Modern portfolio theory is useful for understanding emergent properties of ecological systems and developing effective recovery plans for at-risk species. In particular, there are two general forms of risk with respect to financial markets that are useful in a conservation context. Systematic risk reflects vulnerability to large-scale events that affect entire markets (e.g., government policy). Unsystematic risk is specific to particular securities (e.g., droughts affect commodities like corn with no effect on oil). Portfolio diversification is useful for reducing an investor’s exposure to various unsystematic risks, but only to the extent that the assets are independent from systematic risk. Thus, one should have an understanding of an asset’s sensitivity to systematic risk relative to its return before including it in a portfolio. I adopted a traditional market model to estimate how 21 populations of threatened Pacific salmon (assets) responded to changes in ocean-climate indices (market forces) known to have large-scale effects on them. Specifically, I used multivariate dynamic linear models to examine (1) temporally evolving changes in the magnitude of risk among populations, and (2) shifts in the relationship between risk and normalized returns over time.
Results/Conclusions
I found considerable diversity in population-specific estimates of systematic risk over time. Variability in the returns of some populations was almost always independent of the Pacific Decadal Oscillation (PDO) and the North Pacific Gyre Oscillation (NPGO), whereas other populations had wide swings in their sensitivity to the PDO or NPGO. In addition, the trajectories were highly correlated with one another within two of the pre-determined “major population groups”, but much less so in the other two suggesting greater population diversity. Phase-plane portraits showing one-year ahead forecasts of systematic risk versus forecasted returns revealed three general types of populations: 1) those with consistently low risk and above-average returns; 2) those with consistently low risk but below-average returns; and 3) those with both highly variable risks and returns. This approach offers a framework for prioritizing conservation efforts by distinguishing those populations that might benefit most from restoration actions, from those that may serve as so-called bellwether populations.