Marine Pollution Bulletin 126 (2018) 428–435

Contents lists available at ScienceDirect

Marine Pollution Bulletin

journal homepage: www.elsevier.com/locate/marpolbul

Decarbonizing the international shipping industry: Solutions and policy T

recommendations
⁎
Zheng Wana,b, , Abdel el Makhloufic, Yang Chena, Jiayuan Tangd
a
College of Transport and Communications, Shanghai Maritime University, Shanghai 201306, China
b
Institute of Transportation Studies, University of California Davis, Davis, CA 95616, USA
c
Amsterdam University of Applied Sciences, 1097, DZ, Amsterdam, The Netherlands
d
Cass Business School, 106 Bunhill Row, London EC1Y 8TZ, United Kingdom

A R T I C L E I N F O A B S T R A C T

Keywords: Ship-source greenhouse gas (GHG) emissions could increase by up to 250% by 2050 from their 2012 levels,
International shipping owing to increasing global freight volumes. Binding international legal agreements to regulate GHGs, however,
Greenhouse gas are lacking as technical solutions remain expensive, and crucial industrial support is absent. In 2003, the
International maritime organization International Maritime Organization adopted Resolution A.963 (23) to regulate shipping CO2 emissions via
Policy
technical, operational, and market-based routes. However, progress has been slow and uncertain; there is no
Emissions trading scheme
concrete emission reduction target or definitive action plan. Yet, a full-fledged roadmap may not even emerge
until 2023. In this policy analysis, we revisit the progress of technical, operational, and market-based routes and
the associated controversies. We argue that 1) a performance-based index, though good-intentioned, has loop-
holes affecting meaningful CO2 emission reductions driven by technical advancements; 2) using slow steaming to
cut energy consumption stands out among all operational solutions thanks to its immediate and obvious results,
but with the already slow speed in practice, this single source has limited emission reduction potential; 3)
without a technology-savvy shipping industry, a market-based approach is essentially needed to address the
environmental impact. To give shipping a 50:50 chance for contributing fairly and proportionately to keep
global warming below 2 °C, deep emission reductions should occur soon.

1. Introduction determine measures to facilitate their practical implementation.

To satisfy the lofty goal of the Paris Agreement to limit global
Ocean shipping, the most energy-efficient form of freight transport, warming below 1.5 °C–2 °C, all sectors may be ultimately required to
is the backbone of global trade, but this sector heavily depends on fossil produce zero emissions or develop tools to remove GHGs from the at-
fuel. The lengthy debate on whether ship-source greenhouse gas (GHG) mosphere (Williamson, 2016). Regarding shipping emissions, large
emissions are classified as marine pollution has delayed the interna- shipping nations and the shipping industry are slow and sometimes
tional regulation and subsequent implementation to limit the carbon reluctant to introduce appropriate measures aimed at reducing emis-
emissions from the shipping sector (Shi, 2016a). sions and improve global rules for the industry (Upton, 2016).
Ship-source GHG emissions could increase by up to 250% by 2050 The IMO placed the climate impact of shipping in the agenda in
from 2012 levels, owing to increasing global freight volumes (Fig. 1). 2003. On December 5 of the same year, the IMO adopted Resolution
Unchecked, such emission levels are projected to constitute 17% of the A.963 (23) requiring the Marine Environment Protection Committee
global CO2 emissions by 2050 from the current figure of approximately (MPEC) to regulate shipping CO2 emissions through technical, opera-
2% (Cames et al., 2015). Yet, at the Paris Climate Agreement of 2015, tional, and market-based routes (IMO, 2004). However, stakeholder
the shipping industry was neither included in the global emissions re- values are diverse, leading to slow negotiations and no concrete emis-
duction targets nor mentioned in the agreement (United Nations sion reduction pathways or definitive action plan. A full-fledged
Framework Convention on Climate Change, 2015). Discussions re- roadmap may not even emerge until 2023 (Green4sea, 2016). Many
garding shipping emissions were simply left, like in the Kyoto agree- widely discussed market tools, such as the cap and trade approach,
ment, to the International Maritime Organization (IMO), who is ex- implemented in other sectors are unlikely to be implemented in the
pected to develop regulations, set emission reduction targets, and shipping business soon because they are linked to a fuel data collection

⁎
Corresponding author at: College of Transport and Communications, Shanghai Maritime University, Shanghai 201306, China.
E-mail address: mrwan@ucdavis.edu (Z. Wan).

https://doi.org/10.1016/j.marpolbul.2017.11.064
Received 9 September 2017; Received in revised form 25 November 2017; Accepted 27 November 2017
Available online 22 December 2017
0025-326X/ © 2017 Elsevier Ltd. All rights reserved.
Z. Wan et al. Marine Pollution Bulletin 126 (2018) 428–435

Fig. 1. CO2 emissions in the shipping sector

Illustration based on Buhaug et al. (2009), Smith et al.
(2014), Anderson and Bows (2012), Lloyd's list of in-
telligence data, and OECD analysis. If shipping aims at
delivering its fair and proportionate contribution at 50:50
chance to keep global warming well below 2 °C, a deep
emission reduction up to 85% by 2050 compared with the
2010 baseline is needed. Existing technology and opera-
tional solutions are not enough.

