Atfera:rg xxf Hsrirultnrc^ Xlibsral Hrts Itt^ethnologg EXPERIMENT STATION LIBRARY Bulletin 295 April, 1937 THE MILKSHEDS OF NEW HAMPSHIRE A Study of Their Characteristics and Relationships By ALAN MacLEOD New Hampshire Agricultural Experiment Station, University of p New Hampshire, Durham, N. H. THE MILKSHEDS OF NEW HAMPSHIRE ! A Study of Their Characteristics and Relationships' ' By ALAN MacLEOD, Assistant Economist in Marketing To understand the working of economic forces in a particular mar- ' ket, a knowledge of the physical factors influencing the supply of milk and cream is necessary. This study attempts to supply such basic in- formation regarding the dairy industry in New Hampshire. State-wide maps showing location of producers and their markets have been prepared. For the markets operating under the State Milk Control Board, estimates of sales by type of distributors and other perti- nent data are presented. This combination of graphic and tabular data gives detailed information regarding the milksheds of the State in a manner which should prove of value in the planning, conducting, and interpreting of future studies of milk marketing. The Problem Milksheds differ in their response to changing demand conditions. A small market, isolated geographically and economically from other markets and supplied with fluid milk and cream solely from near-by pro- ducers who have no alternate outlets of any consequence, can ignore to a r considerable extent price fluctuations taking place in other markets. This is not true in the case of a small market situated in the center of the milkshed of a large market. With producers able to change from one market to another in response to price shifts, any conclusion regard- ing supply in the small market must take into account the effects of hap- penings in the large one. Therefore, it is important not to confine a study of physical factors affecting the supply of milk and cream in a particular market to answer- ing questions as to who produces the milk, where he lives, how far he lives from his market, how he gets his milk to town, and whether he dis- tributes it himself or through a dealer 's plant. These tell only a part of the story. Two markets may secure their milk from about the same number of producers living about the same average distance from town with similar methods of bringing the milk to market and distributing it ; yet to conclude that similar price or pro- duction policies will affect both markets in the same way and to the same extent may be widely in error. It is necessary also to consider the in- fluences (both actual and potential) of other markets. Some light upon the extent of such influence is thrown by maps showing size, location, and present market outlets of dairy farms in New Hampshire. Where, within a six-mile radius of a small market, a num- ber of farms are now shipping to a large market (or markets), develop- (1) A "milkshed" is the area or region furnishing milk to a particular com- munity. (2) This is the first New Hampshire publication in the New England-wide milk marketing study, which is sponsored by the New England Research Council. Acknowledgment is made to the State Milk Control Board, local boards of health, milk distributors, and others who supplied data for use in this study. 4 N. H. Agr. Experiment Station [Bulletin 295 ments in the large market must have a much more direct effect upon supply in the small market than would be true, if only a few farms with- in such a radius were shipping outside. Changing market relationships show their first effects in the areas where inter-market competition takes place. By focusing attention up- on such areas the results of price shifts between markets may be discov- ered without wasted time or effort. The application to other markets of conclusions regarding supply drawn from the study of a particular market, may be aided by the maps and tabular data reported in this study. Likewise, fundamental dif- ferences between markets are brought to light. The Milksheds (1) General Maps have been drawn showing the location, size of herd, and mar- ket outlet of all commercial dairy farms in New Hampshire. In their preparation the following methods were used : Town maps on a scale of two inches to the mile showing location and size of herd of each commercial farmer were available from a types-of- farming survey conducted by the Experiment Station in cooperation with the Agricultural Adjustment Administration in 1935. In making this survey, town maps showing roads and residences were copied from < old Geological Survey maps. One or more selectmen (and in most cases 1 several other key men) were visited in each town and asked to assist in '' the work. From the town inventory a list of farmers was prepared and checked against the names on the map. From the State Milk Control Board, local health officers, milk distributors and others, the destination of each farmer's milk was learned. County maps were prepared from these data and, after reducing the scale, a State map was drawn. Supplementing these maps are data regarding sales