1. Introduction
Farmed salmon production is an important segment of the world fish market today. Production takes place in relatively few countries, the main producer being Norway, with around half of all production [
1]. Other important producer countries are Canada, Chile, the Faroe Islands, and the United Kingdom [
2]. Climate and geographical conditions limit the potential areas for this kind of production. Salmon is a popular food choice among consumers worldwide and demand for it has increased. Due to supply limits, there has been a considerable rise in market price in the last decade; for instance, from 2012 to 2019, the price increased from EUR 3.50 to 6.00 per kilo [
3]. During this period, there was a substantial fall in the value of the Norwegian currency, while production costs rose by less than the rise in the prices. As a result, salmon farming has become very profitable. At the same time, considerable concentration occurred primarily through acquisitions. From 1996 to 2018, the ten largest salmon farming companies’ share of the industry’s overall sales volume increased from 18.9 to 67.3 percent [
4]. Many of the big firms are listed on the Oslo Stock Exchange (OSE). There has been a strong rise in the market values of salmon firms since 2012. The salmon stock price index on the OSE increased almost tenfold from 2012 to 2018 [
5].
There are a large number of published articles addressing the market, cost structure, and profitability of Norwegian salmon farming. The study of Asche et al. suggests that volume shares and small companies have a significant positive impact on performance [
6]. There are considerable risks associated with the production of farmed salmon; hence, there is great variation and uncertainty in the link between input and output in the manufacture of this product [
7,
8]. Critical factors for its operation are weather and temperature changes, fish deaths, illnesses, fish escapes, algae growth, and other harmful factors [
6,
9,
10,
11,
12]. These factors can cause sizeable economic losses [
2,
13,
14]. In addition, there are considerable variations in prices and exchange rates that create uncertainty [
15,
16,
17]. Furthermore, there is significant uncertainty related to feed costs, which account for about half of total costs [
2,
4]. Licences and strict public regulations lead to limited opportunities to increase salmon production [
18]. Despite its success, the production of farmed salmon is quite controversial in Norway due to environmental conditions, among other things [
1,
19].
With the characteristics of this sector, the question of whether one company can achieve higher profits than others over a longer period arises. Location, innovation, and good methods of managing risk can help some companies achieve good financial results during many consecutive years. Asche and Sikveland tested this by studying the development of earnings before interest rate and tax (EBIT) [
20]. They concluded that there was little correlation between the profits from year to year and that the hypothesis of a random walk could not be rejected. The salmon-farming sector is heavily reliant on factors outside the specific firms sphere of control, from diseases in the farms to exchange-rate fluctuations.
Salmon farming stands out from most other industries due to its combination of high volatility and profitability. In the literature of profit persistency, there has been little focus on such cases. In this paper, we contribute to the literature by applying the theory of Mueller (1986) and Gibrat (1931) to this unusual industry. By analysing this outlier industry, we can learn about how industries with similar characteristics work. In particular, the results in this article can be applied to other exporting industries with high risk in production [
21,
22]. The year-to-year correlation of within-firm profit rates is measured in deviations from sector mean, meaning that sector-wide yearly variations are controlled for through a simple transformation of the main variable. We apply this variable by using a dynamic panel data methodology to estimate the persistency of profit rates, and thus to analyse the importance of firm-specific factors in salmon farming. More knowledge about the profitability of the sector will be useful for companies engaged in aquaculture, for lenders, and not least for the authorities in the assessment of various forms of regulation and tax design.
Fish farming involves a risk of disease or escape of fish. Increased density of fish farms leads to an increased risk of disease that can have serious consequences for the industry. For this reason, the industry is strictly regulated. It is the central authorities that grant or sell licenses. Each license gives a company a permit for a maximum of tonnes of biomass (the maximum allowable weight of all salmon). In southern Norway, this is 780 tonnes, while in northern Norway, where the risk of disease is lower, each permit is set at 900 tonnes. The number of licenses granted by the authorities thus indicates the capacity limit for the industry.
The fish-farming industry in Norway has an annual production that runs counter to the capacity limit set by the authorities. Increasing production by getting the authorities to issue more licenses is a process that takes a long time. The production is therefore inelastic. Due to inelastic supply, higher demand (see
Figure 1, left) results in higher prices and increased profit among firms producing farmed salmon. Data from public statistics confirm this picture (see
Figure 2, left). From 2012 to 2020, there has been a solid increase in the price of salmon measured in NOK (Norwegian Krone). There is a strong link between price and profit rate. High salmon prices have led to solid profits in the sector. Despite high prices and very good profitability from 2014 to 2020, growth has been limited (
Figure 2, right). The demand has become less price-elastic over time [
23]. Hence, fish farming stands out in profitability compared to many other sectors.
