Group Title: proposed regional model for agricultural adjustment in the west
Title: A proposed regional model for agricultural adjustment in the west
Full Citation
Permanent Link:
 Material Information
Title: A proposed regional model for agricultural adjustment in the west
Physical Description: 11 leaves ; 28 cm.
Language: English
Creator: Hildebrand, Peter E
Publication Date: 1963
Subject: Land use, Rural -- West (U.S.)   ( lcsh )
Farms -- Economic aspects -- West (U.S.)   ( lcsh )
Agricultural extension work -- Management -- West (U.S.)   ( lcsh )
Spatial Coverage: United States of America
Statement of Responsibility: by Peter E. Hildebrand.
General Note: Typescript.
General Note: Caption title.
General Note: "A paper presented at the WFEA annual meetings, July, 1963 in Laramie, Wyoming."
 Record Information
Bibliographic ID: UF00080623
Volume ID: VID00001
Source Institution: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 155218398

Full Text




Peter E. Hildebrand

In recent years, there has been an increasing interest in

studying the allocation of resources in a regional or interregional

framework. Most agricultural economists are familiar with the

methods commonly used to study resource allocation at the farm

level. There is less widespread knowledge, and further, less

agreement and acceptance of methods and models to consider resource

allocation problems for units larger than the firm. Many of the

objections raised to aggregate analysis involve controversy concern-

ing the objectives which are meaningful at the regional or broader

level of investigation. In this paper, after considering a relevant

side issue, several regional models are discussed. The appropriate

use of each and the necessary conditions which restrict the use of

each are covered. Finally, a model to study the allocation of

agricultural resources in the West, and thus, agricultural adjust-

ment in the West, is proposed./

The Concepts of Positive, Normative, and Predictive

It will be useful at this point to discuss the differences

between three concepts about which there appears to be confusion

in the profession. When one sees the statement, "This study is

I/A paper presented at the WFEA annual meetings, July, 1963 in
Laramie, Wyoming.

2/The model has been proposed for use by the W-54 regional research
committee. A subcommittee of W-54, composed of the author, Walter
Butcher, Jay Andersen and Harold Carter devised the proposed
model. The contribution of the subcommittee should be recognized;
however, the presentation in this paper is the responsibility of
the author.


normative and therefore not predictive.", it is obvious that some

confusion exists as to the meaning and the use of a normative

model. In this statement, reference is apparently made to the

distinction between normative and positive methods of analysis.

Positive analysis in this context usually refers to a regression

analysis based on time series data. Such a study is predictive

to the extent that changes in the magnitude in the variables

over the span of the time series continues into the future. The

prediction is based on the assumption that conditions which affected

a variable in the past will continue to have the same effect in

the future and, furthermore, these conditions will continue to

be the same in the future.

The original, and frequently still used, definition of normative

involves the use of subjective considerations. A normative study,

in these terms, considers problems about what ought to be in the

sense of personal issues. It was natural to move from this inter-

pretation to one which could more precisely be called conditional

normative. A model which is conditional normative would ponder

'what would be' if a specified objective were appropriate. The

specified objective, then, represents the conditions under which

the conditional normative model is appropriate.3/

Thus, a normative study, while it must be based on known or

estimated physical and economic relationships, differs from a

positive model in that a specific objective is considered in the

3-An interesting discussion on this point is in: James H. White,
et. al. Influence of selected restraints on normative supply
relationships for dryland crop farms on loam soils, southwestern
Oklahoma. Tech. Bul. T 101, Okla. Agri. Exp. Sta. May, 1963.


model. This objective is either maximized or minimized subject

to some set of restraining conditions. The accuracy and complete-

ness of these restraining conditions as well as the authenticity

of the objective determine the accuracy of prediction of the

normative model. The further the restrictions are from the actual

restraining conditions in the real world, and the less authentic

the objective, the less accurate the model is as a means of prediction.

In summary, to say that a model is normative and therefore not

predictive puts entirely too much restriction on a normative model.

Both positive and normative models can be predictive and for any

given set of circumstances either may be more accurately predictive

than the other.

It should further be noted that a model need not be exclusively

normative nor exclusively positive. As a matter of fact, the model

which will be proposed at a later point in this paper contains

some elements which are positive and some which are normative.

The extent to which the proposed model is predictive will depend

upon the accuracy of the restrictions as well as the appropriateness

of the objectives included in the normative model, in addition to

the accuracy with which the past trends used in the positive portion

of the model predict future conditions.

