Request Info

Who’s Afraid of Input-Output?

December 19, 2008 by Emsi Burning Glass

To effectively understand and respond to economic changes (like job loss) or to justify, promote and market proposed regional developments (like new industries), education, workforce, and economic development professionals must be able to provide a best estimation of the impacts of their proposed activities.

Whether it’s a college developing a new  program, a workforce board transitioning workers to new higher paying occupations, or an economic development council trying to attract a new company – they all need to justify their decisions based on the potential regional impact.

This can be a tall order, especially when staff, time, and money are limited. Many organizations will simply give up at the very thought of trying to come up with those figures, especially at the local level.

Now, colleges, workforce boards and economic development councils need fear not.

With advances in computer technology and data availability access to applied regional input-output modeling that can help estimate the impact of regional changes is readily available. Such analysis could only be dreamed of just 10 years ago.  So why are so many reluctant to make use of this powerful, well recognized tool?

Undoubtedly, much of this has to do with the textbook image of input-output as a complex accounting system requiring advanced mathematics and massive data to apply.  Applied input-output modeling programs as well have a reputation of being difficult to understand and operate, and this must scare off some.  And finally, the difficulty of applied programs has been aggravated by a lack of user support on the part of the input-output software vendors.

We start with a brief review of input-output fundamentals.  We turn next to a short list of the issues addressed with input-output: “Why you need input-output.”  We conclude with a practical look at how to take advantage of recent advances and build your own input-output model and analysis.


Input-output models are constructed on a platform of data indicating interconnectedness of the industries, households and government entities that occupy a given geographic space.  Its name stems from the fact that a portion of the output (i.e., sales) of one industry will appear as the input (i.e., purchases) of other industries.

With origins in the 19th century and before, input-output modeling has roots as old as any in contemporary economics.  It was not until the mid-20th century, however, that input-output appeared in a form specifically tailored to many of the most pressing real world planning issues, and Russian born American economist Wassily Leontief was awarded the Nobel Prize in Economics (1973) for developing that uniquely applicable form.

In the early days, until the early 1960s, only largest of nations could afford the main-frame computing power needed to house and operate an adequately detailed input-output model: the first U.S. input-output model was constructed to address the industrial planning issues raised by World War II.  By the late 1970s and 1980s, applied input-output was routinely available at the state-level.  Today even a small rural community can afford a computationally powerful input-output model.


According to the classic definition, the process of economic development planning has two parts: 1) formulate goals and 2) assess the effectiveness of alternative policy actions in achieving those goals.  No wonder at input-output’s popularity: it directly addresses both parts.

Economic development goals are formed in terms of the basic descriptive categories conveyed in the region’s input-output model.  Common goals for economic development might include, for example, increasing earnings per worker, increasing economic diversity, substituting local for imported goods, expanding various industry clusters, attracting quality of life migrants, and so on.  The current state, or baseline, of each of these measures is contained within the input-output model.  Development planners can thus use the model in a diagnostic exercise, to indicate areas where the economy might benefit from improvements, and conversely, areas were additional attention is likely to be of less importance.

With development goals in place, input-output is used to identify the best approach for their attainment.  Suppose for example there are competing uses for some parcel of industrial land, and suppose also that among community development goals is utilization of some underemployed community resource, perhaps a group of idled occupations.  The input-output model is used to simulate in turn each of the alternative uses and among its many indicators; the impact of each use on the target occupations is examined.  Alternative uses of existing resources, the impact of departing industries, the impacts of new industries, these are but a few of the policy issues addressed by a regional input-output model.


An applied regional input-output model is available from a number of sources, the leading ones being Economic Modeling Specialists, Inc. (the “EI Model,” see: economicmodeling.com), The Minnesota IMPLAN Group (the “IMPLAN Model,” see: implan.com), Regional Economic Modeling Inc. (the “REMI Model,” see: remi.com), and the U.S. Department of Commerce, Bureau of Economic Analysis (the “RIMSII,” see: bea.gov).  The features of these modeling alternatives vary considerably and while it is certainly a worthwhile endeavor, cataloguing these differences is beyond the scope of the present paper.  We rely instead chiefly on access and use of the EMSI EI Model.

The general steps involved in using a regional input-output model for policy analysis are:

1.  First, regional input-output models convey a tremendous amount of baseline information right out of the box. So in the first step of the process, the model’s base data are useful for informing the policy discussion.  As a result, policy makers might want to focus on the current occupational make-up of the region’s workforce, the relative magnitude of residents’ outside versus inside incomes, the importance of trade center industries, the leading industry clusters, and so on.

2.  The next step, which often poses a stumbling block for many, is to translate the policy issue into associated direct effects.  The policy issue might be the impacts of constructing a new building, and their may be interest in the economic impacts during the project’s construction phase.  EMSI’s way of defeating this hurdle is two-pronged.

a.  To begin with, EMSI makes every effort to maintain an up-to-date set of users’ manuals and website FAQs covering much of needed direction.

b.  Second, EMSI maintains a team of knowledgeable economic analysts (the most extensive customer service department in the industry) ready to provide over the phone hands-on assistance and direction with modeling issues.

3.  In the third step, the direct effects are entered into the regional input-output model.  This is fundamentally a mechanical procedure the ease of which hinges of the relative user-friendliness of the modeling software.  The EMSI EI model is web-based and specifically tailored to ease the burdens of the user.

4.  Finally, with direct effects entered, the model quickly produces the total impact of the modeled policy alternative.  For the most part this is a straightforward exercise.  The model will show the change, positive or negative, in regional jobs and earnings by industry (with detail down to the 6-digit NAICS code level).

For more questions about input/output modeling or to see a demonstration of EMSI’s Economic Impact module please contact us.

Want to get notified?