commodity price change across all industries
Hi,
What is the best way to model a change in commodity price across all industries.
I'm using the distribution of commodity use (gross) from the commodity balance sheets to get the use by industry.
At that point it seems like I should be able to import those values into events in an industry change activity and specify the value either from the level or specifying it before hand on the excel sheet.
I may be missing something basic that I should include or do differently (please advise if so). But if not, the one thing that is concerning me is the comment on the import template about budget items. Because it is a commodity purchase it sort of is a budget item and therefore the template says something about zeroing out employment. If that is the case can you explain? If it isn't the case, how are the ratios calculated in the model to match employment compensation with the employment/output. This seems like it should be obvious but I'm getting off in the 3rd or 4th decimal when I try using StudyAreaIndustryData and controlling for deflation industry and gdp. ??
(for calculating other variables in industry change event) (Ok never mind on the calculations of emp comp and other VA items. I think I got what will work. The differences between what I'm getting and what I see in the db are not significant.)
If commodity matters on this analysis it is related to energy pricing.
Thanks,
Dale
-
Hi Dale, Thank you for your post! So that I understand your question, could give us a better idea about what the goal of the analysis is, we might be able to help more. As regards to a Commodity Change Activity Type, Employment is not something you can or should include, and this also holds true for spending patterns. This is because when commodities are calculated they are distributed across Industries and Institutions by market share as the first step. Then values attributed to producing Industries are assessed on the basis of their Output per Worker for determining the Employment in each Industry. Values attributed to Institutions are leaked. I'm not sure if this is the case in your template, but in mine, the tab for Commodity Change Activity Type is called Households <5000. As regards deflators the calculations used involve around seven significant digits, although we only display some of these, so there is virtually no way to get exactly the deflated amount, unless you first solve for the full deflator. Thanks! -
Back at this question. The specific commodity would be a energy source. Change in cost of electric or the cost of natural gas. This change would impact all industries and households. Using the IMPLAN model I can get the use of the commodity by all the industry sectors households by income levels and governments. The change is either an cost increase or reduction in production. There isn't a direct increase observed in output. For households and I think for government I can assume that this savings is just an increase in their income levels and can run an household income change or institution spending pattern change. Therefore, I'm wondering if for industry it is just an industry spending pattern change where each industry is impacted by the change in the change in total savings as distributed across all industries based on the gross input share of that energy commodity used in it's production function. I realize that by default the industry spending pattern is based on one industry's activity but in this case it seems like you could specify the spending vector as a share of 100% of all spending on electricity in each industry. -
Hi Dale, In all cases the key is your assumption on how the savings will be used by the recipient of the savings. For Households this assumption could differ from income group to income group as lower income households may actually spend their savings, while higher income households will probably put back the savings into capital, investment or savings. For governments there is no clear connection between spending and income, so you would have to use you best judgment on how, where and when the savings would be used. (For local governments this would clearly be within the Study Area as long as the Study Area represents the entire local government purvey, but for state and federal, where savings occurs and where money is spent and when is less directly connected). For Industries, savings most likely will go towards profits, but again if you feel that they would buy more or pay more you can certainly make and exercise these assumptions in the Model. In the end, if you do decided to model any of these savings, you would need to determine who was spending and what they were spending the savings on. Then you would impact these Sectors with the value of the savings. If the change is significant, you could consider modifying the spending pattern for Industries, but since you are looking at an across economy change, that could be daunting to consider. Hopefully this helps --Implan Support Staff -
Thank you for the response. I've been thinking a lot about this. Your response matches the direction I've been going as I've thought more about it. The industry is the one that I've been having trouble conceptualizing. The split among who captures these savings is tough. Proprietors yes. But then if they know about them or even if not and one makes some assumptions about the employees, and suppliers some would go to them as well. Thinking about all this across an economy (all inds) gets sort of complex. Given your response, and I assume much of the industry savings goes to profits among industries, how, or rather, what would be the best way to model that? Thanks again! -
Profit itself is not something you can model with IMPLAN. In effect it is like savings and until it is spent, it has no meaning to a production Model. So you would have to have a sense of where, when and how that profit was spent in order to model it. One of the key issues with corporate profits is they are often distributed to owners in vastly different and far flung locations from the company itself. It can be extremely complex to determine how this network plays out. For these reason, IMPLAN leaks profits from the Multiplier matrix. If you feel comfortable making assumptions or have data that could give you information that the company spends profits on certain Capital Expenditures or that a portion of dividend payees are in the Study Area you could certainly model these things. The first would be done either as wholesale or in the production Industry (likely with Margins applied) the latter through a Household Income Change. Hopefully this helps --Implan Support Staff
Please sign in to leave a comment.
Comments
6 comments