Some Possible Regional Models

In studying a single farm unit, the problem is to determine

the allocation of a set of resources variable between enterprises

within the firm but usually fixed to the firm, in producing products

which are determined within the solution. I doubt that a positive

method of analysis has ever been seriously proposed as a means of

predicting the changes in output and resource allocation of a farm

unit. The main reason, of course, is that most farms operate

under one or a set of objectives. A common objective is that of

profit maximization. Thus, studies of a farm unit are normative

in nature.

The purpose of regional studies is similar to the study of

a farm unit in that the end product is to learn something about the

allocation of resources within the region or between regions as

well as the determination of the product outputs. Since objectives

associated with regions are less clearly defined than for farms,

positive analysis has some value in studies of a regional nature.

Barring abrupt changes in the variables affecting the allocation

of resources and product output of a region, it could be expected

that future allocation and production could be predicted reasonably

well from a time-series study. On the other hand, if an economic

system tends toward an equilibrium, then it is possible to consider

the equilibrium conditions as the goal or objective of the system

and a normative study would be possible. To define the equilibrium

conditions as an objective suitable for an economic model, becomes

a major problem in normative regional analysis.

If one were to consider a whole economic system, taking into

account all productive resources and all possible products, it is

reasonable to consider the type of analysis presented by Lefeber
in his book Allocation in Space.- Briefly, the Lefeber model

- L. Lefeber. Allocation in Space. North-Holland Publishing Co.
Amsterdam. 1958.

considers the possibility of producing any commodity in any region

with transportation of mobile resources from region to region

possible, with consumption taking place in all of the regions

and with final products being shipped between any two points as

necessary. The objective is to maximize total national (or world)

product given the price of final commodities in each of the consumption

areas. Costs of production other than the cost associated with

transportation of factors and final products is determined as an

end product in the model. The costs are functions of the shadow-

prices in the direct solution or of the variables in the dual

solution. Given the price of commodities in each of the consuming

regions the Lefeber model is an equilibrium solution for factor

allocation since all factors are price according to their marginal

value product and the marginal value product of each resource is the

same in all alternative uses. However, since product prices are

given, this model is not an equilibrium from the standpoint of

equating supply and demand.

While it is meaningful when considering a total economy to

have as an objective function the maximization of gross national

product, this objective has little meaning when considering only

one industry within the economy. It has even less meaning when

considering only a few of the several commodities produced by the

total industry. When considering agricultural adjustment, only

one industry of the entire economy is under consideration. When,

as is being proposed here, only a few commodities in agriculture

are going to be considered directly, then only a portion of the

industry is being taken into consideration. Obviously, for purposes


of studying adjustment of the producers of a few agricultural

commodities the Lefeber model as discussed, is not appropriate.

Schrader and King,-/ in a paper presented at these meetings

last year, used a model similar to the Lefeber model in order to

study the location of beef feeding activity in the United States.

The objective in their model was to maximize return minus transport

cost (of factors and products). Equilibrium output and prices

were obtained through an iterative process which equated supply

and demand. Cost levels for the factors were obtained by the dual

solution. The iterative procedure to arrive at an equilibrium

solution has merit. Further, if one is interested in studying the

differential between present factor costs and equilibrium factor

costs, using the dual to obtain the level of the supply function

is valuable. However, using the objective of maximizing revenue

net only of transport costs, and not accounting for present factor

prices, represents a major shortcoming of the model. This procedure

considers that factors are available in fixed supply in a region

(or subject to transfer to another region) and have no opportunity

cost, production cost or price. This model would be more meaningful

if fat beef were the only national product and all resources were

used in its production.

Turning from models which maximize gross product or value of

gross product subject to an endowment of transferable resources,

consider the possibility of a model which has as an objective, the

-The model is also presented in: L. F. Schrader and G. A. King.
Regional location of beef cattle feeding. Journal of Farm
Economics, February, 1962.


maximization of profit to the agricultural sector of the economy,

or to certain commodity groups within agriculture. If factor prices

are given, monopoly theory indicates that output would be contracted

until elasticity of demand were greater than unity before a solution

would be obtained. On the surface, since demand for most agricultural

products is so highly inelastic, it would appear that the required

contraction in output would be greater than society would be willing

to accept both from the standpoint of reduction in the quantity

of food and fiber and from the standpoint of the reduction required

in the number of farmers. Obviously, sufficient demand information

is not available to estimate demand functions for any agricultural

commodity at quantities for which such functions would be elastic.

This restriction, in itself, is sufficient to rule this model non


If it were of interest to answer the question, "What would be

the lowest possible cost to society for obtaining some specified

level of farm production and how would resources be allocated to

obtain this production", an appropriate objective is apparent.

