Integrating Consumption-Based Metrics into Sectoral Carbon Budgets to Enhance Sustainability Monitoring of Building Activities

Abstract

Climate policies such as sectoral carbon budgets use national greenhouse gas emissions inventories to track the decarbonization of sectors. While they provide an important compass to guide climate action, the accounting framework in which they are embedded lacks flexibility for activities that are international and at the crossroads of different sectors. The building activities, being largely linked with important upstream emitters such as energy production or industrial activities, which can take place outside of national borders, are such an example. As legislation increasingly addresses the whole-life carbon emissions of buildings, it is vital to develop cross-sectoral accounting methods that effectively measure and monitor the overall impact of buildings. Such methods are essential for creating sound and holistic decarbonization pathways that align with sustainability policies. This article aims to provide a consistent approach for depicting the life-cycle emissions of buildings at the national level, using France as a case study. By integrating the different emission scopes with decarbonization pathways, this approach also enables the creation of comprehensive whole-life carbon budgets. The results show that the French building stock footprint reached 162 MtCO₂eq in 2019, with 64% attributed to operational emissions, primarily from fossil fuel combustion, and the remainder to embodied emissions, mainly from upstream industrial and energy sectors. Overall, 20% of the emissions occurred outside the national borders. Under various global decarbonization pathways, the significance of embodied emissions is projected to increase, potentially comprising 78% of the life-cycle emissions by 2050 under the current policies. This underscores the necessity for climate policies to address emissions beyond territorial and operational boundaries.

1. Introduction

Human activities are warming Earth at an alarming rate that is unprecedented in the last 2000 years [1]. The last decade witnessed the highest global net anthropogenic greenhouse gas emissions (GHGEs) in human history [2]. In order to reduce the risks and impacts of climate change, the next decades are critical in order to pursue the well-below 2 °C objective of the Paris Agreement [3]. The 1992 Rio Conference is one of the first responses from the international community to address climate change. It marks the creation of the United Nations Framework Convention on Climate Change (UNFCCC), whose primary objective is to stabilize the “greenhouse gas concentrations in the atmosphere at a level that would prevent dangerous anthropogenic interference with the climate system” [4].

1.1. GHGE Accounting Systems

In 1997, the Kyoto Protocol operationalized the UNFCCC by establishing legally binding emission reduction targets for the Annex-I countries. Article 5 also introduces an emission accounting system that includes rules for measuring, reporting, and verifying emissions [5]. It follows a production-based accounting (PBA) system, which is framed by the Intergovernmental Panel on Climate Change (IPCC) methodology [6]. In this framework, GHGEs are attributed to the sectors and regions where they physically occur. One of the main limits of a PBA system is that it does not identify the (potentially) increasing imported emissions that are due to globalization. In parallel, other accounting systems have been developed [7], the most widely used being the consumption-based accounting (CBA) system [8], where the impacts are attributed to the region where the final demand occurs and to the sector at the end of the supply chain [9]. As a result, in most developed countries, using a CBA system reveals much larger GHGEs and can change the repartition between sectors [10].

Moreover, another limitation concerning the current accounting methods relates to the sectoral breakdown used in the UNFCCC process. The IPCC guidelines for national greenhouse gas emissions define four main sectors for reporting GHGEs, namely (1) energy, (2) industrial processes and product use, (3) agriculture, forestry, and other land use (AFOLU), and (4) waste, which can be further subdivided (for example, the energy sector is divided into mobile, stationary, including the emissions from housing heating, and fugitive emissions). In France, the SECTEN format used for the Low-Carbon National Strategy (SNBC) uses a slightly modified breakdown based on seven macroeconomic sectors, namely (a) industry, (b) residential and tertiary, (c) energy, (d) transport, (e) agriculture, (f) waste, and (g) land-use, land-use change, and forestry (LULUCF) [11].

While they are straightforward for reporting purposes, these classifications do not enable clearly pointing out and acknowledging high-emission activities for which the emissions are split across different sectors and countries. With 37% of the CO₂ emissions worldwide in 2021 [12], building and construction activities are such an example. Indeed, in PBA systems, buildings’ emissions are the ones associated with the combustion of fossil fuels (e.g., for heating, hot water, or cooling) in the use stage as they are the only type of emissions that physically occur within them. In the UNFCCC format, they appear within the energy sector (under other sectors, which differentiates commercial, institutional, and residential buildings), while the SECTEN format displays a separate residential and tertiary sector, which represents 17.9% of the national emissions in 2021 [11].

1.2. Buildings in the Needs–Activities–Sectors Framework

In building environmental assessments, emissions are analyzed using life-cycle assessment (LCA) methods, which aim to quantify the environmental impacts through the entire life cycle of buildings, from the production of the building materials to the end-of-life stage as well as the operation of the buildings. In France, it has been introduced in the new environmental regulation (RE2020) for new buildings, which integrates limit values for embodied and operational carbon emissions [13].

If a life-cycle perspective would be transposed to a PBA system, the buildings would be associated with the emissions from multiple sectors, in particular, the industry sector (e.g., for the production of buildings’ materials and equipment) and the energy sector (e.g., for electricity and heat production). For this reason, the term activities might be better-suited and has been proposed in recent studies [14,15]. Figure 1 illustrates this framework.

