In a perfect world, economic impact studies would only be conducted to determine the likely effects of positive events in a community: expansion of a local manufacturer, increase in education attainment, or a reduction in the poverty rate. Unfortunately, sometimes negative impacts must also be measured, such as the economic impact of COVID-19.
While the extent of the pandemic continues to play out, many communities are scrambling to assess the economic damage. To ensure that the results of impact studies are accurate, and thus most useful in developing recovery efforts and allocating relief funds, it’s important to be thorough.
One of the first questions to ask is one of definition: what kind of analyses are we running (or want to run)?
Often any assessment of changes or possible changes to an economy are deemed an economic impact study. But there are different types of studies.
- Contributions Analysis – measures any and every dollar that touches a specific industry along with the associated multiplier effect.
- Gross Regional Product (GRP) Analysis – examines the amount of total value added in a region that stems from a particular industry or firm.
- Economic Impact Study – determines the net change to the economic base of a region that can be attributed to a firm or industry that otherwise would not be there.
Hannah Ruffridge, Director of Consulting for Emsi’s Higher Education team, explained the differences this way, “The Contributions Analysis includes the movement of any money because of a firm or event, regardless if that money stays within the regional economy or leaks out to other regions. The GRP Analysis is only looking at the movement of money that occurs within the regional economy. The Economic Impact Analysis takes it a step further and measures a net, instead of a gross impact, that takes into account that even if the event hadn’t occurred, some impact could have been created anyway.”
So to have a true economic impact study, those factors which already existed or were already in motion need to be known and controlled for. Emsi’s Industry Snapshot, part of our labor market research platform, is a great resource for gaining this baseline understanding. The report includes payrolled business locations, regional trends, occupations employed by the industry, and much more. We recently discussed this report and others that are particularly useful in finding answers to questions created by COVID-19.
Depending on the need, all three studies have their place and value, with GRP a subset of contributions, and impacts a subset of GRP. The advantage of an economic impact study is its focus on the net: only monies brought into the region by a firm or industry, or the local dollars retained in the region that otherwise would have been lost in the absence of the firm.
What about the input-output model? Can it be used to measure the economic impact of COVID-19?
One form of economic impact study is the input-output (I-O) model, which assesses how money flows within a region. This includes how money is exchanged between industries, how employees spend their salaries, tax revenue, long-term investment by companies, and many more factors.
Much like it sounds, the I-O model takes a change or shock (input) and estimates its impacts (output). These impacts include:
- Initial – This is the input being measured. A manufacturing plant opens and hires 100 workers, restaurants shut down for a week, etc. This input can be a change in jobs, earnings, or sales, and does not include ripple effects.
- Direct – The effect of new input purchases by the initially changed industries. This change is due to inter-industry effects and captures the supply chain impact. A new manufacturing plant is going to result in increased power consumption, purchases from suppliers, etc.
- Indirect – The subsequent ripple effect in further supply chains resulting from the direct change. Put another way, the sales change in the supply chains of the supply chain as a result of the direct change. This change is due to inter-industry effects.
- Induced – This change is due to the impact of the new earnings, investment, and government created by the initial, direct, and indirect changes. Induced effects enter the economy as employees spend their paychecks in the region, businesses invest to grow their operations, and government spends more to support the changes.
Direct, indirect, and induced impacts are known as multipliers. We sat down with Jonathan Crapuchettes, one of the creators of our I-O model, to discuss these multipliers and other components. (A summary of the video along with additional I-O model details are available in our Knowledge Base.)
The I-O model is a great place to start to get a sense of a shock, such as COVID-19, to an economy. An industry can be selected, and adjustments can be made to jobs, sales, and earnings to model the impact. The ability to run the scenario for different regions allows users to understand regional differences. Unable to perform much of their normal functions, many hospitals are forced to make layoffs, while in some regions the hospital industry may have additional needs and is expanding.
With many state and local governments requiring restaurants to close or only provide takeout, we used the I-O model to get an idea of the impact on the restaurant industry. Here is an example of the model in action: adding 100 jobs to the aircraft manufacturing industry in Tulsa, Oklahoma.
The I-O model sounds pretty in-depth. Does it have any limitations?
The I-O model is a very advanced tool (it took three years to create), and is thus a great way to model and assess changes in an economy. However, it’s important to remember that it is a model, and that it’s not without limitations. The primary one being that the model doesn’t have a time period associated with it.
