ECAM Quick Start Guide

The Economic Consequences Assessment Model (ECAM) is a tool that facilitates indirect economic analysis using a computable general equilibrium (CGE) model. ECAM is used to provide an efficient screening-level analysis of the indirect economics impact of natural catastrophes such as floods, earthquakes or fires. ECAM is presented through a web-based tool with an interactive map. ECAM computes indirect economic impacts on the county level based on user input of capital and labor loss ratios.

Getting Started

To start an assessment, zoom to your area of interest and click on a county. This action creates an input box for labor and capital loss ratios.

Capital Loss Ratio

Capital losses include damages to non-residential buildings and functional structures such as roads, bridges, and industrial or computerized equipment. Provide this input as a ratio: capital lost to initial available capital. When computing the ratio, ensure that the numerator (modeled lost capital) reflects the same data as the denominator (total initial available capital). For example, if the total available capital does not reflect bridges as an asset, do not to use losses to bridges in the direct economic capital lost.

Computing Capital Loss Ratios

The inputs to ECAM are the capital and labor loss ratios reflecting the state of the components of economic production after the shock created by a natural catastrophe. As described above the capital loss ratio should reflect productive (non-residential) capital. The first step in computing the capital loss ratio is to compute the total capital ($//<![CDATA[ \begin{array}{l}TC_c\end{array} //]]>$)of all non-residential capital in a county ($//<![CDATA[ \begin{array}{l}c\end{array} //]]>$).

where:

$//<![CDATA[ \begin{array}{l}TC_c\end{array} //]]>$ is the total non-residential capital for a county ($//<![CDATA[ \begin{array}{l}c\end{array} //]]>$)

$//<![CDATA[ \begin{array}{l}TIC_c\end{array} //]]>$ is the total industrial capital for a county ($//<![CDATA[ \begin{array}{l}c\end{array} //]]>$)

$//<![CDATA[ \begin{array}{l}TCC_c\end{array} //]]>$ is the total commercial capital for a county ($//<![CDATA[ \begin{array}{l}c\end{array} //]]>$)

$//<![CDATA[ \begin{array}{l}TPC_c\end{array} //]]>$ is the total public capital for a county ($//<![CDATA[ \begin{array}{l}c\end{array} //]]>$)

Determining county capital

The total available capital can be determined using county parcel data or other data sources like the HAZUS database or the National Structure Inventory. Data for the entire county (not just the impacted area) must be supplied.

Calculate the total lost capital by summing the damage for each structure category. The damage estimates should be based on the same exposed values used in the $//<![CDATA[ \begin{array}{l}TC_c\end{array} //]]>$ expression above. If content values are used, then content damage should be included in the estimate of damages. If content values are not used, then content loss should not be included in the capital loss ratio. The total lost capital ($//<![CDATA[ \begin{array}{l}LC_c\end{array} //]]>$) for a county ($//<![CDATA[ \begin{array}{l}c\end{array} //]]>$) is expressed as follows:

where:

$//<![CDATA[ \begin{array}{l}LC_c\end{array} //]]>$ is the total non-residential lost capital for a county ($//<![CDATA[ \begin{array}{l}c\end{array} //]]>$)

$//<![CDATA[ \begin{array}{l}LIC_c\end{array} //]]>$ is total lost industrial capital for a county ($//<![CDATA[ \begin{array}{l}c\end{array} //]]>$)

$//<![CDATA[ \begin{array}{l}LCC_c\end{array} //]]>$ is total lost commercial capital for a county ($//<![CDATA[ \begin{array}{l}c\end{array} //]]>$)

$//<![CDATA[ \begin{array}{l}LPC_c\end{array} //]]>$ is total lost public capital for a county ($//<![CDATA[ \begin{array}{l}c\end{array} //]]>$)

Finally, dividing the ($//<![CDATA[ \begin{array}{l}LC_c\end{array} //]]>$) and ($//<![CDATA[ \begin{array}{l}TC_c\end{array} //]]>$) for a county ($//<![CDATA[ \begin{array}{l}c\end{array} //]]>$) computes the capital loss ratio ($//<![CDATA[ \begin{array}{l}CLR\end{array} //]]>$). This ratio, which cannot be greater than one nor less than zero, is calculated as shown below:

Labor Loss Ratio

Labor losses occur due to the displacement of employed individuals or the removal of such individuals from the workforce. Provide this input as a ratio as well, with similar coordination as described with capital loss ratios. The same assumptions used in the generation of the denominator of the labor loss ratio must be carried through in the computation of the numerator of the loss ratio.