The building activities serve the needs of housing and shelter and contain multiple activities that are involved at the different stages of a building’s life cycle. To address these activities, several economic sectors are involved in the supply chain. This framework acknowledges the significant influence that stakeholders from diverse sectors exert on the emissions associated with building activities. Given the limitations of the narrowly defined ’building’ sector in national emission inventories, a comprehensive understanding of emissions and their potential for mitigation is essential.

The footprint of the construction and building activities has been investigated for various environmental impacts and at different geographical scopes. For GHGEs, some studies handle the global scale [8,16,17,18] while others focus on the national scale, for example, in Ireland [19,20], in Australia [21], in Switzerland [22] and for India, Italy, South Africa, and the UK [23]. Nevertheless, it is difficult to compare these results as the scope of what building or construction means is not homogeneous between studies. Most focus on the footprint of construction as an economic sector, which includes buildings and infrastructure according to the Statistical Classification of Economic Activities (NACE) [24], also referred to as the built environment. As such, they target the emissions arising from the construction materials and equipment life cycle but omit integrating the emissions from the different energy carriers used during buildings’ use stage. Ref. [20] gathers in a single methodology the whole-life carbon (WLC) emissions but targeting the entire built environment. They propose two methods for embodied GHGE accounting, the sector summation method, which uses the floor area and average GHGE intensities, and the commodity accounting method, which relies on various national and international statistics. However, they do not use Input–Output Analysis (IOA), which has been largely used to calculate footprint indicators [25,26] and to identify cross-sectoral impacts [27], not only for GHGEs but also for water, resources, or land-use changes. IOA has also played a significant role in quantifying imported emissions, a dimension typically absent from the conventional sectoral carbon budgets. This approach offers additional consumption-based metrics that hold particular relevance in the context of construction activities given the fragmentation of the international supply chains for construction materials and equipment [28].

1.3. The Need for Embodied GHGE Budget

In recent years, there has been a notable shift in focus towards embodied GHGEs, which have been rising in relative and absolute terms, especially for new buildings [29]. While they represent 25% of the WLC emissions globally today, their share is expected to reach 49% by 2050 under a BAU scenario [30]. At the EU level, significant efforts have been deployed to represent WLC baselines and embodied GHGE pathways [31]. In their efforts to downscale the WLC budgets for buildings in Switzerland and Denmark [32,33] both use the results from previous IOA studies to account for the baseline proportion of embodied GHGEs and their imported share. However, they do not provide a detailed breakdown of the contributions from the specific regions and sectors regarding these embodied emissions. Other studies have undertaken supply chain decomposition, in particular [18,23] with the use of Structural Path Analysis (SPA). Nevertheless, they do not project emissions in the future with the combination of international sectoral scenarios.

1.4. Contributions of the Paper

This article builds upon the needs–activities–sectors framework and aims to reconcile the various methods of emissions accounting in order to prepare for appropriate carbon budgets for buildings [14,34]. It intends to clarify the different types of emissions that arise during buildings’ life cycle and highlights the most important contributors in terms of energy carriers, upstream sectors, and geographical regions. Through the integration of a national operational GHGE scenario with scenarios for the decarbonization of the upstream sectors responsible for embodied emissions, this study tends to promote a more comprehensive approach to the decarbonization of building activities. Consequently, the goal is to better link the climate policies and sectoral legislation and to propose additional consumption-based metrics for national building activities, which can help to reduce GHGEs through the supply chain [35].

It raises two main questions:

• At the national level, how do we account for the life-cycle emissions of buildings as a cross-sectoral and cross-border activity?
• How do we apply suitable decarbonization pathways to the different scopes of emissions identified, in particular embodied GHGEs, to complement the coverage of the existing sectoral budgets?

2. Materials and Methods

In France, there has been no precise quantification of the full scope of emissions that occur during buildings’ life cycle. Yet, France is an interesting case as the Environmental Regulation (RE2020) for new buildings now integrates thresholds for embodied and operational carbon [13], while the Law for Ecological Transition and Green Growth (LTECV) [36] aims to achieve a level of energy performance in accordance with ’low-energy building’ standard for the entire building stock by 2050. A deep understanding and follow-up of buildings’ GHGEs are then essential to link buildings’ LCA with a more holistic vision at the building stock level. Expanding the scope of emission considerations in climate policies, e.g., the SNBC, is also needed. This expansion is necessary because they focus on operational GHGEs and do not explicitly provide pathways and reduction strategies for embodied GHGEs, which is a rising concern at planetary [30] and European levels [37].

In order to address these issues, the methodology developed in this article consists of two parts, displayed in Figure 2a. The system boundary of the study is displayed using the EN-15978 stages in Figure 2b while delineating the life-cycle inventory methods applied for both operational and embodied GHGEs.

Firstly, the study introduces an accounting method for separately calculating distinct emission scopes at the national level, using 2019 as the reference year. This method also facilitates the quantification of imported emissions, which are traditionally absent from national statistics and climate policies for building activities. Subsequently, the results are projected until 2050 using different decarbonization pathways. National scenario for operational GHGEs is combined with different scenarios that reflect decarbonization pathways for the supply chain sectors responsible for the embodied GHGEs of the building activities.