in all intra- state markets in which the State Milk Control Board operated. Technical difficulties prevent printed reproduction of any but a few relatively small markets. Hand-colored copies of the State map are available for reference at the Experiment Station, the U. S. Depart- ment of Agriculture, and the New England Research Council office in Boston. Large scale maps of most local markets are on file at the Ex- periment Station. To illustrate the information made available certain milksheds have been reproduced on a small scale in this study. In the three maps shown here, the solid red circles represent pro- ducers selling to the market under consideration ; the solid black circles, sales to other local markets ; the circles with dots, no sale, and the plain circles, sales to small local markets. The green triangles represent pro- ducers selling to out-of-state markets. The symbol used refers to the market in which the milk is consumed. By varying the scale of the circles and triangles, differences in size of dairy enterprise are indicated. The maps are not completely accurate. A farm may be shown on the wrong side of the road or even a short distance away from the road. \ These are not topographical maps. As only farms of three cows or more are shown, some small producers will not appear. Furthermore, this is a cross-section analysis and it had to be done over a period of time. Shifts of producers in and out of production, of producers from one market (or dealer) to another, and of cow numbers are constantly taking place. April, 1937] The Milksheds of New Hampshire 5 A completely accurate picture of milksheds throughout the State could only be obtained by an instantaneous exposure. Instead of that, a suc- cession of pictures was taken and joined together to make the whole. The inaccuracies inevitably introduced by such a procedure are prob- ably not sufficient to affect the value of the maps for the purposes for which they are intended. For 35 markets in which the Control Board operated, over 98 per cent of the production is accounted for, and no single market shows less than 95 per cent located. Judging from markets where 100 per cent of production has been found, no significant change in the picture is introduced when less than five per cent of production is missing. Maps are reproduced in this study of the Claremont, Keene, and Laconia milksheds, and they have been chosen to illustrate the differ- ences and similarities which occur between markets of about the same size. Claremont (population 12,377) is situated in, and Keene (popu- lation 13,794) is close to, the wholesale dairy section of the Connecticut River Valley. Laconia (population 12,471) is in the summer resort section. (2) Total Sales in Various Markets In this study, the term "dealer" is applied to a distributor who purchases the major portion of the milk he sells. The term "producer- distributor'' is applied to a distributor who produces the major portion of the milk he sells. Estimated daily sales of milk and cream (on milk equivalent basis) by dealers, producer-distributors, and stores in all 35 markets are shown in Table I. These estimates have been made from Control Board rec- ords of distributors' average daily sales in the year immediately preced- ing the summer of 1936. These data, in most cases, are producers' esti- mates and are not taken from actual sales records. All of the cities and most of the large towns are included in this table. The amounts listed under the column ' ' Total ' ' represent estimates of net daily sales in each market. Estimates of dealers' and producer- distributors ' sales are given separately. Items under ' ' Stores ' ' are not included in the estimates of net sales, as double counting would be in- troduced. Milk or cream sold by a store has already appeared in deal- ers' or producer-distributors' sales. (3) Importance of Different Types of Distributive Agencies Total daily sales of milk and cream in New Hampshire markets operating under the State Milk Control Board varied from over 30,000 quarts to less than 200 quarts. In the three markets whose maps are reproduced here, Claremont, Keene, and Laconia, sales varied between five and seven thousand quarts. But estimates of total sales conceal many significant market dif- ferences. Some markets (and this is true of the two largest state mar- kets) receive most of their milk from dealers. In such markets, pro- ducer-distributors play only minor roles ; and producers, therefore, have little direct contact with the problems of retail distribution. Markets of this type include Manchester, Nashua, and Portsmouth. Table I shows that in each of these markets, well over 90 per cent of the milk was distributed by dealers. ▲ o © LEGEND PISIMCIPAL- MAl^l^ET OTHEI2. LOCAL Mii^^IZk: EITS OUT OF STATE UOCAl- 3A<1_& NO 3AL.E fEW EA^XEIR KXKEBX CiAEEMONT LEGEND ▲ • I TO S C ows ik • GTo 1 O COWS ▲ • \ I TO 10 COWS A # 2 I TO 4 0 c OWS ▲ • 4-1 TO GO c 0>JVS OVEI? (bO cows NE.NX^ MAMPSHIBE AGftlCUl_TUl2AL_ ExPEI^I M E.NT STATION N. H. Agr. Experiment Station [Bulletin 295 Table I. Daily Distribution