target-path

A company commits to cutting emissions by a percentage by a year. That sentence fixes one point in the future and says nothing about the line that reaches it. Two companies can make the identical commitment, hit it on the same date, and have put very different quantities of carbon into the atmosphere along the way, because the atmosphere responds to the integral of the path and not to its endpoint. Endpoint compliance cannot see the difference, and almost every published target is stated as an endpoint.

There is a second confusion sitting underneath the first. The same headline percentage describes two different shapes. A linear path removes a fixed share of the base year every year, so the annual cut stays the same size. A compounding path removes a fixed share of last year, so the annual cut shrinks every year while the percentage in the headline does not move. These are not the same commitment, and the wording rarely distinguishes them.

The third problem only appears once a company has fallen behind. Progress is usually reported against the base year, which flatters: five years of visible reduction can still be well short of what the path required by the same date. What that shortfall costs is not a percentage of the base year at all. It is the steeper rate now required from today's level, and it is invisible to anyone watching only the target.

Two pathway shapes, one budget and one diagnostic. Absolute contraction removes a fixed percentage of the base year each year, which draws a straight line. Intensity convergence puts that same straight line under an intensity metric and multiplies it by the reader's own activity forecast, so that a company growing its output can watch intensity fall on schedule while absolute emissions rise. The budget is the area under whichever path was drawn, taken by the trapezoid rule over one year steps, which is exact for the straight line. The diagnostic inverts the path: given where you actually are and how many years remain, it solves for the annual cut that still lands on the target.

absolute      E(t) = E0 × (1 - r × (t - t0))
compounding   E(t) = E0 × (1 - r)^(t - t0)
intensity     E(t) = I0 × (1 - r × (t - t0)) × A(t)

budget        the area under E, by trapezoid over one year steps
required      r' = (E_now - E_target) / (E_now × (t_target - t_now))

Four fields draw the path and everything below them recomputes: the endpoint, the total reduction, the cumulative budget, a chart, and a table of every year with its distance from the base year. A second panel runs the same rate as both a linear and a compounding path and prints the gap in percentage points and in extra carbon. A third takes whatever actual emissions the reader can supply, marks each year behind or ahead, and computes the rate required from the last year entered.

It refuses to interpolate a year the reader left blank. A missing actual is a missing actual, and filling it with a straight line between its neighbours would put a number into the cumulative figure that nobody reported. Where the years entered are not consecutive the cumulative panel says so and withholds the total rather than computing one across a gap.

It refuses to present 4.2 percent as a rule. The figure is the floor the SBTi sets under the absolute contraction approach for a near term Scope 1 and 2 target, and the page prints it with its publisher and the date it was read. It also prints what it leaves out: the criteria adjust that floor upward for base years later than 2020, and set a separate and lower rate for Scope 3. The page makes neither adjustment and says so rather than implying a completeness it does not have.

It refuses to call a target met. Nothing in the arithmetic knows whether an inventory was drawn on the same boundary two years running, whether the base year was recalculated after a disposal, or whether a reduction came from decarbonising or from selling the emitting part of the business. Those are the questions that decide whether a number means anything, and the page lists them rather than quietly assuming past them.