2.1. System Boundary

The study adopts a comprehensive cradle-to-grave analysis that covers all the life-cycle stages. For operational GHGEs, the method integrates a top-down statistical approach with elements of process-based LCA. This involves combining detailed information on energy carriers used in both residential and non-residential buildings with life-cycle emission factors. Conversely, the embodied GHGE calculation adopts an Input–Output Life-Cycle Inventory (IO-LCI) approach, leveraging the Exiobase database [38]. The main advantage of this approach is the minimization of truncation errors, which refers to the omission of relevant processes or sectors from the analysis, a significant concern in process-based LCA [39]. For macro-level analysis, IO-based LCI is often preferred to avoid issues with system boundary definitions [16]. One of the main disadvantages of the IO-LCI approach is the aggregation issue, referring to the consolidation of different industries into broader sectors, potentially masking the nuances and specificities of individual industries. Exiobase is chosen for its extensive coverage of products, facilitating an economy-wide view that minimizes aggregation errors. Ultimately, the study aggregates results for operational and embodied GHGEs to represent whole-life-cycle (WLC) emissions, thereby merging the accuracy of national energy use statistics with the comprehensive perspective of IO-LCI for embodied impacts.

2.2. GHGE Accounting Methods

At the building level, the EN-15978 [40] provides a consistent and standardized methodology for LCA calculations. The four life-cycle stages (e.g., modules A–B–C–D) are often grouped into operational and embodied GHGEs. The first ones refer to the B6–B7 stages and can be further decomposed into direct operational GHGEs, which occur at the building site (e.g., fossil fuels and biomass combustion, or gas-leaks from heat pumps), and indirect operational GHGEs, which are the results of electricity and district heating production. On the other hand, embodied GHGEs are associated with the construction materials and equipment’s life-cycle GHGEs [41].

Another standardized methodology for GHGE accounting is the scope 1–2–3 inherited from the GHG Protocol [42]. While primarily designed for organizations, it has been used to classify building life-cycle GHGEs [16,43]. However, the two systems are not necessarily equivalent. Indeed, two types of emission factors (EFs) can be used for energy carriers: direct emission factors (D-EFs)—which calculate GHGEs for the combustion process—and life-cycle emission factors (LC-EFs)—which also include upstream processes, such as transport. When using LC-EFs, the upstream part of the GHGEs could hence be classified as a scope-3 instead of a scope-1 emission.

The relationship between these two classifications and the sectoral approaches presented in the IPCC guidelines and French SECTEN format is shown in

Table 1. Relationships between GHGE accounting methods.

	Direct Operational GHGEs	Indirect Operational GHGEs	Embodied GHGEs
GHG Protocol	Scope 1	Scope 2	Scope 3
EN-15978	B6–B7		A1–A5 B1–B5 C1–C4
IPCC	Residential (1.A.4.b), Commercial /institutional (1.A.4.a)	Public electricity, heat production (1.A.1.a)	All others
SECTEN	Residential Tertiary	Energy	Industry Transport Waste LULUCF Agriculture

.

When looking at national inventories, building activities’ GHGEs are heterogeneous as they are cross-sectoral. Nevertheless, coupling standardized methodology at the building level with the national inventory process can close the gap between climate policies and building environmental assessment methods. Therefore, in this study, the operational–embodied framework is transposed to the national level to depict the whole-life GHGEs of the building activities.

The SECTEN format is not an official UNFCCC reporting format but it is used at national level for climate policies, e.g., the SNBC. It aims to be more understandable by economic stakeholders and is specific to the French context. It covers GHGEs and air pollutants with annual time-series available from 1990. Inside each sector, the source of GHG and air pollutants are reported by Selected Nomenclature for Air Pollutants (SNAP) code that corresponds to a detailed level [44].

2.2.1. Operational GHGE Calculation

The calculation of operational GHGEs starts from detailed knowledge on the energy carriers used in buildings. At national level, the energy balance provides information on the extraction of energy from the environment to its transformation and consumption by the different economic sectors. In France, the energy balance is provided by the Statistical Data and Studies Department (SDES) [45], which specifies energy flows for different sectors, including the residential and tertiary sectors. Additionally, the French electricity transmission system operator (RTE) offers granular data on electricity consumption by use (e.g., heating, domestic hot water, air conditioning, lighting, and other uses) [46]. The energy flows are then converted into GHGEs by applying emission factors from the Base Carbone [47]. Operational GHGEs also include fugitive fluorinated gases (F-gases) (used as refrigerant gas in heat pumps and air conditioning systems). These emissions are taken directly from the national GHGE inventory [11].

Table 2. Emission factors of energy carriers in 2019 in kgCO₂eq/kWh taken from the Base Carbone [47].

	D-EF	LC-EF
Biomass	0	0.0288
Geothermal	0	0.045
Solar thermal	0	0.055
Biogas	0.0428	0.044
Electricity (average)	0.0418	0.0607
Heat	0.107	0.132
Natural gas	0.204	0.227
LPG	0.233	0.272
Oil products	0.272	0.325
Coal	0.345	0.377

provides values for both direct emission factors (D-EFs) and life-cycle emission factors (LC-EFs).

In the method, LC-EFs are employed to include upstream emissions attributed to the entire life cycle of energy infrastructure (e.g., extraction, production, transportation, and losses). Consequently, the utilization of LC-EFs leads to greater operational GHGEs compared to studies that conventionally rely on D-EFs [48]. Compared to other countries, it should be noted that the French electricity emission factor is quite low thanks to a large reliance on nuclear and hydropower production. The average emission factor is provided in Table 2, with specific values for the different usages detailed in Appendix A.2.

To encapsulate the method in a formula, operational GHGEs for a given year t can be calculated as follows:

$$\theta \left(t\right)=\sum _k^n(E_k\left(t\right)*f_k\left(t\right))+\mathrm{\Gamma }\left(t\right)$$

(1)

where $E_k\left(t\right)$ represents the energy consumption of energy carrier k in kWh derived from the energy balance for the year t, $f_k\left(t\right)$ denotes the LC-EF of energy carrier k in kgCO₂eq/kWh as sourced from the Base Carbone for the year t, and $\mathrm{\Gamma }\left(t\right)$ accounts for the F-gases as reported in the national GHGE inventory, measured in MtCO₂eq for the year t.

2.2.2. Embodied GHGE Calculation at National Level

Recently, embodied carbon is receiving growing attention [29,31], and numerous studies are trying to quantify the weight of supply chain emissions. Embodied emissions are generally a blind spot of buildings policies [12]. At the building level, embodied GHGEs are usually quantified using process-based LCA. However, at national level, the process becomes more challenging. Indeed, buildings’ LCAs are only available for a couple of new buildings, with no equivalent for renovation and demolition projects. Another method would be to study the physical flows from sectors that produce the necessary inputs of the building activities. Alas, detailed material flow statistics on the supply and use of different construction materials and equipment are lacking [49], and it is thus difficult to assess the GHGEs of selected materials at national level.

To overcome these limitations, this study follows a top-down approach for the embodied GHGE calculation. It is enabled by the use of Input–Output Analysis (IOA), which has been widely used to estimate scope 3 GHGEs [27]. IOA looks at how different sectors of the economy are interconnected through their production and consumption patterns. It dates back to the 1930s [50] and has been used in environmental assessments since the 1970s [51], giving rise to Environmentally Extended Input–Output Tables. If it is not adequate to study the GHGE impact of buildings’ use stage because of the lack of information on energy carriers used [52], it is frequently used for macro-scale assessments [8], in particular with the use of Multi-Regional Input–Output (MRIO) databases, which have been flourishing in recent decades. Among the different MRIO databases available, Exiobase [38] is used in this study thanks to its large sectoral decomposition (163 sectors, 200 products) and set of environmental extensions [53].

In IOA, direct and total impact multipliers depict, respectively, the direct and total attribution of impacts from production to one unit of final demand [54]. Total impact multipliers are provided by

$$m=(F·\widehat{X^{-1}})·{(I-A)}^{-1}=f·L$$

(2)

where F represents the total impact matrix, A is the inter-industry coefficients matrix, I is the identity matrix, X is the total output, f is a matrix of direct impact multipliers, and L is the Leontief inverse.

In order to better understand the origins of GHGEs along the supply chain, a diagonal matrix of direct impact multipliers is created and multiplied by the Leontief inverse (L). By multiplying it with the vector of final demand for the French construction sector (y), the footprint is calculated:

$$F{P}_C=\widehat{f}·L·y$$

(3)

where $F{P}_C$ is the construction footprint. As such, it is possible to assess the contribution of the various sectors of the supply chain in the GHGE footprint of the construction sector and identify GHGE hotspots [54].

However, this study focuses on buildings’ activities and not the all-built environment. Thus, one last step concerns the subdivision of the construction sector in order to remove the civil engineering GHGEs (e.g., associated with infrastructure such as bridges). As Exiobase does not differentiate the subsectors inside the construction sector, this study relies on a 139 symmetric IO table provided by the French statistical office (INSEE) to obtain additional information. It includes a subdivision of the NACE construction sector in four subgroups, namely Development of building projects (F41.a NACE code, 75th sector in the IOT), construction of residential and non-residential buildings (F41.b NACE code, 76th sector in the IOT), civil engineering (F42 NACE code, 77th sector in the IOT), and Specialized construction activities (F43 NACE code, 78th sector in the IOT). The Z transaction matrix describes the inter-sectoral exchanges in France, where rows represent the supplying sectors and columns depict demanding sectors:

$$Z=\left[\begin{array}{cccc}{Z}_{1,1}& Z_{1,2}& \cdots & Z_{1,139}\\ Z_{2,1}& Z_{2,2}& \cdots & Z_{2,139}\\ ⋮& ⋮& \ddots & ⋮\\ Z_{139,1}& Z_{139,2}& \cdots & Z_{139,139}\end{array}\right]$$

(4)

where each $Z_{i,i}$ (e.g., on the main diagonal) represents intra-sectoral exchanges and each $Z_{i,j}$ represents economic transactions from sector i to j.

The following equations are then employed to obtain the share of the civil engineering subsector in the construction sector in the Z matrix:

$$r={\displaystyle \frac{{\sum }_{k=1}^{139}{Z}_{k,77}}{{\sum }_{k=1}^{139}(Z_{k,75}+Z_{k,76}+Z_{k,77}+Z_{k,78})}}$$

(5)

where r symbolizes the ratio of the total economic transactions involving the civil engineering subsector and all economic sectors to the sum of economic transactions involving the entire construction sector (e.g., containing the four subsectors previously mentioned) and all economic sectors.

Subsequently, this ratio is used as a proxy to remove the civil engineering-associated GHGEs to finally calculate the building activities footprint $F{P}_B$:

$$F{P}_B=\widehat{f}·L·y·(1-r)$$

(6)

The decomposition of regions and sectors in Exiobase helps to provide a more comprehensive and accurate picture of the environmental impacts of upstream sectors across different geographical regions. Although they provide detailed information, aggregations can also be useful to better interpret the results. In this study, aggregation is conducted using concordance matrices (detailed in Appendix B). In particular, the 44 regions × 200 products classification is linked with two formats:

• A 3 regions × 8 sectors classification. It enables to couple the results with the SECTEN format used by the SNBC. An additional services sector and the intra-sectoral exchanges of the construction sector replace the residential and tertiary. France, the European Union, and an aggregated Rest of the World (RoW) are represented in terms of geographical regions.
• A 15 regions × 19 sectors classification. It is inspired from the traditional aggregation in 17 sectors in IO tables, with additional custom sectors for which the IEA scenarios provide detailed pathways by 2050.

2.3. Scenarios and WLC Budgets

Complex and uncertain factors shape the decarbonization of the economy in the next decades. In this context, scenario analysis is a useful tool to address alternative future pathways [55]. In the methodology, scenarios are used to understand the complex and interconnected factors that will shape the decarbonization of buildings. They are not considered as forecasts but rather taken as insights to quantify the possible evolution of the buildings’ GHGEs in the next 30 years. It is then possible to identify over the years and in the different scenarios the contribution of the different scopes of emissions, including the sectors involved in the building activities supply chain, by assuming that the economic structure remains the same.

On one hand, operational GHGEs are regulated by the Low Carbon National Strategy (SNBC), which is the national translation of the Paris Agreement. It aims to reduce GHGEs and monitor the transition to a low-carbon economy to reach a state of net-zero emissions by 2050. In Figure 3, historical GHGEs are provided from 1990 to 2021 along with the SNBC pathways by sectors used in the SECTEN format.

A sharp decline is planned for all sectors. What is more, carbon sinks need to increase in order to achieve a balanced state in 2050. In the study, direct operational GHGEs follow the residential and tertiary sectors pathway, whereas indirect operational GHGEs follow the energy sector pathway. The two sectors have aggressive reduction pathways, with reduction to happen quickly, e.g., in the next years, compared to sector like transport where it is to happen in the next decades. Thus, only the residential and tertiary pathway for direct operational emissions and the energy pathway for indirect operational emissions are considered.

On the other hand, embodied GHGEs occur in different geographical regions and are caused by multiple sectors. In that case, it is interesting to explore scenarios that display different pathways for regions and sectors. In the methodology, three IEA scenarios, displayed in

Table 3. IEA scenarios used for embodied carbon pathways.

	Scenario Category	Scenario Type
Net-Zero 2050 (NZS)	Normative	Transforming
Announced Pledges Scenario (APS)	Predictive	What-if
Stated Policies Scenario (STEPS)	Predictive	Forecasts

, are used.

The scenario typology described by [56] is used to qualify the nature of each scenario. IEA scenarios are fully documented in the 2023 World Energy Outlooks [57]. The rationale behind each of them is the following:

• Net-Zero Emissions Scenario: it reaches a state of net-zero emissions in 2050 globally. It is compatible with a 1.5 °C temperature rise in 2100 with limited overshoot.
• Announced Pledges Scenario (APS): it assumes that the policies and targets announced by countries will be implemented fully and on time, including their long-term Nationally Determined Contributions (NDC) pledges. It is associated with a 1.7 °C temperature rise in 2100.
• Stated Policies Scenario (STEPS): it considers a wide range of policies and measures those that are currently in place or under development in different countries. It is associated with a 2.4 °C temperature rise in 2100.

In practice, the different scenario pathways (in % of reduction compared to 2019) are applied to the 2019 results for operational and embodied GHGEs:

$$GWP\left(t\right)=GWP\left(2019\right)·\alpha \left(t\right)$$

(7)

where $GWP\left(2019\right)$ represents the calculated GHGEs in 2019, $GWP\left(t\right)$ represents the GHGEs at year t, and $\alpha$ represents the annual reduction rate at year t.

The SNBC (used for operational GHGE pathways) provides annual reduction percentages up to 2050, while IEA scenarios (used for embodied GHGE pathways) are only offered for 2030, 2035, 2040, and 2050. To address this gap, linear regression is employed to estimate values for the intervening years. The graphs are generated using the pyam package [58].

To establish decarbonization pathways for WLC assessment, the SNBC operational GHGE pathway is integrated with the pathways of embodied GHGEs from the three IEA scenarios. This integration not only yields a more comprehensive and holistic idea of the future possible GHGEs arising from the building activities but also enables assessing the distribution of the different scopes of emissions across scenarios.

3. Results

According to the method described above, the GHGEs of the French building activities can be estimated for the year 2019. Then, the decarbonization pathways provided by the SNBC and IEA can be applied to deduce the WLC budgets for buildings by 2050.

3.1. GHGE Accounting

3.1.1. Operational GHGEs

In 2019, the French building stock operational GHGEs represented 104 MtCO₂eq, with 79% being direct operational (e.g., 83 MtCO₂eq) and 21% being indirect operational GHGEs (e.g., 21 MtCO₂eq). The results differ compared to the national statistics provided by the SDES, which indicate a value of 55 MtCO₂ for the operational GHGEs [48]. The two main differences are the accounting of all the GHGEs and not just CO₂ and the use of LC-EFs instead of D-EFs. Indeed, using LC-EFs added nearly 20% of the footprint.

Figure 4 illustrates the relative importance of the different energy carriers in the total energy consumption in Figure 4a and the operational GHGEs in Figure 4b of the residential and tertiary buildings from 2011 to 2020.

By juxtaposing the two sub-figures, Figure 4a,b, a comparative analysis of the impact of each energy carrier on the energy consumption and the resultant operational GHGEs can be conducted. For instance, in 2019, electricity accounted for 41% of the energy consumption but only 17% of the operational GHGEs. This disparity can be attributed to the predominantly low-carbon sources used in French electricity production, such as nuclear and hydropower. A similar pattern is observable for the ‘renewable energy and waste’ category, which constituted 15% of the energy consumption (primarily in residential buildings) but only 2% of the operational GHGEs. In contrast, even though natural gas and oil products have modest proportions in terms of energy consumption, at 24% and 12%, respectively, they substantially influence the operational GHGEs. Natural gas contributes to 44% and oil products to 26% of the total operational GHGEs.

In terms of building types, residential buildings emerge as the dominant contributors both for energy consumption and operational GHGEs. In 2019, they accounted for 64% of the energy consumption and 60% of the operational GHGEs due to their reduced emissions of F-gases compared to tertiary buildings. When looking at the trends from 2011, 2013 was the peak year. Then, a decrease is observed, which can be partly explained by the 2012 Thermic Regulation (RT2012) and the climate severity index [44]. The shares of the different energy carriers are quite stable between 2011 and 2020. For the operational GHGEs, the largest differences are for natural gas, which rose from 38% to 44%, while oil products decreased from 30% to 27%. The F-gases fluctuated, reaching a peak in 2014 with 10.2 MtCO₂eq and decreasing since then to reach 6.5 MtCO₂eq in 2020.

3.1.2. Embodied GHG Emissions

The embodied GHGEs represented 57.9 MtCO₂eq in 2019. After calculation (detailed in Section 2.2.2), the results are available in a 200 products × 44 regions format. This detailed disaggregation enables mapping the most impactful sectors and the regions where they occur. The results show that the GHGE footprint of the French construction sector is quite concentrated, with 20 sector–region combinations representing half of the impact.

Table 4. Top ten couples of country–product contributors to embodied GHGEs in 2019.

Region	Sector	MtCO2eq	%
France	Cement, lime, and plaster	9.14	15.8%
France	Construction work	2.14	3.7%
France	Stone	1.74	3%
France	Supporting and auxiliary transport services	1.14	2%
France	Waste for treatment: Landfill	1	1.7%
RoW Africa	Cement, lime, and plaster	0.9	1.6%
China	Basic iron and steel	0.79	1.4%
China	Electricity by coal	0.77	1.3%
France	Basic iron and Steel	0.75	1.3%
France	Transportation services	0.64	1.1%

displays the top ten combinations of regions and sectors, representing their absolute and relative shares of the total embodied GHGEs (in MtCO₂eq and %, respectively).

The French cement, lime, and plaster sector stands out as particularly impactful, with 15.8% of the GHGE footprint alone (e.g., 9.14 MtCO₂eq), while the second-most impactful sector, the French construction sector (e.g., representing the intra-sectoral exchanges), is far behind, representing 3.7% of the embodied GHGEs (e.g., 2.14 MtCO₂eq).

Figure 5 enables further analyzing the global supply chain emissions. The results and aggregation for the sectors are illustrated with a Sankey diagram (Figure 5a), while the outcomes and aggregation for the regions are presented through a sunburst plot (Figure 5b).

In the analysis of the sectoral distribution within a framework encompassing 200 sectors, the primary contributors identified are cement, lime, and plaster, basic iron and steel, and electricity by coal. This finding aligns with the previous research indicating that mineral production—especially cement—and metals such as steel are predominant factors in the construction footprint [59]. Upon consolidating these 200 sectors into broader macro-sectors, the critical role of the industry and upstream emissions related to energy becomes apparent, accounting for approximately 70% of the total embodied emissions. Additionally, the emissions from transport and within the construction sector emerge as notably significant at the national scale. From a geographical perspective, considering a division into 44 regions, the leading international sources of emissions are China, alongside the combined regions of Asia and Africa. When the data are further aggregated, it reveals that 42% of the footprint is located in France, with the European Union and the Rest of the World (RoW) responsible for 22% and 36% of the footprint, respectively.

3.1.3. Whole-Life GHGEs

After the aggregating operation and embodied GHGEs, the whole-life GHGEs of the French building stock emitted 162 MtCO₂eq in 2019. Operational emissions dominate the GHG footprint of buildings at the national scale. However, embodied GHGEs are already quite important, representing one third of the footprint. As a whole, 20% of the French building stock GHGEs are located outside the national borders.

3.2. Scenarios and WLC Budgets

3.2.1. Operation GHGE Budgets

Applying the SNBC pathways, the results show that operational GHGEs should reach 6.6 MtCO₂eq in order to be aligned with the SNBC, mainly represented by direct operational GHGEs with 5.7 MtCO₂eq. It represents a 94% decrease from 2019 to 2050.

3.2.2. Embodied GHGE Budgets

Embodied GHGEs are located in various geographical regions and sectors, as displayed in Figure 5. The IEA scenarios provide detailed pathways for these geographical regions and sectors but not the geographical region–sector pair (e.g., the Chinese cement sector). For this reason, the 2019 results are projected using the sectoral pathways in Figure 6 for the Stated Policies, Announced Pledges, and Net-Zero scenarios. When applying sectoral pathways, there is a high discrepancy between the results because the scenarios do not follow the same objective, as detailed in Section 2.3. In 2030, the embodied GHGEs would reach 60 MtCO₂eq under the STEPS, 52 MtCO₂eq under the APS, and 44 MtCO₂eq under the NZS. The difference is larger in 2050, with 56 MtCO₂eq under the STEPS, 22 MtCO₂eq under the APS, and 2.2 MtCO₂eq under the NZS.

The assignment of specific pathways to each sector is detailed in Appendix B. As they are applied to emissions from sectors within the construction supply chain, the overall embodied pathway (e.g., the ‘total’ black line in Figure 6) is not necessarily equal to the GHGE pathway of each IEA scenario for the whole economy. Indeed, the share of industry-related GHGEs is substantial in the embodied GHGEs of buildings, and such sectors typically have more modest decarbonization pathways. For instance, the pathways for ‘cement’ and ‘iron and steel’ are often less aggressive, mirroring the ‘hard to abate’ character of industrial sectors [60]. Conversely, the ‘electricity and heat’ sector always has steep and rapid decarbonization, even in the Stated Policies Scenario.

3.2.3. WLC Emissions Budgets

In order to provide WLC emission budgets, the SNBC scenario for operational GHGEs is combined with the different decarbonization pathways for embodied GHGEs, resulting in the creation of three combined scenarios, displayed in Figure 7.

In 2030, the WLC emissions would reach 124 MtCO₂eq under the SNBC-STEPS, 116 MtCO₂eq under the SNBC-APS and 108 MtCO₂eq under the SNBC-NZS. In 2050, the figures will decrease, respectively, to 62 MtCO₂eq, 28 MtCO₂eq, and 8 MtCO₂eq. Regarding the distribution of the emission scopes, in both the SNBC-STEPS and SNBC-APS scenarios, the embodied GHGEs are anticipated to approach operational levels by 2030 and reach, respectively, 90% and 78% of the WLC emissions by 2050. Due to the rapid decarbonization of all the economic sectors in the NZS, these proportions differ significantly, with embodied GHGEs accounting for 41% in 2030 and only 26% in 2050.

Considering that the Announced Pledges Scenario aligns with the policy targets of different countries, it may be the most logical scenario to couple with the SNBC, which represents France’s long-term commitment to reach a state of net-zero emissions.

4. Discussion

This study contributes to the body of work on the construction sector’s carbon footprint and global supply chain analysis, discussed in Section 1.2. However, it is important to acknowledge the absence of a formal validation process, which constitutes a key limitation of this study. The results align well with the findings from previous research, including [48], which highlights the predominant role of fossil fuel combustion in operational GHGEs [17], which emphasizes the significant role of indirect emissions within the construction sector [20], which reports that one-third of the whole-life carbon emissions in Ireland are embodied GHGEs, and [30], which projects that embodied emissions could constitute up to 75% of the life-cycle GHGEs in a Business-As-Usual scenario. However, this alignment with the existing studies does not replace the need for rigorous validation. The validation could involve comparing these results with those obtained from other methodologies, such as bottom-up approaches, although such studies are rare at the national level. While incorporating quantitative indicators for measuring errors, such as sensitivity analysis within the MRIO framework, could also be beneficial, this process is complex and resource-intensive as it involves extensive computational work to evaluate the impact of each variable on the results [61]. Additionally, the inherent challenges of sector aggregation, economic data discrepancies, and the assumptions about the global supply chains add further layers of complexity to such an analysis [62]. Given these challenges, the scope of the current study was focused on establishing a robust methodological framework rather than extending into comprehensive validation. Future research could build on this foundation by developing and applying advanced validation techniques to address these challenges and enhance the robustness of the findings.

Despite this limitation, the study’s findings remain valuable for informing policy decisions and advancing the understanding of whole-life carbon emissions in the construction sector by leveraging the strengths of various LCI methods to overcome the limitations inherent in each. It also addresses the noted scarcity of comprehensive macro-level studies, particularly those using MRIO databases, for capturing both direct and indirect impacts, a gap especially pronounced in EU countries [16], with no study specifically addressing France. This shortfall is particularly significant given France’s advanced policy framework, exemplified by the SNBC and the RE2020. Indeed, this study adopts a more policy-oriented perspective since its accounting methodology is designed to broaden the scope of the existing sectoral carbon budgets. In the majority of the studies that focus on establishing WLC budgets for national building activities (such as [32,33,63]), the approach leans towards absolute sustainability. This involves an initial step to downscale the remaining global carbon budgets to the national level using different allocation principles. In this study, carbon budgets are conceived more as a policy target, defined by the SNBC and complemented by consumption-based metrics. As such, this approach does not address the issue of fairness but can be viewed as more relevant for defining environmental policies [64]. Such an analysis can also provide guidance in the context of international climate negotiation by highlighting the necessity of pursuing emission reductions not only within the construction industry itself but also across its global supply chains, thereby identifying diverse reduction opportunities.

In the WLC pathways, the share of embodied GHGEs becomes more and more important and should outpace operational GHGEs in the next decades in the STEPS and APS. Indeed, when dealing with a large transformation of the building stock, either via retrofits or reconstruction, material-related impacts can become more important than energy-related impacts [65,66]. This finding is particularly true for France, where the indirect operational GHGEs are relatively low thanks to a low-carbon electricity mix. Worldwide, this assertion might not stand since indirect operational GHGEs represented 57% of buildings’ life-cycle emissions in 2019 [67]; thus, it is likely that operational GHGEs will remain predominant. However, the importance of tackling embodied GHGEs should be clearly acknowledged by building policies in order to avoid a displacement of the GHGEs from buildings to industrial sectors that produce construction materials and equipment. Using the suggested accounting framework sheds light on the emission reduction potentials across the supply chain and the interconnection between industrial sectors and building demand. The cement and concrete value chain serves as a prime example due to the significant role of the ‘cement, lime, and plaster’ sector in embodied GHGEs that illustrates the strong link between construction activities and these upstream industrial chains. Beyond the technical and upstream decarbonization levers outlined in Appendix C, the necessity of adopting a value chain reduction approach is underscored, proposing strategies from clinker production to structural applications in buildings [68]. This unified approach accentuates the intrinsic link between construction activities and the decarbonization of cement and concrete, highlighting the vast potential for emission reduction through demand-side measures. Construction sector professionals are encouraged to demand the creation of lower-CO₂ content in concrete, necessitating a concerted focus on the emissions during the construction phase that involves a broad spectrum of stakeholders. By mandating progressive reductions in embodied emissions with limit values set for 2031, the new environmental regulation (RE2020) in France is a good example of such policies, even if it applies to new construction only. In terms of the broader policy implications, the findings support the need for additional policies based on sufficiency measures, such as policies that encourage a reduction in the floor area per capita and the optimization of existing spaces by addressing vacant housing [69].

Previous studies have attempted to apply scenarios to IO models by changing various parts of the IO system based on exogenous assumptions [70,71,72]. They have often focused on the electricity sector [73,74]. In this study, the economic structure (e.g., inter-sectoral exchanges) is assumed to remain constant. Indeed, this study applies exogenous scenarios to IO results by using the IEA sectoral CO₂ pathways, similar to how operational results are combined with SNBC pathways. It is worth mentioning that IEA scenarios do not publicly provide prospective data on economic exchanges among sectors. Thus, if the approach is arguably less sophisticated, future studies would need to incorporate additional models to modify the economic structure. For the time being, the study is more concentrated on the ongoing work at the EU level focused on the creation of the WLC and embodied carbon budgets for buildings [31,37], and it represents a novel contribution by depicting budgets that depend on the decarbonization of the upstream sectors within the supply chain. However, this method would not be suitable to examine the effect of certain policies on the structural dependencies between economic sectors, such as circular economy measures or economic incentives, nor to address the effect on the prices in the economy. These aspects are generally the focus of dynamic models, such as Computable General Equilibrium (CGE) models, while IO models are more conventionally regarded as accounting tools [71].

Additionally, while IOA offers a cross-sectoral and international vision of embodied GHGEs, facilitating the minimization of truncation errors, it does not come without limitations [75]. In particular, the issues of sector aggregation and material-specific data are important when it comes to buildings [76]. For example, the construction sector is aggregated in MRIO tables and its decomposition between buildings and civil engineering GHGEs provides additional uncertainties. In this study, it is completed through national statistics by using economic transactions as a proxy. Matrix augmentation, which consists of subdividing an economic sector using process data, can be a more robust option [62]. It should be noted that this issue does not arise when the all-built environment is considered.

Another major issue of IOA regarding embodied GHGE calculation is that it does not distinguish the building use (e.g., residential and non-residential buildings) and typology (e.g., new construction, renovation, or demolition). However, the present work can represent budget-based targets and be coupled with an LCA bottom-up building stock model, which aims to provide a holistic environmental assessment of the different building stages [77]. It would enable the transformation of the building stock through the years with respect to the top-down targets identified. Indeed, the idea of combining the top-down and bottom-up modeling approaches for building activities is a promising idea [78]. Yet, one of the challenges to be addressed is the potential temporal mismatch between the static LCA results and budget-based targets.

5. Conclusions

The development of sound GHGE accounting methods for cross-sectoral activities is urgent to better link climate policies such as sectoral carbon budgets with industrial and public policies. They can enhance the development of scalable carbon budgets that are needed to better shape stakeholders’ decarbonization strategies. The article builds upon the needs–activities–sectors framework to present a methodology that enables capturing the whole-life GHGEs of buildings at the national level. It supplements the traditional sectoral approach of GHGE accounting by bringing a cross-sectoral and international perspective to national buildings’ activity. Alongside the GHGEs of the different energy carriers used during the buildings’ use stage, the geographical and sectoral embodied GHGEs’ hotspots are highlighted. This holistic approach better recognizes the complexity of buildings’ GHGEs and can help to activate the decarbonization levers along the supply chain. The French case study shows that operational GHGEs are dominant today, with 104 MtCO₂eq, representing 66% of buildings’ GHG footprint, and they are mainly caused by gas and oil combustion. Nevertheless, the weight of embodied GHGEs is already significant today and is likely to become predominant in the following decades in almost every studied scenario. Thus, the results show that strict limits on embodied GHGEs should be enforced to better regulate the whole-life GHGEs of buildings and avoid carbon leakages.