An interesting debate arose after our report on the ripple effects of the Norwegian seafood industry (Nyrud, Iversen, Bendiksen, Robertsen, & Steinsbø, 2023). It was claimed that "The impact analysis for 2022 underestimates the financial footprint of the seafood industry". It was further claimed that "The seafood industry has a very low import share, which indicates that a large part of this consumption of goods and services will end up as Norwegian value creation". This led us to dive into the meaning of correct import shares among suppliers to an industry, for determining the full ripple effects from the industry. For a small, open economy, like the Norwegian, and an international, or global, industry like salmon aquaculture, understanding the nature and extent of trade – in both input and output – is vital for understanding the economic importance of an industry. The Norwegian seafood industry exports around 95 % of its output. At the same time, both fisheries and aquaculture import a range of input. For fishing, it is mainly vessels and fishing gear, while for aquaculture, even feed is imported. Around the seafood industry, a huge supplier industry has emerged (Blomgren et al, 2021). Wouldn´t it be reasonable to expect that this would result in most of the ripple effects to be manifest in Norway?
Methods and Results
While it is customary in Norway to use IO- analyses with IO-matrices from Statistics Norway (Brasch et al, 2020; Bruvoll et al., 2013), in line with well-proven methods for IO-analyses (Eurostat, 2008; Miller & Blair, 2009; Sanjiv et al., 2018), our study developed an IO-matrix tailored to actual trade patterns for the seafood industry. For such an IO-matrix, trade between all industries in Norway (as defined by three-digit NACE-codes) is mapped every second year (an example of the methodology is found in Brasch et al, 2020). As industries defined by three-digit NACE-codes may be diverse, our IO-matrix was based on real trade history from the seafood industry. Results show that while it is true that most of the input to the seafood industry is bought from Norwegian suppliers, the suppliers buy a large amount of their input from abroad, effectively reducing further ripple effects. So no, most of the ripple effects are not manifest in Norway. Or at least a much smaller amount than previously estimated (Johnsen et al., 2021, 2022). Previous methods used, with Standard IO, would have overestimated employment by 20.000 and net value creation by over 30 billion NOK. The biggest input factor in salmon production is feed, and this factor alone accounts for much of the estimation error from using standard matrices. While imports of aquaculture nets, fishing gear, fishing vessels et.c also contribute, this is dwarfed by the import leakage of feed. With a salmon production of 1.65 million tonnes a year, approximately 2.1 million tonnes of feed are consumed, at a total purchase price close to 40 billion NOK. For the feed companies, goods purchased accounts for approximately 85% of turnover, and 92% of the feed ingredients are imported (Aas, Ytrestøyl, & Åsgård, 2022). This translates into an import leakage of around 31 billion NOK, explaining most of the difference from earlier studies. This shows that, at least for seafood, IO-matrices not adapted to a “seafood trade pattern” will heavily overestimate ripple effects.