ASAM: Economic sector contribution

Summary

ASAM is an Excel Application that facilitates the analysis of an economic sector contribution to a region. It does this by taking data from a source like IMPLAN (not included) and then calculating base economic output, employment, wages and value added. The code (written in Visual Basic) driving the application is Open Source, licensed GNU GENERAL PUBLIC LICENSE v3.0. The complete application can be downloaded from GitHub.

Understanding the cascading effect in regional-contribution-analysis
An economic sector has a cascading impact on regional economies

Introduction

Economic multipliers

Economists often talk about the “multiplying effect” of a business or sector, e.g: the economic ripple effect of a primary industry by generating incremental demand for supporting businesses as well as induced demand (through its employees) for local retail, hospitality, real estate and other services. It is easy to double count the economic sector contribution to a region, making the “exaggeration” a common and fair criticism of economic impact studies.

Using a SAM – Social Account Matrix

When taking Industry data (Social Accounts) and applying a series of “squared” calculations can isolate a specific industry sector without double count. This provides insight in how economic impact cascades through a region while keeping the total economic impact of a region whole. This is a well researched methodology [1]Watson, Philip, Stephen Cooke, David Kay, and Greg Alward. “A Method for Improving Economic Contribution Studies for Regional Analysis.” Journal of Regional Analysis and Policy, 2015. Available at: www.researchgate.net.

The ASAM excel application

The ASAM excel application is merely a tool to automate the many steps of economic contribution analysis. It significantly reduces the effort and thus the barrier in applying the methodology [2]for example, creating the “square” matrices requires the use of pivoting techniques in MS Access and MS Excel that usually exceeds the know-how of  its users. See also: Doug Olson, “IMPLAN: Excel Pivot to Matrix”, IMPLAN Group LLC, 2011. Available at: YouTube AVBp9Rbiek8:

  • the application provides automated ripple effect (Gross and Base) calculations for Output, Employment, Wages and Value Added
  • all calculations are traceable and transparently reflected in supporting worksheets
  • the application makes data available in user-friendly summary tables, pie charts for both Gross and Base values  and bar charts comparing Gross with Base values (direct & indirect) – see case Illustration below

Like with any analysis, the quality of results relies on reliable data – garbage in, garbage out. The application will run with any regional IxI sector data. Since IMPLAN is the more widely used data tool for calculating economic contribution in literature it specifically facilitates importing data from IMPLAN’s impdb (and earlier revision iap) files.


Illustration

The following case study illustrates how a contribution study using Economic Base analysis provides insight in a local economy using standard output from the ASAM application.
This analysis is of a region where its county seat also happens to be home to the state’s largest college. When asked, many of the county’s residents would answer that agriculture is by far the most important sector for the county’s economy. The analysis shows a significantly different picture. The first pie chart shows that, if we account for the cascading effect of a sector’s economic output (the Base output), the agriculture sector ranks only 5th, with the following three sectors make up 50+% of the regions economic output:

  • the government sector (most of it is, in this case, accounted for by the education sector, i.e. the university)
  • a sector that is usually only talked about, but never shown: households. This sector represents new money that households bring into the local region – investment income, unemployment fees, social security etc. Since households are always an indirect economic output, it does not show up in the traditional output analysis (chart 2, Gross Output).
  • the 3d sector is construction (again, much of it driven by the university).
Pie_Baseoutput3
CHART 1: Base Output reflects the “new” money that a sector brings into the regional economy. This chart reflects the top 3 contributing sectors, with the college (in NAICS 92) representing most of the 30% wedge. ASAM allows quick generation and (PNG) export of pie charts for Output, Employment, Wages and Value Added while specifying the number of wedges.

Chart 2 shows how we traditionally depict an economy by showing only the Gross Output. The College, again, shows up as the primary driver, but, in only looking at gross output, we would not realize how much of the region’s economy relies on the university.

Pie_Grossoutput3
CHART 2: the way we traditionally show (gross) economic output.

The bar charts that the application generates show the same story in a different manner. Charts 3 and 4 show Output and Wages respectively. Since the college wages are generous compared to wages paid in the other sectors it again shows the influence of the college on the local economy.

Bar_wages5
CHART 3: Gross and Base Output compared.
Bar_wages5
CHART 4: Gross and Base Wages compared.

Why ASAM

Economic contribution studies can be quite self-serving; as Watson et al phrase it: “Criticism of [economic contribution] studies focuses on the perverse incentive for publicity and advocacy purposes to double count the contribution of a given sector by making its direct, indirect, or induced effects appear responsible for a larger share of the economy than the observed data can support”[3]Watson et al, 2015.
An accepted solution is to track all economic contributions in a (social accounting) matrix with sector spending entered in rows such that the columns summarize sector receipts. The row total of a sector’s aggregate spending then automatically equals the aggregate receipts in the sector’s column. This type of “square “calculation makes double counting impossible. If, for example, we say that consumption from households is the biggest contributor in a regional economy, we can verify the statement by splitting the claimed household spending over the different sectors and compare it to the total receipts from households in those sectors.

Even though this is a fairly foolproof solution (contingent to reliable data), you may have picked up on the downside: it amounts to an enormous amount of work!!

  • firstly: we need the regional receipts and payments by sector and region
  • secondly: we need to determine the multiplier effect of a sector in order to give it its fair weight. A sector’s economic contribution translates into not only “direct” contribution, but also into “indirect” economic activity (services and supplies that feed the sector and increase or decrease with the sector’s success) and “induced” activity (employee and owner household spending). This type of “credit” is called the **base** economic contribution, giving credit for the cascading effect of a sector, as opposed to the **gross** economic contribution which only looks at face-value.