system that will start only in 2019 (Dufour, 2016). 2. Technical and operational solutions
In this policy analysis, we revisit the progress of technical, opera-
tional, and market-based instruments and their associated con- 2.1. Technical solutions
troversies. Based on existing evidence, we argue the following.
Technical solutions aim at using technical means to improve a ship's
1) Many technical solutions remain too expensive, and crucial in- energy efficiency, thereby reducing the CO2 impact per capacity mile
dustrial support is absent. A performance-based index, though good- (expressed in ton-mile) (IMO, 2012). The IMO has introduced the
intentioned, has loopholes leading to concerns on the meaningful mandatory Energy Efficiency Design Index (EEDI) for newly built ships
effect on CO2 emission reductions. with the hope that such a measure will stimulate a series of technology
2) Using slow steaming to cut energy consumption is apparent among and engineering innovations ranging from optimized hulls and pro-
other operational solutions due to its immediate and obvious results, pellers and improved engine performance to better waste heat recovery
but with the already slow speed in practice, further emission re- systems (MEPC, 2011). Most newly built ships only need to become
duction potential from this single source may be limited. On the 0%–10% (the actual number depends on the vessel type and size) more
other hand, ships could still adopt faster speed when the market and energy efficient between 2015 and 2020, but the index will be tigh-
economic circumstances improve. A potential impact will be exerted tened incrementally every five years.
on fuel use and associated emissions correspondingly. Other types of A performance-based index, though good-intentioned, has loopholes
operational solutions must be incorporated into shipping companies' that led to concerns on the meaningful effect on the CO2 emission re-
energy management strategies for extra reduction potential. ductions driven by technical advancing. In theory, the use of derated
3) Without a technology-savvy shipping industry, a market-based ap- engines with less power can yield significant EEDI reductions at the
proach is essentially needed to address the environmental impact. expense of speed without extra technology improvements (Psaraftis and
The core of the maritime emissions trading system (ETS) system lies Kontovas, 2013). An empirical study found that regulations on EEDI
in the “carbon emissions ceiling” and “trading process.” However, would even result in slight increases in CO2 emissions in large crude
the method of using benchmarking plus grandfathering rights carriers because by limiting the installed power on board, vessels would
adopted by the European Union (EU) aviation industry cannot be induced to operate on higher revolutions-per-minute engines that
simply be applied to the more complex international shipping in- consume more fuel though the EEDI limit is met (Devanney, 2011). In
dustry. practice, EEDI only reflects the efficiency of ship design but totally
4) If shipping has a 50% probability of delivering its fair and propor- neglects the operational variations that determine the real energy ef-
tionate contribution to keep global warming well below 2 °C, a deep ficiency (Cichowicz et al., 2015).
emission reduction should take place soon. Constructive regional Moreover, increasing competition has led to ever increasingly larger
actions must be recognized as they can be both responsive and cost ships with better fuel economy, which translates into a smaller EEDI
effective. (Ozaki et al., 2010), but uncertain demand could complicate the real
performance—ships could consume more energy per goods transported
if half loaded than fully loaded (Wan et al., 2016a). Increasingly larger
ships can yield other unexpected side impacts from concentrated

429
Z. Wan et al. Marine Pollution Bulletin 126 (2018) 428–435

Fig. 2. Potential CO2 reduction potential from an array of technical

and operational solutions
Extracted data (Bouman et al., 2017) from 60 quantitative studies
on the CO2 emissions reduction potential based on the array of
technical and operational solutions listed in the figure. Owing to
data scarcity, it is unlikely to make a meaningful meta-analysis to
reveal statistical inference. The solid bar denotes the whole range of
potential CO2 reduction, but the widest range (i.e., the highest and
lowest reported reduction potential) in each solution suggests poor
agreement in the literature on the reduction potential. Such a high
degree of uncertainty is due to 1) some studies being limited to a
certain type of vessel, meaning that the conclusions cannot be ap-
plied to the entire industry and 2) some studies being based on
different model assumptions. Applying all these types of technical
and operational solutions could yield a high CO2 reduction poten-
tial of 78% or a much lower CO2 reduction potential of 29%. Such a
high degree of performance uncertainty combined with costly in-
vestments lead to lethargic industry action.

