21.9 References
but the possibility of structural change should
be seriously considered.
4. Poor-quality data (Le., lack of precision). Lack
of balance in precision of the many variables
involved in the model.
5. Lack of agreement on appropriate measurement scales and techniques for some variables
is likely to be important. In such cases, there
is a tendency to ignore these variables. Probably the best example of this is the treatment of
nonpriced goods and services.
6. Generally, there are difficulties in measuring
interrelationships falling between academic
disciplines.
7. There may not be enough adequate information about many linkages to develop accurate
forecasts, particularly if magnitudes of statistical and nonstatistical errors are considered. A
major limitation of all approaches is data
scarcity, particularly ecological and social
data. There is a definite tendency in this type
of work to assume that we can quantify impacts more scientifically than is really possible, given the state of the art.
8. Even if we can measure a variable, it is difficult to collect data on responses of species in
reaction to a treatment (or a pollutant or a catastrophe), except under experimental conditions. Because of this, there is a tendency to
retreat to a crude "standards approach," based
largely on laboratory tests or guesses, because
of a lack of knowledge. There is always a question as to the validity of the standard.
9. It is difficult to recognize trade-offs between
environmental media (e.g., between air pollution and water pollution), and comparable valuation is very difficult to achieve.
10. The institutional framework (e.g., ownership
patterns and goals of owners) is generally ignored in regional economic impact analyses.
This aspect is usually left to the implementation phase of environmental planning, but ideally such institutional variables would be recognized in the economic impact assessment.
This is possible, but requires primary data collection by survey.
11. It is difficult to model diffusion processes and
impacts on receptors, even when these can be
identified. For a discussion of these problems,
see Kneese and Bower (1979).
12. Since we are interested in comprehensive assessment of various types of impacts (e.g., ecological, social, and economic), models tend to
be very large and complex. Since it is generally
317
not possible to have a single integrated model,
usually a number of interrelated models must be
developed and linked in some way. However, it
should be noted that highly complex models do
not seem to work very well because of the propagation of error in using data of varying precision (for a classic discussion of this problem in
the context of planning, see Lee, 1973).
21.8 Conclusions
A number of conclusions appear evident from the
preceding exploration:
1. Input-output modeling is generally the best
modeling strategy in ecological assessments.
Also, because of high monetary and time costs,
usually it will be necessary to use one of the input-output modeling systems that includes secondary data on economic linkages.
2. Although input-output multipliers are important, they are not sufficient to quantify economic
impacts in the assessment. It is necessary to apply them to expected changes in the regional
economy, which must be based on ecological
impacts, among others.
3. Forecasts of production levels and distributions
of intermediate and final outputs are highly desirable.
4. Given that the input-output model does not include resource constraints, it is necessary to
carry out side calculations to be sure that forecasted production levels, income, and employment are feasible in terms of the environment
as a source of raw materials.
5. Results of the economic impact analysis may
provide guidelines as to which ecological variables need to be quantified more precisely.
21.9 References
Blair, J. P. 1991. Urban and regional economics. Homeward, IL: Irwin.
Brucker, S.; Hastings, S. E.; Latham, W. R., III. 1987.
Regional input-output analysis: a comparison of five
"ready-made" model systems. Rev. Reg. Studies 17(2):
1-16.
Brucker, S.; Hastings, S. E.; Latham, W. R., III. 1990.
The variation of estimated impacts from five regional
input-output models. Int. Reg. Sci. Rev. 13(1&2):119139.
Bureau of Economic Analysis. 1992. Regional multipliers: a user handbook for the regional input-output
modeling system (RIMS II), 2nd ed. Washington, DC:
but the possibility of structural change should
be seriously considered.
4. Poor-quality data (Le., lack of precision). Lack
of balance in precision of the many variables
involved in the model.
5. Lack of agreement on appropriate measurement scales and techniques for some variables
is likely to be important. In such cases, there
is a tendency to ignore these variables. Probably the best example of this is the treatment of
nonpriced goods and services.
6. Generally, there are difficulties in measuring
interrelationships falling between academic
disciplines.
7. There may not be enough adequate information about many linkages to develop accurate
forecasts, particularly if magnitudes of statistical and nonstatistical errors are considered. A
major limitation of all approaches is data
scarcity, particularly ecological and social
data. There is a definite tendency in this type
of work to assume that we can quantify impacts more scientifically than is really possible, given the state of the art.
8. Even if we can measure a variable, it is difficult to collect data on responses of species in
reaction to a treatment (or a pollutant or a catastrophe), except under experimental conditions. Because of this, there is a tendency to
retreat to a crude "standards approach," based
largely on laboratory tests or guesses, because
of a lack of knowledge. There is always a question as to the validity of the standard.
9. It is difficult to recognize trade-offs between
environmental media (e.g., between air pollution and water pollution), and comparable valuation is very difficult to achieve.
10. The institutional framework (e.g., ownership
patterns and goals of owners) is generally ignored in regional economic impact analyses.
This aspect is usually left to the implementation phase of environmental planning, but ideally such institutional variables would be recognized in the economic impact assessment.
This is possible, but requires primary data collection by survey.
11. It is difficult to model diffusion processes and
impacts on receptors, even when these can be
identified. For a discussion of these problems,
see Kneese and Bower (1979).
12. Since we are interested in comprehensive assessment of various types of impacts (e.g., ecological, social, and economic), models tend to
be very large and complex. Since it is generally
317
not possible to have a single integrated model,
usually a number of interrelated models must be
developed and linked in some way. However, it
should be noted that highly complex models do
not seem to work very well because of the propagation of error in using data of varying precision (for a classic discussion of this problem in
the context of planning, see Lee, 1973).
21.8 Conclusions
A number of conclusions appear evident from the
preceding exploration:
1. Input-output modeling is generally the best
modeling strategy in ecological assessments.
Also, because of high monetary and time costs,
usually it will be necessary to use one of the input-output modeling systems that includes secondary data on economic linkages.
2. Although input-output multipliers are important, they are not sufficient to quantify economic
impacts in the assessment. It is necessary to apply them to expected changes in the regional
economy, which must be based on ecological
impacts, among others.
3. Forecasts of production levels and distributions
of intermediate and final outputs are highly desirable.
4. Given that the input-output model does not include resource constraints, it is necessary to
carry out side calculations to be sure that forecasted production levels, income, and employment are feasible in terms of the environment
as a source of raw materials.
5. Results of the economic impact analysis may
provide guidelines as to which ecological variables need to be quantified more precisely.
21.9 References
Blair, J. P. 1991. Urban and regional economics. Homeward, IL: Irwin.
Brucker, S.; Hastings, S. E.; Latham, W. R., III. 1987.
Regional input-output analysis: a comparison of five
"ready-made" model systems. Rev. Reg. Studies 17(2):
1-16.
Brucker, S.; Hastings, S. E.; Latham, W. R., III. 1990.
The variation of estimated impacts from five regional
input-output models. Int. Reg. Sci. Rev. 13(1&2):119139.
Bureau of Economic Analysis. 1992. Regional multipliers: a user handbook for the regional input-output
modeling system (RIMS II), 2nd ed. Washington, DC: