very odd multiplier

Comments

3 comments

  • Avatar
    DougO
    The "rule of thumb" of 2 for multipliers only apply to output. In this case for timber tracts in 2008 Oregon data, the output multiplier is 1.8. In any sector with high output per worker, the indirect and induced employment will be high relative to the direct employment giving you a, apparently, high employment multiplier. I know enough about foresty to know that a large proportion of timber tract employment is proprietors, ie, the landowners. Some timber tract owners have enough tracts to maintain wage and salary employment to administer and apply silvicultural practices on the lands in a rotating basis. The majority; however, hire forestry services contractors to plant and thin their forests as required every 10 years or so (if they do any management at all). If you look at the industry production function for forestry products (sector 15) you will see that for every dollar of output, 33 cents is used to buy forestry support services - ie, the contracted labor. Using sector 15 essentially assumes a significant number of your new 55 jobs are landowners, introducing huge new tracts of timber. I would consider using forestry support services (sector 19) or maybe a combination of the two depending on your study.
    0
    Comment actions Permalink
  • Avatar
    tbanda
    I'm getting odd multipliers for induced effects on the U.S. national model. I ran the U.S model using 2009 data and I have induced multipliers ranging from 3.6 to 7.2 for the 440 sectors. As a test, I ran the national model using 2008 data, and I again got very high induced multipliers in the range 2.7 to 6.8. Note. Last year, I ran the national model with 2008 data and the induced multipliers ranged from 0.3 to 2.7. Also, a colleague in a different office got similar "normal" results using 2009 data. What could be causing my model to generate such high induced multipliers this time?
    0
    Comment actions Permalink
  • Avatar
    jenny
    If you are talking about Employment multipliers, then your figures are quite normal. The U.S. will nearly always have higher multipliers (of all types) than smaller study areas because there is no leakage to domestic trade - commuting, domestic imports, residents spending their income in other regions - though there will still be some leakage to foreign trade. Industries that are more labor-intensive (lartely services, but also agriculture, construction) will tend to have higher induced multipliers. As for differences across years and between yours and your colleague's models, are the direct effects the same between the models? If so, I wonder if perhaps a comparison is being made between output multipliers and employment multipliers? An easy way to compare multipliers would be to use the Explore > Multipliers feature of IMPLAN.
    0
    Comment actions Permalink

Please sign in to leave a comment.