It is of great interest to gain more knowledge about how profitability is distributed between companies and how it develops over time. See graphs showing the Fish Price Index (
Figure 1, right), Industry Profits (
Figure 2, left) and Industry Growth (
Figure 2, right). Both growth and profit are related to prices, growth being the yearly change in revenue. The years 2010–2012 were characterised by negative growth, low profit, and falling prices. Prices increased after this, with a large jump in 2016, followed by stabilisation. The same jump and stagnation are seen in both industry profits and growth.
In recent years, there has been discussion of imposing a special tax on the aquaculture industry with reference to its high profits. This debate has lacked nuance due to little knowledge. Our contribution is as follows: Although the aquaculture industry can point to many companies with large profits, the financial result is very volatile. In the further debate on taxation, this should be taken into account [
24].
3. Research Questions
Having explained the salmon-farming market structure, we now focus on three key areas: profit, size, and firm-specific factors. By analysing profit persistence in salmon farming, we can measure the ability of firms to keep profits above the sector mean or, equivalently, the difficulty of catching up to mean profit. This parameter is understood as the speed of adjustment to the mean after a deviation from the mean profit rate. As far as we know, this is the first empirical analysis of Mueller’s profit persistence within this sector. Previous studies have found that industries with high levels of profitability tend to have high levels of profit persistency, but this might not be the case here.
Research Question:
Research question 1 is described in Equation (
2) and asks: What is the degree of profit persistence for the Norwegian salmon farming industry?
To further analyse the trends of the salmon-farming industry, we estimate the growth dynamics of the sector’s firms. Whereas profit persistence is likely to be close to zero, the persistence of size is likely to be close to unity, which is what Gibrat’s Law predicts. If Gibrat’s Law holds, there is a random walk and the individual company’s relative growth is independent of the size of the company. In this paper, we apply the method used by Valenta et al. [
33], which uses the level form of the dependent variable, in contrast to Asche and Sikveland [
20], who use the difference. Importantly, the moving average component will be included. If it is not included, the estimated autoregressive parameter will be biased, capturing spillover effects.
Research Questions:
(2a) Does Gibrat’s Law hold for the Norwegian salmon-farming industry? (Equation (
5)).
(2b) Will failure or success be maintained the next period? (Equation (
5)).
The third section of our paper investigates firm-specific variables which are believed to influence the profit rate deviations from the sector mean. Asche et al. found a positive correlation between the firm’s share of total sales and profitability [
6]. There was also a positive link between working capital and performance, while operating leverage had a negative impact on profitability.
In Research Questions 1 and 3, we use the deviation of profit rates from the sector average as the dependent variable. When estimating firm-specific factors, we specify two models due to the large effect of salaries on profit rates. In previous literature, the cost of borrowing money tends to have negative impact on the profitability [
34,
35].
Hirsch and Valenta et al. reported a significant positive link between growth and profit rates [
26,
33]. Companies might take advantage of growth to achieve higher profits [
25].
Controlling for firm-specific factors can have an impact on the level of profit persistence. According to the research of Giotopoulos, the effect might be stronger or weaker [
29]. Based on known research, we postulate the following research questions:
We have chosen to present Research Question 3 in a general form. Some factors can have positive effects, others negative ones, and there may be factors that do not have any effect. According to Yadav et al., there is mixed empirical evidence about the relationship between growth and profitability and between size and profitability [
36]. If there are no economics of scale for salmon farmers, the link between profitability and size should be insignificant. Research has also found that growth is a determinant of profitability, and has a positive relationship [
37]. Furthermore, the level of indebtedness has a negative influence on profit rates [
38].
6. Conclusions, Further Research, and Limitation
This paper investigated the degree of profit persistence among Norwegian salmon farmers. Previous literature showed a substantial variation in the speed of adjustment to normal profit between sectors. Despite the abnormally high profit level for companies within the salmon-farming sector, deviations from the mean value seem to be eliminated quickly.
Salmon farming is risky and the relationship between inputs and outputs is uncertain. With the same input, there can be a substantial deviation in the harvest. High production one year can be followed by low production the next year with the same level of input. The correlation between success in one year and in the following year is significantly negative. This might explain why firms follow a random walk and why Gibrat’s Law is not rejected. Our data show different pathways among the firms depending on the initial profit rates.
The main finding, however, is that even though the industry is highly profitable, the profitability does not persist through years for each company. This is useful information in forming tax schemes. This is an argument for having a flexible taxation which depends on the profitability of firms each year.
For companies belonging to the group with low profit, salary rates and debt ratios are negatively correlated to profitability, while growth is positively linked to this dependent variable.
Further research could analyse why there are substantial differences in the pathways of firms depending on their initial profit rates. A limitation of this paper is the duration of the sample and the lack of price strategy information. There is considerable variation between firms in pricing strategy, with the choice between spot and term prices. This could be a significant factor for explaining stability in profit rates.