Northeast, Northcentral) in each season of the year. It was assumed that any regional differ- ences in onion consumption at the time of the survey (1977-78) were also applicable during the study period. This seemed a reasonable assumption in view of the relatively small differences in regional consumption patterns. The per household estimate for a state within a region was multiplied by the number of households in that state for purposes of estimating statewide fresh onion demand. States were aggregated to form the 20 demand regions identified in Figure 2. Estimated annual consumption of dry onions for each region is exhibited in Table 4. For purposes of model construction, it was necessary to estimate regional consumption per month. Regional Production Costs Monthly average prices received by farmers at the point of first sale, for all grades and qualities of dry onions, were used as a proxy for costs. It was assumed that prices were determined in a competi- tive environment, in which case, the monthly average price for each supply region approximates production costs. The 1983-85 monthly average producer prices for each supplying state is shown in Table 5. These values are included in the trade model. Texas producer prices for 1983-85 are highest in the March, April, and July period (Table 5). Lowest average prices in Texas during the 1983-85 period are in May, June, August, and September. Cali- fornia, one of Texas’ principal competitors, exhibits a similar price level and pattern. Prices in Texas during March, April, and July were substantially above the estimated costs, which range between $10 and $11 per cwt. This situation is due to the relatively high prices which occurred in the spring of 1984 when the March and April prices averaged $25.80 per cwt. Although this is a relatively high price, it seems that similar prices have historically occurred. In 1981, as an example, average monthly prices in Texas ranged up to $27.20/cwt. (April). When the unusually high prices are removed from the data, average prices tend to approximate costs. For example, the average March price for 1982, 1983, and 1984 is $11.50 (range $10.10 to $12.30), a value which approximates costs. Regardless of the divergence between prices and costs, it seems that relative prices of competing regions approximated their relative costs (Table 5) (Fuller, Goodwin, and Shafer 1989). The relative costs of producing regions that compete for a market window partially determine their advantage or disadvantage in that window. Because relative prices seem to parallel costs, it was judged appropriate to use producer prices rather than costs. Accordingly, the monthly average prices for 1983-85 were included in the spatial analyses. Transportation Costs The charge for transporting onions between supply and demand regions may have an important bearing on the ability of a supply region to compete in a national market. Transportation rate infor- mation was collected from Fruit and Vegetable Truck Rate and Cost Summary, a publication prepared by the Agricultural Marketing Service, Fruit and Vegetable Division, of the USDA (1982- 1985). Based on this data source, six linear re- gression equations were estimated to calculate the expected charge for transporting dry onions among the designated supply and consumption areas. \- The estimated equations used t0 calculate rates '\linking Texas with the 20 consumption regions included in the transportation model are shown in Table 6. Both equations show distance of haul (miles) to be the most important factor determining rates, and, to a lesser extent, the month of shipment. The coefficient on the distance variable is similar '\ Table 3. Estimated supplies of fresh dry onions, 1983-85 averages (1000 cwts). in both equations, with an estimated value of $.002/cwt/mile. The estimated equations explain 87 and 99 percent of the variation in rates. The estimated rates linking Texas with the 20 demand regions during the state’s dry onion shipping sea- sons are shown in Table 7. . Yearly Origin Jan Feb Mar Apr May June July Aug Sept Oct Nov Dec Total Arizona - — - - 306 359 - - - - - - 665 California 104 56 39 74 910 1088 910 421 199 195 186 153 4335 Colorado 302 284 193 50 - - - 287 583 440 431 361 2931 Georgia - - - - 157 89 - - - - - - 246 Idaho 366 289 166 1 1 1 - - 103 264 423 390 392 2405 Michigan 263 208 178 37 - - - 20 