pollution at hub ports, such as Hong Kong and Shanghai, to under- than 400 GT for international voyage. Operational solutions are typi-
regulated ship scrapping of smaller vessels in developing countries, cally linked to the shipping companies' energy management strategies,
such as Bangladesh and Pakistan. This may lead to public health dis- including slow steaming, optimized ship trim, enhanced network
asters and biodiversity crises owing to the heavily polluting ship- routing, hull cleaning, and engine maintenance. In general, these
breaking process. measures require low operating costs and do not involve hefty initial
An array of technical solutions to serve as alternative power have investment but can achieve promising energy savings.
emerged, including fuel cells, waste heat recovery, solar or wind power, An extensive review shows that slow steaming is the most sig-
and shore-to-ship power (see Fig. 2 for detailed information and com- nificant reduction strategy among all major optimization initiatives
ments); however, they struggle to steer the shipping industry toward a (Armstrong, 2013). Fuel consumption is directly linked with engine
low‑carbon direction because they require not only engineering power impacted by operational speed. A medium-sized containership
breakthroughs in an economically feasible way but also rapid and that reduces speed by 30% could use 55% less fuel (Cariou, 2011).
transformative adoption by an industry that is always slow to react. As Thanks to the widespread adoption of slow steaming in response to
shown in Fig. 3, many technical solutions carry a hefty price tag—the industrial overcapacity after the financial crisis, CO2 emissions from
average CO2 reduction cost ranges between 50 USD/ton and 200 USD/ international shipping fell by around 10% during 2007–2012 (Smith
ton (Eide et al., 2011), which is far more expensive than the emission- et al., 2014). However, slower speed will inevitably increase transit
trading price of $5–$15 per ton in the United States. time and other operating costs (e.g., on-board labor costs) and reduce
Technical solutions also include capital-intensive options to power on-time performance possibly interrupting logistics reliability. Given
vessels with cleaner fuels or even nonfossil fuels, but so far the re- that many oceangoing vessels already operate at the speed of
sponses are polarized. A theoretically sound technology to reduce CO2 15–18 knots, further fuel savings and emission reduction potential from
may be difficult to engineer and economically unfeasible in real-world slow steaming alone may be limited (Woo and Moon, 2014). Even in-
settings. For example, converting a vessel to be propelled by liquefied corporating slow steaming into future forecasts, only two scenarios out
natural gas (LNG) can significantly reduce airborne pollutants and GHG of sixteen have CO2 emissions declining to nearly 2012 levels by 2050,
emissions, but this requires millions of U.S. dollars in investment, sa- with the rest continuing to grow (Bows-Larkin et al., 2015). Another
crificing precious on-board storage space (Verbeek et al., 2011). Such concern is that when the transport market reverts to booming, shipping
converted ships also rely on a supporting LNG-charging infrastructure companies tend to revert to faster steaming speeds to improve vessel
network that barely exists today. utilization, posing a doubt whether this option can last long.
Other operational solutions have received less industrial attention
2.2. Operational solutions because they often require a skilled crew to respond quickly to a rapidly
changing, and often sophisticated, environment rather than a conven-
Operational solutions are usually implemented under the Ship tional “follow the protocol” approach on board. For example, opti-
Energy Efficiency Management Plan (SEEMP) targeting vessels greater mizing the trim of the ship can lead to 0.5% and 2.3% fuel savings

430
Z. Wan et al. Marine Pollution Bulletin 126 (2018) 428–435

Fig. 3. Average marginal CO2 reduction cost per option

Figure adapted from the study by Eide et al. (2011). Cost-effective solutions are mainly operational ones, such as slow steaming and weather routing. In contrast, many technical solutions
carry a hefty price tag that offers no incentive to the industry to invest—the average CO2 reduction cost is 50–200 USD/ton, which is far more expensive than the emission-trading price of
$5–15/ton in the United States. Carbon trading is favored by many economists and industry because it will lower the compliance cost while meeting the emission reduction target.

(Coraddu et al., 2018); however, this has to be done using complicated apply a “three-step approach.” (Report of the Marine Environment
data analysis techniques. Not done properly, it will endanger naviga- Protection Committee on Its Sixty-Eighth Session, 2015) This approach
tional safety. comprises 1) collection of global data for the fuel consumption of ships,
2) data analysis, and 3) decision making on possible resources. The
shipping industry expects that it will take many years before the first
3. Market-based solutions collected global data will be analyzed. Measures such as carbon pricing
will take a long time before implementation owing to the three-step
3.1. The surging of market-based solutions approach agreement, e.g., no decision making can take place before
data collection and analysis.
Indeed, if it works as planned, the combination of EEDI and SEEMP The MEPC agreed to propose an initial GHG reduction roadmap
can substantially reduce business-as-usual CO2 emissions; however, covering the 2017–2023 period, and a sketchy mitigation strategy is
increasing trade volume on a global scale would still bring total emis- expected to be announced in 2018 (Green4sea, 2016). However, as
sions in 2050 at twice the 2007 level (Bazari and Longva, 2011). If mentioned earlier, many widely discussed market tools that have been
shipping aims at delivering its fair and proportionate contribution at a implemented in other sectors are unlikely to be applied to the shipping
50:50 chance to keep global warming well below 2 °C, a deep emission business soon because they are linked to an IMO-led fuel data collection
reduction up to 85% by 2050 compared with the 2010 baseline is system that will start only in 2019. Industry analysts believe that a full-
needed (Anderson and Bows, 2012). We believe that technical and fledged plan may only emerge in 2023 with actual implementation
operational measures alone will not suffice. schedules many years later (Merk, 2017). Having said that, it could take
Increasing calls for market-based measures have been emerging another six years, with a list with recommended options, and manda-
(Shi, 2016b). The IMO received several proposals from its member tory requirements could take another ten years to come into effect.
states (mostly advanced industrialized nations) on possible emission-
trading systems to tackle CO2. Representative proposals include MEPC/
60/4/22 by Norway, MEPC/60/4/26 by the United Kingdom, MEPC/ 3.2. EU's pioneering response
60/4/41 by France, and MEPC 60/4/54 by Germany (their features will
be mentioned later). At the MEPC 63rd session, measures to reduce As early as 2004, the EU Commission Decision 2004/156 was
GHG emissions by market mechanisms have received widespread at- adopted and guidelines for monitoring and reporting CO2 emissions
tention and were discussed; however, opinions differed widely between were established during the first phase of the implementation of the
developed and developing countries, with the latter worrying about the European Union Emission Trading Scheme—the world's first and largest
unknown economic impact and the ripple effect on the export sector. international carbon emission-trading market (Ellerman and Buchner,
Some member states emphasized that “the necessary financial, tech- 2007). However, the shipping industry was not included in these
nological and capacity-building support for developing countries by guidelines because the shipping community believes that the IMO is the
developed countries,” “the principles of common but differentiated recognized international entity to act. Frustrated at the slow progress in
responsibilities” must be incorporated into a future resolution (MPEC, achieving international commitment, the EU was determined to act on
2012). its own first. In 2015, the EU adopted the MRV system for the shipping
In subsequent MEPC sessions, the IMO members finally agreed to industry; it stipulated that all vessels over 5000 gross tonnage calling

