📄 earth_science

Who Generates the Carbon Emissions Embedded in Tourism Demand? Uncovering Hub Sectors and Strategic Roles in China’s Intersectoral Emission Network

This study integrates environmental input-output analysis with complex network theory to reveal that China's tourism-driven carbon emission network has evolved into a scale-free structure dominated by a few key hub sectors, suggesting that effective decarbonization policies must target these specific "receiver" and "carrier" sectors based on their distinct structural roles.

Original authors: Xichang Li, Zongpeng Tang, Keji Liu, Ye Zhan

Published 2026-06-30

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Xichang Li, Zongpeng Tang, Keji Liu, Ye Zhan

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

The Big Picture: Tourism is a "Shadow" Emitter

Imagine the tourism industry not as a single building, but as a giant, invisible spiderweb connecting to almost every other part of the economy. When you buy a plane ticket, stay in a hotel, or buy a souvenir, you aren't just paying for that one service. You are pulling on a string that vibrates the entire web.

This paper argues that while tourism itself doesn't burn a lot of coal directly (like a power plant does), it is a massive indirect polluter. Every time a tourist eats a meal, the restaurant needs electricity, the food needs to be grown and transported, and the plates need to be made. All those steps create carbon emissions "embedded" in the supply chain.

The authors wanted to map this invisible web to see: Who is actually generating the pollution that tourism demands?

The Tools: A Map and a Network

To solve this, the researchers used two main tools:

Environmental Input-Output Analysis (EIOA): Think of this as a giant financial ledger that tracks every dollar and every ton of carbon moving between industries. It traces the money (and the pollution) from the final customer all the way back to the raw materials.
Complex Network Theory: This is like looking at a social network (like Facebook or LinkedIn). Instead of people, the "nodes" are industries (like farming, manufacturing, or hotels), and the "lines" connecting them are the flow of goods and carbon.

The Journey: What They Found

The researchers looked at data from China spanning 2007 to 2023. Here is what the map revealed:

1. The Web is Changing Shape (From a "Small World" to a "Star")

In the past, the carbon network looked like a "small world" where everyone was connected to everyone else in just a few steps.

The Change: Over time, the web has become more like a star-shaped galaxy. Most connections now funnel through a few "super-hubs."
The Analogy: Imagine a town where everyone used to walk to the local store. Now, almost everyone drives to one massive, central mega-mall. If you want to stop the traffic, you don't tell everyone to stay home; you fix the roads leading to that one mega-mall.
The Result: A few heavy industries (like making steel, chemicals, and electricity) have become the "hubs." They are the main generators of the pollution that tourism eventually uses.

2. The Roles: Receivers vs. Carriers

The study sorted the industries into different roles based on how they handle carbon flows:

The Suppliers (The Upstream Generators): These are the heavy industries (mining, electricity, chemicals). They are at the top of the supply chain. They make the raw materials that everyone else needs. They are the "factories" of the pollution.
The Receivers (The Downstream Users): Most tourism sectors (hotels, restaurants, shopping, entertainment) fall here. They are at the bottom of the chain. They don't generate much pollution themselves; they receive the pollution embedded in the products they buy. They are the "customers" of the carbon.
The Carriers (The Middlemen): Some tourism sectors, specifically Transportation (air, road, rail, water), act as bridges. They take raw materials and move them to the final tourist. They are the "delivery trucks" of the network.

3. The "Shopping" Surprise

One interesting finding was about the Shopping sector. In 2007, it was a major "bridge" (a key connector) in the network. By 2023, its importance dropped significantly. It's no longer a central hub for moving carbon around; the network has reorganized around other sectors.

The Main Takeaway

The paper concludes that tourism is not the main factory of carbon; it is the main consumer of carbon.

If you want to reduce tourism's carbon footprint, you can't just tell hotels to turn off the lights or tourists to fly less. That's like trying to stop a flood by telling the people at the bottom of the hill to hold back the water.

Instead, you have to look at the hubs at the top of the hill (the heavy manufacturing and energy sectors). Because tourism is so deeply connected to these hubs, making those specific industries cleaner will automatically clean up the tourism supply chain.

Summary in One Sentence

Tourism is like a customer ordering a custom cake; the paper shows that to reduce the mess (carbon emissions), you don't just tell the customer to eat less cake, you have to help the baker (the heavy industries) use cleaner ingredients and ovens.

Technical Summary: Who Generates the Carbon Emissions Embedded in Tourism Demand?

Problem Statement
While tourism is recognized as a significant driver of anthropogenic greenhouse gas emissions (accounting for an estimated 5%–8% of global emissions), existing research has largely focused on accounting for direct emissions or total embodied emissions within supply chains. A critical gap remains in understanding the structural relationships and network mechanisms between tourism-related sectors and the broader economy. Previous studies often fail to clarify the specific positions and roles of tourism sectors within the inter-sectoral carbon emission flow network (CEFN), leading to a lack of targeted policy interventions. Furthermore, traditional analytical methods often suffer from "double counting" when tracing total emission flows, obscuring the true topological structure of emission linkages.

Methodology
This study proposes a novel hybrid analytical framework integrating Environmental Input-Output Analysis (EIOA) with Complex Network Theory to trace tourism-driven inter-sectoral carbon flows in China.

Data Sources: The analysis utilizes China's Input-Output Tables for 2007, 2012, 2017, and 2023. To ensure consistency across years with varying sector classifications (ranging from 135 to 211 sectors), all sectors were re-aggregated into 51 consistent sectors. These were categorized into 18 tourism-related sectors (based on the Statistical Classification of National Tourism and Related Industries (2018)) and 33 non-tourism sectors.
Carbon Accounting: The study employs a consumption-based accounting approach. Direct carbon emission intensities were estimated using energy balance sheets and IPCC emission factors.
Network Construction: A critical methodological distinction is the construction of the network based on first-layer emission flow relationships rather than total cumulative flows. The authors argue that using total flows (Leontief inverse) leads to double counting in network analysis. Instead, they utilize the formula $SC = \hat{c} \cdot A \cdot \hat{f}$ (where $\hat{c}$ is direct emission intensity, $A$ is the direct input coefficient matrix, and $\hat{f}$ is final demand) to define the weighted intensity of emission flows between sectors.
Network Metrics: The constructed CEFN (51 nodes) is analyzed using:
- Small-world properties: Average path length and clustering coefficient.
- Centrality: Degree centrality (in-degree and out-degree) and betweenness centrality to identify hubs and intermediaries.
- Block Modeling: The CONCOR (CONvergence of iterated CORrelations) algorithm is used to cluster sectors into blocks based on structural equivalence, identifying roles such as "receivers," "suppliers," and "carriers."

Key Results

Network Evolution: The CEFN has gradually lost its "small-world" properties (characterized by short path lengths and high clustering) between 2007 and 2023. It has evolved into a pronounced scale-free network, where the distribution of connections is highly uneven. Most linkages are dominated by a few highly connected hubs, while the majority of nodes have few connections.
Sectoral Roles: Tourism-related sectors exhibit heterogeneous roles within the CEFN:
- Receivers: The majority of tourism-related sectors (e.g., Accommodation, Shopping, Cultural Activities, Financial Services) fall into "Block I" and act as receivers. They are located downstream in the supply chain, consuming intermediate inputs from other sectors rather than supplying them.
- Carriers: A subset of tourism sectors, primarily transportation (Air, Road, Water, Rail) and Food/Beverage services, act as carriers (Block III). They function as intermediate links, receiving emissions from upstream sectors and transmitting them to downstream sectors.
- Suppliers: Tourism sectors generally do not act as primary suppliers of carbon emissions to the network; this role is dominated by heavy manufacturing and energy sectors (Block IV).
Centrality Findings: Manufacturing sectors (e.g., Chemical Raw Materials, Non-metallic Minerals, Electricity) dominate the network in terms of out-degree and betweenness centrality, acting as the core sources and bottlenecks of emission flows. Among tourism sectors, transportation modes (T1–T4) and Accommodation (T5) show higher centrality, indicating their strategic importance in emission transmission.
Intensity Trends: The weighted intensity of carbon emission flows has shown a declining trajectory across the economy, driven by efficiency improvements. However, the structural evolution toward a scale-free network suggests that emissions are becoming increasingly concentrated in specific key sectors.

Significance and Contributions
The paper claims two primary contributions:

Methodological Innovation: By combining EIOA with complex network theory and specifically utilizing first-layer emission flows to avoid double counting, the study provides a more accurate structural map of inter-sectoral carbon relationships. This allows for the identification of "hub" sectors and the topological evolution of the network, which previous total-flow models obscured.
Policy Relevance: The study demonstrates that tourism is not a monolithic generator of emissions but a complex inducer of demand-driven indirect emissions. The findings suggest that effective decarbonization policies cannot be uniform across the tourism sector. Instead, policies must be tailored to the specific structural roles of tourism sub-sectors:
- For receivers, policies should focus on managing demand and collaborating with upstream heavy industry sectors.
- For carriers (like transport), decarbonization is critical as these sectors act as intermediaries; their reduction efforts can yield synergistic emission reductions across the entire supply chain.

Limitations
The authors acknowledge limitations regarding data timeliness, as the analysis relies on input-output tables released every five years (latest being 2023), which may not capture the most recent dynamic shifts. Additionally, the study focuses on national-level aggregates and does not yet account for regional heterogeneity within China, which may yield different network characteristics in different provinces.