217 259 270 244 1696 New Mexico - - - - - 473 435 325 58 - - - 1291 New York 432 353 359 239 - - - 84 390 408 390 366 3021 Oregon 488 416 356 86 16 - 7 155 359 511 449 455 3298 Texas - - 127 1307 1341 351 596 465 43 - - - 4230 Utah 43 - - - - - - - 42 78 81 57 301 Washington 122 116 117 49 1 - 280 127 120 144 143 142 1361 Total U.S. 2120 1722 1535 1853 2732 2360 2228 1987 2275 2458 2340 2170 25780 % 8.3 6.7 6.0 7.3 10.1 8.9 8.7 7.8 8.9 9.6 9.2 8.5 100 Mexico 136 148 484 274 84 34 - - — - 16 75 1251 \Total 2256 1870 2019 2127 2816 2394 2228 1987 2275 2458 2356 2245 27031 Source: USDA, AMS, Fruit and Vegetable Division. Fresh Fruit and Veg States and Months. FVUS-7. 1977-86. Table 4. Dry onion demand regions and the associated annual consumption. etable Shipments by Commodity, Estimated Annual Demand Regions‘ Consumption (1000 cwt.) 1. Maine, New Hampshire, Vermont, Massachusetts, Connecticut and Rhode Island 1765 2. New York 2774 3. Pennsylvania, Maryland, Delaware and New Jersey 3273 4. Ohio and West Virginia 1384 5. Virginia, NorthCarolina and South Carolina 1560 6. Kentucky and Tennessee 914 7. Alabama and Georgia 997 8. Florida 1372 9. Wisconsin, Illinois and Indiana 2125 10. Arkansas, Louisiana and Mississippi 914 \ 11. Minnesota, North Dakota and South Dakota 492 , 12. Nebraska, Kansas, Iowa and Missouri 1110 13. Oklahoma and Texas 2249 14. Montana and Idaho 167 15. Colorado and Wyoming 338 16. Arizona and New Mexico 448 17. Washington and Oregon 813 18. Nevada and Utah 233 \ 19. California 2887 20. Michigan 926 ‘See Figure 2 to identify geographic areas included in the respective demand regions. Table 5. Monthly average onion prices by state, 1983-85 ($/cwt). State Jan Feb Mar Apr May June July Aug Sept Oct Nov Dec Average‘ Arizona3 — - - - 8.82 8.29 - - - - - - 8.61 California 15.621 14.861 16.39115.0013 12.473 10.8023 14.4012 13.231 12.471 11.621 12.131 16.971 13.83 Colorado1 13.56 12.63 13.75 13.34 - - - 12.66 10.07 10.48 11.60 14.83 12.55 Idaho1 11.80 12.20 14.97 14.52 15.87 - - 9.10 7.63 7.67 9.63 13.93 11.73 Michiganl 8.73 9.37 10.77 11.23 - - 13.67 9.90 8.33 8.37 F, 8.80 9.91 New Mexico2 - - - - - 9.20 14.80 10.43 8.77 - - ’ - 10.80 New York1 13.07 13.12 14.30 13.85 - - 16.67 14.30 12.53 13.27 13.53 13.84 Oregon1 9.63 9.00 11.00 11.00 11.66 - 9.50 8.17 6.38 7.97 9.00 11.00 10.73 Texas - - 16.033 15.833 13.303 13.6023 15.232 12.002 11.232 - - - 13.89 Utah1 g 9.04 9.07 8.91 - - - - - 6.13 6.30 6.50 9.70 7.95 Washington 11.051 10.671 12.871 3.731 15.021 - 12.322 10.592 8.271 7.071 7.931 10.881 10.96 1Summer storage 2Summer non-storage 3Spring Source: U.S. Department of Agriculture, Agricultural Statistics Board, Annual Price Summary, National Agricultural Statistical Service, June 1986. Table 6. Estimated linear regression coefficients on equations used to calculate motor carrier transportation rates linking Texas with onion demand regions ($/cwt).1 Origin of Haul McAllen, Texas Hereford. Texas Variable Coefficients t-ratio Coefficients t-ratio , Intercept ($/cwt) $0.347 95 4.640 » $030298 2.508 T Distance (SS/cwt/mile) $0.00195 45.724 $0.00206 24.312 April ($/cwt) $001130 2.005 NA NA May ($/cwt) -$0.00640 -0.127 NA NA July ($/cwt) NA NA $011555 1.569 August ($/cwt) NA NA -$0.01820 -0.873 September ($/cwt) NA NA -$0.28100 -0.924 R-squared 0.9528 .8783 1The months of March and June are excluded from the respective McAllen and Hereford. Texas equations because they were considered base months. As such, their estimated parameters are included in the intercept. NA=Not applicable shipment periods. Table 7. Estimated transportation rates linking Texas with the transportation mode1’s twenty demand regions during Texas’ dry onion shipping season, 1983 ($/cwt). Texas Shipping Demand Regions‘ Month 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 March 4.68 4.29 3.91 3.36 3.42 2.63 2.69 3.04 3.23 1.93 3.51 2.38 1.43 4.25 275 2.53 4.98 3.44 4.22 3.61 April 4.56 4.17 3.79 3.24 3.30 2.51 2.57 2.92 3.11 1.81 3.39 2.27 1.31 4.14 263 2.41 4.86 3.32 4.10 3.49 May 4.57 4.18 3.79 3.25 3.30 2.52 2.58 2.93 3.12 1.82 3.39 2.27 1.32 4.14 2.64 2.42 4.87 3.32 4.11 3.49 June 4.51 4.12 3.81 2.99 3.47 2.40 2.70 3.29 2.76 1.91 2.77 1.79 1.08 3.12 1.54 1.71 3.89 2.31 3.37 2.80 July 4.51 4.12 3.81 2.99 3.47 2.40 2.71 3.30 2.76 1.92 2.77 1.79 1.08 3.12 1.54 1.71 3.90 2.32 3.37 2.81 August 4.40 4.01 3.69 2.88 3.36 2.29 2.59 3.18 2.64 