When Efficiency Improves but Total Emissions Still Rise

Efficiency is a rate; strategy is accountable for the total outcome.

A business can become more efficient and still move further away from its strategic objective.

That sounds contradictory only when leaders confuse intensity with impact.

A factory can reduce energy per unit while producing so many more units that total energy consumption rises. A construction sector can improve energy intensity while expansion in demand drives total embodied emissions higher. A digital platform can reduce compute per transaction while transaction volumes grow faster than the savings. A fleet can improve fuel efficiency while total kilometres increase.

The improvement is real. The overall result can still be worse.

This distinction matters because many sustainability and productivity programs are governed through ratios: energy per tonne, emissions per square metre, cost per transaction, waste per unit, labour hours per output. Ratios are valuable, but they can create a false sense of progress when scale is moving in the opposite direction.

The Strategic Context

Shi, Chen and Shen's 2017 structural decomposition study of the Chinese construction industry provides a strong historical example. Using 1995–2009 data, the researchers decomposed changes in construction-related carbon emissions into five effects: carbonisation, energy intensity, production structure, final-demand ratio and total final demand.

Their result was not a story of failed efficiency.

Energy intensity produced a substantial counter-effect against emissions growth. But the increase associated with total final demand was even larger. Over the full period, the study attributed a 131% increase effect to total final demand and a 108% offsetting effect to improved energy intensity. Production structure and the final-demand ratio also contributed upward pressure, while changes in the carbonisation factor produced only a relatively small downward effect.

Those numbers belong to a historical Chinese industry context and should not be treated as current benchmarks. The strategic insight is durable: improving the efficiency of a growing system does not guarantee improvement in the system's absolute outcome.

What Leaders Commonly Misread

The most common mistake is treating an intensity metric as if it were the strategic objective.

If the objective is to reduce total carbon, "carbon per unit" is a driver or diagnostic measure. It is not the final outcome.

The second mistake is assuming operational efficiency neutralises demand growth. It may not. If output, floor area, transactions, customers or kilometres grow faster than the efficiency improvement, aggregate consumption and emissions can continue rising.

The third mistake is treating indirect effects as someone else's problem. Construction illustrates why this is weak systems thinking. Much of the environmental burden associated with construction occurs in upstream industries producing steel, cement, chemicals, energy and other inputs. The project boundary does not define the physical system.

A similar error occurs when a company outsources an energy-intensive process and reports the resulting reduction inside its own operations. The internal result may improve while the system-level burden moves to a supplier.

Reframing the Issue

A mature performance system distinguishes at least three layers.

Intensity asks how much resource or impact is associated with a unit of activity.

Scale asks how much activity the organisation or economy is creating.

Structure asks how the system is configured: what inputs are used, where they come from, how demand is distributed, and which upstream activities are induced.

The total outcome emerges from all three.

That means a sustainability strategy cannot be written solely as an operational-efficiency program. It also requires strategic choices about demand, product mix, asset utilisation, sourcing, service models, material substitution, reuse and the business model itself.

This is where sustainability stops being a technical agenda and becomes an enterprise agenda.

The Rebound Problem in Management Terms

Leaders do not need to adopt a specific academic rebound model to understand the managerial problem.

Imagine a hypothetical engineering business that reduces electricity consumption per manufactured unit by 20%. If production increases 35% over the same period, electricity consumption can still rise even though the process is materially more efficient.

The business has not failed operationally. It has failed only if leadership expected the intensity improvement by itself to deliver an absolute reduction.

The correct response is not to dismiss efficiency. Efficiency remains valuable. It reduces the impact that would otherwise have occurred and can improve margin and capacity. The mistake is asking one lever to perform the work of an entire strategy.

This distinction is equally important in public infrastructure. A building standard can improve the energy performance of each new building while aggregate demand for floor space, materials and infrastructure continues to expand. Both facts can be true.

Growth Changes the Decision Logic

When demand is expanding, leaders need to distinguish between avoided growth in impact and absolute reduction in impact.

These are different benefits.

If a program reduces expected emissions from 150 units to 120 while the baseline was 100, the intervention has avoided 30 units of future growth but total emissions are still 20 units higher than the starting point.

That may still be strategically valuable. The important requirement is honest accounting.

Boards should know whether an initiative:

lowers absolute impact;
lowers impact intensity only;
prevents some expected future increase;
shifts impact elsewhere;
or changes the timing rather than the total.

Without those distinctions, good technical work can be over-claimed and poor strategic choices can remain hidden.

Structure Can Defeat Local Optimisation

The construction study is also useful because its decomposition went beyond direct energy use. Production structure affected the result.

This matters for organisations operating through complex supply chains.

A company may improve its own process while shifting toward a product mix that requires more energy-intensive materials. It may source cheaper components from a location with a more carbon-intensive energy system. It may redesign a service to reduce internal labour while increasing customer travel or supplier activity.

Systems thinking requires asking where the burden goes after the intervention.

A reduction inside the accounting boundary is evidence of local improvement. It is not automatically evidence of system improvement.

Decision Framework

A practical executive review can use five questions.

1. What is the absolute outcome?

Start with the total quantity that ultimately matters: total emissions, energy, waste, water, cost, defects or another enterprise outcome.

Do not allow the headline metric to be only a ratio.

2. What is the intensity?

Measure impact per meaningful unit of output.

This separates process performance from scale and shows whether operations are genuinely becoming more efficient.

3. What is driving scale?

Identify the causes of increased or decreased activity:

customer growth;
capacity expansion;
product mix;
policy;
infrastructure investment;
price;
utilisation;
geographic expansion;
acquisition.

Scale is often a strategic choice, not an uncontrollable external fact.

4. What structural changes are occurring?

Examine supply-chain composition, energy mix, material inputs, make-or-buy decisions and technology architecture.

A lower-impact operating process can be overwhelmed by a higher-impact structural configuration.

5. What was displaced?

Test whether the intervention eliminated impact, reduced it, deferred it or transferred it to another part of the system.

This question is particularly important for outsourcing, electrification and digitalisation.

From Strategy to Execution

Immediate action: place absolute and intensity metrics side by side in executive reporting. Where an intensity KPI is improving while the total outcome worsens, require an explanation of scale and structural drivers.

Medium-term capability building: develop decomposition capability. Leaders do not necessarily need a complex econometric model for every decision, but they do need the ability to separate volume, efficiency, mix, sourcing and technology effects. Finance, operations and sustainability data should be reconciled around the same causal story.

Long-term strategic positioning: incorporate demand and business-model choices into sustainability strategy. This may include designing for longer asset life, higher utilisation, reuse, shared capacity, lower material intensity, service substitution or portfolio changes that reduce dependence on high-impact growth.

The goal is not indiscriminate contraction. It is to understand which forms of growth create enterprise value after their full system consequences are considered.

Portfolio Implications

This reasoning changes project selection.

A portfolio containing many efficiency projects can appear strongly aligned with sustainability while leaving the main growth driver untouched. If total impact is dominated by demand, product mix or asset expansion, the highest-value intervention may sit outside operations.

Portfolio leaders should therefore classify initiatives according to the mechanism by which they influence the outcome:

intensity reduction;
demand reduction or avoidance;
structural substitution;
circularity and reuse;
energy-source change;
behavioural change;
measurement and governance.

If almost every funded initiative sits in the first category, the portfolio may be optimising what is easiest to measure rather than what most strongly determines the result.

Signals to Monitor

Watch for:

improving unit metrics alongside deteriorating totals;
rapid volume growth that is absent from sustainability forecasts;
major shifts in product or customer mix;
increasing upstream material intensity;
outsourcing that reduces internal footprint but not system footprint;
energy-source improvements being offset by larger activity volumes;
business cases claiming "reduction" when they actually mean reduced growth against a baseline;
efficiency programs with no connection to absolute enterprise targets.

A useful leading indicator is the efficiency-growth gap: compare the rate of intensity improvement with the rate of activity growth. If growth consistently exceeds efficiency improvement, absolute reduction will remain difficult without additional structural levers.

Questions for the Leadership Team

Are our headline sustainability measures absolute outcomes or only efficiency ratios?
Which growth assumptions could overwhelm the gains in our current improvement portfolio?
What portion of impact arises outside our direct operating boundary?
Are we eliminating environmental burden or transferring it to suppliers, customers or another geography?
Which portfolio initiatives act on demand and structure rather than only operational efficiency?
How would our strategic conclusion change if output grew 20%, 50% or 100% faster than planned?

Closing Perspective

Efficiency is essential because waste should not be defended.

But efficiency alone is not a strategy for a system whose scale and structure are changing.

Leaders need to know whether they are making each unit cleaner, making the total system cleaner, or merely slowing the rate at which the total problem grows. Each can be valuable. They are not the same achievement.

The quality of strategy begins with telling the difference.

Source basis: This article is an original ERANORTH synthesis principally informed by Qian Shi, Jindao Chen and Liyin Shen (2017), Driving factors of the changes in the carbon emissions in the Chinese construction industry, Journal of Cleaner Production, 166, 615–627, with supporting system-boundary insight from Rui Hu and Chao Zhang (2017).

About EraNorth Insights
EraNorth Insights publishes practical analysis on strategy, projects, operations, transformation and decision intelligence for professional and organisational use. About EraNorth.