INNOVATION AND PRODUCTIVITY GROWTH IN NORWEGIAN PRODUCTION OF JUVENILE SALMONIDS

Abstract

The literature on productivity growth in the grow-out phase for salmon is substantial, yet little attention has been given to productivity growth of the input factors. This is despite the fact that a number of innovations have improved quality and reduced prices for many input factors that contribute to the competitiveness of the industry. This article provides an analysis of productivity growth for one key input factor in salmon farming, juvenile salmon. We estimate translog cost functions on salmon hatcheries for the period 1988 to 2010. The econometric analysis shows that innovations and productivity growth have led to a reduction of unit costs in production of salmon juveniles, particularly at earlier stages, but the rate of technical progress has declined in recent years.

Keywords:

• juvenile production
• productivity
• salmon aquaculture
• technical change
• translog cost function

ACKNOWLEDGMENTS

The authors would like to thank two reviewers for helpful comments.

Notes

Data from the Norwegian Directorate of Fisheries (1988–2010).

TC = TC _pure +TC _factor +TC _scale .

Fry is sold to other hatcheries for further growth in fresh water and smolt is sold to grow-out farms for further growth in salt water. In 1988, 38% of the hatcheries sold both fry and smolt. In 2010, the number of hatcheries that sold both fry and smolt was reduced to 27%. Hence, smolt is the main product of interest for the hatcheries.

It is not possible to separate costs directly related to the number of sold fry. This means that the costs per unit of fish presented in Figure 1 are all costs related to produce smolt, including the fry.

Price variability is substantial also for smolt, but no tool to handle price risk has been developed as has been the case for the grow-out farms (Oglend, 2013; Oglend & Sikveland, 2008; Solibakke, 2012).

The farming of brood stock and production of offspring are done at separate locations. Large companies have their own units for breeding and brood stock, but smaller independent hatcheries still buy the roe from external suppliers.

Some of the new hatcheries have all their production facilities indoors.

Chu et al. (2010) provide a more general discussion of the impact of regulations for the development of aquaculture.

This is a pilot project from May 1, 2012.

Marine Harvest is Norway's and the world's largest salmon producer. The organization produce more than 20% of all Norwegian salmon as well as more than 20% of all Atlantic salmon globally (Asche, Guttormsen, & Nielsen, 2013).

Hence, several of the arguments provided by Asche et al. (2013) in the case of grow-out plants also seem to apply here. However, the plants are still well spread out, and as such, the risk diversification arguments of Oglend & Tveteras (2009) also apply here.

Asche (1997) shows how diseases can have cost consequences on the scale of the industry, and Asche, Roll and Tveterås (2009) and Hansen and Onozaka (2011) discuss the consequences of the recent disease crises in Chile.

This is a main reason why the supply of Atlantic salmon is more evenly distributed over the year than coho and salmon trout, as one is not able to influence the smoltification process for these species. This may also be an important factor in explaining why the share of Atlantic salmon production is increasing in salmonid aquaculture (Asche et al., 2013).

Optimal rotation and effective use of capacity is becoming more and more important in grow-out farms because the production systems are becoming more expensive.

Abundant water resources could be the reason why the recirculation technology was introduced later in Norway compared to other countries as Denmark and the United States.

Influencing flesh color using of astaxanthin that makes up over 10% of the feed cost in grow-out farms (Forsberg & Guttormsen, 2006) is not done in juvenile production. Torrissen et al. (2011) provide a more general overview of feed ingredients.

From 1986 to 1996 the cost share of feed increased from 27% to approximately 50%. In the period from 1996 to 2008 the use of feed has been relatively constant (Asche, Guttormsen, & Nielsen, 2013).

The hatcheries do not report the weight of the sold fish (only number of units sold) on the questionnaire. However, industry sources indicate that the average weight of a smolt has not changed very much. The increased growth has primarily lead to earlier release time and thereby increased production capacity.

A non-parametric method like the Total Factor Productivity (TFP) index could also have been used to investigate this issue. There are two reasons for choosing the parametric approach. First, The TFP approach is deterministic, and as salmon farmers are exposed to shocks from biophysical factors, allowing for stochastic noise seems appropriate. The error term in the model allowed for statistical noise as temperature, light and diseases. Second, the data set gives information of prices and costs, which means that we are able to use a specific functional form representing the production process. Follow this, the technical change gives an estimation of the productivity after adjusted for the prices.

Guttormsen (2008) discusses how growth is a function of temperature, and Hermannsen and Heen (2012) discuss how larger temperature changes influence salmon production.

Color versions of one or more of the figures in the article can be found online at www.tandfonline.com/uaqm.