--- title: nature already runs a portfolio canonical_url: https://ensurance.app/guide/nature-already-runs-a-portfolio markdown_url: https://ensurance.app/guide/nature-already-runs-a-portfolio.md subtitle: several hundred salmon populations keep one fishery open. a network of projects does the same job for the money category: natural-capital --- # nature already runs a portfolio *several hundred salmon populations keep one fishery open. a network of projects does the same job for the money* A portfolio approach means holding many things, so that one bad year in one of them does not close the whole. Harry Markowitz put the idea on formal footing in the *Journal of Finance* in 1952: spread capital across holdings and the swings of the whole fall for the same expected return, as long as the holdings do not all move together. Every allocator knows that sentence. Fewer know that some of its clearest field evidence swims. In Bristol Bay, Alaska, sockeye salmon come back each summer up the rivers and into the lakes and gravel streams where they hatched — not as one run, but as several hundred. The fishery that families and towns in the region eat from and earn from — and the bears, gulls, and streamside forests fed by the same fish — stays open because those runs do not all fail in the same year. The headwater forest, the floodplain, the meadow, and the run of fish exist — and already trade water, shade, soil, and food with each other — whether or not anyone funds a project on them. Ensurance gives each one an account so the people who depend on it can pay into the network. The network is not the forest. ## what a portfolio approach to conservation is **A portfolio approach to conservation** is funding many projects on named places as one holding instead of betting on one site, so that a bad year at one site does not end the work at all of them. In a network of projects, each is paid by the people who depend on that place, and each can route a share of what it receives to the projects it depends on. The salmon were doing it first. Ecology borrowed finance's word for it. ## several hundred runs, one fishery Daniel Schindler and colleagues put numbers on it in *Nature* in 2010. Bristol Bay's sockeye are several hundred discrete populations, spread across rivers, lakes, and tributary streams, with fish that spend different lengths of time in fresh water and at sea before they come home. The finding: variability in annual returns is **2.2 times lower** than it would be if the system were a single homogeneous population. In fishery terms, today's diversity means complete closures **fewer than four times per century**. Without it, closures would come **every two to three years** — ten times more frequent. A closed year in Bristol Bay is not a drawdown. It is a season without the fish that people in the region eat and earn from, and a thinner year for the bears and gulls the run feeds. The paper finds the same effect in the watershed food web: the diversity of runs stabilizes and extends the season over which predators can reach the fish. No single run is reliable. Some streams boom while others crash, and fish return at different ages, so one bad ocean year does not land on a whole generation at once. What holds the fishery open is not that every run is strong. It is that the runs are different from each other. ## the weak runs are the point This is the part that matters for anyone running projects. The paper's management reading is not "harvest the strongest runs and let the rest go." It is close to the opposite: protect weak stocks from overharvest in mixed-stock fisheries, maintain intact habitat networks, and reduce the homogenizing effect of hatcheries on genetic diversity. Each of those keeps the underperformers in the system. A run that is weak this decade may carry the fishery in the next, when conditions shift toward its stream. Lose it and the whole gets thinner, more synchronized, and more likely to close. Hatcheries that replace many different fish with copies of one would do the same damage from the other side, which is why the paper names them. The portfolio holds because the weak stay in it. That is the entire argument for carrying new and underperforming projects. :::johnson **the fishery stays open because the struggling runs are kept, not written off.** A network of projects earns its resilience the same way — by keeping its weakest places linked, paid, and alive. [see how projects route to each other →](/guide/projects-that-fund-each-other) ::: ## what this means for a network of projects Much conservation and restoration work is funded like a single homogeneous population: one project, one site, one funder, one output, one grant cycle. When that one thing has a bad year — a fire, a flood, a funder who moves on — the work closes. Not because it was bad work. Because nothing else was holding it up. A network of projects is built the other way. Each project sits on a named place: a headwater forest, a floodplain, a meadow, a reach of river. Each is paid by the people who already depend on it — the utility below the forest, the county whose road sits under the slope, the town that draws from the river. And each can route a share of what it receives to the projects it depends on. A watershed council running four projects already holds a small portfolio. What it usually lacks is the routing between them. | in bristol bay | in a network of projects | |---|---| | Several hundred discrete populations | Many projects, each on a named place | | Runs that move only weakly together | Projects spread across link types and regions | | Weak stocks protected, not culled | New and underperforming projects carried | | Intact habitat networks | Paid links between projects that depend on each other | | Hatcheries that homogenize the stock | One funder, one template, one output for every site | ### a new project enters with the network behind it A new project is the riskiest holding in any portfolio: alone, it has to find a funder, prove itself, and survive its first bad year with no track record. In a network, it can start with routing already pointed at it: the projects that depend on it can set a share of what they receive to flow to it, so their payors reach it without writing a check to a stranger. ### an underperforming project is carried while it recovers A burned slope. A meadow that did not rewet. Standing alone, that is the end of the funding case. In a network, it is a weak stock: still linked to the places that depend on it, and still able to receive a share from the projects having a good year while it recovers. Strong projects carrying weak ones is not charity inside the network. It is the same move Bristol Bay's managers make when they protect a weak stock: keep the difference that keeps the whole open. The routing has a name. Each place or purpose holds an account in its own name, and **proceeds** are the routing that sends a set share of what one account receives on to the accounts it depends on. The operator of each account sets the split. Nothing rebalances itself, and no engine decides who gets carried. Read the live splits on [the proceeds view](/proceeds?from=guide). Volumes are small today. What is live is the structure, and the structure is what a project alone lacks. ## the limit: nature is not uncorrelated An honest portfolio note names its correlation. Nature is not a set of independent draws. One drought hits every node in a basin at once: the forest, the meadow, the floodplain, the run. A regional fire season does not check property lines. Bristol Bay's runs are not independent either; they share one ocean and one climate. The portfolio works because the runs are only loosely linked, not because they are unlinked. So a network of projects in one basin, all on water, is a concentrated position wearing a diversified label. Resilience comes from three things: 1. **Spread across link types.** Water, fire, corridor, and soil links fail for different reasons. A forest that sits in a water network and a fire network is held up by both. The purpose accounts that coordinate those links are live: [water cycle](/water-cycle.syndicate?from=guide), [wildfire resilience](/wildfire-resilience.syndicate?from=guide), [wildlife corridor](/wildlife-corridor.syndicate?from=guide), and [erosion control](/erosion-control.ensurance?from=guide). 2. **Spread across regions.** A drought in one basin is not a drought in every basin. Networks in different regions do not all share the same bad year. 3. **Keep the weak nodes alive.** The project having a bad year is the one most likely to be cut, and the one whose loss makes the rest of the network move more in lockstep. None of this repeals correlation. It manages it. For the finance side, read [the fine print of the free lunch](/guide/the-fine-print-of-the-free-lunch) and [what asset correlation actually measures](/guide/what-asset-correlation-actually-measures). ## your own small portfolio of the region Once the network is clear, the instrument is simple. A **certificate** is a share of one named place — one forest, one floodplain, one reach. Hold several certificates across places that are linked to each other and to you, and you hold your own small portfolio of the region: not nature in general, but the specific places your water, your road, or your lake depend on. Named places are on [specific ensurance](/specific?from=guide). The price on a certificate is a bridge, not the worth. The forest is not a return stream, and the fish are not a yield curve. The portfolio language is borrowed because it describes something real: many living things, different from one another, holding a place steady together. ## frequently asked questions ### what is a portfolio approach to conservation? A portfolio approach to conservation funds many linked projects on named places as one network, instead of relying on one site. Each project is paid by the people who depend on it and can route a share to the projects it depends on, so one bad year at one site does not end the work. ### what is the portfolio effect in ecology? The portfolio effect is the steadying of a whole system because its parts do not all rise and fall together. In Bristol Bay, Alaska, several hundred sockeye populations make annual returns 2.2 times less variable than a single homogeneous population would be (Schindler et al., *Nature*, 2010). ### can a network carry an underperforming project? Yes, if the network routes to it. The projects that depend on the struggling one can send it a share of what they receive while it recovers. Each account's operator sets that split. It is a choice, not an automatic rebalance — the same choice Bristol Bay's managers make when they protect a weak stock. ## read next The next post shows the region: how water, fire, animals, and soil cross the property line, and how one forest project in California grew into a partnership because the people downstream paid for it. Read [landscape-scale conservation needs a way to pay across the fence](/guide/landscape-scale-conservation-needs-a-way-to-pay-across-the-fence). ## sources [Schindler, D. E. et al., "Population diversity and the portfolio effect in an exploited species," *Nature* 465, 609–612 (2010)](https://ui.adsabs.harvard.edu/abs/2010Natur.465..609S/abstract) — Bristol Bay sockeye: several hundred discrete populations; variability in annual returns 2.2 times lower than a single homogeneous population; complete closures fewer than four times per century instead of every two to three years; the management reading on weak stocks, habitat networks, and hatcheries. Markowitz, H., "Portfolio Selection," *The Journal of Finance* 7(1), 77–91 (1952) — diversification lowers the variance of the whole for the same expected return when holdings do not all move together. ## the series 1. [what a networked project actually is](/guide/what-a-networked-project-actually-is) 2. [why conservation projects fail on their own](/guide/why-conservation-projects-fail-on-their-own) 3. [nature already runs a portfolio](/guide/nature-already-runs-a-portfolio) 4. [landscape-scale conservation needs a way to pay across the fence](/guide/landscape-scale-conservation-needs-a-way-to-pay-across-the-fence) 5. [how to fund the project you depend on](/guide/how-to-fund-the-project-you-depend-on)