The model essentially captures all the industry activity within an economy and assumes the economy is in a steady state, then estimates the end ripple effects of a shock. If you ever took an economics class, when your professor was drawing a diagram on the whiteboard, he probably at some point said “…everything else held constant…” To perform an economic analysis, it’s often useful to hold variables constant, and that’s what the I-O model does.
But within an I-O model there are different levels, known as types. Depending on the scenario being run, these types can be used to help account for the model’s lack of a time period.
Type I – Looks at industry-to-industry transactions and the initial, direct, and indirect impacts. That is, it closes the model to just the industry and doesn’t include induced effects. Because Type I looks at inter-industry flows (placing less emphasis on exports), it is best used at the national level where exports are less important than at the local and regional level.
Type II – Adds to Type I by introducing the effects of households. As businesses grow or contract, the resulting changes in income (positively or negatively) impact the economy. This is borne out in the induced effects.
Type Emsi – Takes Type II and factors in additional investment concepts that impact the economy. These include business purchases such as new equipment or buildings, and spending by government back into the economy from business tax revenue, like on schools and infrastructure. (This has been the default in Emsi’s input-output model; we are updating the tool to allow you to toggle between these three types.)
How should temporary layoffs be addressed when running economic impact studies?
Communities are in a constant state of change, this is especially true with layoffs. Many layoffs in the current crisis are hopefully temporary. Thus when running an I-O assessing the economic impact of COVID-19, we can’t simply take away a certain number of jobs from an industry, as it’s likely some of those jobs will eventually return. Additionally, the Coronavirus Aid, Relief, and Economic Security (CARES) Act will inject monies into local economies via households, counteracting some wage loss, as well as through small business loans, counteracting some business slowdown. It’s these externalities that limit the ability of the I-O model in the specific case of COVID-19.
In discussing these externalities, Ruffridge advised, “Negative (and positive) externalities should be considered when looking at layoffs. For example, how will this affect their household spending, thus affecting other industries. Will some areas increase, such as alcohol, whereas leisure and pleasure purchases will decrease? What about relocation? Will those who are permanently laid off be more likely to relocate than temporary layoffs? How will that affect the local economy in the long run?”
While the I-O model can’t account for all the externalities associated with COVID-19, it is still one of the best tools available and can serve communities, businesses, and higher education very well, especially if the right type is used. If conducting a national level estimation, a Type I model should be used, looking only at initial, direct, and indirect impacts. For shorter-term and state or regional models, Type II should be used. This is because people will be impacted, but government tax revenue and spending will not yet be altered in the short term.
Ok, so the I-O model can provide a good initial estimation. But to take a deeper dive and begin strategizing our recovery, how should we proceed?
The economic impacts of COVID-19 are unique, to say the least. One look at unemployment claims over the last few weeks reveals the uncharted territory we are in. To begin the conversation about industry impact, and where immediate attention should be placed, the I-O model is a great tool. And coupling I-O with business surveys as well as real-time and traditional labor market data can provide a holistic assessment of COVID’s impact, allowing for data-driven recovery efforts.
Regional nuances and the lifting of restrictions will result in varying impacts across communities. But as grant and other relief funding becomes available and communities begin pivoting their economic and workforce development strategies, economic impact studies will be a vital decision-making guide.
Ruffridge had this advice for those setting up a study of COVID’s impact: “I would suggest researchers come up with a defensible methodology and then use that methodology to measure multiple scenarios of the impact of COVID-19. The situation is changing daily, thus no one can pinpoint exactly how it will impact the economy. However, by providing different scenarios, it will provide readers with a more comprehensive and complete picture of what impact COVID-19 could have on the regional economy.”
There are different types of economic activity studies (contributions, GRP, and impact). The I-O model is an impact tool that can quickly estimate the results of an economic shock based on industry jobs, sales, and earnings. It can be used to get a very good picture of the economic impact of COVID-19 restrictions and closures on an industry, either at the national, state, or regional level. Because of the uniqueness of the COVID-19 pandemic and the many externalities, special attention should be given to the three types of models which can be run: Type I, Type II, and Type Emsi.
If you’re considering a COVID specific study, we’re happy to discuss your project and answer any questions you may have. Emsi’s team of economists has extensive experience in conducting economic impact studies.
Emsi provides the best labor market data to help communities, educators, and people make better decisions. Please let us know how we can help your organization or contact your account manager if you’re in need of guidance regarding a COVID-19 impact study.