Backcasting

Thinking Tools4.3 Crafting Solutions for the Future StateAlso known as Natural Step Backcasting, The Natural Step

From the Natural Step framework of Karl-Henrik Robèrt (1989), drawing on backcasting as John B. Robinson named it (Energy Backcasting, 1982) and the earlier energy work of Amory B. Lovins (1976).

Defines the end state that must be reached, measures today's gap from it, lays milestones backward from that future, and screens each near-term move against it.

The path drawn backward from a zero-emission fleet in 2035

Planning starts at the end state and steps back toward today: each milestone is what must already be true for the one after it. Play walks the path from 2035 back to this year; select a milestone to read why it sits where it does. Illustrative.

Select a stage to read it, or play the sequence.2035: zero emissions. The end state, defined precisely: no van burns fossil fuel, and the depots buy only renewable power. Everything before it is worked out from here.2031: half electric. Vans run about ten years and a tenth of the fleet is replaced each year, so buying only electric from this year puts about half the fleet on electric by 2031.2028: chargers at depots. Grid upgrades take two to three years to approve and build, so the power must be applied for now to be ready before most of the fleet turns over.This year. A diesel van bought now would still run in 2035, so diesel buying stops now. The first moves are the ones every later milestone depends on: apply for depot power and lease a first batch of electric vans to learn on.
  1. 2035: zero emissions. The end state, defined precisely: no van burns fossil fuel, and the depots buy only renewable power. Everything before it is worked out from here.
  2. 2031: half electric. Vans run about ten years and a tenth of the fleet is replaced each year, so buying only electric from this year puts about half the fleet on electric by 2031.
  3. 2028: chargers at depots. Grid upgrades take two to three years to approve and build, so the power must be applied for now to be ready before most of the fleet turns over.
  4. This year. A diesel van bought now would still run in 2035, so diesel buying stops now. The first moves are the ones every later milestone depends on: apply for depot power and lease a first batch of electric vans to learn on.

Four moves for this year, screened against the 2035 end state

Each possible first move faces the three questions: does it head toward the end state (a must), is it a platform for later steps, and does it return enough to fund the next one? A move that heads the wrong way is out, however cheap. Change a score or a weight and the ranking follows. Illustrative.

Heads toward zero emissions by 2035

Must40 cheap diesel vansLease 12 electric vansChargers for all depotsRenewable diesel blend

A platform for later steps

Weight640 cheap diesel vans2Lease 12 electric vans7Chargers for all depots9Renewable diesel blend3

Returns enough to fund the next step

Weight540 cheap diesel vans8Lease 12 electric vans6Chargers for all depots4Renewable diesel blend6

Can start this year

Weight340 cheap diesel vans9Lease 12 electric vans7Chargers for all depots6Renewable diesel blend9

Weighted total

40 cheap diesel vans79OutLease 12 electric vans93OutChargers for all depots92OutRenewable diesel blend75Out

First moveNo option passes every must: rescope or find another.40 cheap diesel vansLease 12 electric vansChargers for all depotsRenewable diesel blend

In order

  1. 9340 cheap diesel vansLease 12 electric vansChargers for all depotsRenewable diesel blendOut: fails a must (Heads toward zero emissions by 2035).
  2. 9240 cheap diesel vansLease 12 electric vansChargers for all depotsRenewable diesel blendOut: fails a must (Heads toward zero emissions by 2035).
  3. 7540 cheap diesel vansLease 12 electric vansChargers for all depotsRenewable diesel blendOut: fails a must (Heads toward zero emissions by 2035).
  4. 7940 cheap diesel vansLease 12 electric vansChargers for all depotsRenewable diesel blendOut: fails a must (Heads toward zero emissions by 2035).

An example

A regional parcel carrier had promised its largest customers a zero-emission delivery fleet by 2035. The fleet plan, built by extending the past, replaced vans as they wore out and assumed electric vans would get cheaper; it reached about 40 percent electric by 2035.

The operations director started from the other end. She defined the end state precisely: all 300 vans electric, depot power from renewable sources. Then she worked back. Vans run about ten years, so any diesel van bought from now on would still be running in 2035: diesel buying had to stop this year. Grid upgrades at the depots take two to three years, so the power had to be applied for now to have chargers by 2028.