431
Z. Wan et al. Marine Pollution Bulletin 126 (2018) 428–435

EU ports should monitor, report, and verify the CO2 emissions from high and made reverse incentives to enterprises that took pioneering
2018 on per voyage basis (Verifavia-shipping, 2015). action to reduce emissions. In addition, unlike land assets, ships are
In a recent update, the EU's environment committee issued a mobile around the world, which will affect the applicable area of
statement on December 16, 2016, vowing to bring shipping into the quotas. For example, Region A integrates the shipping industry into its
EU–ETS system from 2023 unless the IMO has adopted a comparable local ETS. A ship carried out transport activities in Region A, and its
scheme by 2021. European parliament adopted the proposal with a few route was changed into Region B without ETS. Then, according to the
rhetoric modifications (Spurrier, 2017). Ships that may call at the EU historical emissions method, this particular ship can still receive free or
ports are now officially counting down to embrace market instruments discounted carbon quotas from Region A in the next trading cycle,
aiming to curb the intensive emission growth. which could affect the carbon emission-trading market of Region A.
Constructive regional actions such as EU's MRV system, a pre- Some currently notable carbon trading markets include the EU's
requisite for including shipping into the current carbon emissions ETS, the California and Quebec shared carbon market, and China's pilot
trading market, play a vital and catalytic role in promoting global ETS (Xiong et al., 2017). Other nations have stated that they are
policy action. In fact, the EU's strategy echoes many proposals (e.g., planning or considering carbon pricing under their own jurisdiction but
Norway, the United Kingdom, France, and Germany) submitted to the none includes shipping. The environmental committee of the European
IMO. They share the following highlights. Parliament has proposed to include shipping in the EU's ETS that would
apply to all ships calling at the EU ports irrespective of origin. However,
(1) Under the total amount control strategy, set a carbon emissions other nations following the EU's proposal would induce economical
ceiling for a period of time. complications as well as political resistance because all ports within
(2) “Ships” are objects subject to regulation. Ship operators, flag states, different jurisdictions can demand carbon tax even for a single voyage.
and port states should jointly validate or supervise the real energy Such a proposal is doomed to be unwelcome, as evidenced by the
performance of ships. widespread boycott of the EU's aviation carbon tax scheme (Liang and
(3) During the initial implementation phase, relevant parties in the Zhang, 2014). We believe that integrating the fragmented carbon
carbon emission-trading market can freely gain or purchase quotas trading systems, at least for some global business operating with great
in the primary market corresponding to their bunker consumption mobility (in our case, the shipping industry), should be carefully eval-
share. They can then trade the quota in the secondary market or uated to close the loopholes and minimize business complications.
access emission allowances from other sectors to offset carbon
emissions. Ships must ensure that their possessed emission quotas 4. Policy implications
can cover the actual emissions during the period; otherwise, they
will incur a penalty. 4.1. Limitations of technical solutions
(4) The auction of these emission allowances shall be organized by a
recognized national or international entity. EEDI has pros and cons and is a flawed proxy to be relied upon to
(5) The revenues generated by the initial auction of the emission al- regulate emissions reduction. The focus should be on real-world op-
lowances are used for climate change funds to support mitigation eration performance rather than this theoretical number for char-
and adaption efforts. acterizing ship design. Decreasing the EEDI number does not auto-
matically translate into equally lower energy consumption in real-world
3.3. Maritime ETS system operations. In certain circumstances, the pursuit of lower EEDI number
will sacrifice navigational safety under extreme weather and sea con-
In our view, the core of the maritime ETS system lies in the “carbon ditions due to the derated engine power.
emissions ceiling” and “trading process.” Setting a reduction target is Technological advances have long been hailed as promising solu-
vital to the success of the scheme. We argue that the uncertainty of tions to fight climate change but do not count on them to transform the
emission allowances trading under a floating limit is less than that shipping emissions in the near future. The early adoption phase in the
under the fixed one (see Appendix for proof). Setting a reasonable shipping industry can be very expensive and requires significant pre-
floating limit on carbon emissions can encompass the uncertainties in mium from early adopters with unguaranteed investment returns. The
the future maritime trade volumes, thereby reducing the impact from cost would eventually relay to shippers, and consequently, clients will
dramatic emission quota price volatility. shift to other low-cost carriers. Subsidies or tax rebates from govern-
Notably, building consensus over the allocation method of the ments could help, but they can create allocative inefficiencies during
shipping carbon emissions is challenging. The method of using bench- the process and impose huge financial burden on governments (Chen
marking plus grandfathering rights (González, 2006). adopted by the et al., 2017). Some technological advancements could alter the ship
EU aviation industry cannot be applied to the more complex interna- design and reconfigure the ship-building process, but duplicating lab
tional shipping industry. Cargo ships have a high degree of customi- results in massive ship-building projects will require excessive patience
zation; various types and sizes could yield a combination of hundreds of and risk taking that most ship builders would not want to exercise ra-
emission levels per unit good transported. If a benchmarking method pidly; they would rather adopt a “wait-and-see” approach until a
similar to that used in aviation is adopted, the emission levels of each technology pathway is confirmed.
shipping company will then simply be decided by multiplying a fixed In many cases, economic sustainability is achieved at the cost of
emission performance indicator, freight turnover, and length of voyage environmental sustainability. For example, economic gains lead to en-
(see https://ec.europa.eu/clima/policies/ets/allowances/aviation_en vironmental damage and societal effects at city and regional levels. Ship
for detailed explanations). This is problematic. According to the demolition markets are a good example that shows the difficulty in
IMO's third GHG research report (Smith et al., 2014), dry bulk carriers balancing economic benefits (such as employment, incomes, and taxa-
emit 8.4 g carbon dioxide per km ton, whereas refrigerated ships emit tion) and environmental and societal effects (such as pollution, con-
80.4 g per km ton, a difference of almost 10 times. Developing the tamination of water resources, and health risks). New vessels are built,
quota system without distinguishing the ship types and sizes is defi- and the old ones are taken apart in areas that are ecologically vulner-
nitely unfair. able (Choi et al., 2016). Therefore, regulatory actions should be de-
Grandfathering means that shipping companies with high historical signed appropriately on the basis of full benefit and cost analyses to
emissions will have more free or discounted quotas for present use, understand the best policy strategy. It is necessary to balance the in-
which is contrary to the common but differentiated principle. Such a terests and ensure that mitigation strategies adopted by the shipping
quota system would reward companies whose prior emissions were industry are sustainable, not at the expense of other, particularly less-

432
Z. Wan et al. Marine Pollution Bulletin 126 (2018) 428–435

developed, economies.

4.2. Expansion of operational solutions

Using slow steaming to cut energy consumption stands out among

all operational solutions due to its immediate and obvious results;
however, with the already slow speed in practice, further emission re-
duction potential from this single source is limited. Other types of op-
erational solutions should not be neglected as they can provide op-
portunities for extra reduction at a low cost.
Operational solutions such as ship trimming require sophisticated
data-based optimization. Casual relationships between each type of
operation and subsequent energy performance need to be carefully
evaluated using real-time data analytics. While sensor technologies
generate vast quantities of information, they are usually difficult for
researchers outside the company to access for fear that commercial
confidentiality is compromised. It is possible to develop a protocol for
pooling and utilizing vessel performance data with the research com-
munity without compromising business secrets. Doing so would open
up more fuel saving potentials with additional analysis that can yield
important operational insights (Poulsen and Johnson, 2016). A good
practice in this respect is the US-based smart manufacturing leadership
coalition and Germany's platform industries 4.0 initiative for re-
searchers to overcome barriers to discovery and access and use factory-
level data to make manufacturing processes more energy efficient,
profitable, and sustainable (Kusiak, 2017).
Fig. 4. Top 15 ship-owning countries and their flag configuration
Data from the UNCTAD (2016) review of maritime transport. The 2016 UN statistics
4.3. Introduction of maritime carbon trading suggested that the top 15 ship-owning countries register the majority of their capacity
under foreign flag, which often lack incentive to embrace environmental conventions. It
Without a technology-savvy shipping industry, a market-based ap- could take a number of years for complex negotiations among nations to develop a new
proach is essentially needed to address the environmental impact. Some convention, and it will have taken many more years for the convention to enter into force.
regional carbon trading markets are functioning well and are evolving
to meet targets (Keohane and Morehouse, 2017). Including shipping in down approach) to regulate seaborne pollution, the IMO should re-
the scheme is necessary. cognize and encourage constructive regional actions (patchwork ap-
Carbon trading is favored by many economists and industries be- proach) to address GHG mitigation issues. Regional actions should not
cause it lowers the compliance cost while meeting the emission re- be equated to illegitimate unilateralism; rather, they can play a vital
duction targets. Much can be learned from decarbonizing the auto- catalytic role in promoting global policy action (Shaffer and Bodansky,
mobile industry in California, with an array of policy and market tools 2012). Appropriately designed and administered regional actions—for
to address climate change. Based on the Air Resources Board cost and example, the EU and California's climate policy for transportation—can
benefits analysis (California Air Resources Board, 2015), $83.4 million be both responsive and cost effective and of potentially great value to
rebates for “clean vehicles” (such as electric vehicles) resulted in 2.2 the world by more directly engaging various stakeholders and raising
million tons of carbon reduction for a cost per ton over $37 (three times environmental awareness (Sperling and Eggert, 2014). Patchwork ap-
more costly than the carbon trading). Investment in a hybrid and zero- proaches can sometimes be superior because they advance the regions
emission truck and bus program is even worse, with a cost per ton of beyond the current status quo of policy inaction. More importantly,
$139 (11 times most costly than the carbon trade). Emission reductions they may allow the rest of the world to explore the dynamic process of
under technology advancement in the transport sector alone are costing policy action and industrial reaction; therefore, future large-scale policy
3–11 times more than those under carbon trading, yet electric vehicles diffusion could be more data-driven and evidence-based.
are only able to capture approximately 2% of the market share in Ca- The need for global cooperation to tackle shipping emissions is
lifornia—the largest “clean vehicles” sales market in the United States. difficult because of the diversity in institutional framework and ar-
Carbon trading has been applauded as the best approach to achieving rangements for environmental and marine pollution that involve many
cost-effective CO2 reductions (Stavins and Schatzki, 2014). Adding a different agendas, strategies, ambitions, and goals (industry, states,
high carbon tax (15USD/ton) for the shipping industry will only raise nonstates, and global and regional organizations) (Roe, 2012). In this
the commodity price by less than 1% (MEPC, 2010) —unlikely impact respect, improving multilateral cooperation and technical assistance
the global trade given that all other transportation modes are subjected among countries is crucial. This is because many developing countries
to additional regulations. lack the (economic and organizational) resources and technical capa-
city to fully participate in international conventions and implement the
4.4. Regional actions and global cooperation obligations and regulations effectively.