The objective of minimizing total cost of production of a specified

output subject to some level of factor prices and regional quantities

of factors would answer several meaningful questions. Henderson,

in a study of the coal industry, reports one of the most complete

investigations of this nature.-/ As explained in his book, a cost

minimizing study is most useful when determining the possible

- James M. Henderson. The efficiency of the coal industry. Harvard
University Press. Cambridge, Mass. 1958.


allocation of resources to produce past levels, or some other fixed

level, of production. In using this approach, factor prices are

of basic importance, but prices of products have no bearing and

equilibrium of supply and demand is not an appropriate consideration.

In summarizing the preceding discussion, keep in mind that it

is desired to develop a model which will be useful in studying

agricultural adjustment, presumably toward a condition of supply-

demand equilibrium. A positive model of supply has been ruled out

because, in agriculture, conditions of the future are not approximated

close enough by the past. In addition, a positive model has no

value when studying questions of what would be if specified objectives

were optimized. A model to maximize gross national product does

not apply to a single industry. A model which maximizes gross

return to a single commodity fails to consider alternative uses

of factors by ignoring either opportunity costs or prices. If

prices are considered and output is flexible, a profit maximizing

solution would result in drastic curtailing of output. A cost

minimization objective results in a model which gives direct

allocation solutions while avoiding the necessity to account for

supply-demand equilibrium conditions.

The Proposed Model

The attributes which are desirable in a regional model may

be specified as follows. It is desirable to have a model which

will result in a supply-demand equilibrium. The model must

specify resource allocation within the region under this equilibrium

condition. In specifying the allocation of resources, it would be

particularly helpful to know something about the excess capacity

which exists in the region. In particular, emphasis should be put

on labor and land resources. Initially, since prices of factors

of production are given, it would be desirable to consider them as

such and account for them in the model. Finally, the objectives

of the firms operating within the region should be considered,

particularly since regional objectives are difficult to specify and

are less operational.

In brief, a proposal which meets these requirements is this:

use parametrically programmed, individual firm supply response curves

as inputs for the regression procedure from which smooth, regional

supply functions can be derived. Using these supply functions,

and a set of demand functions, determine the equilibrium output for

the region. Once this is known the required adjustments of the

firms within the region are specified.

Actual operating procedure of the W-54 committee would require

that each cooperating state be assigned responsibility for determine

the response of farmers within corresponding states to various

price combinations for the commodities to be considered. In the

northern tier of states the commodities would be wheat, the so-

called feed grains and beef. In the southern tier, cotton would

replace wheat. Representative farming situations would be specified

by each state and within the areas specified, goals relevant to

individual firms would be incorporated into the program either as

objectives or restrictions. By keeping responsibility for program

construction within the state, objections to unrealistic interpreta-

tions of the aggregate solution can be minimized.


Solutions would be obtained for selected combinations of

prices for each of the three commodities. The proposal is to consider

three (or four) prices for wheat in combination with three (or four)

prices for feed grain. This would produce 9 (or 16) separate programs.

In the case of three prices of each commodity, a maximum of three

of the nine programs could be eliminated since for these the feed

grain price would exceed the price of wheat--a price relationship

which is highly unlikely. For each of the separate programs

remaining, solutions would be obtained with the price of beef--

in those cases where beef production is considered as an alternative--

being varied parametrically. This procedure will produce a supply

response of the type with which we are all familiar. Evenly

spaced points on these response surfaces will be interpreted as

observations and used to obtain continuous supply functions by

regression methods.

The regional aggregate quantity of each commodity supplied

would be a function of the price of all three commodities:

Q = fl (Pw Pf, b)

Q = 2 (, Pf Pb)

Qb = (Pw' Pf Pb)
These supply functions, along with the demand functions relating

quantity demanded to the price of each commodity:

Qd =
f L (pw)
Q d
Q = f5 (Pf)
Qbd = f6 (Pb)


and the three conditions stipulating equilibrium:

s = Qd
Q Qf

would yield a system which could be solved for the equilibrium

prices and regional output of each commodity. Given average regional

prices and area price differentials, output for each area in the

region can be determined. On the basis of the indicated adjust-

ment of representative farms, regional resource allocation would

be specified.

It was mentioned that the model should indicate the degree of

excess capacity in land and labor. A possibility for considering

the labor resource would be to superimpose a variable price program

involving opportunity cost of labor onto the other programs in

the model. It may be necessary to do the same for land and possibly

other capital resources. Including all possibilities increases the

number of programs geometrically. The difficulty is not in visualizing

the requirements but rather in the magnitude of the task.

It is obvious that many details need to be specified and agreed

upon by the W-54 committee if an effective procedure is to be followed.

Input-output data must necessarily be consistent between farms,

areas and states. Coordination will be a necessity if this proposed

model is to be effective. If it is effective, the benefits which

would accrue would be more than sufficient to offset the efforts


-< >

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs