**Hypotheses 1 (H1).** *Company growth is independent of their size (β = 1).*

This hypothesis is the original hypothesis of Gibrat's Law (P1), and it refers to the autoregressive process of the sector. If it holds, the firms follow a random walk. Due to the natural and government restrictions that the sector faces, we expect that the individual campsites have a 'natural' or steady-state size that they revert to in the long run. Therefore, if H1 fails, we expect that the autoregressive component to be below unity for the sector size (equivalent to being below zero for sector growth).

The second hypothesis is one of the two additions of the stricter version:

**Hypothesis 2 (H2).** *Success or fiasco one year has no effect on growth in the subsequent year (ρ = 0).*

If this hypothesis does not hold, shocks to revenue are followed by additional shocks in subsequent years. That is, there is inertia in shocks to revenue.

The last hypothesis connected directly to Gibrat's Law is:

**Hypothesis 3 (H3).** *There is no link between growth volatility and firm size (δ = 0).*

The third hypothesis is most often seen to fail, as small firms tend to have relatively higher volatility in size in comparison with larger firms.

Furthermore, we extend the model to investigate whether there is any link between the exchange rate, the level of debt, and the growth of campsites in Norway. This constitutes two models, the first one being represented in Equations (2) and (3). We may call this the restricted model, whereas the unrestricted model is as follows:

yit = αit + β yi,t−<sup>1</sup> + γ<sup>1</sup> x1t + γ<sup>2</sup> x2,it + εit, where εit = ρεi,t−<sup>1</sup> + ut, uit ~ N (0, σ 2 ) (4)

where x1,t and x2,it represent the log of the exchange rate (NOK/Euro) and the log of debt, respectively. The exchange rate is defined as a predetermined variable, which means that it is not allowed to be affected by the other variables. This assumes that the campsite sector does not affect the exchange rate, which seems realistic for the Norwegian economic structure.

Recent research has shown that a depreciation in the Norwegian currency has increased the inflow of foreign visitors to Norwegian campsites (Idsø and Opstad 2021; Opstad et al. 2021a). We postulate the following hypothesis:

**Hypothesis 4 (H4).** *There is no correlation between currency rate and growth (γ<sup>1</sup> = 0).*

If H4 does not hold, this inflow of foreign visitors translates into higher/lower growth for the campsites, indicating that campsite revenue is sensitive to macroeconomic conditions. If this is the case, we would expect the variable to be positively significant.

Lastly, we include the debt level as a firm-specific variable with the hypothesis:

**Hypothesis 5 (H5).** *Firms' debts are not associated with growth (γ<sup>2</sup> = 0).*

Hypothesis five (H5) assumes that growth is independent of the level of firm debt. However, limited access to capital can prevent businesses from growing, which means that firms need capital to grow; again, we expect that if the assumption does not hold, it will be positively significant, indicating that firms that have more access to capital (through debt) have higher rates of growth.

First, we test these hypotheses for the whole sample, and then investigate further splitting the sample into three. By splitting instead of including size dummies and/or interaction terms, we can apply the complete analysis to each size segment without unnecessary complications.
