By David Kay, IMPLAN
IMPLAN’s new I-RIMS products have been developed as an alternative to the BEA’s RIMS II products. This document is designed to describe the differences between the RIMS II multiplier methodology and IMPLAN’s multiplier methodologies.
Previous users of RIMS II multipliers will notices that the I-RIMS multipliers differ from the RIMS II multipliers, even for the same industry and study area. While RIMS II Type II multipliers and IMPLAN Type SAM multipliers both include induced effects, the RIMS II induced effects differ from IMPLAN’s induced effects in three main ways:
- RIMS II uses a single household type for induced personal consumption while IMPLAN uses 9 household types.
- RIMS II maintains the national ratio of Value-Added to Intermediate Outlay regardless of the study area. IMPLAN adjusts the national coefficients to reflect state- and county-level value-added data.
- RIMS II uses location quotients to regionalize the national technical coefficients; a method which underestimates inter-regional trade and overestimates regional multipliers when cross-hauling is present. IMPLAN uses trade flows based on a gravity model. For more on IMPLAN’s gravity model, see here.
Please also note the following definitional differences between RIMS II multipliers and IMPLAN multipliers:
Technically, a multiplier is unit-less because it is calculated as follows:
- Output Multiplier = Total Output / Direct Output
- GDP Multiplier = Total GDP / Direct GDP
- Employment Multiplier = Total Employment / Direct Employment…etc.
This is why the IMPLAN multiplier reports differentiate between effects, which are on a per-million-dollar basis (e.g., a direct employment effect of 7.5 indicates that 7.5 direct jobs are needed for $1 million worth of production in that industry), and multipliers, which are unit-less (e.g., an employment multiplier of 2.5 indicates that 1.5 additional indirect and induced jobs in a variety of industries are needed for every direct job in that industry).
RIMS II’s final demand ‘multipliers’ are all on a per-dollar of final demand basis, and thus are more like ‘response coefficients’ than multipliers. Because IMPLAN’s new I-RIMS products are designed as replacements for RIMS II, I-RIMS uses the same definitional (if not methodological) procedures as RIMS II. That is, the I-RIMS final demand multipliers are also on a per-dollar of final demand basis. Similarly, the I-RIMS direct-effect multipliers follow the same definition as the RIMS II direct-effect multipliers.
Each set includes six types of multipliers—four final-demand multipliers and two direct-effect multipliers:
The final-demand multipliers are all on a “per dollar of output” basis. To use these multipliers, you will need an estimate of the value of the output purchased by the final user (also known as the final demand change). Multiplying the final-demand change valued in “producer prices” by the final-demand multiplier for that sector will provide you with an estimate of total gross output, total earnings, total jobs, and total value added associated with that change in final demand, depending on the final-demand multiplier that you use (i.e., using the output final-demand multiplier will yield total output impact, while using the employment final-demand multiplier will yield total jobs, etc.).
The direct-effect earnings multiplier is an “earnings-per-earnings” multiplier, and the direct-effect employment multiplier is a “jobs-per-jobs” multiplier. Thus, to use the direct-effect earnings multiplier, you will need an estimate of the initial change in earnings. Similarly, to use the direct-effect employment multiplier, you will need an estimate of the initial change in employment. The initial change in earnings and/or employment should factor in only those employees the industry producing the final demand output who resides in the region. This is an important consideration when defining the study region. Also note that jobs are not full-time equivalents (FTE), but rather annual average full-time and part-time jobs.
Description of I-RIMS Multipliers:
- Final-demand output – Total industry output per $1 change in final demand
- Final-demand earnings – Total household earnings per $1 change in final demand
- Final-demand employment – Total number of jobs per $1 million change in final demand
- Final-demand value added – Total value added per $1 change in final demand
- Direct-effect earnings – Total household earnings per $1 initial change in household earnings
- Direct-effect employment – Total number of jobs per initial change in jobs
I-RIMS multipliers are available as either regional or industry multipliers. The regional product provides multipliers for all industries within a user-defined region. Regions can be defined at the county, MSA, or state level (or any contiguous combination therein) and can be provided for 62 aggregated industries or 406 detailed industries. The industry product provides multipliers for a single industry across all 50 states, plus D.C. The industry can be selected from a list of 62 aggregate industries or 406 detailed industries.
