22 N. H. Agri. Experiment Station [Station Bull. 307 



Wliile the range of charges of siniihir routes shows only the differ- 

 ence between the lowest and highest route and is therefore easily af- 

 fected by abnormal cases, it does give a rough measure of the degree of 

 variation present within the group. 



Considering only those groups containing five or more routes, none 

 show a range as low as might be expected under freely competitive con- 

 ditions. The ranges most frequently found are 10 and 15 cents. (The 

 range of 30 cents found in one group would be lowered to 20 cents if 

 one truck route, on whicli a temporary situation had lowered the rate 

 to five cents, were omitted.) 



The standard deviation is a much more accurate measure of variation 

 than is the range. Unlike the range, it takes into account the variation 

 of all the charges and not only the lowest and highest. The standard 

 deviations for groups containing five or more routes were found to be 

 about five cents. Broadly interpreted, this means that about two- 

 thirds of the routes in each group show charges within five cents above 

 and five cents below the average charge for the group (a total variation 

 of ten cents), and that one-third of the routes have charges outside 

 these limits. 



To compare the variation within different groups, the standard de- 

 viation is expressed as a percentage of the average for each group, and 

 the result is termed the co-efficient of variation. Most of the groups 

 show co-efficients of variation" in the neighltorliood of 20 per cent. 



Judging from these measures of variation, charges for routes carry- 

 ing similar volumes of milk and traveling similar distances show much 

 greater variation than would be expected if freely competitive condi- 

 tions existed and charges were based upon services rendered. If those 

 routes with rates near the lower limit of the group are charging enough 

 to cover their costs and allow for replacements, then routes within the 

 same group but with charges near the upper limit rnnst be receiving 

 more than the income necessary to provide and maintain efficient truck- 

 ing service. That many of the routes charging low rates have been in 

 operation for some years ; that, in general, trucks are comparatively 

 new (about one-half of the trucks were 1936 or 1937 models) ; and that 

 there appears to be no correlation of rate with age of truck — these facts 

 suggest that most operators liave found the rates sufficiently remuner- 

 ative for them to continue in business. 



II. Extent to whicli individual charges paid by producers vary with 

 distance from market : 



It has l)eeii seen above tluit under conditions of perfect competition 

 charges paid by producers should vary with their distance from mar- 

 ket, density of jji-oduction i-cmaiiiing constant. An attempt to ascer- 

 tain the individual rate structure al)out typical markets is presented in 

 Figure and Appendix Table HI. 



In tiie jireparation of this chart and tahle. tlie rates paid by individual 

 pi'oducers shipping 1o Lancaster and Nashua have been si)ot1ed on maps 

 and the average rates paid l)y shippers within various radii of tlie mar- 



' Appendix Tabic II .shows standard deviations and co-efficients of va- 

 riation for each of the jjroups containing five or more routes. 