There are good sources to provide us with reliable source data. Sector data is available from the Bureau of Economic Analysis as well as from companies like MIG, inc.: a company that grew from a University of Minnesota project and provides regional data by sector, which is commonly used in economic impact studies.

Creating the base output by sector, accounting for the direct, indirect and induced output, is less simple. Companies like EMSI collect SAM data and solve the second problem by doing the calculations for us. They do so in a proprietary manner, however, whereas academic studies prefer a more transparent method. Researchers often use MS Excel’s pivot tables to do the calculations, but doing so is very time consuming. Aggregating the data slightly differently, or comparing historic data sets requires running the matrix calculations again and multiplies the amount of work. The ASAM project was born to cut that time from days into hours…

How did it come about…

I took a graduate class in economic modelling from University of Idaho’s Phil Watson, a young energetic professor well versed in economic contribution analysis. Is UI good at this stuff? Remember EMSI, the consulting firm that provides contribution analysis nation-wide? That company is located in Moscow, a tiny town in the Idaho’s panhandle. Moscow happens to be home to Idaho’s Land Grant university, and to prove its agricultural teaching worthy of investment, UI became an expert in regional economic contribution analysis. EMSI is simply a spin-off from that know-how, staffed with UI economists going private.

Phil Watson makes his students go through all of the calculations of base economic analysis by manipulating IMPLAN data pasted into excel. I was intrigued by the different insights a base economic analysis gave of an economy (see charts), and wanted to do historical and regional comparisons. The amount of work was staggering. So I automated the work. Then I shared the work with Phil, who shared it with Abelardo Rodriguez – a professor in UI’s extension organization. Abelardo pushed to make it into a stand-alone application that could be used by local extension educators, keen on showing the nuts and bolts of local economies to community and business leaders in a region (Rodriguez et al, 2011), and provided a download location for the ASAM application on UI’s website. Abelardo left UI in 2009, however, and the webpage is no longer maintained with that ASAM revision frozen in time.

I still get questions from individuals that use the original application or report bugs. So I created a GitHub repository for those that want to use the work, and/or build on it (‘fork’ it)…

Strengths and Limitations of Economic Base Analysis

Strengths: Before I did Watson’s course on Economic Base analysis I had managed factories & companies in Europe and the US, and was in the ideological camp of many of my business peers: local economic growth was simply the aggregate of the success of its local businesses. Understanding the cascading effect of economic sectors, both private and public, made me realize that this was a gross oversimplification. Economic Base analysis, for which ASAM is an enabling tool, helps to visualize that cascading effect. It made me realize that regions become sustainably prosperous only when collaboration and interaction builds from both private and public initiative.

Limitations: There are two that stand out. The first weakness is in the use of standard reported IMPLAN or BEA data – they do not always reflect reality. It is very worthwhile to sit down with local stakeholders and screen the data for obvious discrepancies.
The second limitation is the method’s narrow focus on export (“new” money brought in by companies or tourism). This focus tempts many economists and consultants to promote export over other means of economic growth. Sure, export dollars are important, especially in rural regions dependent on natural resources. Rather than concentrating on more of the same (Chambers of Commerce calling for ever-more visitation), however, a region should also focus on wealth creation (“growth from within”) and create a deeper local economy by substituting imports.

Download & Installation

GitHub-Mark-32px Download from GitHub

The application folder on GitHub (link) contains:

  • a small-footprint application file named ASAM_shell.xlsm
  • a module folder with all VBA code modules; this code will be loaded upon launching the application for the first time
  • a data-sets folder with two IMPLAN data-sets (a 2006 .iap file and a 2010 .impdb file). Since IMPLAN is a paid service, these are “play files” for instructional use.
  • licence, readme and revision_history files

Installation

You need to have Excel installed on your computer. Note that the Excel file needs to run macros (scripts) to do its work, making the file suspicious to virus checkers etc. That is why all the macros are available as open-source code on GitHub for anyone to browse, check and research.

NOTE: if you do not see this security warning, then your security settings disables macros without notification. You will need to change the setting in your Trust Center to “Disable all macros with notification”. See Microsoft’s instructions for recent Excel versions: https://support.office.com/ or Excel 2003 and older: www.mdmproofing.com/iym/macros.php.

Use

This linked article in the Journal of Extension provides a compact description of use [4]Rodriguez, Abelardo, Willem Braak, and Philip Watson. 2011. “Getting to Know the Economy in Your Community: Automated Social Accounting.” Journal of Extension, August 2011. Available at: https://archives.joe.org/joe/2011august/iw3.php. The application itself has a and step-by-step user interface and a FAQ section (also available as web-page[5]Frequently asked user questions).

About the code

It’s not pretty. Most of it is spaghetti code from the initial design (I often just used the VB recorder to automate some calculations) loosely organized in VBA modules. I added an export class to separate the Visual Basic BAS modules from the spreadsheet in order to make version control possible in GIT, but that’s about all the prettifying you’ll find. Any improvements are welcome!

 

References & further reading

References & further reading
1 Watson, Philip, Stephen Cooke, David Kay, and Greg Alward. “A Method for Improving Economic Contribution Studies for Regional Analysis.” Journal of Regional Analysis and Policy, 2015. Available at: www.researchgate.net
2 for example, creating the “square” matrices requires the use of pivoting techniques in MS Access and MS Excel that usually exceeds the know-how of  its users. See also: Doug Olson, “IMPLAN: Excel Pivot to Matrix”, IMPLAN Group LLC, 2011. Available at: YouTube AVBp9Rbiek8
3 Watson et al, 2015
4 Rodriguez, Abelardo, Willem Braak, and Philip Watson. 2011. “Getting to Know the Economy in Your Community: Automated Social Accounting.” Journal of Extension, August 2011. Available at: https://archives.joe.org/joe/2011august/iw3.php
5 Frequently asked user questions