Historically, there is a big gap between pledges and actions in the

shipping community regarding environmental regulations (Tan, 2005). 5. Conclusions
The endeavor to tackle a single maritime pollution could take dozens of
years from initial discussion to action because the agenda of the IMO Global antimarine pollution efforts led by the IMO emerged in the
conventions is largely impacted by the regions where the ships are re- 1970s, and some vessel-source pollutants have been gradually regulated
gistered—typically with lax environmental and financial regulations since the 1980s. Binding international legal agreements to regulate
(Wan et al., 2016b) (Fig. 4). GHGs, however, are evolving slowly, as technical solutions remain
Rather than solely relying on universal or majority consensus (top- expensive, and crucial industrial support is absent.

433
Z. Wan et al. Marine Pollution Bulletin 126 (2018) 428–435

Though being the most energy-efficient form of freight transport, Synergies, led by the top maritime commander, the IMO, must speed up
the shipping industry is not immune from the global efforts to dec- cooperation in the global shipping community, linking scientists, en-
arbonize economic growth. A pathway goal for the shipping industry is gineers, businesses, and policymakers to honor the centuries-old in-
prominently needed to ensure projected reductions. A laissez-faire ap- dustry in the “spirit of Paris.”
proach or slow mitigation efforts will have long-lasting transboundary
effects. Overcoming the barriers of mitigation currently in place Acknowledgement
throughout the shipping industry should be of high priority. Current
and possible future mitigation strategies commendably include trans- We would like to thank the reviewers and editor for their insightful
disciplinary solutions covering technical, operational, and market as- comments to improve this article. We also thank Dr. Mo Zhu for her
pects; however, widespread application requires them to be evidence- careful fact checking. Z.W. is supported by National Natural Science
based and economically viable. Their success also hinges on the con- Foundation of China (71704103 and 71402096), Y.C. is supported by
tinuous assessment and refinement in real-world applications. National Natural Science Foundation of China (71402093).

Appendix A

The cointegration analysis by Xu et al. (2014) suggests that the world's maritime freight turnovers are positively linked to the global shipping CO2
emissions. Specifically, when the world's maritime transaction volume changes by 1%, the international shipping carbon emissions will corre-
spondingly change by 0.85%. Inspired by Jotzo and Pezzey (2007), it can be proven that a floating limit on the total amount of CO2 emissions has
fewer uncertainties compared with a fixed quota system. Setting a reasonable floating limit on the total amount of CO2 emissions considers the
uncertainties in future maritime trade volumes, thereby reducing the uncertainty of carbon trading and making the target reflect actual conditions
better. Consequently, the supply of carbon quotas can be adjusted according to their demands, and carbon prices can be stabilized.
First, assume that when there is no restrictive policy to control the CO2 emissions, the fluctuations in maritime trade correspond to fluctuations in
carbon emissions
∼
EC = EC (1 + εY) (1)
∼
where EC is the observed value of CO2 emissions when there is no restrictive policy; EC is the expected value of CO2 emission when there is no
restrictive policy, and εYis the prediction error of maritime freight turnover.
Assume that εY represents random error and obeys the normal distribution. The above terms are independent of each other and we get
E[εY] = 0 (2)

E[εY 2] = σY 2 (3)
The upper emissions limit is set based on the amount of carbon emissions generated when no restrictive policy is implemented, specifically shown
in the following function:
∼
X = μEC (1 + βεY) (4)

X
μ=
EC (5)
∼
where X is the observed upper limit of floating CO2 emissions; X is the expected upper limit of floating CO2 emissions; μ is the ratio of the upper
limit of carbon emissions under the business-as-usual situations (i.e., without restrictive policy). The stricter the policy, the smaller is μ and the less
are the total allowable emissions (0 ≤ μ≤ 1). β is the floating degree of the upper limit of CO2 emissions (0 ≤ β ≤1).
The correlation between the upper limit of CO2 emissions and freight turnover depends on β, particularly when β = 0,the upper limit of CO2
emissions, is fixed and has nothing to do with the freight turnover, so that
X 0 = μEC (6)
where X0 is the fixed upper limit of CO2 emissions.
∼
Based on Eqs. (1) and (4), the difference between the total amount of CO2 emissions and its limit is obtained—the emissions reduction Q is
∼ ∼ ∼ ∼
Q = EC − X = EC − X + N (7)
∼ ∼ ∼
N = (EC − X) − (EC − X) = EC [(1 − βμ)εY] (8)
∼
where N denotes the difference between the observed and expected values of the emissions reduction considering all the uncertainties under the
floating CO2 emissions quota.
∼
The variance of N is
∼ ∼2 ∼
D(N) = E[N ] − E[N]2 = EC2 (1 − βμ)2σY 2 (9)
∼
Similarly, according to Eqs. (6)–(8), the variance of N0 under a fixed CO2 emissions reduction quota is
∼
D(N0) = EC2σY 2 (10)
and because βμ ≥ 0, so
(1 − βμ)2 < 1 (11)
From Eqs. (9)–(11), the following relation is obtained:
∼ ∼
D(N) < D (N0) (12)