1.80 2.65 1.67 0.96 3.01 142 1.59 3.78 2.20 3.25 2.69 a September 4.55 4.16 3.84 3.03 3.51 2.44 2.74 3.33 2.79 1.95 2.81 1.83 1.12 3.16 157 1.75 3.93 2.35 3.41 2.84 ‘See Figure 2 for geographic location of demand regions. l‘ To develop confidence in the transportation model’s ability t0 project actual flows, an effort was made t0 compare model-generated flows with historical flows (Phillips et al. 1976). Unfortunately, there is little onion trade flow data; therefore, validation could not be directly accomplished. An effort was made, however, to compare city terminal market share data with projected flows (shares). Assuming that Texas’ share of a city terminal market is analogous to the region where the city is located, then some intuitive evidence is offered‘ by a comparison. The projected flows reflect the earlier finding that Texas’ principal markets are in the east, in particular, the southeast, and in the midwest and south central regions (Fuller, Goodwin, and Shafer 1989). In addition, Texas’ projected market share often approximates the historical information. The historical unload data shows Columbia, South Carolina to receive about one-third of its onion unloads from Texas, a market share which closely corresponds to the estimated share (38 percent). In general, the model-projected share in the northeast U.S. closely parallels the historical share. For j example, in the cities New York, Philadelphia, and ‘a ittsburgh, Texas’ market share historically ranged “' between 7-19 percent, 13-22 percent, and 16-23 percent, respectively. These compare with projected shares of 13 percent, 18 percent and 18 percent for Validation of Model these respective cities. In general, the model underestimated flows to midwest regions and showed no flows to the west. Historical shares at midwest locations ranged from 19 percent to 30 percent while Texas’ projected flows generally represented less than 20 percent of a terminal market. Even though Texas’ market share in the western U.S. is small (e.g., Los Angeles is 3 percent), the model underestimated this flow by projecting no movement to this area. ' There are numerous factors which contribute to the descrepancy between projected and historical trade patterns. Clearly, fresh onions are a hetero- geneous product which go to satisfy a variety of different demands. Because the model fails to recognize product qualities and associated de- mands, discrepancies exist. Further, it is difficult to estimate representative transportation charges because of the unregulated nature of these hauls. Because the transcontinental flow of commodities and products from west to east exceeds the east to west flow, very low transportation charges often exist for hauls to West Coast locations. This situation may partially account for the observed small flow of onions from Texas to western U.S. locations. In spite of descrepancies, the model seemed to cor- rectly project the major flows and accordingly, was judged appropriate to determine the competitive- ness of Texas in the national fresh onion market. Results Three scenarios are developed and analyzed with the validated trade model for purposes of evaluating Texas’ ability to compete in the national dry onion market. First, an effort is made to determine Texas’ ability to displace competing regions and expand its market share, i.e., to be cost-competitive in the national market. Then, Texas’ costs are increased to identify the sensitivity of Texas’ ex- panded market share to these unfavorable cost adjustments. The second scenario focuses on com- ,.peting regions and their ability to reduce Texas’ market share through incremental cost reductions. (Finally, the trade model is used to identify those regions which would produce if total production and transportation costs were minimized. l Ability of Texas to Expand Its Market Share '3. The purpose of this scenario is to measure Texas’ ~ability to displace competing dry onion producing regions based on possible cost advantages that Texas may possess during its market window (March through September). The validated trade model determines those flow patterns which mini- mize total producer and transportation costs given producer costs (prices), transportation charges, fixed regional damands, and fixed regional supplies (1983-85 average). By relaxing Texas’ historic monthly supply constraints, the least-cost model simultaneously determines whether the additional Texas supplies would be shipped and which regions may lose as a result of Texas expanded market share. Next, the model is used to determine whether Texas can hold this expanded market share as its costs are incrementally increased. The analyses show Texas’ cost advantage to be substantial during its market window. When Texas’ historic supply constraints (1983-85) are relaxed, the trade flow model projects shipments to increase from 4230 to 7833 million pounds, an 85 percent increase.