Modifying Trade Flows

Comments

3 comments

  • Avatar
    IMPLAN Support
    Hi Nicholas, We would appreciate it if you could give us a "big picture" idea of what you are trying to accomplish here. From reading your notes it sounds like you are anticipating that IMPLAN is providing information on forward linkages (i.e. the impact of increased logging on sawmills) rather than backward linkages (i.e. the impact of increased sawmill production on logging). The IMPLAN Model is only designed to examine backward linkages. Perhaps though we are misunderstanding, which is why we think getting an idea of overall what you are trying to accomplish may be helpful. There are a number of reasons and factors that keep the Model from being a true forward-linked Model, and if you are modifying backward linkages with the assumption of creating a forward linked Model, this unfortunately will not work. RPC's are only Commodity based. This is because the annual regional supply can come from multiple Industries and well as from Institutional production and inventories. Thus the RPC can only be accurately reflected on a Commodity basis. In the Trade Flow screen when you are looking at individual organizations and their RPC, you are seeing the rate at which those organizations purchase the selected Commodity locally. Which is presumed to be based on the amount of local supply to meet demand, thus all the coefficients are the same. You would only alter this if you knew that a specific Industry bought a higher than average proportion of that Commodity's local production. Again, modifying this screen does not sound like it will accomplish the purpose you are intending, if we are correctly understanding your concerns. Again, we would appreciate any additional information you can provide so that we can assist you in trying to accomplish your goal. Thanks
    0
    Comment actions Permalink
  • Avatar
    ndennis
    I'm not sure what information I provided led you to think I anticipated that IMPLAN provided forward linkages. That's not the case. I'm customizing a model of the Siskiyou forest sector to increase is accuracy. Most of the stock IMPLAN data on the TIO and commodity purchases by key forest sector industries are substantially different from the data I've obtained through a survey of local businesses. So I'm changing the TIO and production functions for timber tracts, logging, sawmills, veneer mills, and miscellaneous wood products. In addition, the stock IMPLAN data on the shares of key commodities that these industries purchase from within the region (i.e., county) do not comport well with my data, so I'm customizing the RPCs for these key commodities. To do this, I used the procedures described in my previous message. Sound OK?
    0
    Comment actions Permalink
  • Avatar
    IMPLAN Support
    Hi Nicholas, There is nothing wrong with the methodology that you have described in order to have Sector 96 purchase locally produced logging at a higher rate. However, if you are looking to have them purchase a higher rate of logs, rather than the same ratio of logs but more locally, this would be done through modifications to the coefficients in the spending pattern. That said here are a couple of thoughts/questions 1. Are you just interested in the Multipliers, or are you looking at analysis? Analytically, you will only see the change you've made if you impact Sector 96. This won't really affect the logging Sector. 2. If you are doing analysis with your modifications there are a couple of things to keep in mind. > The methodology you are describing changes the rates for every purchase (Direct, Indirect, Induced) of Sector 96. If you are not intending to change the entire Sector then the preferable way to make this modification is using Analysis-by-Parts. This is also true of any other Sectors that you modify production functions or RPC's for using the Customization methodologies. This may be your intent, but we just want to ensure that you are aware of that caveat of Customizing the Model. >When doing analysis with multiple Sectors along the supply chain, it is important to avoid double counting. You describe in your post that you have data for "timber tracts, logging, sawmills, veneer mills, and miscellaneous wood products" which describes a series of forward-linked Industries: timber goes to logging, logging to sawmills, then to veneer and miscellaneous. Each of these steps will make backward purchases from the Industry that occurs before it. So if you are doing analysis of all parts of this supply chain you will want to remove the purchases of the backward linked Commodities from your production functions either through the methodology you have earlier described or by means of Analysis-by-Parts. Thanks!
    0
    Comment actions Permalink

Please sign in to leave a comment.