Haute Lumière
Commerce · VII.05 · MMXXVI · daylight
TEN CONCEPT BRIEFS · Chapter VII.05 — Bioregional Economics
One page each. A reader who reads only these ten pages has the chapter.
BRIEF 1 — The Bioregion, and the Level
The idea. A bioregion is not a mystical boundary and it is not a marketing radius. It is a published polygon drawn where several measurable things change together — geology, landform, soils, potential natural vegetation, climate, hydrology, land use and wildlife. Where those patterns shift in concert there is an edge; where they do not, there is not.
The part everyone misses: it nests. The Commission for Environmental Cooperation publishes 15 regions at Level I over North America, 50 at Level II and 182 at Level III. Globally, Dinerstein and colleagues map 846 terrestrial ecoregions.
North America, 24.71 million km2
Level I 15 regions 1,647,267 km2 each on average
Level III 182 regions 135,764 km2 each on average
ratio of grain 12.1 xWorked example. A regional food strategy written at Level I covers an area twelve times larger than one written at Level III. Since production scales with area and consumption scales with people, the same place is food-secure at one level and food-dependent at the other. Neither answer is wrong. Only the unstated level is.
Why it matters. The boundary is a resolution setting, not a discovery. Stating which level you used is what makes your number comparable to anybody else's — and the years that bioregional programmes have spent arguing about the true boundary were spent on a question the classification had already answered by declining to have one answer.
You already know this because you have seen two teams produce opposite answers about the same market and discover, an hour in, that one was counting the metropolitan area and the other the travel-to-work region.
BRIEF 2 — The Foodshed
The idea. A foodshed is the area that supplies a population's food, drawn the way a watershed is drawn: by following the flows back to where they start. As an analytical object it is a capacity map, not a behaviour map — it asks what a region could supply if distribution were optimised, not what it does supply.
Worked example. Peters and colleagues built the foodshed of New York State and found that, with all agricultural land in use and distribution optimised to minimise distance, the state could meet 34 percent of its total food needs from within the state, at an average distance of 49 kilometres. Underneath that average: the smallest population centres could meet virtually all their needs from about 25 kilometres, and mid-sized urbanised areas about 84 percent from about 51 kilometres.
The figure to carry. The state-level deficit is not general. It is metropolitan, and in New York it is one city.
Why it matters. Foodshed capacity and foodshed behaviour are different measurements and they are constantly confused. A region that could supply a third of its food and does supply a twentieth has a logistics problem. A region that could supply a twentieth has a latitude problem. The two require entirely different instruments and the aggregate percentage tells you nothing about which one you have.
You already know this because you have watched a capacity study and a utilisation study get quoted against each other in the same meeting.
BRIEF 3 — The Density Ceiling
The idea. Self-sufficiency has a hard arithmetic ceiling and it is a population density.
Peters and colleagues' biophysical model puts the annual land requirement of the diet Americans actually eat at 1.08 hectares per person — 0.18 ha of cultivated cropland, 0.16 ha of perennial cropland, 0.74 ha of grazing land. One square kilometre is 100 hectares. So:
self-sufficiency requires density <= 100 x f / 1.08 persons per km2
f = the farmable fraction of the region's land
f = 1.00 every hectare farmed 92.6 persons per km2
f = 0.38 the global agricultural share 35.2 persons per km2Worked example. A region of 20,000 km² with 25 percent of its land in farms can feed 20,000 × 100 × 0.25 / 1.08 = 463,000 people at that diet. If 900,000 live there, it is at 51 percent and no amount of enthusiasm moves it. What moves it is the diet: at 0.25 ha per person — a healthy omnivorous scenario with modest meat — the same region feeds two million.
Why it matters. The lever with the most travel in it is not the boundary and not the logistics. It is the composition of the diet, because 1.08 falls to 0.13 across the published scenarios, a factor of eight.
You already know this because you have noticed that the countries that talk least about food security are the ones with the fewest people per hectare.
BRIEF 4 — The Commodity Balance, Never the Aggregate
The idea. A single self-sufficiency percentage averages a large surplus against a structural zero and returns a number that describes neither.
Worked example. New York City, 8,804,190 people, claims 23.5 million acres of agricultural land at the baseline diet. New York State's entire land in farms is 6,502,286 acres — 28 percent. But the commodity table says something the 28 percent cannot:
corn for silage 316 % wheat for grain 11 %
milk cows 255 % hogs and pigs 2 %
forage 115 % rice 0 %
vegetables 96 % peanuts 0 %
orchards 66 % cane sugar 0 %Why it matters. Those rows call for four different actions. The surpluses are export businesses. The 96 percent is a logistics and contracting problem. The 11 percent is a partial substitution. The zeros are a latitude and will not move. A programme that acts on the 28 percent acts on none of them.
You already know this because you would never manage a P&L off a single blended margin, and a regional balance is a P&L.
BRIEF 5 — Value Density and the Freight-Share Test
The idea. Whether a good should travel is decided by the share of its delivered value that freight consumes — not by the distance it covers.
freight share = freight cost per tonne / delivered value per tonne
above about 10 % the good is regional by nature
below about 2 % the good is traded by natureWorked example. Ricardo's own 1817 numbers set the tolerance exactly. England needs 100 man-years for its cloth and 120 for its wine; Portugal 90 and
- Specialisation saves 30 man-years on four units — 7.7 percent. Two units
cross the water, so the gains vanish at 15 man-years of freight per traded unit: 15 percent of the cloth's cost and 18.75 percent of the wine's.
Ocean freight on manufactures runs far below that. Road freight on cheap bulk does not: crushed stone, ready-mixed concrete, fluid milk, beer and animal feed have run inside a haul radius of a few tens of kilometres for a century, set by nothing but this arithmetic.