‘ Because the model fails to consider ‘The methodology (network flow) does not allow for the incorporation of upward-sloping supply functions and, as such, the solution overstates the production potential. Regardless, it offers some measure as to the competitive- ness of Texas production. biological constraints, however, some 0f the ex- panded production is not attainable. For example, in March it is projected that shipments from Texas could be increased from 12.7 to 61.2 million pounds -- an unrealistic projection in view of the difficulty associated with producing a high quality product in large volume during this early period. Regard- less, the projected volume is a substantial increase over the average 1983-85 shipments, and this implies that Texas producers are not at a cost disadvantage during their market window. A closer examination of the solution to the trade model offered two additional findings regarding the competitiveness issue. First, most of the pro- jected increase in onion shipments by Texas was at the expense of California; i.e., California shipments were reduced to relatively low levels when Texas supplies were unconstrained. This implies that Texas producers have an unexploited cost advan- tage relative to California. Second, the projected increase in shipments differs by month. The model projects that Texas’ cost advantage yields modest increases in shipments during April and June (1() percent) but more substantial increases in May (41 percent), July (112 percent), August (106 percent), and September (105 percent). To determine the sensitivity of Texas’ expanded market share to unfavorable cost movements (in- creases in production and transportation costs), the trade model was used to project monthly shipments from Texas as costs were increased 5, 10, 15, and 20 percent, or an average of about $.85, $1.70, $2.55, and $3.40 per cwt, respectively. If the unfa- vorable cost adjustments have little effect on Texas shipments, then it is appealing that the cost ad- vantage is meaningful; conversely, if shipments are substantially reduced, then the cost advantage is modest and possibly insignificant. The analyses show a 5, 10, 15, and 20 percent increase in Texas costs to reduce the expanded shipment volume (7833 million pounds) to 7071, 5000, 4579, and 4523 million pounds, respectively. Shipments in the March, April, May, and June window are reduced to historic 1983-85 levels with a 10 percent increase in production and transportation costs in Texas. July shipments decline to the historic level with a 15 percent increase in costs. Only projected August and September shipments exceed the historic level when costs are increased to 20 percent. Thus, the analysis shows Texas monthly shares to be sensitive to adverse cost movements except in August and September. During August it is projected that Texas could increase shipments from 465 to 614 million pounds and in September from 43 to 231 millions pounds, a projected increase in total shipments of about 8 percent. The expanded ship- ments are to Arkansas, Louisiana, Mississippi, Oklahoma, and Texas markets and are at the expense of California. Vulnerability of Texas’ Market Share California and Arizona are important dry unitm shippers during 'l‘exas’ market wimltnv. To lest the sensitivity of 'l‘exas' market. share. ('aliforni:t’s W l4 11 '7 — 4,4 ‘I Iwggi’ Figure 2. Demand regions included in spatial model. Q- t 10 historic supply constraints were relaxed and their costs subsequently reduced. The trade model shows \the removal of California’s historic supply con- straint to have little effect on California or Texas. This finding is in contrast to the earlier scenario which showed shipments from Texas to dramati- cally expand when its supply constraints were removed and California’s shipments to decline as a result of Texas’ market expansion. This analysis shows the market share held by Texas to dramati- -\ cally decline only if California’s costs are reduced (Figure 3). When California costs are reduced by 10 percent, Texas share of the market is reduced to 60 percent of its historic level. Texas shipments in June, April, and May are most vulnerable to Cali- fornia’s cost reductions while March, August, and September shipments are little affected. These findings reinforce earlier results regarding the opportunity to expand Texas shipments in August and September. Arizona is an important dry onion shipper in May and June. Based on the trade flow model. Arizona has the ability to displace nearly all com- peting regions during this period. The exception is New Mexico which retains its historical shipments in June. Location of Production Based on Least-Cost Criteria To gain more insight into regional cost advan- tages and/or disadvantages, historical supply con- straints for all producing regions were relaxed. Based on this modification, the trade model projects least-cost dry onion production in the United States. This analysis shows onion production to be concen- trated in six states (Michigan, Oregon, Utah, Washington, Arizona, and New Mexico). Summer onion producers (Michigan, Oregon, Utah, and Washington) supply 81 percent of the onion output while Arizona and New Mexico generate the remaining supply. Storage onion stocks are the source of supply in March and April while during May, June, July, and August onion supplies originate in Arizona, New Mexico, and Oregon. Even though the unconstrained trade model in- cludes simplifying assumptions, it shows there would be a dramatic relocation of dry onion production if cost were the only factor determining the national production pattern. Clearly, other variables such as quality have an important role in explaining the geographic location of production. Million Cwts 5 O 6 1O Percent reduction in California costs K 15 2O Figure 3. Annual shipments from Texas in view of reduced costs in California. 11 Summary and Conclusions Historically, Texas producers supplied about 20 percent of the national onion market. In recent years, however, their market share has declined to about 14 percent. Total shipment of fresh onions by all U.S. producers has trended upward over time while shipments from Texas edged downward; consequently, a declining market share for Texas producers. Texas’market window extends from March through September but is concentrated in the April-May window when about sixty percent of the state's fresh onion production is marketed. Texas’ April onion production is the nation’s first new crop shipments, and, as such its competition is primarily limited to carryover stocks and imports from Mexico. In May-August, new crop production in California, Arizona, New Mexico, and other Summer producing states offer competition. His- torical data shows Texas’ share of the fresh market in each shipping month to be trending downward except in April. Much of Texas’ declining market share in May, June, July, and August has been claimed by California producers. The objective of this study was to determine the cost-competitiveness of Texas in the national onion market and identify windows where Texas ship- ments may be increased. Although the analysis was accomplished with a national model, the focus of the research was on Texas and its competitors. An inter-regional trade model was developed and validated to address the competitiveness issue. The analysis shows Texas to be cost-competitive during the market window, i.e., based on the cost parameters included in the trade model, the decline in Texas’ market share is not attributed to un- favorable costs. This outcome is appealing in view of the earlier finding that Texas production costs are similar to major competitors during its window and Texas is closer to many of the major markets than its principal competitors. Consider that much of the market for Texas onions is located in the eastern half of the United States and most of the competing onion production is located in the western United States. It is estimated that Texas’ transport cost advantage over California is near $2.50/cwt in the northeast U.S. markets and in excess of $3.00/cwt in southeastern markets. 