Computing Labor Loss Ratios

Labor loss ratios should reflect the reduction in the workforce as a percentage of the workforce prior to the natural catastrophe. One way to compute labor losses would be to assume the labor force is uniformly distributed across the entire population. To carry out this computation, utilize the population for all structures within the structure inventory that are impacted by the natural catastrophe; the population in both residential and non-residential structures will be utilized to calculate labor loss. Using the entire population for labor loss ratios implies that the distribution of laborers and non-laborers within the study area is homogeneous, both geospatially and temporally.  This method assumes that for a given county, the percentage of workers in the population during the day is the same as at night (the workers mostly live in the county they work in) and the percentage of workers are evenly distributed in structures in and out of the area of impact.

The first step is to determine how long the event impacts the workforce. This impact duration ($//<![CDATA[ \begin{array}{l}ID\end{array} //]]>$) should reflect the duration of the event ($//<![CDATA[ \begin{array}{l}e\end{array} //]]>$) itself, the time to clean up post event, and the time required to reconstruct any damaged infrastructure (or housing) limiting workforce participation.

where:

$//<![CDATA[ \begin{array}{l}ID_e\end{array} //]]>$ is the duration of the impact to the workforce for an event ($//<![CDATA[ \begin{array}{l}e\end{array} //]]>$) in hours

$//<![CDATA[ \begin{array}{l}D_e\end{array} //]]>$ is the duration of the event ($//<![CDATA[ \begin{array}{l}e\end{array} //]]>$) in hours

$//<![CDATA[ \begin{array}{l}C_e\end{array} //]]>$ is the time to cleanup from the event ($//<![CDATA[ \begin{array}{l}e\end{array} //]]>$) in hours

$//<![CDATA[ \begin{array}{l}R_e\end{array} //]]>$ is the time required for reconstruction from the event ($//<![CDATA[ \begin{array}{l}e\end{array} //]]>$) in hours

$//<![CDATA[ \begin{array}{l}ID_e\end{array} //]]>$ may not exceed 8,766 hours (equivalent to one year, or 365.25 days).

Compute the workforce ($//<![CDATA[ \begin{array}{l}WF\end{array} //]]>$) for the county ($//<![CDATA[ \begin{array}{l}c\end{array} //]]>$) by determining the population ($//<![CDATA[ \begin{array}{l}P\end{array} //]]>$) in all structures ($//<![CDATA[ \begin{array}{l}N\end{array} //]]>$). 

where:

$//<![CDATA[ \begin{array}{l}WF_c\end{array} //]]>$ is the workforce in the county ($//<![CDATA[ \begin{array}{l}c\end{array} //]]>$)

$//<![CDATA[ \begin{array}{l}P_n\end{array} //]]>$ is the population in the nth structure in the county

$//<![CDATA[ \begin{array}{l}N\end{array} //]]>$ is the total count of structures in the county

Determining county labor

The total available labor (Workforce) can be determined using U.S. Census information directly, or other derivative products like the Longitudinal Household Employer Dynamic (LEHD) data . Data for the entire county (not just the impacted area) must be supplied. Data must be disaggregated to the individual structure to evaluate the reduction caused by the natural catastrophe.

Compute the population displaced ($//<![CDATA[ \begin{array}{l}PD\end{array} //]]>$) for the county ($//<![CDATA[ \begin{array}{l}c\end{array} //]]>$) by determining the population ($//<![CDATA[ \begin{array}{l}P\end{array} //]]>$) in all structures ($//<![CDATA[ \begin{array}{l}N\end{array} //]]>$) impacted by the event ($//<![CDATA[ \begin{array}{l}e\end{array} //]]>$). If it is possible for the user to compute life loss from the event, the life loss population would need to be removed from the population displaced. 

where:

$//<![CDATA[ \begin{array}{l}PD_c\end{array} //]]>$ is the population displaced in the county ($//<![CDATA[ \begin{array}{l}c\end{array} //]]>$)

$//<![CDATA[ \begin{array}{l}P_n\end{array} //]]>$ is the population in the nth structure in the county