of Milk and Cream' in New Hampshire Markets (1935-1936) MHk and Cream Distribution i Producer- City or Towna Total Dealers Dis tribute rs Stores^ No. Quarts No. Quarts No. Quarts No. Quarts Manchester 42 33,515 26 32,175 16 1,340 318 6,793 Nashua 30 15,805 20 15,215 10 590 158 4,024 Concord 79 13,143 7 7,410 72 5,733 95 2,964 Berlin 47 7,174 16 4,554 31 2,620 74 1,871 Portsmouth 18 7,727 3 7,050 15 677 68 2,788 Keene 66 6,421 9 3,180 57 3,241 43 1,518 Dover 78 6,186 6 1.948 73 4,238 55 1,278 Laconia 70 7,006 6 2,795 64 4,211 60 1,452 Claremont 89 5,117 4 675 85 4,442 37 962 Rochester 77 4,199 3 383 74 3,816 56 1,029 Lebanon 23 2,226 3 770 20 1,456 15 361 Franklin 46 2,798 , , ■ • • • 46 2,798 20 508 Somersworth 30 1,647 3 209 27 1,438 18 352 Exeter 31 3,622 1 245 30 3,377 18 420 Newport 51 2,590 1 422 50 2,168 16 349 Littleton 32 2,099 2 535 30 1,564 17 438 Suncook 25 1,393 2 417 23 976 8 202 Hanover* 18 3,498 1 810 17 2,688 7 227 Lancaster 22 1,383 , , > • • • 22 1,383 10 284 Gorham 14 1,047 1 200 13 847 6 135 Farmington 40 1.159 1 46 39 1,113 12 329 Hampton and Rye Beach^ 13 8,929 5 5,525 8 3,404 23 1,035 Newmarket 30 1,131 , , .... 30 1,131 4 124 Northumberland 17 927 , , 17 927 9 136 Lisbon 20 976 .... 20 976 4 68 Walpole 10 666 1 250 9 416 7 377 Colebi-ook 16 710 1 20 15 690 5 55 Whitefield 24 802 2 70 22 732 8 298 Bristol 25 1,058 4 301 21 757 15 317 Newington, Stratham and Greenland 11 231 1 112 10 119 Belmont 6 287 2 168 4 119 6 '187 Henniker 12 528 , , .... 12 528 4 93 Stewarts town 4 170 .... 4 170 3 50 North Stratford 2 165 .... 2 165 1 34 BethlehemS 18 1,837 3 970 15 867 4 94 1 Cream expressed as milk equivalent. 2 Roughly in order of population, 1930 Census. Non-coincidence of Control Board areas with census divisions makes exact ranking impossible. 3 Not included in Total. * College town. ^ Summer only. 8 Resort town. Markets dominated by neither dealers nor producer-distributors include Concord, Keene, and Laconia. And markets served largely by producer-distributors include Claremont, Rochester, and most of the smaller communities. Table I shows the wide variation in number of dealers and producer- distributors from market to market. Store sales are included to make the picture complete. In the larger cities, the number of stores handling milk may greatly exceed the number of other distributive agencies. This situation is reversed in the smaller markets, where producer-distribu- tors ordinarily outnumber stores. (4) Extent of Milksheds Milksheds differ in extent. Production densities being equal, the larger the market the farther must it reach out in order to satisfy its requirements. Where, as is frequently the case, production densities also vary, another factor must be considered. The Keene and Berlin markets illustrate strikingly the effect of different production densities ril, 1937] The Milksheds of New Hampshire 'able II. Approximate Numbers of Cows on Farms Supplying Local Markets from Various Distances Together with Comparative Numbers of Cows Supplying "Other Large Markets" Destination (ilkshed of Milk Within 3 miles Dista 3 to 6 miles nee from 6 to 9 miles Local Market 9 to 12 miles Over 12 miles Out-of- state ichester Local Other 231 668 503 129 408 507 2,217 •• hua Local Other 222 16 587 476 289 817 142 1,000 865 10 cord Local Other 300 15 713 445 272 152 597 33 •• in Local Other 133 159 247 75 122 250 ismouth Local Other 430 119 335 185 40 652 10 Many 10 10 ae Local Other 508 322 172 190 30 Many 10 •• er Local Other 608 40 140 114 71 Many 7 Many ... 60 ania Local Other 324 81 363 189 46 235 11 Many 40 •• emont Local Other 510 246 158 477 Many Many ... •• hester Local Other 400 100 154 50 Many Many 36 40 inon Local Other 374 347 59 Many Many Many ... 5 nklin Local Other 253 42 90 283 3 Many Many ... • • lersworth Local Other 494 205 Many Many Many ... 135 ter Local Other 122 335 179 Many 9 Many Many ... •• /port Local Other 231 31 75 278 22 Many Many ... •• leton Local Other 284 262 20 642 Many Many ... 10 lover Local Other 130 32 41 668 Many Many ... 250 caster Local Other 90 723 54 Many Many Many ... •• cook Local Other 138 102 57 Many Many Many ... •• ham Local Other 80 2 15 61 Many Many 25 2 mington Local Other 129 19 17 166 1 Many Many 5 •• vmarket Local Other 123 211 28 916 Many Many ... jon Local Other 93 798 32 Many Many Many ... •• Ipole Local Other 83 401 15 Many 5 Many ::: 5 jbrook Local Other 57 593 15 Many Many Many ... 20 itefield Local Other 70 285 28 671 Many Many ... 5 3tol Local Other 75 44 20 Many 5 Many Many 25 •• mont Local Other 30 253 4 Many Many Many ... •• iniker Local Other 64 295 19 185 Many Many ... •• warts town Local Other 2 651 12 Many Many Many ... 6 •th Stratford Local Other 27 95 224 Many Many ... •• hlehem Local Other 109 87 15 225 8 Many Many ... 