Board of Regents of the University of Wisconsin System

"The Impact of Urban-Industrial Development on Agricultural Incomes and Productivity in

Finland": Comment
Author(s): Wilhelm Peters
Source: Land Economics, Vol. 53, No. 2 (May, 1977), pp. 253-255
Published by: University of Wisconsin Press
Stable URL: http://www.jstor.org/stable/3145930
Accessed: 11-08-2016 06:38 UTC
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted
digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about
JSTOR, please contact support@jstor.org.

Board of Regents of the University of Wisconsin System, University of Wisconsin

Press are collaborating with JSTOR to digitize, preserve and extend access to Land Economics

This content downloaded from 115.248.64.169 on Thu, 11 Aug 2016 06:38:22 UTC
All use subject to http://about.jstor.org/terms

"The Impact of Urban-Industrial Development

on Agricultural Incomes and Productivity
in Finland": Comment
Wilhelm Peters*

In an article appearing in this journal, initial insight into the real situation. CorJuhani Tauriainen and Frank W. Young relation and multiple regression analyses
[ 1976] examine the impact of urban-in- based on fifty rural communities (11
dustrial development on agricultural in- percent sample stratified according to
comes and productivity in Finland, with four patterns of population developthe additional aim of contributing "new ment) complete the empirical part of the

material in two areas: the effect of re-

paper.

gional centers and of national policy on Correlation coefficients are estimated

agricultural incomes and productivity" between the proportion of industrial
labor of the total labor force in 1970

[p. 192].

They begin with a discussion of the and various agricultural indices for 1960

reasons for differences in regional agri- and 1970. Multiple regression analysis is
cultural development. Two main groups used to explain agricultural income, the
output of milk, meat and crops per agriof factors are deduced to explain dependencies between interregional differ- cultural worker and tractors per 100
farms, by regional differences in the perences in agricultural incomes and
productivity and distance from urban-centage of industrial labor, distance to
industrial centers. One group of factorsHelsinki, number of retail trade establishments, and development of
is based on the impact of industrial development on the markets for agricul-regional services [p. 200].
tural products and factors of production;

The authors conclude that the "urban-

the other is based on the urban impact industrial impact" hypothesis can be
falsified neither for 1960 nor for 1970.
and its implications for a sufficient supply of agriculture with a "package ofBeyond this they see the need to "measure this impact at the regional level and
practices" [p. 193].
The theoretical discussion of the pos-with measures that are sensitive to the
urban aspects of central places as well as
sible dependencies is followed by an atto
tempt to empirically test the deduced industrial development" [p. 204]. An
hypotheses. A description of four large
regions (South, West, Central and North, *Institut fur Strukturforschung, Forschungsanstalt
arranged according to decreasing level of
fir Landwirtschaft Braunschweig-Volkenrode, West
Germany.
industrial development) provides an
Land Economics * 53 * 2 * May 1977

This content downloaded from 115.248.64.169 on Thu, 11 Aug 2016 06:38:22 UTC
All use subject to http://about.jstor.org/terms

Land Economics

254

196 f]. The percentage of labor engaged

intertemporal comparison of the results
indicates that, between 1960 and 1970,
in industry reflects rather incompletely
the impact of the independent variables
the income opportunities of agricultural
on agricultural income, milk and cropworkers outside agriculture; these opporproduction per head has decreased, and
tunities also depend on individual educahas increased for meat production per
tion, industrial wage level, number of
head. The authors attribute this imporvacant jobs, structure of the industries,
tant change to effects of settlement
etc. Nor is the "urban" term completely
policies, concentrated especially in
described by the variables. There is, e.g.,
a lack of variables to characterize the
northern areas [p. 199].
This empirical test of the "urbanregional supply with infrastructure, disindustrial impact" hypothesis contance to regional centers, etc.
3. In addition, the econometric
tributes to the general validity of the
hypothesis.1 Results from such studies model does not take into consideration
sectoral determinants of agricultural incan provide criteria for developing
rational policies with the aim of reducing comes and productivity. With regard to
soil quality and climate, the authors exinterregional disparities.
To reach their aim the authors apply plain their exclusion by the close correlawidely used statistical methods. Never- tion of these variables with distance
theless, additional comments seem neces- from Helsinki [p. 203]. But other variables such as quality of "labor" and farm
sary.
size structure are neglected. Without a
1. The sample communities, for
which the correlation and multiple re- doubt there is a close correlation begression analyses are being carried out,tween these variables and those emare randomly drawn from four patternsployed in the model. But it is not imposof population development. It may be sible that these excluded variables have
possible that, due to the sample stratifi-additional effects. And, if the model
does not include all those variables
cation and/or the limitation of the
which may influence the target variable
sample to certain parts of Finland, the
"real" correlations are not fully repre- the residual variances of an LSQ estimation are not random, and the estimated
sented. This would, for instance, hold
true if the variance of regions according parameters, therefore, are biased.
4. Accordingly, it may not be admisto population development were significantly different from the variance ac- sible to explain the residuals and their
cording to agricultural incomes and pro- intertemporal change simply in terms o
ductivity, or if distribution of the sample a political impact [pp. 199, 202]. The
between the four patterns were signifi- computed residuals are based on unspecicantly different from that of the total. fied nonpolitical (see above), political,
These problems are not mentioned in the and random variances. This means that
study.
2. The choice of independent vari-

the political determinants of agricultural

incomes and productivity are not iso-

ables also poses some problems. It must lated in the study. The parameters of the
be assumed that the variables used by

the authors are not suited to completely

describe the complex determinants of
'See, for example, an analysis
agricultural incomes and productivity [p. the Federal Republic of Germany

This content downloaded from 115.248.64.169 on Thu, 11 Aug 2016 06:38:22 UTC
All use subject to http://about.jstor.org/terms

by this author for

[Peters 1975].

Peters: Urban-Industrial Development in Finland: Comment

specified variables, therefore, are presumably biased.

5. The intertemporal change of some
special parameters can be explained not
only by "political" influences, but also

by means of the hypothesis of urban-

industrial impact. One of the reasons for

the observed shifts of milk production to
remote areas [Tables 2-4] is the cost of
labor. In remote areas labor costs are

255

pirical part of the paper, Tauriainen and

Young's study contains an important
analysis. Of particular importance is the
authors' conclusion, deduced from the
"urban-industrial impact" hypothesis, to
turn away from sector-oriented policy
toward an integrated regional policy,
concentrated in a network of regional
centers. Such a policy offers the opportunity for better coordination of hither-

relatively low compared with urbanto different partial policies for rural deindustrial areas. Consequently, the relavelopment.
tive labor-intensive milk production
shifts to remote areas, where labor costs
References
(mixed labor and especially opportunity
costs of self-employed farmers) are
lower. Considering this consequence of
Peters, Wilhelm. 1975. "Ausmass und Bestimmungsgriinde der interregionalen Einkomthe tested urban-industrial hypothesis,
mensverteilung
in der Landwirtschaft der
the intertemporal change in location of
Bundesrepublik Deutschland." Agrarwirtagricultural production can be explained
schaft 62.
without explicit political factors. AcTauriainen, Juhani, and Young, Frank W. 1976.
cordingly, the observed income adaption
"The Impact of Urban-Industrial Development on Agricultural Incomes and Proseems partly plausible without taking
ductivity in Finland." Land Economics 52
into consideration possible effects of
(May): 192-206.
political measures.
Despite some weak points in the em-

This content downloaded from 115.248.64.169 on Thu, 11 Aug 2016 06:38:22 UTC
All use subject to http://about.jstor.org/terms