434
Z. Wan et al. Marine Pollution Bulletin 126 (2018) 428–435

References MEPC, 2011. Report of the Marine Environment Protection Committee on its 62nd
Session, MEPC 62/24, London, UK.
Merk, O., 2017. Shipping and emissions trading: in or out? http://shippingtoday.eu/
Adoption of the Paris Agreement FCCC/CP/2015/L.9/Rev1, 2015. United Nations shipping_emissions_trading/ (access on Sep. 1, 2017).
Framework Convention on Climate Change. MPEC, 2012. Report of the marine environment protection committee on its 63rd session.
Anderson, K., Bows, A., 2012. Lloyd's list intelligence. Executing a Scharnow turn: http://www.dco.uscg.mil/Portals/9/DCO%20Documents/5p/5ps/International
Reconciling shipping emissions with international commitments on climate change. %20Maritime%20Organization/MEPC/mepc63-report.pdf?ver=2017-06-29-
Cardiol. Manag. 3, 615–628. 130038-667 (access on Nov 7, 2017).
Armstrong, V.N., 2013. Vessel optimisation for low carbon shipping. Ocean Eng. 73, Ozaki, Y., Larkin, J., Tikka, K., Michel, K., 2010. An evaluation of the Energy Efficiency
195–207. Design Index (EEDI) baseline for tankers, containerships and LNG carriers. In:
Bazari, Z., Longva, T., 2011. Assessment of IMO Mandated Energy Efficiency Measures for Climate Change and Sips: Increasing Energy Efficiency a SNAME and Marine Board
International Shipping: Estimated CO2 Emissions Reduction from Introduction of Symposium, USA.
Mandatory Technical and Operational Energy Efficiency Measures for Ships. Poulsen, R.T., Johnson, H., 2016. The logic of business vs. the logic of energy manage-
International Maritime Organization, London, UK. ment practice: understanding the choices and effects of energy consumption mon-
Bouman, E.A., Lindstad, E., Rialland, A.I., Strømman, A.H., 2017. State-of-the-art tech- itoring systems in shipping companies. J. Clean. Prod. 112, 3785–3797.
nologies, measures, and potential for reducing GHG emissions from shipping–a re- Psaraftis, H.N., Kontovas, C.A., 2013. Speed models for energy-efficient maritime trans-
view. Transp. Res. Part D: Transp. Environ. 52, 408–421. portation: a taxonomy and survey. Transp. Res. Part C Emerg. Technol. 26, 331–351.
Bows-Larkin, A., Anderson, K., Mander, S., Traut, M., Walsh, C., 2015. Shipping charts a Report of the Marine Environment Protection Committee on Its Sixty-Eighth Session,
high carbon course. Nat. Clim. Chang. 5, 293–295. 2015. MEPC 68th Session, Agenda Item 21, IMO Doc MEPC 68/21. (29 May 2015).
Buhaug, Ø., Corbett, J., Winebrake, J., Mjelde, A., Lee, D., Hanayama, S., ... Yoshida, K., Roe, M., 2012. Maritime governance and policy-making. (SSBM).
2009. Second IMO GHG Study 2009 Update of the 2000 GHG Study: Final Report Shaffer, G., Bodansky, D., 2012. Transnationalism, unilateralism and international law.
Covering Phrases 1 and Phrases 2. IMO, London. TEL 1, 31–41.
California Air Resources Board, 2015. Greenhouse gas quantification methodology for the Shi, Y., 2016a. Are greenhouse gas emissions from international shipping a type of marine
air resources board low carbon transportation program. https://www.arb.ca.gov/cc/ pollution? Mar. Pollut. Bull. 113 (1), 187–192.
capandtrade/auctionproceeds/arbldprogramsqm.pdf. Shi, Y., 2016b. Reducing greenhouse gas emissions from international shipping: is it time
Cames, M., Graichen, J., Siemons, A., Cook, V., 2015. Emission reduction targets for in- to consider market-based measures? Mar. Policy 64, 123–134.
ternational aviation and shipping. available at. http://www.europarl.europa.eu/ Smith, T.W.P., et al., 2014. Third IMO Greenhouse Gas Study 2014 (International
RegData/etudes/STUD/2015/569964/IPOL_STU(2015)569964_EN.pdf. Maritime Organization).
Cariou, P., 2011. Is slow steaming a sustainable means of reducing CO 2 emissions from Sperling, D., Eggert, A., 2014. California's climate and energy policy for transportation.
container shipping? Transp. Res. Part D: Transp. Environ. 16, 260–264. Energ. Strat. Rev. 5, 88–94.
Chen, C.W., Hu, W.M., Knittel, C.R., 2017. Subsidizing Fuel Efficient Cars: Evidence from Spurrier, A., 2017. European Parliament votes to bring shipping into EU emissions trading
China's Automobile Industry (No. w23045). National Bureau of Economic Research. scheme. https://fairplay.ihs.com/safety-regulation/article/4282156/european-
Choi, J.K., Kelley, D., Murphy, S., Thangamani, D., 2016. Economic and environmental parliament-votes-to-bring-shipping-into-eu-emissions-trading-scheme (access on Nov
perspectives of end-of-life ship management. Resour. Conserv. Recycl. 107, 82–91. 7, 2017).
