Decline in multipliers and Local purchase %

Hi Benjamin, Thank you for your question. The change in the Shannon-Weaver Diversity Index does not directly affect estimated LPPs or multipliers. There are several reasons why LPPs and multipliers might change between data years, but it is always worth checking a few alternative explanations for significant differences in results between years: 1) Did you select the appropriate event years for activities in this study and in the last study? On a similar note, are the dollar year for view values consistent across both studies? For background, see http://implan.com/index.php?option=com_content&view=article&id=603:603&catid=248:KB28 and http://implan.com/index.php?option=com_content&view=article&id=529:529&catid=233:KB16 2) Are retail sales being entered as sales or margin values consistently in both of your studies? implan.com/index.php?option=com_content&view=article&id=594:594&catid=244:KB24 3) Is the same Trade Flows Method used in each model (e.g. "Trade Flow Model" or "Econometric", seen on the Model Overview page)? If you need more explanation about these three points, please ask and we will elaborate. If none of those is at issue here, then LPPs can change based on estimated shifts in trading patterns. If, for example, there are new competing sellers nearby, but outside of your study area, that could result in lower LPPs since purchases made by local entities would be less likely to be made within the study area. No major methodological changes have been made in how IMPLAN estimates trade flows from 2010 data to 2012 data, so the differences are largely the result of changing economic conditions. If you have reason to believe that certain LPPs should not have changed, you can manually adjust the model to reflect your own estimates for LPPs. Could you please tell us which region(s) you are using and the type of study so that we can look into the particular LPP shifts? Thanks!

Thanks for your response. It confirms what I thought was happening in the model. For both studies, we are using a nine county region is Southeast Georgia including: Bryan, Bulloch, Candler, Chatham, Effingham, Emanual, Evans, Jenkins and Screven Counties. Any other insights you could provide to us would be most helpful.

Hi Benjamin, If you look at Explore>Social Accounts>View By: Commodity Summary, you will see regional purchase coefficients (RPCs - the same as initial estimates for LPPs) as well as regional supply coefficients (RSCs), which estimate how much of local supply is purchased by local demand. For the sectors you identified as having a decrease in LPP, you will see that they tend to have very high RSCs. If an RSC is 100%, it means that 100% of the local output of a commodity is estimated to be bought by local demand (not counting exports to foreign countries). So, RPCs below 100% combined with RSCs near 100% implies that local demanders are purchasing as much as they can of these commodities from suppliers in the local area; however, there is not enough local supply to meet all of the local demand, so demanders must make purchases from outside of the study area. One possible explanation for this is that local demand for these commodities has increased, but that local supply has not correspondingly increased (and/or that local supply has decreased). Thanks!

The research team in my office still has questions about the RPCs adjustments. Using the Commodity Summary I have found that all 5 of the IMPLAN sectors shares 125 product lines. From this 125 we have found 7 products that declined by more than 10 points from the 2012 data to 2010 data. These product lines are: 3033 Water, Sewage treatment, and other utility services 3432 Products and services of State & Local Govt enterprises (except electric utilities) 3407 Fitness and recreational sports center services 3410 Other amusements and recreation 3362 Automotive equipment rental and leasing services 3319 Wholesale trade distribution services 3360 Real estate buying and selling, leasing, managing, and related services Why did IMPLAN decrease the RPC in these and other product lines?

Hi Ben, I apologize, could you please advise us as to which region you have built this model on. We can then determine the change. Thank you!

We are using a nine county region is Southeast Georgia including: Bryan, Bulloch, Candler, Chatham, Effingham, Emanual, Evans, Jenkins and Screven Counties. Any other insights you could provide to us would be most helpful.

Hi Ben, In addition to the previous post about the reason for the change: There are several reasons why LPPs and multipliers might change between data years, but it is always worth checking a few alternative explanations for significant differences in results between years: 1) Did you select the appropriate event years for activities in this study and in the last study? On a similar note, are the dollar year for view values consistent across both studies? For background, see http://implan.com/index.php?option=com_content&view=article&id=603:603&catid=248:KB28 and http://implan.com/index.php?option=com_content&view=article&id=529:529&catid=233:KB16 2) Are retail sales being entered as sales or margin values consistently in both of your studies? implan.com/index.php?option=com_content&view=article&id=594:594&catid=244:KB24 3) Is the same Trade Flows Method used in each model (e.g. "Trade Flow Model" or "Econometric", seen on the Model Overview page)? If you need more explanation about these three points, please ask and we will elaborate. If none of those is at issue here, then LPPs can change based on estimated shifts in trading patterns. If, for example, there are new competing sellers nearby, but outside of your study area, that could result in lower LPPs since purchases made by local entities would be less likely to be made within the study area. Other factors that might affect RPCs: 1. New Census of Agriculture will affect county-level output for agriculture sectors. 2. Updated inter-county impedance values, IMPLAN’s metric of the relative difficulty of trade between two counties, will affect RPCs somewhat (assuming they are comparing to R1 of the 2012 data). 3. In Release 1 of 2012 we used a $1 million trade minimum (i.e., demands of less than $1 million were met by local supply if possible). In Release 2 2012 and in 2013 we reduced the cutoff to $0.5 million, which will lower RPCs in cases affected by this. (This is a moot point if they are using 2012 Release 2 and/or if their local demand > $1 million) 4. New enterprise vs. admin definitions may also affect RPCs somewhat (i.e., local demand vs. local supply for those that got reassigned). See 2013 release notes for more If this does not address your concern, would you be able to specifically detail for us what the concern is about the change. Thank you!

Sorry, just missed your reply