Description of I-RIMS Product Types:
- Regional Multipliers ($270/region) || Available Geography – Any contiguous combination of Counties, MSAs, States || Available Sector Scheme – 62 aggregated sectors, 406 detailed sectors.
- Industry Multipliers ($70/industry) || Available Geography – All 50 states + DC || Available Sector Scheme – Industry selection from one of the following: 62 aggregated sectors, 406 detailed sectors
The I-RIMS multiplier reports include both Type I and Type SAM multipliers. Type I multipliers include direct and indirect effects, while Type SAM multipliers include direct, indirect, and induced effects. Indirect effects stem from input purchases (also known as intermediate expenditures), while induced effects stem from the spending of labor income (also known as household expenditures). Non-local purchases (i.e., purchases outside of the chosen study area, as determined by IMPLAN’s trade flow model) are excluded from the indirect or induced effects. Savings and income earned by in-commuters are excluded from the induced effects.
Converting Purchaser Prices to Producer Prices
The multipliers are based on producer prices; thus, the final-demand changes to be applied to the multipliers must be also valued in producer prices. Unlike purchaser prices, producer prices exclude wholesale and retail margins and transportation costs. For example, in the 62-sector scheme, the household margins for sector 1 (Crop and animal production) are as follows:
Sector | Margin Sector | Margin Value
- 1 | 1 | 0.5277
- 1 | 27 | 0.1112
- 1 | 28 | 0.2804
- 1 | 29 | 0.0027
- 1 | 30 | 0.0215
- 1 | 31 | 0.0145
- 1 | 32 | 0.0420
What this means is that for every dollar’s worth of crop and animal products that a household purchases in a retail store, only 53 percent of that dollar (53 cents) goes to the producer itself (sector 1), while the retail store (sector 28) gets 28 percent of the dollar (28 cents), the wholesale sector (sector 27) gets 11 percent of the dollar (11 cents), and the transportation sectors (sectors 29-32) all together get the remaining 8 percent of the dollar (8 cents).
Any sector whose products are sold at retail or wholesale outlets will have margins. These margins allow the purchaser price (what the consumer pays at the retail or wholesale store) to be split out amongst the retailer, wholesaler, transporters, and producer. If we did not apply margins and simply multiplied the household’s expenditure amount by the retail sector’s multiplier, we would vastly overstate the impact to the retail sector and completely miss the impact to the wholesale, transportation, and producer sectors. Conversely, if we simply multiplied the household’s expenditure amount by the producing sector’s multiplier, we would vastly overstate the impact to the producer sector and completely miss the impact to the retail, wholesale, and transportation sectors.
Thus, when you model anything that is purchased through a retail or wholesale outlet, you must first apply margins to the purchaser value before using the multipliers. There are two ways to use the margins, depending on the amount of information you have about the purchase.
- If you know the exact item being purchased (e.g., crop and animal products), you would use the margins for the sector that produces the item (e.g., sector 1) to split out the purchaser price amongst all the margin sectors present in the region (e.g., sectors 1, 27, 28, 29, 30, 31, and 32) and then apply each of those sector’s values to their own multipliers. The resulting final-demand change for crop and animal products, transportation services, wholesale services, and retail services can be applied to the multipliers for these industries only if these goods and services are provided by firms in the study region.
- If, on the other hand, you do not know the exact item that was purchased, but only that a certain amount was spent at a retail store, for instance, then you would use the margin for the retail sector to calculate the amount of the purchaser price that is actually kept by the retail sector (31.09% in this case) and then apply this value (0.3109 * purchaser value) to the multiplier for sector 28.
The I-RIMS product comes with 5 types of margins for the various types of purchases that may be made through a retail or wholesale outlet: household purchases, industry purchases, federal government purchases, state and local government purchases, and investment purchases.