$//<![CDATA[ \begin{array}{l}N_e\end{array} //]]>$ is the total count of structures impacted by the event ($//<![CDATA[ \begin{array}{l}e\end{array} //]]>$) in the county

Next compute the number of labor hours reduced ($//<![CDATA[ \begin{array}{l}LHR\end{array} //]]>$) for the working population across the county ($//<![CDATA[ \begin{array}{l}c\end{array} //]]>$) by computing the ratio of labor hours in a year, using that ratio to calculate the fraction of the duration of impact ($//<![CDATA[ \begin{array}{l}ID_e\end{array} //]]>$) that affects labor hours, and multiplying through by the displaced population:

where:

$//<![CDATA[ \begin{array}{l}LHR_c\end{array} //]]>$ is the labor hours reduced in the county due to the event ($//<![CDATA[ \begin{array}{l}e\end{array} //]]>$) in labor hours

$//<![CDATA[ \begin{array}{l}ID_e\end{array} //]]>$ is the duration of the impact to the workforce for an event ($//<![CDATA[ \begin{array}{l}e\end{array} //]]>$) in hours

2000 is the number of working hours per laborer per year

365.25 is the number of days per year

24 is the number of hours per day

$//<![CDATA[ \begin{array}{l}PD_c\end{array} //]]>$ is the total population displaced in the county ($//<![CDATA[ \begin{array}{l}c\end{array} //]]>$)

Finally, compute the labor loss ratio ($//<![CDATA[ \begin{array}{l}LLR_c\end{array} //]]>$) for the county ($//<![CDATA[ \begin{array}{l}c\end{array} //]]>$)  by dividing the labor hours reduced ($//<![CDATA[ \begin{array}{l}LHR_c\end{array} //]]>$) by the product of the workforce ($//<![CDATA[ \begin{array}{l}WF_c\end{array} //]]>$) for the county ($//<![CDATA[ \begin{array}{l}c\end{array} //]]>$) and the number of working hours for a year. This ratio, which cannot be greater than one nor less than zero, is shown below:

where:
$//<![CDATA[ \begin{array}{l}LLR_c\end{array} //]]>$ is the Labor Loss Ratio for the county ($//<![CDATA[ \begin{array}{l}c\end{array} //]]>$)

$//<![CDATA[ \begin{array}{l}WF_c\end{array} //]]>$ is the workforce for the county ($//<![CDATA[ \begin{array}{l}c\end{array} //]]>$)

$//<![CDATA[ \begin{array}{l}LHR_c\end{array} //]]>$ is the labor hours reduced for the county ($//<![CDATA[ \begin{array}{l}c\end{array} //]]>$)

2000 is the number of working hours per laborer per year

Computing ECAM

Once labor loss and capital loss ratios are entered, select the "Compute ECAM" button. This button will trigger an ECAM compute. When ECAM completes the compute, a list item will be added in the side-bar panel on the right-hand side of the page. The result will contain a high level summary of the output or an error message if something went wrong during the compute. Results are stored based on the County FIPS (Federal Information Processing Standard) ID and the labor and capital loss ratios input by the user. The results are not stored permanently, and are not accessible once the user leaves or refreshes the page. The user must explicitly save results to their computer using the "Save Results" button to keep a copy of the results.

Viewing Results

To view results, click on a list item in the "ECAM Results" side-bar panel. Hovering over the result item will highlight the item and the county that the item represents.

Clicking on the item shows the results in a modular panel. The results are organized by a "Labor" and "GDP" tab. Labor losses are expressed in full-time equivalent jobs and GDP results are expressed in millions of USD losses. Results are reported for the county by the sectors available in that county. At the bottom of each tab, the results are summed for a total loss.

If you wish to save the results, press the "Save Results" button to save the results as a text file to your own device.

You will be requested to select a location to store the result. The default name is based on the county name, adding the scenario. Including labor and capital loss ratios of a given scenario to the file name is helpful to separate multiple runs of the same county.

When you click save after selecting a file name and location, the file will be saved and you will be presented with an opportunity to open the file.

Your computer will open the file with the default text editor and the results are stored inside the file.

References

HEC-FIA 3.0 Technical Manual Chapter 5

Using HEC-FIA to compute Indirect Economic Losses (2015)