That week a dealer offered forty diesel vans at a deep discount. Screened against the end state, the offer failed the first question outright. The board turned it down, applied for power at all four depots, and leased twelve electric vans to learn on.

Why it works

A forecast starts from today and asks where current trends lead. When the trends lead somewhere unacceptable, the forecast still recommends adjusting the trend a little. Starting from a fixed end state asks a different question: what must already be true each year for the end state to be possible? That reveals deadlines that are much nearer than they look, such as the last year a long-lived asset can be bought.

The screen on present actions keeps the path honest. A move that is cheap and pays back fast can still head the wrong way, and the three questions catch it: does it move toward the end state, does it leave room for later steps, and does it return enough to fund the next one? Redrawing the path as understanding improves keeps the destination fixed while the route changes.

The thinking behind it

Karl-Henrik Robèrt Karl-Henrik Robèrt, a Swedish cancer scientist, circulated a paper on sustainability to fifty scientists in 1989 and, after 22 drafts reached consensus, founded The Natural Step. Its framework plans by backcasting: start from principles of a sustainable society and work back to the steps that lead there. He told the story in The Natural Step Story: Seeding a Quiet Revolution (2002).

John Holmberg John Holmberg, a sustainability researcher, set out the method with Robèrt in Backcasting from Non-Overlapping Sustainability Principles: A Framework for Strategic Planning (2000), in the International Journal of Sustainable Development and World Ecology. Success is defined first, as a set of principles that do not overlap, and planning runs backward from that end state to the steps an organization can take now.

John B. Robinson John B. Robinson, an environmental policy scholar, coined the term in Energy Backcasting: A Proposed Method of Policy Analysis (1982), in Energy Policy. Define a desirable future first, then work backward to the policies and steps that would connect it to the present, the reverse of forecasting from current trends. The Natural Step also drew on earlier energy work by Amory B. Lovins (1976).

How to run it

  1. Define the end state precisely. Write it as a condition that can be checked, such as all 300 vans electric by 2035, with the principles behind it.
  2. Measure today against it. List each gap between where the organization is now and that end state.
  3. Work backward to the milestones. For each milestone, ask what must already be true the year before, using real lead times such as asset lives and approvals.
  4. Find the nearest deadline. Look for the step that has to happen this year or next for any later milestone to be possible.
  5. List candidate actions for now. Gather every move on the table this year, including the ones that look cheap and easy.
  6. Screen each action three ways. Ask whether it heads toward the end state, whether it is a platform for later steps, and whether it returns enough to fund the next.
  7. Redraw the path each year. Keep the end state fixed, and revise the milestones as costs, technology, and lead times change.

Where it goes wrong

The end state is vague. A goal such as being greener cannot be worked back from, because no milestone follows from it. Define it so that anyone could check in 2035 whether it was met.

The backward path ignores lead times. Milestones spaced evenly on a chart look plausible and miss the long ones, such as asset lives and permits, which set the real deadlines. Ask, for each milestone, how long its slowest part takes.

The screen gets dropped when a bargain appears. A discount on the wrong asset is the moment the plan is tested. Run every offer through the first question before discussing price, and record the answer.

When to use something else

  • Design Fiction. When the end state itself is unclear and people need something concrete from that future to argue about first.
  • Scenario Planning. When the future cannot be chosen and the plan must hold up across several possible ones.
  • Means-Ends Analysis. When the goal is near and the work is closing the gap step by step from where things stand.

Where it sits in SOLVE eX

Go to the source

  • The Natural Step framework, founded in Sweden; The Natural Step Story, New Society Publishers, 2002, 1989
  • Backcasting from Non-Overlapping Sustainability Principles: A Framework for Strategic Planning, International Journal of Sustainable Development and World Ecology (John Holmberg and Karl-Henrik Robèrt), 2000
  • Energy Backcasting: A Proposed Method of Policy Analysis, Energy Policy, 1982
  • Energy Strategy: The Road Not Taken?, Foreign Affairs, 1976

Back to the catalog