Changes in inventories in the Norwegian National Accounts

Hence, the changes are a mix of actual changes in inventories and statistical errors, which makes the figure hard to interpret. In the NNA, changes in inventories have been positive every year since 1970. Changes in inventories and statistical errors as a proportion of gross domestic product (GDP) have been large and increasing since 2004.

This implies that the figures of the changes should be examined and, preferably, improved. This report aims to examine these changes in inventories and statistical errors, including considering whether we can use accounting data retrieved from corporations as a source to changes in inventories.

Ideally, we would like to estimate changes in inventories and statistical errors directly. This requires having one or several good and reliable sources to calculate these changes. Good sources for changes in inventories in Norway have been missing. However, the income statement for corporations (“Næringsoppgave” in Norwegian, referred to as “NO”) provides us with value data for changes in inventories. If we want to calculate the direct changes of inventories, using an available source in Norway, we must use these “NO” data. This report examines whether the quality of the “NO” figures have improved in a way that enable us to use it as a source for changes in inventories.

The report includes theory concerning how the inventory and changes in inventory should be valued in the NNA and in Norwegian accounting, and how similar countries calculate the figures. In our analysis, we start by presenting how the NNA figure has evolved in recent years. Further we compare these with the relevant figures retrieved from the “NO” data. The figures for changes in inventories retrieved from “NO” data can explain only a small fraction of the total changes in inventories and statistical errors in the NNA. Additionally, because changes in inventories may be due to both changes in the number of goods as well as price changes we split the changes in inventories in these two. This implies that an even smaller fraction of the NNA changes in inventories can be explained by “NO” data. Additionally, there are several other issues with the “NO” data that need to be considered.

Even though our analysis still leaves a lot of the NNA figures for changes in inventories and statistical errors unexplained, our analysis concludes that the “NO” figures can, at least, be an estimate for some of the changes in inventories, an estimate that has been missing until now.