May 30, 2001

 

 

 

Mr. Daniel Smith

Executive Director

Northeast Dairy Compact Commission

64 Main Street, Room 21

Montpelier, Vermont  05602

 

Dear Mr. Smith:

 

Thank you for your letter as of May 21, 2001 requesting assistance in reviewing the recent study by Professors Ronald Cotterill and Andrew Franklin entitled, “The Public Interest and Private Economic Power:  The Case of the Northeast Dairy Compact.” 

 

I am more than willing to assist the Compact Commission with an unbiased assessment of this study.  In fact, I also asked Professor James Dunn from Penn State to assist me in this review. 

 

My overall observation is that Drs. Cotterill and Franklin devised an unorthodox methodology merely to prove their assumption that the Compact had little impact on retail milk prices in New England.  Once this assumption was made, they then concluded that the balance of the observed retail milk price increases was due to market power and “tacitly collusive pricing at the retail level.”

 

This is a very strongly worded study.  My concern is that the data and methodology does not support the conclusions reached.  In other words, I don’t think the Agricultural Economics profession will support the conclusions Drs. Cotterill and Franklin reached regarding the retail impact of the Compact and his theories on tacit collusion.

 

First, Drs. Cotterill and Franklin assumed that the farm-to-retail margin for milk would decrease once the Compact was enforced.  This was based on two assumptions:  1) asymmetric price transmission and 2) input price risk reduction.  Let me deal with the second assumption.  The authors state on pg. 6 of their report that, “the standard deviation of the farm price fell from 0.102 prior to the Compact to zero after the Compact.” 

 

 

PAGE 2

DAN SMITH

May 30, 2001

 

 

Unfortunately, this is not true.  I measured the standard deviation for the farm price—equal to the Class I cost of milk plus coop premiums plus the Compact over-order obligation—both before and after implementation of the Compact.  I found it fell from 1.03 during the period January 1994 – June 1997 to 0.99 during the period July 1997 – December 2000.  The point is, Drs. Cotterill and Franklin were wrong to assume that the Compact had reduced the variation in the Class I cost of milk to zero.  Therefore, their assumption that this reduction should translate into a tighter margin is also wrong.

 

The authors made a rather strong observation that “there is absolutely no relationship between farm and retail prices” for milk.  Therefore, they assumed by this statement that the Compact had little or no impact on retail milk prices.  In the process they rejected a rather large body of economic literature that suggests there is in fact a very strong relationship between farm and retail milk prices.  But this is an important point in this study since it justifies the next step.

 

Without a classic farm-to-retail model to rely on, the authors “created” an unorthodox methodology to evaluate the impact of the Compact on retail milk prices.  This methodology was supposed to use a “before and after” approach in analyzing the Compact.  In fact, there was no comparison between the farm-to-retail margin before the Compact and after the Compact.  Rather, an index was created.  This index compared the trend in retail milk prices before the Compact to the retail price increase during one month in July 1997.  Again, this is an unorthodox methodology that is clearly not supported by the Agricultural Economics literature.

 

The author’s methodology depends critically on the assumption that the Compact elevated retail milk prices 6 cents per gallon in 30 of the 40 after-Compact periods, for an average increase of 4.5 cents per gallon.  This assumption/conclusion was derived from an observation that the Class I farm price of milk (Class I price plus Compact premium) rose 6 cents per gallon from June 1997 to July 1997. 

 

My own computation shows the Class I cost of milk to processors, which includes the announced minimum federal order price plus the cooperative premiums plus the Compact over-order obligation, averaged 12 cents per gallon over the period July 1997 through July 2000.  Thus my figure conflicts with the authors claim that only a 4.5 cents per gallon elevation in the retail milk price is justified.

 

Let’s get back to the authors Compact damages model.  Essentially they used a retail price trend with a 6-cent per gallon increase in the retail milk price due to the Compact and a slight elevation for other marketing costs.  He then compared this index to the actual retail milk prices.  Thus, the authors never examined the actual farm-to-retail milk

 

PAGE 3

DAN SMITH

May 30, 2001

 

 

 

margin before and after the Compact implementation.  They also fixated on just one period of the historical data:  a comparison of the months prior to the Compact to July 1997.  It is this comparison and the use of the index that I am strongly suggesting reflects an “unorthodox methodology.”

 

Finally, the author’s conclusion lacks a sense of logic that is necessary when reaching such strong conclusions.  First, the authors only examined supermarket scanner data representing about 40 percent of the retail milk market.  The other 60 percent was not studied.  They conclude that the Compact raised $128.5 million over the Compact period studied (July 1997 – July 2000).  That translates to $51.5 million higher costs for processors on sales at the supermarket level (0.40 * $128.5).  Retail milk prices at the supermarket level rose $130 million.  However, of this amount, $19 million was due to the Compact, $49 million was due to market power, and the rest, $61.6 million, was due to higher processing costs and Class I price spikes.

 

If only $19 million was passed on to consumers and processors paid $51.5 million higher costs for milk due to the Compact, I would conclude that processors/retailers “ate” the difference of $32.5 million.  Given their study, I would argue that processors/retailers lacked the market power to pass these higher costs on to consumers.  However, Cotterill and Franklin assumed just the opposite.  Again, their numbers don’t add up.

 

To conclude, Cotterill and Franklin used an unorthodox methodology to reach unsubstantiated claims regarding the retail price impact of the Compact.  Their methodology is simply not supported by the Agricultural Economics literature.  In addition, their claims of market power and tacit price collusion are not supported by the data.  One has only to ask whether a retail milk price of $2.78 per gallon in the Northeast supports such claims.  Logic will tell you it does not.

 

Sincerely,

 

 

 

Kenneth W. Bailey