Integrated Economic and Environmental Assessment of Regional Development Projects

                                                               Method

The methodological basis of the WIT is input-output (IO) analysis. Leontief proposed input-output analysis in 1936, for which he won the Nobel Prize in Economics. Since then it has evolved as a standard tool for estimating the economic impact of public policy choices at the community, regional, state, national or international level.  Most countries, including the U.S., periodically produce national IO tables for use in IO analysis.

IO models are derived from a set of linear equations representing sales transactions between hundreds of industry sectors (e.g., agriculture, utilities, mining, manufacturing, etc.).  The definition of industry sectors in the U.S. is based on the North American Industry Classification System (NAICS).  Transactions between sectors include purchases of inputs from other sectors and sales of output to other sectors and final consumers.

The core IO table is the technical coefficients matrix (denoted as a) where each cell (a_ij) represents the dollar value of input required from one industry sector i to produce one dollar’s worth of output in another sector j  ( i = 1...n , and  j = 1...n ). The total output of each industry is represented by the vector x.  The total output sold to final consumers is represented by the vector f.

Since the total output of an industry sector is the sum of final demand f and intermediate demand ax, the input-output system can be written: 

                                                        x - ax = f                                                   (1)

The vector of sectoral outputs required to meet a given exogenous demand f is obtained by pre-multiplying (1) by [ I - a ]^-1  

                                                      x =  [ I - a ]^-1  f.                                             (2)

Thus, it is possible to examine how changes in final demand affect industry output in different sectors.  (For a more detailed description of input output analysis, underlying assumptions about the structure of the economy and limitations, refer to Miller and Blair 1985; and U.S. Commerce 1994). For example, suppose that consumer demand for recreational fishing increased.  The increased final demand directly affects the industries that provide inputs to the tourism sector (e.g., energy, water, food), which in turn will have indirect effects on other industries.

The U.S. Department of Commerce Bureau of Economic Analysis produces IO tables for the U.S. economy using inter-sector transaction data from its periodic surveys.  These tables form the core of  IO models that examine economic impacts at the national level. To carry out economic analysis at the regional level, IO analysis techniques have been modified to reflect local production technologies and inter-regional trade (Miller and Blair 1985, Isard 1998). Normally, the national technical coefficient matrix is adjusted using a vector of regional purchase coefficients that take into account the fraction of commodity demand met from local production.

The IO model can be extended for environmental analysis by adding a matrix of environmental burden coefficients.  Suppose r is a k*n  matrix of environmental burden coefficients, where r_{k j} is environmental burden k (e.g., carbon monoxide emissions) per dollar output of sector j; and e is the vector of total environmental burdens, then the economy-wide total (direct and indirect) environmental burden associated with an exogenous demand vector f  becomes

                                                            e  = rx  =  r [ I - a ]^-1  f                                            (3)

The environmental burden matrix r can include coefficient vectors for any environmental impact of interest such as energy use, non-renewable resource use, greenhouse gas emissions, etc.  The contribution of individual industry sectors to the total environmental burden can be estimated by replacing each of the environmental burden coefficient vectors in r by its diagonal matrix.

Data sources

The regional input-output model for MRW is based on datasets obtained from MIG Inc. To enable environmental assessment, we augmented the IO model with sector level environmental burden matrices. Table 1 summarizes the environmental burden matrices included in the model and the data sources. These environmental data were compiled in collaboration with researchers at the Green Design Initiative, Carnegie Mellon University, Pittsburgh.

                                  Home            Method          Tutorial          Model           Register           Contact 

               © 2004, Dr. Satish Joshi, Department of Agricultural Economics, Michigan State University