Observations, Implications and Limitations

As anticipated, the communities most severely affected by the Mifflin Plan were generally located in the West Coast of Vancouver Island and the North Coast.  Also, it was found that the Metropolitan Region experienced the lowest decrease in commercial salmon licences.  More specifically, it was observed that Klemtu (North Coast), Sayward (Central Island), Hartley Bay (North Coast), Ahousat (West Coast Island), Bella Bella (North Coast) and Kyuquot (West Coast Island) lost the most licences as a percentage of their community employment.  Additionally, it was determined that the communities with the largest gains in licences were Port Simpson and Area (North Coast), Alert Bay (North Island), Sointula and Area (North Island), Maple Ridge (Metropolitan), Abbottsford (Metropolitan) and Coquitlam (Metropolitan).
    Our first analysis was analyzing the correlation between licence changes and average dwelling value.  The rational for this analysis was based upon Ecotrust’s (2004) argument that fishers without adequate collateral to use in obtaining a loan will generally be less able to purchase expensive commercial salmon licences.  Given that one’s dwelling is generally the best collateral available, it would be expected that as a fisher’s dwelling value decreases, so would their ability to purchase a commercial salmon licence, or hold on to it for an extended period of time.  With mixed success, our analysis seems to reflect this trend (R-squared=11%); however, it is difficult to draw strong causal relations between the two variables.  Firstly, it is misleading to suggest that on-reserve aboriginals have a dwelling value of 0 (as is the case in Statscan’s census) and thus cannot use their dwelling for collateral.  Indeed, these aboriginal individuals qualify for government assistance that compensates for their low dwelling value.  Secondly, the average dwelling value of an area leads to the ecological fallacy where what is true of the whole CSD is mistaken for what is true of the individual fishers living in the area.  The case of Tofino is a prime example.  The proliferation of tourism and influx of wealthy ‘outsiders’ skews the average dwelling value higher than would be the case if just salmon fisher’s houses were taken into consideration.  As a side note, using median dwelling value produced nearly identical results.  Nevertheless, as illustrated in Figure 3, the West Coast of Vancouver Island has lost a higher than predicted percentage of licences based on the regression model.  Clearly other factors at work.
    Closely associated with the average dwelling value analysis is our second analysis that looks at the predictive quality of the % of Aboriginal population on licence change.  The rational for this analysis was that aboriginal people, on average, tend to have lower than average dwelling values and in many cases live in isolated communities with little access to jobs or alternative sources of revenue.  Once again, our regression model did not illustrate the robust relationship that we had predicted.  The R-squared value between the two variables was a lackluster 15%.  But, a quick glance of the map showed that two particular coastal communities were outlying the linear plot.  Sayward and Port Simpson defied our predictive model.  Sayward, with a low aboriginal population would not be expected to have a large decrease in salmon licences, yet it lost the second most with respect to its community employment.  Port Simpson on the other hand, with a high aboriginal population, actually gained licences over the period of 1995 to 1999.
    Removing these two cities from the next regression illustrated two things.  First, it showed that there is a strong relationship between the % of aboriginal population and licence change (R-squared=63%).  Second, it illustrates that something peculiar is going on in two of BC’s coastal communities- namely, Sayward and Port Simpson.
    Our next analysis helps to potentially explain Sayward’s large loss in licences.  Given Sayward’s reliance on the primary sector, and the transition that this industry has been going through recently, it is not surprising that the city has lost a disproportionately high number of licences.  When a community relies heavily on the primary industry, it treads a fine line from falling into a staples trap.  The violent ups and downs of the forestry and fishing industry, in particular, not only impact their own profession but also the whole community employment profile.  Sayward’s reliance on the primary sector, coupled with the recent downturn in the forestry industry, helps to explain the high loss in salmon licences as a percentage of its community employment.
    Our final analysis (Figure 9), helped address some of our lingering questions, while also raising more.  First off, it should be noted that our analysis is not a ‘perfect’ depiction of isolation in the real world.  Galiano Island, which showed up as an isolated community, is a good example of the shortcomings of our analysis.  It is true that Galiano is relatively isolated, but it is not true that the island suffers from the economic constraints that we associate with isolation.  With easy boat access to and from the island and a relatively affluent residency, Galiano Island is not a coastal community at risk from the 1996 commercial salmon restructuring.  However, aside from Galiano, the buffer analysis was quite successful.  In fact, it shows three of the highest licence losers illustrated in Figure 1, namely Klemtu, Hartley Bay and Sayward, further explaining these communities licence losses.  It should also be noted that community isolation does not only help predict licence change, it also creates unique problems for communities that have already lost licences.  In particular, commercial salmon fishing is the lifeblood of the aboriginal communities.  As Gislason (1996) notes, in some isolated communities commercial fishing vessels are additionally used for transportation for grocery and supply shopping.  The potential hazards of community isolation coupled with large licence losses cannot be overstated.
    However, despite our best efforts and multiple analyses, we were not able to explain the phenomenon that is occurring in Port Simpson.  All three variables tested in this report would indicate that Port Simpson should lose a disproportionately high number of licences as a percentage of its community employment.  Yet, Port Simpson has gained licences, and at a disproportionately high level as compared to the other coastal communities.  While this observation does not fit well into our report’s analysis it opens the door to further research on the restructuring of the Pacific salmon fleet and its impact on coastal communities.  This observation has unique implications for policy as well.  It illustrates that one size fits all solutions do not necessarily apply to real world problems.  Indeed, Port Simpson does not need the support services on the same level as a city such as Sayward, despite its socioeconomic characteristics and racial profile that might dictate otherwise.