Why it matters. The construction-aggregates industry is the most completely bioregional sector in the economy and has never once used the word. It got there by freight cost alone, which is the strongest available evidence that the test is real.
You already know this because nobody ships sand across an ocean and everybody ships microchips.
BRIEF 6 — Load Factor
The idea. Transport emissions are decided by the vehicle, not by the distance. A full lorry carries a kilogram at a small fraction of the cost of a car carrying ten.
The figure. Weber and Matthews' measured intensities, per tonne-kilometre:
international water, container 14 g CO2e
rail 18 g
truck 180 g 12.9 x container
air 680 g 48.6 x containerWorked example, and it is the one that overturns the intuition. Take a car at 6.0 litres per 100 km. Petrol carries 2.31 kg of CO₂ per litre — chemistry, not policy — so the car emits 0.1386 kg per kilometre. Put a ten-kilogram basket in it:
0.1386 kg / 10 kg = 13.86 g CO2 per kg per km
container ship = 0.014 g CO2 per kg per km
-------------------------------------------------
one kilometre by car = 990 kilometres by container shipColey, Howard and Winter measured the same effect in the field and found the crossover at a 6.7 km round trip: beyond that, driving to the farm shop emits more than the whole packing-hub-doorstep chain of a large box scheme. In this arithmetic, 6.7 km is 6,633 km of sea freight.
Why it matters. A regional supply chain assembled out of half-empty vans does not merely fail to beat the long chain. It loses to it.
You already know this because you have watched a two-pallet delivery arrive in a forty-tonne lorry and felt that something was wrong with it.
BRIEF 7 — Comparative Advantage, Stated Fairly
The idea. Two regions gain from trade even when one is better at everything, because what matters is the ratio of their costs, not the levels.
Worked example. Ricardo, 1817:
cloth wine opportunity cost of wine
England 100 120 1.20 cloth
Portugal 90 80 0.89 cloth
autarky, four units 390 man-years
specialised, four units 360 man-years
the gain 30 man-years = 7.7 %Portugal is better at both and still buys English cloth, because a bottle of wine costs it less cloth than it costs England.
Why it matters. This is the strongest argument against localisation and it should be stated at full strength before it is examined, because a bioregional economics that has not understood it will localise the one commodity where it has no advantage. Honour the argument. Then ask what it leaves out.
You already know this because you have hired someone to do a job you could have done yourself faster, and been right to.
BRIEF 8 — The Freight-Share Ceiling
The idea. The maximum emissions benefit available from perfect localisation is the share that transport occupies in the first place — and it is small.
The figure. Across the average US household's food footprint of 8.1 tonnes CO₂e a year, Weber and Matthews find production 83 percent, all transport 11 percent, final delivery producer-to-retail 4 percent.
Worked example. A regional grower whose emissions per kilogram are 15 percent above the distant grower's cannot be rescued by proximity: the production penalty of 15 percent exceeds the entire 11 percent that transport could ever have saved, before a single lorry moves. The comparison only reverses where freight is not 11 percent but the majority — which is exactly the case for anything flown. A good flown 9,000 km carries 6.12 kg CO₂e per kilogram in freight alone, which exceeds most foods' whole production footprint.
Why it matters. It gives you a single, checkable admission test: is local production within about 11 percent of the distant producer's efficiency? If not, localise for resilience, for employment, for taste or for control — all defensible — but do not claim the carbon.
You already know this because you have seen an efficiency programme aimed at the smallest line on the cost sheet.
BRIEF 9 — Resilience as a Priced Option
The idea. A good that is cheap to import and slow to substitute is not a localisation problem. It is a hedging problem, and hedges are priced.
buy the regional premium when premium per tonne <= p x L
p annual probability of interruption on this line
L loss per tonne if it occursWorked example. A premium of £42 per tonne is justified on resilience grounds alone when p × L ≥ 42. If the loss on interruption is £400 per tonne, the implied probability is 10.5 percent a year. That is a number a risk committee can argue about, accept or reject — which is the whole point of writing it down.
Why it matters. Most bioregional proposals assert resilience and never price it. Pricing it does two things at once: it wins the cases that deserve to be won, and it stops the programme buying acreage when what it needed was an option.
You already know this because you insure a building you do not expect to burn down, and you can say what the premium is.
BRIEF 10 — The Offtake with a Load-Factor Covenant
The idea. The instrument that makes a bioregional supply line real is a multi-year fixed-volume offtake between an anchor buyer and a producer cluster inside a named ecoregion at a stated level — with the regional premium made conditional on a measured delivery covenant.
The four terms that carry it.
- Perimeter — the ecoregion by code, level and boundary version. Not "local".
- Volume and tenor — a nominated annual tonnage over three to five years, take-or-pay on around 70 percent. Three years is the minimum that changes a planting decision.
- Price — a published regional cost-of-production index plus a margin, collared against the incumbent's delivered landed cost.
- The load-factor covenant — the premium is payable only in periods where the weighted average vehicle load factor clears a stated threshold, verified from weighbridge tickets or telematics.
Worked example. The covenant turns the finding in Brief 6 from a criticism into a contract term. Below the threshold, price reverts to the incumbent's landed cost. The cluster's consolidation problem becomes the cluster's revenue problem — which is the one place it can actually be solved.
Why it matters. For the buyer this is an executory purchase commitment, disclosed under IAS 37 or ASC 440-10 rather than recognised as a liability. For the grower it is bankable. Two settled accounting questions: whether the take-or-pay tranche creates an onerous-contract provision if the collar binds, and whether any named asset makes it a lease.
You already know this because you have signed a power purchase agreement, or read one, and this is the same animal pointed at a field.