State-level VA data for agricultural sectors
Dear IMPLAN Moderators,
I'm hoping to get a better understanding of how accurate IMPLAN's VA:IE ratio is for agricultural sectors at the state level. I see that at the state-level the total industry output is obtained from ERS Cash receipts data and here (http://implan.com/index.php?option=com_content&view=article&id=766:speciality-sectors&catid=269&Itemid=112) it says that the data has been "adjusted for local Output-to-Value Added ratios and local RPCs".
I will be changing the national average production functions to account for irrigation practices and other purchases using Census of Ag data and will be customizing employee compensation and proprietor income to match ERS and BEA statistics, but I am trying to decide whether it is appropriate to customize my model (completely) and change the VA:IE ratio for state-level agricultural sectors.
I believe that I have the data necessary to update TOPI (from ERS - gov't payments, property taxes and other fees) but due to Census of Ag data non-disclosures I'm not certain that I can accurately distribute them amongst the agricultural sectors. Which would be more appropriate?
1. Only customizing EC and PI and keep the lock box checked so that the existing IE:VA ratio remains the same, or
2. Use the data I have to approximate TOPI for each ag sector and keep the lock box unchecked thus changing the VA:IE ratio.
Thanks!
-
Hi Ashley- On the topic of accuracy: ERS agriculture output data are released annually and are usually current with respect to the IMPLAN data year; that is, for a 2012 IMPLAN dataset, the ERS data were from 2012. Additionally, agriculture output is published at the state level. Value-added data for agriculture is published at the state level and also annually, but frequently with a 1-year lag with respect to the Implan dataset (i.e. value-added data from 2010 for 2011 IMPLAN data), requiring IMPLAN to project the lagged data to the year of the IMPLAN dataset. Additionally, state-level value-added data for agriculture are published at a high level of aggregation: “farms” and “forestry, fishing, and related activities” are the two constituent sectors of agriculture for which we obtain VA data. We do have compensation data (a component of value-added) for distinct agricultural sectors at finer levels of geographic and industry detail, which we use (along with other information) to estimate variables ratios of VA:IE across states, counties, and sub-sectors within the agricultural sector. Intermediate expenditures, by definition, can be calculated by subtracting value-added from output. So, the accuracy of the ratio of VA:IE relates directly to the accuracy of our output estimates – which, at the state level, require minimal estimation beyond what is directly reported by the USDA – and the accuracy of our value-added data – which, as noted, are reported directly at the state level, but require some estimation to allocate among agricultural sub-sectors and to project by 1 year. If you are using the second release of 2012 data, however, IMPLAN was able to use current-year state-level VA estimates and did not need to project them by 1 year. On the topic of adjusting your model: Before making a recommendation, it would help to know more specifics of which Census of Ag and which ERS data you would like to use for income and TOPI. It might be useful to compare the two different models, too. Also, I recommend consulting USDA ARMS data, which, for certain states, will provide a breakdown of expense types and income types by the production specialty (e.g. wheat, cattle, etc.) of farms. The sectors there do not align perfectly with IMPLAN sectors, though. See here http://www.ers.usda.gov/data-products/arms-farm-financial-and-crop-production-practices/tailored-reports-farm-structure-and-finance.aspx#P5ae0b3a8059e4fc1ad6c627c3df5948e_7_66iT0R0T0R0x2. Thanks!
Please sign in to leave a comment.
Comments
1 comment