12 It was reasoned that Texas’ greatest opportunity for increasing shipments and market share would be in periods when its greatest cost advantage existed. This was determined with the trade model by removing Texas’ historic supply, constraints, incrementally increasing Texas’ costsl; then solving the associated models and identifying flows. This analysis shows little opportunity for Texas to expand its market share during its peak shipment period (April-May), and a modest ability to increase July shipments with any increase in Texas ship- ments coming at the expense of California. The most promising window for expansion would appear to be in August and September, a period when Texas’ cost advantage is substantial. Although the absolute increase would be relatively small (4000 cwts), it does represent a 70 percent increase in shipments for Texas during the August-September window. Although Texas is a cost-competitive producer during its window, it generally has higher costs than many of the regions which harvest storage onions in the late summer. Consequently, if the location of the nation’s onion production were based on cost, Texas would be an insignificant supplier. Onions consumed in the spring and early summer would come from storage stocks. This implies that Texas’ current role as a major supplier is based on its ability to offer a high quality product in the early spring, a product which is preferred to an onion coming from storage. Regardless, it is im- portant that Texas be cost competitive during its window. California, a major competitor, has the ability to dramatically reduce Texas’ market share if it were to modestly lower cost. Thus, cost-reducing innovations in combination with quality improve- ments are important for Texas producers. In summary, the proximity of Texas to the major eastern markets gives it a cost advantage relative to the principal onion producers in the western United States. If high-quality onions can be pro- duced at costs which are comparable to those of competing regions, then there would seem to be opportunities to expand shipments, in particular, in the August-September window when substantial cost advantages exist for Texas producers. N Acknowledgments Countless individuals have made substantial con- tributions to this research. Great appreciation is extended to Grant Vest, head, Department of Horticulture, Texas A&M University (TAMU), who made resources available to carry out this effort. Others making substantial contributions include Tom Longbrake, extension horticulturist, Texas \ Agricultural Extension Service; Haruna Bello, graduate assistant; Melissa Huie Schneider, grad- uate assistant; Doo Bong Han, graduate assistant; Oral Capps, professor in the TAMU Department of Agricultural Economics, and Merritt Taylor, extension economist, Texas Agricultural Extension Service. References Bressler, R. B., and R. A. King. Markets, Prices and Interregional Trade. New York: John Wiley and Sons. 1970. Fuller, Stephen, H. L. Goodwin, and Carl Shafer. Trends of the Dry Onion Industry in Texas and the United States. Texas Agricultural Experiment Station. MP- 1672. July 1989. Phillips, D. T., A. Ravindran, and James T. Solberg. Operations Research: Principles and Practices. New York: John Wiley and Sons. 1976. South Texas Onion Committee Report, Mercedes, Texas, 1975-1985. Texas Department of Agriculture, Texas Agricultural x Statistics. Texas Vegetable Statistics. 1976-1986. J. S. Department of Agriculture, Agricultural Market- ing Service, Fruit and Vegetable Division. Fruit and Vegetable Truck Rate and Cost Summary. 1982-1987. 13 U. S. Department of Agriculture, Agricultural Market- ing Service, Fruit and Vegetable Division. Fresh Fruit and Vegetable Shipment Totals by Commodity, States and Months. FVUS-7. 1977-1987. U. S. Department of Agriculture, Agricultural Statistics Board. Annual Price Summary. National Agricultural Statistical Service. June 1986. U. S. Department of Agriculture, Economic Research Service, Vegetable Situation and Outlook Yearbook. TVS-Series. 1976-1987. U. S. Department of Agriculture, Human Nutrition Information Service, Consumer Nutrition Division. Nationwide Food Consumption Servey, 1977-78. Report No. H-9. August 1983. U. S. Department of Commerce, Bureau of Census. 1980 Census of Housing, Detailed Housing Characteristics, U. S. Summary. December 1983. [Blank Page in Original Bulletin] l '1“: ‘- [Blank Page in Original Bulletin] “ a "'4 i‘ Q W