220 -thumberland Local Other 110 205 20 Many Many Many 6 10 N. H. Agr. Experiment Station [Bulletin 295 on the extent of the milkshed. Keene, situated in the center of a j^ood dairy section, receives most of its milk from within nine miles of the center of town. Berlin, on the other hand, being located in a wooded mountainous section must go from 15 to 18 miles to get a part of its milk. And yet consumption in the two markets, as indicated by sales, does not differ greatly. Maps show best the extent of milksheds. To assist the eye, circles with radii of three, six, nine, and twelve miles have been drawn around the centers of the various markets. To supplement, and to some extent to substitute for, the maps, a method of tabular presentation has been adopted. Within each three-mile radius marked off by the circles, the approximate numbers of cows on farms selling to the local market and to other large markets are indicated. In markets where comparisons have been made of daily sales of milk and of cow numbers, the two have been in rough agreement. The number of cows is also helpful in show- ing market inter-relationships. Table II shows, for most of the market areas, the number of cows within each three-mile radius up to 12 miles from the market. Over 12 miles only three markets — Berlin, Manchester, and Nashua — get ap- preciable amounts of milk. Where purchases are made out-of-state, no attempt has been made to estimate the distances from which the milk comes. The cow numbers estimated in each area under "Other Large Markets" represent estimates of cows on farms shipping to markets other than that under consideration. (5) Relationship of Markets to One Another Well over half of the milk produced in New Hampshire is con- sumed within the State. Table III, which includes only farms with three or more cows, shows the number of cows in each county on farms shipping out-of-state and supplying local needs. Were one- and two- cow farms included, the estimates of the percentage of production con- sumed within the State would be even higher. Another factor which, if taken into consideration, would increase still further the percentage of local consumption is that, on farms reported shipping out-of-state, some milk is retained for home use. The data in Table III should be read with these two modifying factors in mind. State and county estimates of the importance of out-of-state mar- kets, while valuable for some purposes, do not show the influence of other Table III. A-pproximate Number of Cows on Farms Shipping Out-of-State and Supplying Local Needs County Percentage of Number of Cows on Farms Cows Supplying Supplying Shipping Local Needs Local Needs Out-of-state Total Belknap 77 2,372 693 3,065 Carroll 90 2,202 237 2.439 Cheshire 60 3,033 2.020 5,053 Coos 28 2,616 6,611 9,227 Grafton 37 5,373 8,984 14.357 Hillsborough 72 5,638 2,203 7.841 Merrimack 83 5,993 1,270 7,263 RockinKham 52 3.759 3,439 7,198 Strafford 86 2,994 490 3,484 Sullivan 44 1,947 2,460 4,407 state 56 35,927 28,407 64,3341 1 82 per cent of 1935 Census. Does not include farms with very small herds. April, 1937] The Milksheds of New Hampshire 11 markets upon a particular local market. In most of New Hampshire by "other markets" is meant the Boston market. This is largely true of the three milksheds reproduced in this study. The Claremont, Keene, and Laconia milksheds appear similar as lojig as attention is focused upon the circles. But now, look at the tri- angles. Keene, though near the Connecticut Valley with its well de- veloped wholesale milk business, is comparatively isolated. No milk goes to other large markets from within six miles of Keene and only a small amount from within a radius of nine miles. The Claremont milk- shed, on the contrary, shows that within three miles farms shipping out of state have about half as' many cows as those selling locally. In the three- to six-mile area the producers selling on the Boston market great- ly outnumber those selling locally. Laconia, while not as isolated, eco- nomically, from the Boston market as Keene, is not as exposed as Clare- mont. Considerable milk from within six, and even three, miles of La- conia goes to other large markets. Table II presents in tabular form similar material for 34 New Hampshire markets. In each milkshed the numbers of cows within cer- tain distances whose milk is consumed in that market are compared with the numbers whose milk goes to other large markets. These compari- sons are made for the areas marked off by concentric circles three miles apart. In summary form, much the same sort of information can be ob- tained from use of this tabular material as is supplied by the maps. The comparison of the Claremont, Keene, and Laconia markets indicates the use of these data. As stated earlier in the study, this factor of ' ' other market ' ' influ- ence is one of great importance. There are few markets of any size in New Hampshire which are not influenced to some extent by conditions in the Boston market. Usual description of the Boston milkshed in- cludes the whole of New Hampshire. Within the Boston milkshed are large secondary milk markets such as Manchester and Nashua, which in turn exert an influence upon smaller communities. How direct and how strong are these outside influences upon any particular market are pertinent questions and upon them this study should throw some light. I 08p> PAS con ryp .-4K>i385 306 >• 1