Cichowicz, J., Theotokatos, G., Vassalos, D., 2015. Dynamic energy modelling for ship Stavins, R., Schatzki, T., 2014. Beyond AB 32: Post-2020 climate policy for California.
life-cycle performance assessment. Ocean Eng. 110, 49–61. http://www.analysisgroup.com/uploadedfiles/content/insights/publishing/post_
Coraddu, A., Oneto, L., Baldi, F., Anguita, D., 2018. Vessels fuel consumption: A data 2020_ca_ghg%20policy.pdf, Accessed date: 15 July 2017.
analytics perspective to sustainability. In: Soft Computing for Sustainability Science. Tan, A.K.J., 2005. Vessel-source Marine Pollution: the Law and Politics of International
Springer, Cham, pp. 11–48. Regulation. Vol. 45 Cambridge University Press.
Devanney, J., 2011. The impact of the energy efficiency design index on very large crude UNCTAD, 2016. Review of maritime transport. http://unctad.org/en/pages/
carrier design and CO2 emissions. Ships Offshore Struct. 6, 355–368. PublicationWebflyer.aspx?publicationid=1650, Accessed date: 7 November 2017.
Dufour, J., 2016. EU MRV vs. IMO fuel consumption data collection system. https:// Upton, J., 2016. Shipping industry mulls modest steps to slow warming. http://www.
www.green4sea.com/eu-mrv-vs-imo-fuel-consumption-data-collection-system/ (ac- climatecentral.org/news/shipping-industry-modest-response-to-climate-change-
cess on Nov 7, 2017). 20811 (access on Nov 7, 2017).
Eide, M.S., Longva, T., Hoffmann, P., Endresen, Ø., Dalsøren, S.B., 2011. Future cost Verbeek, R.P., Kadijk, G., van Mensch, P., Wulffers, C., van den Beemt, B., Fraga, F.,
scenarios for reduction of ship CO2 emissions. Marit. Policy Manag. 38, 11–37. Aalbers, A.D.A., 2011. Environmental and Economic Aspects of Using LNG as a Fuel
Ellerman, A.D., Buchner, B.K., 2007. The European Union emissions trading scheme: for Shipping in The Netherlands. TNO, Delft.
origins, allocation, and early results. Rev. Environ. Econ. Policy 1 (1), 66–87. Verifavia-shipping, 2015. As a first step towards cutting greenhouse gas emissions from
González, P.D.R., 2006. Harmonization versus decentralization in the EU ETS: an eco- maritime transport, the EU requires operators of ships exceeding 5,000 GT to monitor
nomic analysis. Clim. Pol. 6 (4), 457–475. and report their carbon emissions and transport work on all voyages to, from, and
Green4sea, 2016. Industry stakeholders react on the EU ETS decision. https://www. between the EU ports. http://www.verifavia-shipping.com/shipping-carbon-
green4sea.com/industry-stakeholders-react-on-the-eu-ets-decision/ (access on Nov 7, emissions-verification/shipping-mrv-regulation-overview-6.php, Accessed date: 7
2017). November 2017.
IMO, 2004. Resolution A.963(23) Adopted on 5 December 2003, (Agenda item 19). In: Wan, Z., Zhu, M., Chen, S., Sperling, D., 2016a. Three steps to a green shipping industry.
IMO Policies and Practices Related to the Reduction of Greenhouse Gas Emissions Nature 530 (275–227).
From Ships, . http://www.imo.org/blast/blastDataHelper.asp?data_id=26597& Wan, Z., Chen, J., Makhloufi, A.E., Sperling, D., Chen, Y., 2016b. Four routes to better
filename=A963(23).pdf, Accessed date: 7 November 2017. maritime governance. Nature 540, 27–29.
IMO, 2012. EEDI - rational, safe and effective. http://www.imo.org/en/MediaCentre/ Williamson, P., 2016. Emissions reduction: scrutinize CO2 removal methods. Nature 530,
HotTopics/GHG/Pages/EEDI.aspx (access on Nov 7, 2017). 153–155.
Jotzo, F., Pezzey, J.C., 2007. Optimal intensity targets for greenhouse gas emissions Woo, J.K., Moon, D.S.H., 2014. The effects of slow steaming on the environmental per-
trading under uncertainty. Environ. Resour. Econ. 38, 259–284. formance in liner shipping. Marit. Policy Manag. 41, 176–191.
Keohane, N., Morehouse, E., 2017. Carbon markets: Extend, don't limit. Nature 544, 297. Xiong, L., Shen, B., Qi, S., Price, L., Ye, B., 2017. The allowance mechanism of China's
Kusiak, A., 2017. Smart manufacturing. Int. J. Prod. Res. 1–10. http://dx.doi.org/10. carbon trading pilots: a comparative analysis with schemes in EU and California.
1080/00207543.2017.1351644. Appl. Energy 185, 1849–1859.
Liang, W., Zhang, L., 2014. Legal issues concerning the EU unilateral aviation ETS: a Xu, H., Liu, W., Xu, M., 2014. Study on the relationship among carbon emission from
Chinese perspective. SCJ Int. L. & Bus. 11, 1. international shipping. In: World Seaborne Trade Amounts and Global Economic
MEPC, 2010. Report of the Marine Environment Protection Committee on its 60th Activity. 13. Science and Technology Management Research, pp. 219–223.
Session, MEPC 60/4/54, London, UK.

435