RIMS Employment Multipliers versus IMPLAN

For the multipliers, the main differences (as of 2009) are: • BEA uses location quotients to derive trade flows, whereas IMPLAN uses RPCs (based on econometric equations or IMPLAN's national tradeflows model, depending on your study area) • BEA uses a more textbook type II multiplier, whereas IMPLAN uses a true type SAM multiplier (allows leakage of employer-paid taxes and in-commuters). • IMPLAN provides a complete set of social accounts data, allowing the user to introduce new industries and to modify existing industries to better fit the data for the specific firm being studied. Click here for a comparison a chart: https://implan.com/v4/index.php?option=com_content&view=article&id=665%3A665&catid=253%3AKB33&Itemid=1

There are a number of additional possibilites (I assume you are referring to the RIMS II multipliers). 1) Are you comparing similar aggregations, as a rule of thumb, aggregation tends to increase multipliers. Also, the output per worker for a given sector can be quite different than the sector average which would have a huge effect on the employment multiplier. 2) I know that the Benchmark I/O internalizes most administrative government services which increases the multipliers. 3) Conversely, IMPLAN treats proprietor income as a driver of induced effects while the benchmark has all operating surplus (proprietor income plus other property income) as a leakage which increases Implan multipliers relative to the BEA. 4) Be sure to compare similar years. Employment multipliers have been consistently decreasing over time as productivity increases. Also, the US economy is becoming more open (more imports and exports) over time which decreases all multipliers, not just employment. RIMS multipliers are not broken out to direct, indirect, and induced which be a tremendous help in determining the cause of the differences.