Deliverable 2.3

From PrimeFish Wiki
Jump to: navigation, search

Report on the economic performance of selected European and Vietnamese farmed species

Executive Summary

Global production of the main farmed species consumed in the EU has increased drastically in recent years. Production of Atlantic salmon is estimated to have grown by 157% during 2000-2016 and exports of pangasius from Vietnam increased from 700 tonnes in 2000 to 660 thousand tonnes a decade later, with a quarter of those exports finding its way to EU markets. Production of sea bass and bream increased by 259% between 2003 and 2016. But not only has the volume increased, prices of salmon and sea bass and bream have become higher, up approx. 100% and 10%, respectively, while pangasius prices have fallen.

Fish farmers within the EU face competition from many directions. They must compete with wild capture fisheries within and outside the EU, aquaculture firms from outside Europe, as well as other food products.

The aim of this deliverable is to use firm level data to analyse and compare the economic performance of aquaculture firms within and outside the EU. For this purpose, it was decided to base the analysis on two key fish farming activities within the EU - Scottish salmon firms and Mediterranean sea bass and sea bream firms – and two important international competitors – Norwegian salmon firms and Vietnamese pangasius firms.

The economic performance of firms is here gauged in terms of changes in efficiency and productivity. Using Data Envelopment Analysis (DEA), an efficiency frontier, which is made up of the most efficient firms, is constructed for each of the four case studies. The position of each firm relative to the frontier is then used to calculate efficiency scores, which are then decomposed into pure technical efficiency and scale efficiency. Technical efficiency indicates how well firms update existing production technology and how they improve their production management, whereas scale efficiency is an indication of how well firms have managed to take advantage of the existing economies of scale. DEA also makes it possible to estimate shifts in the efficiency frontier which are taken to represent changes in technology. An outward shift will then signify technical progress and an inward shift technical regress. Productivity growth is then analysed in terms of these two factors, changes in technical efficiency and technology.

The data at hand differs slightly between case studies, both in regard to the input variables available and time dimension. The output variable is the same for all cases, output revenue. The Norwegian study uses costs of employment and materials, current and fixed assets and shareholders’ funds as inputs, while the Scottish data includes observations on current and fixed assets, current liabilities and the number of employees. The Vietnamese data includes current and fixed assets as well as current and non-current liabilities, and the data on the EU sea bass and bream industry has information on these same four variables as well as the number of employees. For Norway, the data covers the period 2006-2015, for Scotland 2008-2015, and for the EU sea bass and bream and Vietnamese pangasius firms the study covers the years 2009-2014. Despite these differences, there is both sufficient overlap in time period and in information available, to compare the four different sectors.

The salmon firms in Norway and Scotland were on average more efficient than the other aquaculture firms, in regard to both technical efficiency and the ability to take advantage of the scale efficiency at hand. Technical efficiency under the assumption of variable-returns-to-scale averaged 0.962 for Scottish salmon firms and 0.947 for their Norwegian counterparts, but was only 0.794 for Vietnamese pangasius firms and 0.72 for sea bass and bream firm in the EU. Firms on the efficiency frontier are assigned a score of 1.0. The results thus show that Scottish salmon firms could on average reduce their input utilization by 3.8% (1-0.962) without reducing their level of output, and Norway could produce the same amount of salmon while using 5.3% less inputs. By contrast, Vietnamese firms could reduce their input utilization by 20.6% and Mediterranean firms by 28%.

Salmon firms in Norway and Scotland enjoyed scale efficiencies of 0.949 and 0.933, while the estimated scale efficiency of Vietnamese firms was 0.855 and only 0.605 for EU sea bass and bream firms.

However, comparison of productivity performance yields a completely different picture. Here, Vietnamese pangasius firms show a remarkable performance, with average productivity of 16% per year, with the EU sea bass and bream firms also showing strong productivity growth of 9% per year. Both Norway and the UK experienced a productivity decline during this period. The productivity growth of the Vietnamese firms can both be attributed to improvements in technical efficiency and improved technology, while better efficiency explains most of the growth of the EU firms. The UK salmon firms have also become more technically efficient, but technical regress has a negative impact on their productivity growth. Norwegian firms have seen their technical efficiency decline slightly and have also experienced a slight technical regress.

Using data at firm level has advantages for understanding the competitiveness of EU aquaculture, as it provides valuable insight into the industry structure; that enables us to understand better the overall trends in productivity and efficiency of the entire sector as well as also for individual firms, and to compare the performance between sectors as regards of utilisation of specific inputs at firm level. The results of this deliverable therefore are useful for discussion with industries regarding areas for improvement, and of course for the development of the simulation model and DSS tool within the project, i.e. in WP5 and WP6, respectively. However, the analysis provided in this deliverable is based on limited data, and the number of firms, period of data, and input variables used in analysis for four case sectors are not identical. In addition, the results are based on the application of a single method, DEA, and may not be robust to the use of different methodological approach. The interpretation and implications of the results should acknowledge those limitations.

Download Full Report