June, 1941] Local Structure of Milk Prices 31 



which is shown the net price advantage to Manchester shippers over Bos- 

 ton shippers. 



No attempt at establishing a relationship between the relative advan- 

 tage of Manchester over Boston and the per cent of producers shipping 

 to Manchester in each particular month has been made. It is most unlikely 

 that producers who shift markets as seldom as do most milk producers 

 would respond to any particular month's differential and even though a 

 desire to change markets has been felt, the obstacles in the way of such 

 change may prevent for a considerable time or indefinitely, the actual 

 change. Also, the effects of other factors than price 'have not been eUm- 

 inated, and any conclusions based on correlation between the two series 

 under consideration, even assuming various time lags, would be logicallv 

 difficult to justify. However, if a price differential is continued over a 

 long period of time, it may have a cumulative effect on producers. Just 

 how long a differential must be in effect before it produces a shift, there 

 is no way of knowing. However, if instead of pay-periods, a comparison 

 of the same two series is made by yearly averages, with differing lags of 

 the percentage shipping to iManchester behind that of the price differen- 

 tial series, a positive correlation might be found. 



A more likely explanation of shifts between two markets, than that 

 they take place in accordance with the differential price of one market 

 over another, allowing for some definite time lag, is that following a 

 I-Tice differential which has been sustained for a "sufficient" time, a re- 

 serve of producers is created who, given the opportunity, would shift to 

 the higher priced market. But institutional factors may prevent for 

 months, or even years, such shifts being made, though from the stand- 

 point of price differentials they would be desired by many producers. 



This is just another way of saying, that in this particular area, mar- 

 kets were not all free, and producers were unable for many reasons to 

 respond to price differentials, even though such differentials may have 

 been in effect for a number of years. 



Of course, other factors mentioned above, have influenced producers 

 to ship to particular markets. Such considerations as financial security, 

 methods of sale, personal loyalties, etc., have been important. To some 

 extent producers may have been unacquainted with price relationships 

 between the two markets, though it is doubtful whether sustained differ- 

 entials existed in this section without being apparent to most producers. 

 But as long as prices paid are not available freely to any seller, long con- 

 tmued differentials between the two markets can continue and the work- 

 ing of free competition to bring prices (taking into consideration other 

 factors of sale) into a stable competitive relationship may be indefinitely 

 postponed. 



The net shift of producers in the Haverhill area of only 3 per cent 

 is in spite of the fact that in the first half of the period under study, Man- 

 chester prices were at a significant discount under Boston (without taking 

 into consideration Grade A premiums), and in the second half, Manches- 

 ter prices went to a large premium over Boston. 



If judgments of future prices are made on the basis of past experi- 

 ence, the continued advantage of Boston over Manchester for 1931-35, 



